L-2014-109, 120-Day Response to Request for Additional Information Re License Amendment Request for Transition to 10 CFR 50.48(c) - NFPA 805 Performance-Based Standard for Fire Protection for Light Water Reactor Generating Plants.

120-Day Response to Request for Additional Information Re License Amendment Request for Transition to 10 CFR 50.48(c) - NFPA 805 Performance-Based Standard for Fire Protection for Light Water Reactor Generating Plants.

ML14135A395
Person / Time
Site:	Saint Lucie
Issue date:	04/25/2014
From:	Jensen J Florida Power & Light Co
To:	Document Control Desk, Office of Nuclear Reactor Regulation
Shared Package
ML14135A394	List: L-2014-109, 120-Day Response to Request for Additional Information Re License Amendment Request for Transition to 10 CFR 50.48(c) - NFPA 805 Performance-Based Standard for Fire Protection for Light Water Reactor Generating Plants.
References
L-2014-109
Download: ML14135A395 (108)

Text

Enclosure 2 contains Security Related information - Withhold under 10 CFR 2.390.

Upon removal of Enclosure 2, this document is decontrolled.

April 25, 2014 L-2014-109 10 CFR 50.90 U.S. Nuclear Regulatory Commission Document Control Desk Washington, D.C. 20555-0001 Re: St. Lucie Units 1 and 2 Docket Nos. 50-335 and 50-389 120-Day Response to Request for Additional Information Regarding License Amendment Request for Transition to 10 CFR 50.48(c) - NFPA 805 Performance-Based Standard for Fire Protection for Light Water Reactor Generating Plants (2001 Edition)

References:

1. FPL Letter L-2013-099 dated March 22, 2013, Transition to 10 CFR 50.48(c) -NFPA 805 Performance-Based Standard for Fire Protection for Light Water Reactor Generating Plants (2001 Edition).

2. Email from Siva Lingam, NRC, to Ken Frehafer, FPL, dated June 7, 2013 , St. Lucie NFPA-805 LAR Acceptance Review Clarification Questions.

3. FPL Letter L-2013-193 dated June 14, 2013, Transition to 10 CFR 50.48(c) -NFPA 805 Perfornance-Based Standard for Fire Protection for Light Water Reactor Generating Plants (2001 Editions) Acceptance Review Clarification Response.

4. St. Lucie Plant Units I and 2 Request for Additional Information on License Amendment Request to Adopt National Fire Protection Association Standard 805 Performance-Based Standard for Fire Protection (TAC Nos. MF 1373 and MF 1374) dated December 26, 2013.

Per Reference 1 above, Florida Power and Light Company (FPL) requested an amendment to the Renewed Facility Operating License (RFOL) for St. Lucie Units 1 and 2. The License Amendment Request (LAR) will enable FPL to adopt a new fire protection licensing basis which complies with the requirements in 10 CFR 50.48(a) and (c) and the guidance in Revision I of Regulatory Guide (RG) 1.205.

Per Reference 3 FPL responded to NRC LIC- 109 acceptance review questions received by FPL via Reference 2 to clarify aspects of the LAR submittal.

By letter dated December 26, 2013 (Reference 4) NRC Staff requested additional information regarding the LAR. Based on discussions with the NRC Staff, the additional information Florida Power & Light Company 6501 S. Ocean Drive, Jensen Beach, FL 34957

L-2014-109 10 CFR 50.90 requested was prioritized and the response to the request for additional information will be provided in three separate submittals. The attachments to this letter provide the 120-day response to the request for additional information.

The information provided in this submittal does not impact the 10 CFR 50.92 evaluation of "No Significant Hazards Consideration" previously provided in FPL letter L-2013-099.

FPL requests that Enclosure 2 to this letter, which contains sensitive security-related information.

be withheld from public disclosure in accordance with 10 CFR 2.390. Upon removal of , this document is uncontrolled.

This letter makes new commitments and changes existing commitments. The commitment revisions are included in Enclosure 2 as mark-ups to Attachment S, Table S-2, Implementation Items. Changes to Attachment S that are related to specific RAI responses are included with the respective RAI separately.

Should you have any questions regarding this application, please contact Mr. Eric Katzman, Licensing Manager, at 772-467-7734.

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

Executed on April ,, 2014.

Respectfully submitted, tp ensen Site Vice President St. Lucie Plant JJ/rcs

Enclosures:

1. FPL's St Lucie Units 1 and 2 NFPA 805 LAR 120-Day RAI

Response

2. FPL's St Lucie Units 1 and 2 NFPA 805 LAR 120-Day RAI Response - Withheld from Public Disclosure cc: Ms. Cindy Becker, Florida Department of Health

Enclosure I to L-2014-109 Page 1 of 106 Enclosure 1 120-Day Response to Request for Additional Information Regarding License Amendment Request for Transition to 10 CFR 50.48(c) - NFPA 805 Performance-Based Standard for Fire Protection for Light Water Reactor Generating Plants (2001 Edition)

PSL FM RA1 Oa PSL PRA RAI 0]a PSL FM RAI1 1b PSL PRA RAI 01b PSL FM RAI O0c PSL PRA RAI 01c PSL FM RA1 O1d PSL PRA RAI 01 d PSL FM RAI O1e PSL PRA RAI 01e PSL FM RA1 1f PSL PRA RAI 0If PSL FM RAI O1g PSL PRA RAI O0g PSL FM RAI O0h PSL PRA RAI 01h PSL FM RAI 01i.i PSL PRA RAI O1j PSL FM RAI O1i.ii PSL PRA RAI 01k PSL FM RAI 01i.iii PSL PRA RAI 011 PSL FM RAI O0i.iv PSL PRA RAI 01m PSL FM RAI O1j PSL PRA RAI 0In PSL FM RAI O1k PSL PRA RAI 01o PSL FM RAI O1I PSL PRA RAI 01p PSL FM RAI 01rn PSL PRA RAI 08 PSL FM RAI O1n PSL PRA RAI 09 PSL FM RAI O1p PSL PRA RAI 16 PSL FM RAI 04 PSL FM RAI 06a PSL FM RAI 06b

Enclosure I to L-2014-109 Page 2 of 106 PSL RAI FM 01a 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 consolidated fire growth and smoke transport (CFAST) model was used to calculate control room abandonment times, and in the hot gas layer (HGL) and multi compartment analyses (MCA).

" Heskestad's plume temperature correlation was used to determine Severity Factors.

" The generic fire modeling treatments (GFMTs) approach was used to determine the ZOI in all fire areas throughout plant.

" FLASH-CAT for calculating fire propagation in stacks of horizontal cable trays.

" HEATING 7 computer code was used in the assessment of the relative safety margins of cable damage for thermoplastic cables.

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 acceptability of the fire models that were used.

Specifically regarding the acceptability of CFAST for the main control room (MCR) abandonment time calculations:

a. The MCR abandonment analyses indicates that parts of the walls of the MCRs are covered with wood paneling. Provide technical justification for not considering fire scenarios that involve this wood paneling.

RESPONSE

The Shift Supervisor's Office in the Unit I Main Control Room (MCR) is the only area in the Unit 1 and Unit 2 MCR envelope that has a wood panel wall cover. The control room abandonment calculation for the Unit 1 MCR (Report 0027-0009-014-001, Rev. 0) has been updated to include a sensitivity fire scenario that addresses the effect of the wood paneling in the Shift Supervisors office on the predicted abandonment time and the probability of control room abandonment. It is shown that the abandonment time is approximately the same as a workstation fire in an office without wood paneling on the walls. In addition, it is shown that the total probability of control room abandonment is bound by an open location transient ignition source because of the limited area over which a fire in the Shift Supervisor's Office occurs relative to the total floor area of the MCR. LAR Attachment J has been updated to reflect the changes to the reports documenting the control room abandonment times and due to its length, is provided at the end of Enclosure 1.

Enclosure I to L-2014-109 Page 3 of 106 PSL RAI FM 01b NFPA 805, Section 2.4.3.3, states: "The PSA [probabilistic safety assessment] approach, methods, and data shall be acceptable to the AHJ ....." The NRC staff noted that fire modeling comprised the following:

• The consolidated fire growth and smoke transport (CFAST) model was used to calculate control room abandonment times, and in the hot gas layer (HGL) and multi compartment analyses (MCA).

" Heskestad's plume temperature correlation was used to determine Severity Factors.

" The generic fire modeling treatments (GFMTs) approach was used to determine the ZOI in all fire areas throughout plant.

• FLASH-CAT for calculating fire propagation in stacks of horizontal cable trays.

" HEATING 7 computer code was used in the assessment of the relative safety margins of cable damage for thermoplastic cables.

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 acceptability of the fire models that were used.

Specifically regarding the acceptability of CFAST for the main control room (MCR) abandonment time calculations:

b. Provide the basis (e.g. data from fire drills) for the assumption in the MCR abandonment time calculations that the fire brigade is expected to arrive within 15 minutes. Describe the uncertainty associated with this assumption, discuss possible adverse effects of not meeting this assumption on the results of the FPRA and explain how possible adverse.

RESPONSE

The PSL fire brigade response time of 15 minutes assumed in the MCR abandonment calculations is based on typical fire brigade response times to nuclear power plant areas and is considered to be bounding for the PSL Unit 1 and Unit 2 control rooms. This was validated by a review of a portion of the fire brigade drills conducted between November 25, 1981 and December 26, 2013 in various PSL plant areas, including the Reactor Auxiliary Building and the Turbine Building. The fire brigade response times for these drills are summarized in Table FMO0b-1. The plant areas listed in Table FMOI b-1 consist of spaces near the control room as well as spaces that are farther from the muster area than the control room and thus provide an indication of the control room fire brigade response time. The times shown in Table FMO Ib-1 represent the time interval between the fire brigade page and the arrival of the last required fire brigade member and include both announced and unannounced fire drills. The average response time is 9.8 minutes for the announced fire drills and 10.2 minutes for the unannounced fire drills. The minimum response time among all drills is five minutes and the maximum time response time is fifteen minutes. The key aspect of the response with respect to the control room abandonment calculation is the potential for the ventilation conditions to change via an open door; as such, the times shown in Table FMO lb-I are conservative insofar as they are based on the arrival of the last team member.

Enclosure 1 to L-2014-109 Page 4 of 106 Table FMO1b Fire Brigade Arrival Times at Various PSL Plant Areas Time on Time on Scene - First Scene - All Brigade Member Brigade Members Announced Fire Drills 3/4/98 Cable Spreading Room 6 minutes 8 minutes 3/6/98 Cable Spreading Room 3 minutes 6 minutes 2/15/00 1B Switchgear Room 2 minutes 5 minutes 2/22/00 1B Switchgear Room 1 minute 7 minutes 2/29/00 1B Switchgear Room 4 minutes 12 minutes 3/7/00 1B Switchgear Room 3 minutes 13 minutes 6/12/00 1A Switchgear Room 2 minutes 14 minutes 5/7/13 SL2 H2 Seal Oil Unit 7 minutes 11 minutes 8/29/13 SL2 C AFW Feedwater Pump 6 minutes 10 minutes 9/14/13 SL1 DEH Platform 5 minutes 11 minutes 12/4/13 SL1 DEH Platform 7 minutes 10 minutes 12/26/13 SL1 Cold Chem Lab 6 minutes 11 minutes Unannounced Fire Drills 11/25/81 Cable Spreading Room 3 minutes 5 minutes 10/25/82 Cable Spreading Room 4 minutes 11 minutes 1/15/92 1A Switchgear Room 2 minutes 6 minutes 2/12/92 1B Switchgear Room 1.5 minutes 5.5 minutes 6/14/00 1A Switchgear Room 2.5 minutes 11 minutes 1/24/13 SL2 19.5' Hot Locker Rm 7 minutes 10 minutes 1/28/13 SL2 19.5' Hot Chem Lab 6 minutes 11 minutes 2/22/13 SL1 19.5' Hot Locker Rm 11 minutes 14 minutes 2/25/13 SL1 43' B SWGR Room 9 minutes 15 minutes 3/29/13 SL2 TGB 39.5 "D" Batt Rm 5 minutes 9 minutes 6/24/13 SL2 B Main Feed Pump Mtr 6.5 minutes 11 minutes 6/7/13 So. Service Bldg Swgr 6 minutes 11 minutes 5/16/13 SL2 TGB Swgr Batt Chrgr 5 minutes 11 minutes 4/24/13 SLU Condensate Polisher 8 minutes 12 minutes 7/31/13 SL2 B ICW Pump Motor 5 minutes 10 minutes 8/23/13 SL2 B CCW Motor 5 minutes 10 minutes 9/11/13 SL2 C AFW Feedwater Pump 7 minutes 10 minutes 11/20/13 SL2 2B Inst. Air Compressor 3.5 minutes 11 minutes The fire brigade response time is incorporated into the control room abandonment calculation models via a change in the status of the boundary doors (closed to open), though credit for manual suppression is independent of this assumption and the fire heat release rates in the CFAST models are not reduced at the brigade arrival time. The MCR boundary doors may open for reasons other than fire brigade arrival, such as operator actions or occupant egress, so a value of fifteen minutes was selected as an intermediate value between always closed and always open within the twenty-five minute interval considered in the calculation. The FPRA uses the natural ventilation configuration that produces the minimum abandonment time as a representative value to define the baseline abandonment scenarios. Because the most adverse abandonment time is used for the range of natural ventilation conditions, the uncertainty in the door open time is bound by the use of the data provided in the control room abandonment calculation.

For completeness, the control room abandonment calculations have been updated to include a sensitivity assessment of the model results to the time the boundary doors are assumed to open (see Section A2.2.15 in Report 0027-0009-014-001, Rev. 0 and Section A2.2.13 in Report 0027-0009-0014-003, Rev. 0). The sensitivity assessment considers the effect of opening the boundary door to

Enclosure I to L-2014-109 Page 5 of 106 the control room between ten and twenty minutes on both the calculated abandonment times and the total probability of control room abandonment. It is shown in Section A2.2.15 of Report 0027-0009-014-001, Rev. 0 that opening the doors at ten minutes can affect the total probability of abandonment time by -5.95 to +1.8 percent over the ten minute interval for Unit 1. Similarly, it is shown in Section A2.2.13 of Report 0027-0009-014-003, Rev. 0 that opening the doors at ten minutes can affect the total probability of abandonment time by -0.041 to +0.61 percent over the ten minute interval for Unit 2. Note that given that the maximum time at which abandonment can affect the non-suppression probability is 20.9 minutes, the scenarios for which the door opens at twenty minutes are nearly the same scenarios as the closed door baseline scenarios.

Based on the actual response times of the fire brigade, the use of the abandonment times in the FPRA, and the sensitivity of the abandonment times to uncertainty in the fire brigade arrival time, there are no known adverse effects associated with not meeting this assumption. LAR Attachment J has been updated to reflect the changes to the control room abandonment reports and is provided with the response to PSL RAI FM 01f.

PSL RAI FM 01c NFPA 805, Section 2.4.3.3, states: "The PSA [probabilistic safety assessment] approach, methods, and data shall be acceptable to the AHJ ....." The NRC staff noted that fire modeling comprised the following:

" The consolidated fire growth and smoke transport (CFAST) model was used to calculate control room abandonment times, and in the hot gas layer (HGL) and multi compartment analyses (MCA).

• Heskestad's plume temperature correlation was used to determine Severity Factors.

• The generic fire modeling treatments (GFMTs) approach was used to determine the ZO1 in all fire areas throughout plant.

• FLASH-CAT for calculating fire propagation in stacks of horizontal cable trays.

• HEATING 7 computer code was used in the assessment of the relative safety margins of cable damage for thermoplastic cables.

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 acceptability of the fire models that were used.

Specifically regarding the acceptability of CFAST for the main control room (MCR) abandonment time calculations:

c. Explain how the different fires (six in Unit 1 and five in Unit 2) and ventilation conditions (six in both units) were weighted in terms of their contribution to the probability and associated risk for MCR abandonment and provide technical justification for the approach that was used. In addition, for the electrical panel fire scenarios that were considered in the FPRA, provide technical justification for the location of the fire (inside or outside the MCR) in the CFAST analysis and explain up to what extent fire spread to adjacent panels was assumed.

Enclosure I to L-2014-109 Page 6 of 106

RESPONSE

The updated MCR abandonment time analysis, revised to address various associated FM RAIs, includes the following cases:

1. Single cable bundle electrical panel fire in the MCR.

2. Multiple cable bundle electrical panel fire in the MCR.

3. Multiple cable bundle electrical panel fire (spreading) in the MCR.

4. Open electrical panel fire (spreading) in the MCR.

5. Open location transient fuel package fire in the MCR.

6. Wall location transient fuel package fire in the MCR.

7. Corner location transient fuel package fire in the MCR.

8. Self-ignited cable tray fire in the MCR.

9. Open location transient fuel package fire in the computer room - staff support area.

10. Wall location transient fuel package fire in the computer room - staff support area.

11. Corner location transient fuel package fire in the computer room - staff support area.

The bounding panel configuration, transient location and HVAC configuration, associated with the above cases is used to calculate the control room abandonment frequency.

The CFAST model considers two volumes, the Main Control Board (MCB) area and the support areas surrounding the MCB areas. All panels are located in the MCB area and are evaluated as such in the CFAST analysis.

The analysis of the control room fires for evaluation of abandonment frequency assumes fire spread (Case 3 above for panels).

The Fire PRA has been re-quantified, incorporating the impact of all RAIs, with the associated results (Attachment W) provided in Enclosure 2 to this RAI transmittal letter.

PSL RAI FM Old NFPA 805, Section 2.4.3.3, states: "The PSA [probabilistic safety assessment] approach, methods, and data shall be acceptable to the AHJ ...."The NRC staff noted that fire modeling comprised the following:

• The consolidated fire growth and smoke transport (CFAST) model was used to calculate control room abandonment times, and in the hot gas layer (HGL) and multi compartment analyses (MCA).

" Heskestad's plume temperature correlation was used to determine Severity Factors.

• The generic fire modeling treatments (GFMTs) approach was used to determine the ZOI in all fire areas throughout plant.

• FLASH-CAT for calculating fire propagation in stacks of horizontal cable trays.

" HEATING 7 computer code was used in the assessment of the relative safety margins of cable damage for thermoplastic cables.

Enclosure I to L-2014-109 Page 7 of 106 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 acceptability of the fire models that were used.

Specifically regarding the acceptability of CFAST for the main control room (MCR) abandonment time calculations:

d. For the case when cables in an adjacent electrical cabinet are in direct contact with the separating wall, NUREG/CR-6850, Appendix S recommends a fire spread time of 10 minutes. Provide technical justification for using the assumption in the MCR abandonment time calculations that fire spreads to adjacent cabinets in 15 minutes.

RESPONSE

The original control room abandonment calculations for Unit 1 and Unit 2 postulated a fifteen minute propagation between adjacent panels assuming the conditions of NUREG/CR-6850, Appendix S for fifteen minute propagation would generally be met. However, in order to demonstrate that this is technically justified, detailed walkdowns are required to document the internal panel configuration. In lieu of performing these walkdowns to justify the propagation time for each specific panel configuration, the conservative generic value often minutes per Appendix S of NUREG/CR-6850 has been used in place of the fifteen minutes in the updated control room abandonment calculations for both Unit I and Unit 2. LAR Attachment J has been updated to reflect the changes to the control room abandonment reports and is provided with the response to PSL RAI FM 01 f. The Fire PRA has been re-quantified, incorporating the impact of all RAls, with the associated results (Attachment W) provided in Enclosure 2 to this RAI transmittal letter.

PSL RAI FM Ole NFPA 805, Section 2.4.3.3, states: "The PSA [probabilistic safety assessment] approach, methods, and data shall be acceptable to the AHJ ..... " The NRC staff noted that fire modeling comprised the following:

" The consolidated fire growth and smoke transport (CFAST) model was used to calculate control room abandonment times, and in the hot gas layer (HGL) and multi compartment analyses (MCA).

• Heskestad's plume temperature correlation was used to determine Severity Factors.

" The generic fire modeling treatments (GFMTs) approach was used to determine the ZOI in all fire areas throughout plant.

• FLASH-CAT for calculating fire propagation in stacks of horizontal cable trays.

• HEATING 7 computer code was used in the assessment of the relative safety margins of cable damage for thermoplastic cables.

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 acceptability of the fire models that were used.

Enclosure 1 to L-2014-109 Page 8 of 106 Specifically regarding the acceptability of CFAST for the main control room (MCR) abandonment time calculations:

e. LAR Attachment H, Table H-I indicates that frequently asked question (FAQ)-08-0052, "Transient Fires - Growth Rates and Control Room Non-Suppression" (ADAMS Accession No. ML092120501) was used in the LAR submittal. Provide justification for using transient fire growth rates in the Units 1 and 2 MCR abandonment time calculations that are different from those specified in FAQ-08-0052.

RESPONSE

The heat release rate growth rate for transient fuel package fires was evaluated in the original PSL control room abandonment calculations as 'Medium' t' fires based on data provided in the SFPE Handbook of FireProtection Engineering,Section 3-1, which is different from the guidance provided in NUREG-6850, Supplement 1 (i.e., FAQ 08-0052). The heat release rate during the growth stage is defined by the following equation:

0()=Oat2 t < J(*/a (FMle-I)

Where 0(t) is the heat release rate (kW [Btu/s]) at time t (s), a is the heat release rate growth constant (0.0117 kW/s 2 [0.0111 Btu/s3 ]), and Op is the peak heat release rate for the fire scenario (kW [Btu/s]). The duration of the growth stage for transient fuel packages with heat release rates ranging from 22 - 578 kW (21.2 - 548 Btu/s) is 0.7 - 3.7 minutes. By contrast, recent guidance in NUREG/CR-6850, Supplement I recommends that a constant growth time should be assumed: two minutes for loose or unconfined transient material and eight minutes for transient material located within containers or bins. Because the growth rate varies with the NUREG/CR-6850, Supplement 1 approach, the method assumed in the control room abandonment calculation can be more conservative or less conservative depending on the particular heat release rate bin considered and the type of transient fire scenario postulated. Specifically, the 'Medium' t 2 growth rate is conservative and bounding for transient fires that are postulated to have an eight minute growth rate per NUREG/CR-6850, Supplement 1. The 'Medium' t' growth rate is conservative and bounding for fires that are postulated to have a two minute growth rate per NUREG/CR-6850, Supplement 1 when the peak heat release rate is less than 168 kW (159 Btu/s). This heat release rate is greater than NUREG/CR-6850 transient heat release rate Bin 5 but less than NUREG/CR-6850 transient heat release rate Bin 6.

In lieu of quantifying the effect of the conservative/non-conservative bias for each transient fire scenario relative to the FAQ 08-0052 guidance, the characterization of the transient fire scenarios has been revised in updated control room abandonment calculations for the Unit 1 and Unit 2 MCR such that they are consistent with the guidance provided in NUREG/CR-6850, Supplement

1. The baseline fire scenarios postulate a confined transient fuel package fire having an eight minute growth rate and soot production fuel properties consistent with a poorly-ventilated fire (see Sections 5.1.3.2 and 5.1.4 in both Report 0027-0009-014-001, Rev. 0 and in Report 0027-0009-014-003, Rev. 0). Section A2.2.8 in Report 0027-0009-014-001, Rev. 0 and Section A2.2.6 in 0027-0009-014-003, Rev. 0 consider the effect of a transient fuel package fire having a two minute growth rate and fuel properties consistent with a soot production ftiel properties consistent with a well-ventilated fire (i.e., unconfined fuel package) on the predicted abandonment times and the

Enclosure I to L-2014-109 Page 9 of 106 total probability of control room abandonment. It is shown that the two minute growth rate transient fire scenarios in the MCR having well-ventilated fuel properties produces a 56.04 percent lower probability of abandonment in Unit 1 and a 55.74 percent lower probability of abandonment in Unit 2 than the eight minute growth rate transient fire scenarios in the MCR having poorly-ventilated fuel properties. The effect is similar but diminished for transient fire scenarios postulated in the support areas. Accordingly, the baseline fire scenarios use the eight minute growth rate transient fire scenarios having poorly-ventilated fuiel properties. LAR Attachment J has been updated to reflect the changes to the control room abandonment reports and is provided with the response to PSL RAI FM 01 f. The Fire PRA has been re-quantified, incorporating the impact of all RAIs, with the associated results (Attachment W) provided in Enclosure 2 to this RAI transmittal letter.

PSL RAI FM 01f NFPA 805, Section 2.4.3.3, states: "The PSA [probabilistic safety assessment] approach, methods, and data shall be acceptable to the AHJ ....." The NRC staff noted that fire modeling comprised the following:

• The consolidated fire growth and smoke transport (CFAST) model was used to calculate control room abandonment times, and in the hot gas layer (HGL) and multi compartment analyses (MCA).

" Heskestad's plurme temperature correlation was used to determine Severity Factors.

" The generic fire modeling treatments (GFMTs) approach was used to determine the ZO1 in all fire areas throughout plant.

" FLASH-CAT for calculating fire propagation in stacks of horizontal cable trays.

" HEATING 7 computer code was used in the assessment of the relative safety margins of cable damage for thermoplastic cables.

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 acceptability of the fire models that were used.

Specifically regarding the acceptability of CFAST for the main control room (MCR) abandonment time calculations:

f. Some fire model parameters that were assessed in the sensitivity analyses for the Units 1 and 2 MCR abandonment time calculations appear to have a significant effect on the abandonment times. For example, an elevated ambient temperature in the MCR may significantly reduce the abandonment time under certain conditions. Explain how the results of the sensitivity analyses were used in the FPRA.

RESPONSE

The parameter sensitivity analysis provided in Appendix B of the original control room abandonment calculation was not used by the fire PRA because of the difficulties in propagating the parameter uncertainty into the fire PRA model. The parameter sensitivity analyses in the control room abandonment calculations have been updated to provide a conservative basis for the baseline fire scenarios (see Attachment 2 of both Report 0027-0009-014-001, Rev. 0 and Report

Enclosure I to L-2014-109 Page 10 of 106 0027-0009-014-003, Rev. 0) in lieu of incorporating the parameter sensitivity results directly into the FPRA. The conservative basis for the assumed input parameters is assessed by the absolute effect on the control room abandonment time and the cumulative integrated effect on the total probability of abandonment as computed using the methods described in NUREG/CR-6850.

The revised parameter sensitivity analysis provided in the updated control room abandonment calculations shows that the assumptions on parameter input values can be separated into one of three groups:

" The parameter does not significantly affect the analysis results over the potential range of values that could be assigned to the parameter (Parameter Sensitivity Group 1);

" The parameter does affect the analysis results, but value selected for the baseline case is conservative (Parameter Sensitivity Group 2); and

" The parameter does affect the analysis results significantly and the value selected for the baseline case is not conservative (Parameter Sensitivity Group 3).

A significant effect is defined in the sensitivity analysis as a fifteen percent variation in the probability of control room abandonment as summed over three heat release rate bins. This significance threshold is consistent with the theoretical and observed uncertainty in calorimeter heat release rate measurements as described in the SFPE Handbook of Fire Protection Engineering,Section 3-2. Because the heat release rate is a primary input parameter provided by NUREG/CR-6850, the uncertainty analysis for other parameters is resolved to a level comparable with this parameter. It is also important to note that the uncertainty threshold for the parameter sensitivity analysis is significantly narrower than the uncertainty in the probability of control room abandonment as computed using the fifth and ninety-fifth percentile suppression rate parameters (i.e., A) as provided in NUREG/CR-6850.

The sensitivity parameters that fall into the first and second group provide a basis for a conservative parameter assumption and baseline model configuration with respect to the parameter. The sensitivity parameters that fall into the third group are of interest because they indicate that the baseline configuration is not conservative with respect to uncertainty in the parameter. However, the process used in developing and documenting the parameter sensitivity analysis is iterative and, in most cases, a result that falls into the third category initiates a revision to the baseline assumption such that the updated result falls into the first or second groups.

There are two key exceptions to this iterative update process. The first is the assumed initial ambient temperature and the second is the effect of a large transient fuel package fire scenario. In the case of the assumed initial ambient temperatures, the sensitivity analysis is used to establish an upper limit on the results applicability. This limit is 29.9°C (85.8°F) for Unit 1 (see Section A2.2.10 in Report 0027-0009-014-001, Rev. 0) and 29.5°C (85.1°F) for Unit 2 (see Section A2.2.8 in Report 0027-0009-014-003, Rev. 0). In both cases, the temperature limits are greater than the maximum normal temperature conditions of 23.8'C (75°F) for Unit I as documented in DBD-HVAC-1, Rev. 2 and 26.7 0 C (80'F) for Unit 2 as documented in DBD-HVAC-2, Rev. 2 and therefore represent off-normal conditions. Off-normal conditions are not postulated for the baseline FPRA fire scenarios concurrently with a fire event. In the case of the large transient fuel package fire scenarios, which include workstation and wood panel-lined office fires, a detailed assessment of the total probability of control room abandonment indicates that these scenarios are

Enclosure Ito L-2014-109 Page 11 of 106 comparable to or bounded by an open location transient ignition source as a result of the limited area over which the severe fires occur relative to the area available for a transient fuel package.

Based on the revised sensitivity analysis, the baseline results are considered conservative over the range of parameter uncertainty because the probability of abandonment is maximized, which in turn maximizes the CDF and LERF. LAR Attachment J has been updated to reflect the changes to the control room abandonment reports, due to its length, is provided at the end of Enclosure 1. The Fire PRA has been re-quantified, incorporating the impact of all RAIs, with the associated results (Attachment W) provided in Enclosure 2 to this RAI transmittal letter.

PSL RAI FM 01g NFPA 805, Section 2.4.3.3, states: "The PSA [probabilistic safety assessment] approach, methods, and data shall be acceptable to the AHJ .... The NRC staff noted that fire modeling comprised the following:

• The consolidated fire growth and smoke transport (CFAST) model was used to calculate control room abandonment times, and in the hot gas layer (HGL) and multi compartment analyses (MCA).

" Heskestad's plume temperature correlation was used to determine Severity Factors.

• The generic fire modeling treatments (GFMTs) approach was used to determine the ZOI in all fire areas throughout plant.

" FLASH-CAT for calculating fire propagation in stacks of horizontal cable trays.

HEATING

- 7 computer code was used in the assessment of the relative safety margins of cable damage for thermoplastic cables.

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 acceptability of the fire models that were used.

Specifically regarding the acceptability of the GFMTs approach:

g. In GFMTs, Section 2.4 describes the critical heat flux for a target that is immersed in a thermal plume. Explain how the modification to the critical heat flux was used in the ZOI determination.

RESPONSE

The modified critical heat flux is a means of accounting for both elevated temperatures and flame heat fluxes and was implemented using either a two or a three point (i.e., temperature) treatment in the fire PRA. When the modified heat flux is used to establish an ignition source Zone of Influence (ZOI) in an enclosure with an elevated temperature, the ZOI is larger than an ambient temperature based ZOI. Most plant areas use the two point treatment of the modified critical heat flux. The first point corresponds to temperature conditions between ambient and 80'C (176°F) and represents the temperature interval in which the ZOIs such as those documented in the "Generic Fire Modeling Treatments" report (1 SPH02902.030) are applicable. The second point corresponds to temperature conditions greater than 80'C (176°F) and is conservatively characterized in the fire PRA as a full-room bumrout. This applies to both targets located in the thermal plume region and to targets that are located outside the thermal plunne region.

Enclosure I to L-2014-109 Page 12 of 106 Several plant areas - Unit 1 Cable Spread Room, Unit 2 "A" Cable Loft and Unit 2 Cable Spread Spread Room (Fire Zones 1_57, 2_51X, and 2_52 respectively) use a three point treatment for greater resolution on the risk characterization. The first point corresponds to temperature conditions between ambient and 80'C (176 0 F) and represents the temperature interval in which the ZOIs for thermoplastic cable targets are applicable. The second point corresponds to temperature conditions greater than 80 0 C (176°F) but less than 13 I°C (268°F) and represents the region where the hot gas layer can produce a heat flux up to 3.0 kW/m2 (0.26 Btu/s 2). The ZOIs for sensitive components, which have a heat flux threshold of 3.0 kW/m 2 (0.26 Btu/s) are applicable in this temperature range when used to identify thermoplastic cable targets because the total heat flux at the ZOI boundary is 6 kW/m 2 (1.0 Btu/s 2 ), the generic threshold for thermoplastic cables per NUREG/CR-6850. The third point corresponds to temperature conditions greater than 131 °C (268 0 F) and is conservatively characterized in the fire PRA as a full-room burnout. This applies to both targets located in the thermal plume region and to targets that are located outside the thermal plume region.

PSL RAI FM 01h NFPA 805, Section 2.4.3.3, states: "The PSA [probabilistic safety assessment] approach, methods, and data shall be acceptable to the AHJ [...]."The NRC staff noted that fire modeling comprised the following:

• The consolidated fire growth and smoke transport (CFAST) model was used to calculate control room abandonment times, and in the hot gas layer (HGL) and multi compartment analyses (MCA).

• Heskestad's plume temperature correlation was used to determine Severity Factors.

" The generic fire modeling treatments (GFMTs) approach was used to determine the ZOI in all fire areas throughout plant.

" FLASH-CAT for calculating fire propagation in stacks of horizontal cable trays.

• HEATFNG 7 computer code was used in the assessment of the relative safety margins of cable damage for thermoplastic cables.

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 acceptability of the fire models that were used.

Specifically regarding the acceptability of the GFMTs approach:

h. Provide technical justification to demonstrate that the GFMTs approach as used to determine the ZOI of fires that involve multiple burning items (e.g., an ignition source and an intervening combustible such as a cable tray) is conservative and bounding.

RESPONSE

The Zones of Influence (ZOIs) applied at PSL for fires that involve multiple burning items have been updated to specifically include the effects of secondary combustibles as documented in Report 0027-0009-014-004, Revision 0. The methods follow the guidance provided in NUREG/CR-6850 and NUREG/CR-7010 for the determination of the extent of fire spread in the secondary combustibles and the overall heat release rate of the ignition source - secondary

Enclosure I to L-2014-109 Page 13 of 106 combustible fuel packages. A number of conservative assumptions are provided with the ZOI calculations to ensure the application is conservative and bounding. These include the following:

" Cable trays are assumed to be frilly loaded and capable of sustaining a fire at a given location for the duration of the fire scenario.

• Fire spread along a cable tray is assumed to occur in two directions.

• A sufficient length of cable tray is assumed for supporting fire spread in a given direction for the duration of the fire scenario.

• Cable trays are assumed to be vertically stacked so that the emitting area is maximized.

• All cables are assumed to be thermoplastic; thermoset cable materials and coated thermoplastic cables are not considered when determining the flame spread rates.

" The ignition delay for the lowest cable tray is assumed to be one minute, regardless of the actual tray height above the ignition source.

The conservative implementation of the NUREG/CR-6850 and NUREG/CR-7010 guidance is considered to be bounding and provides the technical justification for the approach. The response to PSL RAI FM 0l.i.iv provides additional details on the effect of the secondary combustibles on the ZOI dimensions. LAR Attachment J has been updated to reflect the changes to the reports documenting the cable tray fire propagation models and due to its length, is provided at the end of . The Fire PRA has been re-quantified, incorporating the impact of all RAIs, with the associated results (Attachment W) provided in Enclosure 2 to this RAI transmittal letter.

PSL RAI FM Oli.i NFPA 805, Section 2.4.3.3, states: "The PSA [probabilistic safety assessment] approach, methods, and data shall be acceptable to the AHJ [ ... ]." The NRC staff noted that fire modeling comprised the following:

" The consolidated fire growth and smoke transport (CFAST) model was used to calculate control room abandonment times, and in the hot gas layer (HGL) and multi compartment analyses (MCA).

• Heskestad's plume temperature correlation was used to determine Severity Factors.

• The generic fire modeling treatments (GFMTs) approach was used to determine the ZO in all fire areas throughout plant.

• FLASH-CAT for calculating fire propagation in stacks of horizontal cable trays.

" HEATING 7 computer code was used in the assessment of the relative safety margins of cable damage for thermoplastic cables.

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 acceptability of the fire models that were used.

Enclosure Ito L-2014-109 Page 14 of 106 Specifically regarding the acceptability of the GFMTs approach:

i. Regarding the flame spread and fire propagation in cable trays:

i. Describe how the flame spread and fire propagation in cable trays and the corresponding heat release rate (HRR) of cables was determined.

RESPONSE

The flame spread and fire propagation within an arrangement of secondary combustibles (i.e.,

cable trays) follows the guidance provided in Appendix R of NUREG/CR-6850 and Section 9 of NUREG/CR-70 10, Volume I for thermoplastic cables as documented in Report 0027-0053-000-002, Revision 0, Report 0027-0009-014-004, Revision 0, and Report 0027-0009-014-005, Revision 0. Specifically, the vertical fire propagation within a vertical stack of cable trays postulates that the first cable tray ignites one minute after the ignition source, the second cable tray four minutes after the first cable tray ignites, the third cable tray ignites three minutes after the second cable tray ignites, the fourth cable tray ignites two minutes after the third cable tray ignites, and all subsequent cable trays ignite one minute after the cable tray immediately below it ignites.

The initial segment of bottom cable tray ignited is equal to the characteristic dimensions of the ignition source, and the initial ignition length for additional cable trays within a stack is determined using an inverted frustum based at the lowest cable tray in the stack and having thirty-five degree sides. The fire spread along an ignited cable tray is in two directions with a constant spread rate of 0.89 mm/s (0.035 in/s) and the heat release rate per unit area is 250 kW/m 2 (22 Btu/s), the recommended value for thermoplastic cables per NUREG/CR-7010, Volume 1. LAR Attachment J has been updated to reflect the changes to the reports documenting the cable tray fire propagation models and due to its length, is provided at the end of Enclosure 1. The Fire PRA has been re-quantified, incorporating the impact of all RAIs, with the associated results (Attachment W) provided in Enclosure 2 to this RAI transmittal letter.

PSL RAI FM 01i.ii NFPA 805, Section 2.4.3.3, states: "The PSA [probabilistic safety assessment] approach, methods, and data shall be acceptable to the AHJ [...]."The NRC staff noted that fire modeling comprised the following:

" The consolidated fire growth and smoke transport (CFAST) model was used to calculate control room abandonment times, and in the hot gas layer (HGL) and multi compartment analyses (MCA).

• Heskestad's plume temperature correlation was used to determine Severity Factors.

" The generic fire modeling treatments (GFMTs) approach was used to determine the ZOI in all fire areas throughout plant.

" FLASH-CAT for calculating fire propagation in stacks of horizontal cable trays.

• HEATfNG 7 computer code was used in the assessment of the relative safety margins of cable damage for thermoplastic cables.

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 V&V,"

for a discussion of the acceptability of the fire models that were used.

Enclosure Ito L-2014-109 Page 15 of 106 Specifically regarding the acceptability of the GFMTs approach:

i) Regarding the flame spread and fire propagation in cable trays:

ii. Provide technical justification for the assumption that the lowest cable tray in a stack located above an ignition source will not ignite unless the tray is located below the flame tip of the ignition source fire.

RESPONSE

The full scale test data available for cable tray ignition and propagation data is the physical basis for the assumption that the lowest cable tray in a stack located above an ignition source ignites if it is at or below the flame tip. The empirical flame spread model for cable tray fires as provided in NUREG/CR-6850 and validated in NUREG/CR-7010, Volume I is based on the following three basic test series:

" EPRI-NP-1 881 (Sumitra tests);

" NUREG/CR-0381 (Klamerus tests); and

• NUTREG/CR-7010, Volume I (NIST tests).

The test reports for these test series document the results of about thirty-five to forty open configuration, unprotected cable tray fire tests. In all cases, the initial ignition source for the lowest cable tray within a stack is a gas burner or liquid fuel pan fire that causes flame impingement on the lowest cable tray in the stack. Further, there are no cases presented in which the thermal plume above the flame tip alone was sufficient for igniting a cable tray. An indication of this effect may be observed in the test series presented in NUREG/CR-0381, Test 28, which was a two tray stack with ceramic blanket on the top of both trays. The propane burner was sufficiently large to ignite the lower tray and to expose the upper tray to the thermal plume during the exposure fire cycle, but the upper tray did not ignite. A quantitative indication of the conditions necessary for fire ignition and surface spread is provided in NUREG/CR-5384. Bum mode evaluations for both non-rated (thermoplastic) and low flame spread (thermoset) cables are presented and indicate that for thermoplastic cables, a surface temperature of 538°C (1,000°F) and an internal fuel temperature of 577°C (1,070°F) are necessary for surfaceflames to develop (see Figure FM 01 i.ii-l). Smoldering and pyrolysis occur at lower temperatures, and a deep seated fire may result if the internal fuel temperature is approximately 538°C (1,000°F) regardless of the surface temperature.

Enclosure I to L-2014-109 Page 16 of 106 IwoI 1400

!1200 MW 1000 -,

~60

• aZLL.A-_

200 ,**-"-*- **

-0 200 400 000 80010001200140016001800 FiEAL T orE f NnRe C Figure 3. 5: Burn Mode Analysis of Non-Rated Cable Fire Figure FM 01 i.ii Burn Mode Analysis of Thermoplastic Cables per NUREG/CR-5384.

It can be shown using the Heskestad flame height correlation, the Heskestad virtual origin correction, and the Heskestad plume centerline temperature correlation that the temperature at the flame tip is approximately equal to 528°C (983°F), which is lower than the minimum temperature needed for either deep seated burning or for surface flame spread per Figure FM 01 i.ii-l. The Heskestad flame height correlation is given as follows per Section 2-1 of the SFPE Handbook of FireProtectionEngineering:

04 Lf = -1.02D + 0.235Qý (FM O i.ii-1)

Where Lf is the flame height (m), D is the effective fire diameter (m), and Q is the total heat release rate of the ignition source (kW). The Heskestad plume centerline temperature correlation is given as follows per Section 2-1 of the SFPEHandbook of Fire Protection Engineeringand "Fire Plumes and Ceiling Jets" in the Fire Safety Journal, Vol. 11, Nos. I & 2:

2 13 - Zo)-5/3 T, = T.o + 220 /(Z (FM 01 i.ii-2)

Where Tc is the plume centerline temperature (°C) at an elevation Z (mn)above the fire base, T., is the initial temperature (°C), and z, is the height of the virtual origin below the fire base (m). The virtual origin height is given by the following equation per Section 2-1 of the SFPEHandbook of Fire ProtectionEngineering:

0 4 zo = -1.02D + 0.083W (FM Oli.ii-3)

Where all terms have been defined. At the flame tip, the height above the fire base Z, is equal to the flame height Lf. Combining Equations FM 01 i.ii-1, FM 01 i.ii-2, and FM 01 i.ii-3 result in the following:

T, = T.o + (22)(23.1)J2/3ý2/3 = Too + 508 (FM 0 1i.ii-4)

Enclosure Ito L-2014-109 Page 17 of 106 Where all terms have been defined. The plume centerline temperature is thus independent of both the fire diameter and the heat release rate at the flame tip and is equal to 528°C (983°F) for an ambient temperature of 20'C (68°F).

These calculations are consistent with the observation provided in Section 7.2 of NUREG/CR-7010, Volume 1 that the damage threshold for cables as characterized by a heat flux is not a good indicator of ignition. Based on the cone calorimeter tests summarized in N UREG/CR-7010.,

Volume 1, a heat flux of exposure of 25 kW/m2 (2.2 Btu/s-ft2) is minimally sufficient to cause ignition and sustained burning for all classes of cables considered, including the thermoplastic cables. Data provided in Section 2-14 of the SFPE Handbook of Fire Protection Engineering indicates that the net heat flux to an object immersed in the fire plume at the flame height as determined from the stagnation point is between 5 - 15 kW/Min 2 (0.44 - 1.32 Btu/s-ft2), which is significantly less than the minimum heat flux necessary to cause sustained ignition per NUREG/CR-7010, Volume 1. This is further supported by the test data for thermoplastic cables provided in Figures 10- 12 of NUREG/CR-6931, Volume 3. A shroud temperature (exposure temperature) of about 300 - 330'C (572 - 626°F) is used to heat various types of thermoplastic cables in order to determine the damage times. Although the focus of the tests was not on the ignitability of the cables, the temperature profiles provide an indication of the cable response to the temperature exposure. The figures indicate that the cables do not ignite over the ten to twenty minute exposure interval and typically show the cables reach a steady state temperature close to 300'C (572°F) even though damage via electrical short occurs around 200'C (392°F).

The requirement for flames to impinge on the lowest cable tray before ignition is assumed is considered reasonable and supported by the available documents. A summary of the basis is as follows:

• All cable tray fire test data involves an ignition source that results in flame impingement on the lower cable tray in a cable tray stack

• A burn mode analysis of thermoplastic cables suggests that the minimum temperature required for surface flames or deep seated burning to develop is approximately of 538 0 C (1,000°F)

" The temperature at the flame tip is about 528°C (983°F) in the models used to develop the Zones of Influence (ZOls) for the PSL PRA, which is lower than the value required for surface flames to develop per the burn mode analysis Note that the flame height is used as an indicator of ignition and the ZOI dimensions are used as an indicator of damage. Consequently, cable trays located above an ignition source may be above the flame tip but still be within the ZOI for cable damage.

Enclosure I to L-2014-109 Page .18of 106 PSL RAI FM 01i.iii NFPA 805, Section 2.4.3.3, states: "The PSA [probabilistic safety assessment] approach, methods, and data shall be acceptable to the AHJ [...]."The NRC staff noted that fire modeling comprised the following:

" The consolidated fire growth and smoke transport (CFAST) model was used to calculate control room abandonment times, and in the hot gas layer (HGL) and multi compartment analyses (MCA).

" Heskestad's plume temperature correlation was used to determine Severity Factors.

" The generic fire modeling treatments (GFMTs) approach was used to determine the ZOI in all fire areas throughout plant.

" FLASH-CAT for calculating fire propagation in stacks of horizontal cable trays.

• HEATING 7 computer code was used in the assessment of the relative safety margins of cable damage for thenroplastic cables.

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 acceptability of the fire models that were used.

Specifically regarding the acceptability of the GFMTs approach:

i. Regarding the flame spread and fire propagation in cable trays:

iii. Explain how cables coated with Flamemastic 77 and covered cable trays were treated in these calculations.

RESPONSE

The detailed fire modeling calculations in which flame spread and fire propagation in cable trays is addressed (Report 0027-0053-000-002, Revision 0, Report 0027-0009-014-004, Revision 0, and Report 0027-0009-014-005, Revision 0) do not credit passive fire protection features such as Flamemastic 77 or cable tray covers. All cable trays within the defined stack are assumed to contain exposed thermoplastic cables subject to ignition, fire propagation, and flame spread (see Assumption 9 in Report 0027-0009-014-004, Revision 0, for example). This is conservative treatment for cables that are coated with Flamemastic 77 or are covered and included in the stack arrangement because there is no credit for the mitigating features provided by these elements. LAR Attachment J has been updated to reflect the changes to the Generic Fire Modeling Treatment reports documenting the cable tray fire propagation models and due to its length, is provided at the end of Enclosure 1. The Fire PRA has been re-quantified, incorporating the impact of all RAIs, with the associated results (Attachment W) provided in Enclosure 2 to this RAI transmittal letter.

Enclosure Ito L-2014-109 Page 19 of 106 PSL RAI FM 01i.iv NFPA 805, Section 2.4.3.3, states: "The PSA [probabilistic safety assessment] approach, methods, and data shall be acceptable to the AHJ [...]."The NRC staff noted that fire modeling comprised the following:

• The consolidated fire growth and smoke transport (CFAST) model was used to calculate control room abandonment times, and in the hot gas layer (HGL) and multi compartment analyses (MCA).

" Heskestad's plune temperature correlation was used to determine Severity Factors.

" The generic fire modeling treatments (GFMTs) approach was used to determine the ZOI in all fire areas throughout plant.

" FLASH-CAT for calculating fire propagation in stacks of horizontal cable trays.

" HEATING 7 computer code was used in the assessment of the relative safety margins of cable damage for thermoplastic cables.

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 acceptability of the fire models that were used.

Specifically regarding the acceptability of the GFMTs approach:

i. Regarding the flame spread and fire propagation in cable trays:

iv. Describe how the flame spread, fire propagation and HRR estimates affect the ZOI determination and HGL temperature calculations.

RESPONSE

The flame spread and fire propagation within an arrangement of secondary combustibles (i.e.,

cable trays) result in an increase in the total heat release rate for the ignition source fire scenario relative to a scenario in which secondary combustiblesare not ignited. Specifically, the flame spread aspects result in an increasing surface area of cables that are ignited and a physical displacement of the flame front. The fire propagation aspects result in additional cable trays within a stack igniting and a larger ignition length as the cable tray distance above the ignition source increases. The combined effects of the flame spread along a cable tray and fire propagation through a stack result in an increased total heat release relative to a scenario in which there are no secondary combustibles. The conservative aspects of the inputs used to characterize the flame spread and fire propagation through a cable tray stack are described in the responses to RAI FM 01h and RAI FM Oli.i.

The increased heat release rate results in a shorter time to reach threshold hot gas layer temperatures. Report 0027-0053-000-002, Revision 0 and Report 0027-0009-014-005, Revision 0 provide tabulations of the time the hot gas layer reaches a threshold value and demonstrate that the times decrease as the amount of secondary combustibles increases. Report 0027-0009-014-005, Revision 0 also provides tabulations of the time the hot gas layer reaches a threshold value for scenarios that do not involve secondary combustibles for comparison. Comparing any of these tables with a corresponding table for a secondary combustible configuration provides a quantitative basis for the effect on the time the hot gas layer temperature reaches a threshold value and shows the time decreases when secondary combustibles are included.

Enclosure I to L-2014-109 Page 20 of 106 The increased heat release rate and physical displacement of the flame front result in a larger Zones of Influence (ZOIs) relative to a case in which secondary combustibles are not ignited.

Report 0027-0009-014-004, Revision 0 provides tabulations of the ZOI dimensions for ignition source - secondary combustible configurations and demonstrates that the ZOI dimensions increase as the amount of secondary combustibles increases, both for fixed times and different numbers of cable trays evaluated and for a fixed number of cable trays and different times (i.e., a spreading fire within a stack). Report 0027-0009-014-004, Revision 0 also provides tabulations of the ZOI dimensions for scenarios that do not involve secondary combustibles for comparison. Comparing any of these tables with a corresponding table for a secondary combustible configuration provides a quantitative basis for the effect on the ZOI dimensions shows the ZOI dimensions increase when secondary combustibles are included. LAR Attachment J has been updated to reflect the changes to the reports documenting the cable tray fire propagation models and due to its length, is provided at the end of Enclosure 1. The Fire PRA has been re-quantified, incorporating the impact of all RAIs, with the associated results (Attachment W) provided in Enclosure 2 to this RAI transmittal letter.

PSL RAI FM 01j NFPA 805, Section 2.4.3.3, states: "The PSA [probabilistic safety assessment] approach, methods, and data shall be acceptable to the AHJ [...]."The NRC staff noted that fire modeling comprised the following:

• The consolidated fire growth and smoke transport (CFAST) model was used to calculate control room abandonment times, and in the hot gas layer (HGL) and multi compartment analyses (MCA).

• Heskestad's plume temperature correlation was used to detenrmine Severity Factors.

" The generic fire modeling treatments (GFMTs) approach was used to determine the ZOI in all fire areas throughout plant.

• FLASH-CAT for calculating fire propagation in stacks of horizontal cable trays.

• HEATING 7 computer code was used in the assessment of the relative safety margins of cable damage for thermoplastic cables.

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 acceptability of the fire models that were used.

Specifically regarding the acceptability of the GFMTs approach:

j. Describe how transient combustibles in an actual plant setting are characterized in terms of the three fuel package groupings in the GFMTs Supplement 3 (Transient Ignition Source Strength). Identify, areas, if any, where the NUREG/CR-6850 transient combustible HRR characterization (probability distribution and test data) may not encompass typical plant configurations. Finally, explain if any administrative action will be used tb control the type of transient in a fire area at St. Lucie Units 1 and 2.

Enclosure Ito L-2014-109 Page 21 of 106

RESPONSE

The transient fuel packages are categorized as miscellaneous materials (trash configurations) that do not contain acetone or other combustible liquids. This corresponds to the Group 3 and Group 4 transient fuel packages described in Supplement 3 of the "Generic Fire Modeling Treatments" report (Report 0027-0053-000-003, Rev. 0). The 9 8th percentile transient fuel packages are considered a special case of the Group 3 and Group 4 transient fuel packages with a specific heat release rate per unit area as described in the Supplement 3 of the "Generic Fire Modeling Treatments" report (Report 0027-0053-000-003, Rev. 0).

The transient fire heat release rate distribution specified in NUREG/CR-6850 as a 317 kW (300 Btu/s) 9 8 th percentile peak heat release rate fire is considered to be generically applicable to nuclear power plants. The PSL plant does not differ in any significant manner with respect to its transient combustible controls to warrant a significant increase or decrease in the applicable heat release rate profile. However, for areas that have been designated as "no transient combustible areas", to address the potential for violation of these controls, a 69 kW (65 Btu/s) 98th percentile peak heat release rate fire was applied. This heat release rate is considered appropriate given the unlikely event that transients are stored in these areas contrary to the controls imposed. The 69 kW (65 Btu/s) heat release rate was defined based on the heat release rate specified in NUREG/CR-6850 for a motor fire given that the most likely transient fire in a zone with limited transients would be associated with temporary cabling because this configuration would provide both the ignition source (energized temporary cabling) and combustible (cable insulation). The motor configuration would resemble such a transient fire. See the response to RAI PRA 04, submitted on February 24, 2014, for additional discussion on credit for reduced heat release rates for transient fires and associated administrative controls.

It is noted that there are two cases considered in the Unit 1 and Unit 2 control room abandonment calculations in which a transient fuel package fire scenario is characterized using a heat release rate profile that is more adverse than the standard NUREG/CR-6850, Appendix E, Case 8 transient fuel package fire scenario. Specifically, an office type fuel arrangement is postulated and characterized using a heat release rate profile applicable to such fuel packages, which have a peak heat release rate of 1,800 kW (1,710 BtuL/s) and 6,800 kW (6,450 Btu/s), depending on the number of panels that are assumed. This configuration is unique to the control room area among risk significant plant areas.

The control of combustibles will be ensured Linder NextEra Energy Nuclear Fleet Administrative Procedure FP-AA-101-1004, which limits the accumulation and composition of materials in plant areas.

Note that LAR Attachment J has been updated to reflect the changes to the Unit I and Unit 2 control room abandonment calculations referenced in this RAI response. The revised LAR Attachment J is provided with the response to PSL RAI FM 0If. The Fire PRA has been re-quantified, incorporating the impact of all RAIs, with the associated results (Attachment W) provided in Enclosure 2 to this RAI transmittal letter.

Enclosure I to L-2014-109 Page 22 of 106 PSL RAI FM 01 k NFPA 805, Section 2.4.3.3, states: "The PSA [probabilistic safety assessment] approach, methods, and data shall be acceptable to the AHJ [...]."The NRC staff noted that fire modeling comprised the following:

" The consolidated fire growth and smoke transport (CFAST) model was used to calculate control room abandonment times, and in the hot gas layer (HGL) and multi compartment analyses (MCA).

• Heskestad's plume temperature correlation was used to detennine Severity Factors.

• The generic fire modeling treatments (GFMTs) approach was used to determine the ZOI in all fire areas throughout plant.

" FLASH-CAT for calculating fire propagation in stacks of horizontal cable trays.

• HEATING 7 computer code was used in the assessment of the relative safety margins of cable damage for thermoplastic cables.

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 acceptability of the fire models that were used.

Regarding the FLASH-CAT model:

k. Describe how the ignition time of the bottom tray in the stack was determined.

RESPONSE

The bottom cable tray in a cable tray stack is ignited one minute after the ignition source ignites in all FLASH-CAT calculations (see Assumption 7 in Report 0027-0053-000-002, Rev. 0, for example). This corresponds to the minimum damage time for thermoplastic cable targets listed in Tables H-6 and H-8 of NUREG/CR-6850 and is more conservative than the generic value of five minutes that is assumed in NUREG/CR-7010, Volume 1.

PSL RAI FM 011 NFPA 805, Section 2.4.3.3, states: "The PSA [probabilistic safety assessment] approach, methods.

and data shall be acceptable to the AHJ .... " The NRC staff noted that fire modeling comprised the following:

• The consolidated fire growth and smoke transport (CFAST) model was used to calculate control room abandonment times, and in the hot gas layer (HGL) and multi compartment analyses (MCA).

" Heskestad's plume temperature correlation was used to determine Severity Factors.

" The generic fire modeling treatments (GFMTs) approach was used to determine the ZO1 in all fire areas throughout plant.

• FLASH-CAT for calculating fire propagation in stacks of horizontal cable trays.

" HEATING 7 computer code was used in the assessment of the relative safety margins of cable damage for thermoplastic cables.

Enclosure I to L-2014-109 Page 23 of 106 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 acceptability of the fire models that were used.

Regarding the FLASH-CAT model:

1. Explain why the approach to model fire propagation in a vertical stack of two cable trays (or less) described in Supplement 2 of the GFMTs bounds the FLASH-CAT model in light of the fact that the approach assumes ignition at a single point as opposed to ignition over the characteristic length in FLASH-CAT.

RESPONSE

The cable tray fire propagation model provided in Supplement 2 of the GFMTs has been replaced by the cable tray fire propagation model provided in Report 0027-0009-014-004, Rev. 0, "Combined Ignition Source - Cable Tray Fire Scenario ZOIs for St. Lucie Nuclear Power Plant Applications" and Report 0027-0009-014-005, Rev. 0, "Evaluation of the Development and Timing of Hot Gas Layer Conditions in Generic PSL Fire Compartments with Secondary Combustibles." The cable tray fire propagation model provided in Report 0027-0009-014-004, Rev. 0 and 0027-0009-014-005, Rev. 0 incorporate the recommendations of NUREG/CR-6850 and NUREG/CR-70 10, Volume I with regard to the initial length of cable trays ignited. LAR Attachment J has been updated to reflect the changes to the reports documenting the cable tray fire propagation models and due to its length, is provided at the end of Enclosure 1. The Fire PRA has been re-quantified, incorporating the impact of all RAls, with the associated results (Attachment W) provided in Enclosure 2 to this RAI transmittal letter.

PSL RAI FM 01m NFPA 805, Section 2.4.3.3, states: "The PSA [probabilistic safety assessment] approach, methods, and data shall be acceptable to the AHJ ....." The NRC staff noted that fire modeling comprised the following:

" The consolidated fire growth and smoke transport (CFAST) model was used to calculate control room abandonment times, and in the hot gas layer (HGL) and multi compartment analyses (MCA).

• Heskestad's plume temperature correlation was used to determine Severity Factors.

• The generic fire modeling treatments (GFMTs) approach was used to determine the ZOI in all fire areas throughout plant.

• FLASH-CAT for calculating fire propagation in stacks of horizontal cable trays.

• HEATING 7 computer code was used in the assessment of the relative safety margins of cable damage for thermoplastic cables.

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 acceptability of the fire models that were used.

Enclosure I to L-2014-109 Page 24 of 106 Regarding the acceptability of the approach, methods, and data in general:

m. The NRC staff identified the possibility that non-cable intervening combustibles were missed in fire areas of the plant. For example, during the audit walkdown, the NRC staff observed some combustible pipe insulation in the Unit 2 CSR. Provide information on how non-cable intervening combustibles were identified and accounted for in the fire modeling analyses.

RESPONSE

A confirmatory walkdown was performed of in-situ combustibles identified in the Fire Hazards Analysis resulting in a new transient scenario for the piping insulation located in the Unit 1 cable spreading room and two new transient scenarios for the piping insulation in the Unit 2 cable spreading room. These new transient fire scenarios were analyzed using an HRR contribution of the insulation which was conservatively postulated to be equivalent to that of thermoplastic cable insulation in a single cable tray. These new scenarios were required in order to address a transient fire impacting the insulation and spreading to nearby cable targets. During this walkdown, no other secondary combustibles were identified which impact fire PRA targets.

The criteria used for this walkdown was an evaluation of exposed combustibles. Combustibles contained within a pump (oil/grease) or enclosed within a cabinet (e.g., class A combustibles in a closed cabinet) were not considered to be impacted by a fire given the enclosed nature of the associated combustibles.

PSL RAI FM 01n NFPA 805, Section 2.4.3.3, states: "The PSA [probabilistic safety assessment] approach, methods, and data shall be acceptable to the AHJ ..... " The NRC staff noted that fire modeling comprised the following:

• The consolidated fire growth and smoke transport (CFAST) model was used to calculate control room abandonment times, and in the hot gas layer (HGL) and multi compartment analyses (MCA).

" Heskestad's plume temperature correlation was used to determine Severity Factors.

" The generic fire modeling treatments (GFMTs) approach was used to determine the ZOI in all fire areas throughout plant.

" FLASH-CAT for calculating fire propagation in stacks of horizontal cable trays.

• HEATING 7 computer code was used in the assessment of the relative safety margins of cable damage for thermoplastic cables.

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 acceptability of the fire models that were used.

Enclosure I to L-2014-109 Page 25 of 106 Regarding the acceptability of the approach, methods, and data in general:

ni. During the onsite audit the licensee discussed fire location factors for transient fires and stated that, due to the conservatism associated with the HRRs used and the low likelihood that a fixed ignition source is close enough to a wall or corner and orientated such as to be significantly influenced by the presence of a wall or corner, no specific increases of HRRs were incorporated in the fire modeling. Provide technical justification for not accounting for wall and corner effects in the case of transient fires, combustible liquid fires and electrical cabinet fires.

RESPONSE

The initial Fire PRA supporting the LAR submittal included location factors for transient fires (see Scenario Report, Tables 8-2 and 8-3). A walkdown of other fire scenarios to assess the need for application of location factors due to proximity (within 2 feet) of walls or corners has been performed and the results have been incorporated into the updated Fire PRA which has been re-quantified to incorporate the impact of all RAIs.

PSL RAI FM 01p NFPA 805, Section 2.4.3.3, states: "The PSA [probabilistic safety assessment] approach, methods, and data shall be acceptable to the AHJ [ ... ]." The NRC staff noted that fire modeling comprised the following:

" The consolidated fire growth and smoke transport (CFAST) model was used to calculate control room abandonment times, and in the hot gas layer (HGL) and multi compartment analyses (MCA).

• Heskestad's plume temperature correlation was used to determine Severity Factors.

" The generic fire modeling treatments (GFMTs) approach was used to determine the ZOI in all fire areas throughout plant.

" FLASH-CAT for calculating fire propagation in stacks of horizontal cable trays.

• HEATING 7 computer code was used in the assessment of the relative safety margins of cable damage for thermoplastic cables.

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 acceptability of the fire models that were used.

Specifically regarding the acceptability of the HGL and MCA calculations:

p. Explain how high energy arcing fault (HEAF) initiated fires were addressed in the HGL analysis and provide technical justification for the approach that was used to calculate HGL timing. More specifically, provide technical justification for not using the guidance in NUREG/CR-6850, page 11-19, fourth bullet regarding the fire growth and page M-13, sixth bullet regarding delay to cable tray ignition. Also, considering the energetic nature of the HEAF event, provide justification for the HRR to be used in the HGL calculations for electrical cabinet fires following a HEAF.

Enclosure I to L-2014-109 Page 26 of 106

RESPONSE

The Hot Gas Layer (HGL)/Multi-Compartmnent Analysis (MCA) methodology supporting the LAR submittal included the fire growth curve as defined in NUREG/CR-6850 Section G.3.1 for the post HEAF. The revised analysis will incorporate a maximum cabinet heat release rate in conjunction with the HEAF in accordance with NUREG/CR-6850 page 11-19. The delay time for fire spread between cable trays will be addressed in accordance with NUREG/CR-6850 page M-13, 6 t"bullet.

The post HEAF Heat Release Rate (HRR) probability distribution will be based on that specified for the ignition source in accordance with NUREG/CR-6850 Appendices E and G (as noted in NUREG/CR-6850, p. M-13, 4t" bullet.

PSL RAI FM 04 NFPA 805, Section 2.7.3.3, "Limitations of Use," states: "Acceptable engineering methods and numerical models shall only be used for applications to the extent these methods have been subject to verification and validation. These engineering methods shall only be applied within the scope, limitations, and assumptions prescribed for that method."

LAR Section 4.7.3, "Compliance with Quality Requirements in Section 2.7.3 of NFPA 805," states that "Engineering methods and numerical models used in support of compliance with 10 CFR 50.48(c) were applied appropriately as required by Section 2.7.3.3 of NFPA 805."

Regarding the limitations of use, identify uses, if any, of the GFMTs (including supplements) outside the limits of applicability of the method and for those cases, explain how the use of the GFMTs approach was justified.

RESPONSE

There are two general categories in which fire models are applied at PSL: the application of the Generic Fire Modeling Treatments (GFMT) approach and the calculation of the abandonment times in the main control room. The use of fire models outside the model limitations for each category is described in this RAI response.

1. Generic Fire Modeling Treatments Approach (ZOI)

The GFMT approach is primarily documented in Report 1SPH02902.030, Rev. 0 ("Generic Fire Modeling Treatments"); Report 0027-0009-014-004, Rev. 0 ("Combined Ignition Source - Cable Tray Fire Scenario ZOls for St. Lucie Nuclear Power Plant Applications"); Report 0027-0009-014-005, Rev. 0, ("Evaluation of the Development and Timing of Hot Gas Layer Conditions in Generic PSL Fire Compartments with Secondary Combustibles"); Report 0027-0053-000-003, Rev. 0,

("Supplemental Generic Fire Modeling Treatments: Transient Fuel Package Ignition Source Characteristics"); and Report 0027-0053-000-002, Rev. 0 ("Evaluation of the Development and Timing of Hot Gas Layer Conditions for Fire Scenarios Involving Secondary Combustible Materials at PSL"). A sixth document, "Supplemental Generic Fire Modeling Treatments: Closed Electrical Panels", Rev. B addresses the maximum heat release rate development within limited ventilation panels and has no specific limitations other than being applicable to closed door electrical panels.

The GFMT approach is intended to provide conservative ZOI dimensions and Hot Gas Layer (HGL) timing for various types of ignition sources when used within the stated limitations. There

Enclosure I to L-2014-109 Page 27 of 106 are six basic limitations that should be considered when applying the original GFMT approach as documented in 1SPH02902.030 to determine ZOls and HGL temperatures. The six limitations represent conditions or configurations for which the Generic Fire Modeling Treatment ZOI data may potentially be non-conservative if applied outside the particular limitation:

" The application of the generic ZOI data in compartments in which the hot gas layer temperature exceeds 80'C (176°F).

• The application of the generic ZOI data to fire scenarios in wall and corner configurations.

" The application of the generic ZOI data for panel ignition sources with panels having plan dimensions greater than 0.9 X 0.6 in (3 X 2 ft.).

" The application of the generic ZOI data to scenarios that result in flame impingement to the ceiling.

• The application of the generic hot gas layer data to configurations in which secondary combustibles (cable trays) are ignited.

Application of the GFMT CFAST fire modeling results to spaces that exceed the limitation for CFAST.

Supplemental analysis has been developed to address a number of these limitations under various circumstances that arise at PSL. These documents include Report 0027-0009-014-004, Rev. 0; Report 0027-0009-014-005, Rev. 0; Report 0027-0053-000-002, Rev. 0 and Report 0027-0053-000-003, Rev. 0.

ZOIs in Elevated Temperature Enclosures The fire scenarios considered in the PRA limit the use of ZOIs to situations in which the maximum HGL temperature is 80'C (176°F). If the temperature exceeds 80'C (176°F), a full room burnout condition will be conservatively assumed. There are a limited exceptions to this in which the PRA applies expanded ZOls in enclosures with temperatures tip to 13 IPC (268°F). Refer to the response to RAI FM Olg for additional details on this limit.

ZOIs in Wall and Corner Locations Ignition source fire scenarios are postulated in open, wall, and corner configurations. Although the original GFMT report is limited to open configurations, wall and corner effects are determined using the 'Image' method, which postulates an equivalent plume by increasing the fire size and enclosure volume by a factor of two or four depending on the fire location. In addition, wall and comer effects on both the ZOI and the HGL are specifically addressed in Report 0027-0009-014-004, Rev. 0, Report 0027-0009-014-005, Rev. 0, Report 0027-0053-000-002, Rev. 0, and Report 0027-0053-000-003, Rev. 0 for transient, electrical panel, and motor ignition sources.

ZOIsfor Large Dimension ElectricalPanels The original GFMT report (1SPH02902.030) was derived for panels having plan dimensions up to 0.9 X 0.6 m (3 X 2 ft.). The dimensions primarily affect the extent of the horizontal component of the ZOI that is below the top of the panel. This ZOI component is calculated from an energy balance at the panel surface, and the target exposure mechanism is a heated radiating vertical plane. Consequently, changes in the panel dimensions affect the dimensions of the radiating plane, which in turn affects the geometry configuration factor between the target and the radiating plane.

Enclosure I to L-2014-109 Page 28 of 106 The lower horizontal ZOI dimension is the limiting horizontal ZOI dimension and is used in the fire PRA as the basis for determining the affected target set.

An approximate upper limit for the ZOI dimensions based on the conservative 0.9 X 0.6 111 (3 X 2 ft.) plan dimensions may be estimated by comparing against a limiting open panel configuration.

In this case, the maximum heat transferred across one boundary would be given through the definition of the emissive power and a radiation area as follows:

OQmax = AbE (FM 04-1)

Where Ob,,nax is the maximum heat that can be transferred across a vertical boundary of an electrical panel (kW [Btu/s]), Ab is the area of the boundary (in2 [ft2]), and E is the flame emissive power (kW/m 2 [Btu/s-ft2]). Assuming the maximum average flame emissive power over the panel boundary is 120 kW/m 2 (10.6 Btu/s-ft2) based on Section 3-10 of the SFPE Handbook of Fire ProtectionEngineeringand data provided in Combustion and Flame, No. 139, pp. 263-277, the maximum heat that could be transferred across a vertical boundary via thermal radiation is about 235 kW (223 Btu/s) if the heat transferred across an open boundary is considered to be an Lipper limit on the boundary heat losses in any one direction. To link this heat loss to the postulated fire size, the radiant fraction is used, which is reasonably approximated as 0.3 for enclosure fires per Section 3-8 of the SFPEHandbook of FireProtectionEngineering.Dividing the maximum boundary heat loss of 235 kW (223 Btu/s) by the radiant fraction (0.3) results in the largest fire size for which the lateral ZOI dimensions would be conservative, or 783 kW (742 Btu/s). This value exceeds the severe fire heat release rate used to characterize both the multiple bundle (717 kW [680 Btu/s] based on the Bin 8 heat release rate) and single bundle (211 kW [200 Btu/s])

electrical panels. This result is based on a radiant fraction of 0.3; if a value at the upper end of the often cited range 0.3 - 0.4 is assumed per Section 3-8 of the SFPEHandbook of FireProtection Engineering,the largest fire size for which the lateral ZOI dimensions would be conservative, or 588 kW (557 Btu/s). However, this would be based on all heat losses being directed toward the target. The internal temperature during a fully developed enclosure fire would be greater than 600'C (1,1 12'F), which suggest the heat losses from all boundaries, except the open boundary, would be on the order of 110 kW (104 Btu/s). This means that the maximum total energy that could radiate toward the target via thermal radiation would be about 600 kW (253 Btu/s) X 0.4 or 240 kW (227 Btu/s). This is comparable to the maximum boundary heat loss via thermal radiation (235 kW [223 Btu/s]), which indicates the conclusion applies over a wider range of radiant fractions when the additional boundary heat losses are included. The limiting fire size (and plan dimension for the panels) for wall and corner locations is increased by a factor of two and four due to the symmetry planes assumed in the 'Image' method and applies when the lower ZOI dimension is limiting. There are no electrical panel ignition sources evaluated at PSL using the GFMT approach with a heat release rate greater than 783 kW (742 Btu/s) in an open configuration, 1,566 kW (1,484 Btu/s) in a wall location, or 3,132 kW (2,970 Btu/s) in a comer configuration.

Therefore, although the specific limitation in the GFMT report is exceeded, there is no adverse effect on the ZOI dimensions for the PSL applications.

Flame Height Limitationfor ZOIs The original Generic Fire Modeling Treatment report limits the application of the ZOIs to situations in which the flames remain lower than the ceiling height. Subsequent analysis presented in Report 0027-0009-014-004, Rev. 0 and Report 0027-0053-000-003, Rev. 0 indicates that the ZOls remain conservative provided the ceiling jet temperature at the ZOI boundary remains less

Enclosure I to L-2014-109 Page 29 of J06 severe than the threshold damage temperature for the cable target. The minimum ceiling height above the fire base is listed in Report 0027-0053-000-003, Rev. 0 for transient ignition sources and in Report 0027-0009-014-004, Rev. 0 for fixed ignition sources and ignition sources with secondary combustibles. The minimum ceiling height above the fire base is generally in the 0.91 -

1.5 in (3 - 5 ft.) range, but it does vary with the scenario and in cases that evolve with time, the time after ignition. If the flame height constraint is not met, there are two options available to address the limitation. The first is to assume a hot gas layer condition develops that damages all targets within the enclosure, which avoids the need to apply the ZOI. The second option is to use an increased ZOI dimension to account for the ceiling jet extension. There are no known situations at PSL in which the second option would be applicable.

ZOIs andHot Gas Layer Temperatluresfor Scenarios with Secondao, Combustibles The ZOIs for configurations involving secondary combustibles have been developed using the methods described in NUREG/CR-6850 and NUREG/CR-7010, Volume 1. The ZOI and HGL tables are provided in Report 0027-0009-014-004, Rev. 0, Report 0027-0009-014-005, Rev. 0, and Report 0027-0053-000-002, Rev. 0. The ZOI and HGL data provided in these reports are applied to scenarios at PSL in which secondary combustibles are involved. As such, the original limitation in the GFMT report (1SPH02902.030) does not apply when the updated documentation is applied.

Application of GFMT CFAST Results The Generic Fire Modeling Treatments approach involves CFAST calculations for Generic enclosures that minimize the heat losses to the boundaries. The key CFAST model limits that apply to the PSL CFAST evaluations as identified in NIST SP 1026 and NUREG-1 824, Volume 5 are as follows:

• Maximum vent size to enclosure volume ratio should not exceed 2 m-' (0.61 ft-')

• Maximum heat release rate per unit volume of I MW/rn 3 (0.027 Btu/ft3 )

• Maximum enclosure aspect ratio of five (length to width)

The approach adopted in both the generic enclosure analysis is to evaluate a range of ventilation fractions, from 0.001 to 10 percent of the enclosure boundary. Given that the width is set equal to the length in both the generic evaluations, the maximum vent size to enclosure volume ratio is given by the following equation:

W+2H (FM 04-2) 5HW Where W is the enclosure width (in [ft]), and H is the enclosure height. Based on the definition of the generic volume, the enclosure height is one-half the enclosure width, so that the vent size to enclosure volume ratio can only exceed 2 m-' (0.61 ft-1) if the ceiling height is 0.2 m (0.7 ft.) or less. Because the minimum ceiling height considered is 1.4 in (4.5 ft.), this condition is necessarily met in the tabulated data. Further, there are no spaces at PSL for which the GFMT approach is applied that has an actual ceiling height that is lower than 0.2 m (0.7 ft.), thus this limitation is met in practice as well.

The maximum heat release rate per unit volune for CFAST is I MW/mi (0.027 Btu/ft3 ), above which the two layer approach breaks down. The smallest volume considered in the GFMT reports is 10 ni3 (353 ft3) and the largest fire size considered is 10 MW (9470 Btu/s). As such, the heat release rate per unit volume limitation is met in all cases considered.

Enclosure I to L-2014-109 Page 30 of 106 A third CFAST model limitation of the Generic Fire Modeling Treatments approach relates to the maximum aspect ratio of an enclosure for which hot gas layer data is applied. The hot gas layer information is provided for enclosures having an aspect ratio up to five, per NUREG-1824, Volume 5, Section 3.2. In situations where the model is applied to enclosures having a larger aspect ratio, the behavior transitions to a channel flow typical of a corridor configuration.

Localized effects in the vicinity of the fire could be more severe than the average conditions throughout the enclosure length, and thus a non-conservative result could be generated. NUREG-1934 describes a method to apply a fire model in a conservative manner under these conditions.

This method involves the modification of the enclosure dimensions such that the application falls within the model limitation and the hot gas layer temperature results are conservative. This modification has been applied to fire scenarios postulated in spaces having an aspect ratio greater than five at PSL. As such, the three CFAST limitations are met for the PSL GFMT applications.

2. Main Control Room Abandonment Calculation The key CFAST model limits that apply to the PSL control room abandonment calculations as identified in NIST-SP-1026 and NUREG-1824, Volume 5 are as follows:

" Maximum vent size to enclosure volume ratio should not exceed 2 m-1 (0.61 ft-1 )

• Maximum enclosure aspect ratio of five (length to width)

" Maximum heat release rate per unit volume of 1 MW/m 3 (0.027 Btu/ft3 )

The maximum vent sizes considered in the control room abandonment calculation consist of a single open door and minor boundary leakage areas. The total vent size to enclosure volume ratio for this opening combination remains much less 2.0 m- (0.61 ft1 ) and indicates this limitation is met for all CFAST evaluations.

The overall volume of the enclosures considered range from 21 m 3 (740 ft3) for an office space to 1,595 m3 (56, 300 ft3) for the Unit I control room. The maximum fire size considered in any area is less than 5 MW (4,740 Btu/s); as such, the maximum heat release rate per unit volume constraint is met.

There are two primary spaces used to evaluate the baseline fire scenarios in each of the control rooms: the control room proper and the surrounding staff or support areas. The approximate aspect ratio for each of these spaces is as follows:

" Unit 1 control room proper: 1.16

" Unit 2 control room proper: 1.28

• Unit 1 staff support area: 1.18

• Unit 2 support area: 1.39 These aspect ratios are less than the CFAST limit of five and indicate this limitation is met for all baseline fire scenario CFAST evaluations in the control room abandonment calculation.

Enclosure I to L-2014-109 Page 31 of 106 PSL RAI FM 06a NFPA 805, Section 2.7.3.5, "Uncertainty Analysis," states: "An uncertainty analysis shall be performed to provide reasonable assurance that the performance criteria have been met."

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."

Regarding the uncertainty analysis for fire modeling:

a. Describe how the uncertainty associated with the fire modeling input parameters was accounted for in the fire modeling analyses.

RESPONSE

Fire model uncertainty associated with the fire model input parameters was not explicitly accounted for in the fire PRA development at PSL. However, the uncertainty associated with specific fire modeling parameters is addressed through the use of a conservative and bounding analysis and sensitivity studies are provided in the various documents that demonstrate this. There are five primary areas at PSL in which fire modeling parameter uncertainty is applicable:

" The control room abandonment analyses (Report 0027-0009-014-001, Rev. 0 and Report 0027-0009-014-003, Rev. 0);

• The hot gas layer (HGL) tabulations as contained in 1SPH2902.030, Rev. 0, Report 0027-0009-014-005, Rev. 0, and 0027-0053-000-002, Rev. 0;.

• The ZOI tabulations as contained in I SPH2902.030, Rev. 0, Report 0027-0009-014-004, Rev. 0, and Report 0027-0053-000-003, Rev. 0;

" The closed panel heat release rate estimates as contained in "Supplemental Generic Fire Modeling Treatments: Closed Electrical Panels", Rev. B; and

• The calculation of the cable tray heat flux in the containment building as contained in Report 6372.

The input parameter uncertainty for each group of calculations is described separately in this RAI response.

MCR Abandonment Calculation.The updated control room abandonment calculations (0027-0009-0 14-001, Rev. 0 and Report 0027-0009-014-003, Rev. 0) are structured to provide a reasonably conservative abandonment time for a given heat release rate input over a range of potential input parameter values. The MCR abandonment calculation provides baseline cases for six forced and natural ventilation combinations in each control room and effectively provides a sensitivity assessment on these parameters. Specifically, for a given fire scenario considered in the fire PRA, the shortest abandonment time among the various natural ventilation configurations is selected (see response to RAI FM 0 lb). In order to ensure the analysis results are conservative relative to the uncertainty in other parameters, a fire modeling sensitivity analysis is provided in Attachment 2 of Report 0027-0009-014-001, Rev. 0 and Attachment 2 of Report 0027-0009-014-003, Rev. 0. The sensitivity analysis is used to justify the selection of the input parameter values for the baseline cases using both an absolute abandonment time variation criterion (fifteen percent) and a variation in the total probability of abandonment criterion (fifteen percent). The total probability of

Enclosure I to L-2014-109 Page 32 of 106 abandonment is defined as a product of the severity factor for a particular heat release rate bin and the probability of non-suppression summed over the applicable number of heat release rate bins. A value of fifteen percent is selected as a basis for determining a significant effect because it is consistent with the theoretical and observed uncertainty in calorimeter heat release rate measurements as described in the SFPEHandbook of FireProtectionEngineering, Section 3-2 and provides an uncertainty range in the output parameters that is significantly narrower than the uncertainty in the total probably of control room abandonment as computed using the fifth and ninety-fifth percentile suppression rate parameters (i.e., A) as provided in NUREG/CR-6850 (see RAI FM 01 f response).

A three bin summation over the heat release rate distribution is used in the sensitivity analysis whereas a fifteen bin summation is used when comparing the baseline fire scenarios. The sensitivity analysis demonstrates that the parameter sensitivity may be grouped as follows over the range of parameter uncertainty:

" The parameter does not significantly affect the analysis results over the potential range of values that could be assigned to the parameter (Sensitivity Group 1);

" The parameter does affect the analysis results, but value selected for the baseline case is conservative (Sensitivity Group 2); and

" The parameter does affect the analysis results and the value selected for the baseline case is not conservative (Sensitivity Group 3).

The sensitivity parameters that fall into the first and second group provide a basis for a conservative parameter assumption and baseline model configuration with respect to the parameter. The sensitivity parameters that fall into the third group were reviewed further because they indicate that the baseline configuration is not conservative with respect to uncertainty in the parameter. However, the process used in developing and documenting the parameter sensitivity analysis is iterative and, in most cases, a result that falls into the third category initiates a revision to the baseline assumption such that the updated result falls into the first or second groups. Cases in which the baseline is not revised to eliminate a parameter assumption that falls into Sensitivity Group 3, such as the initial ambient temperature, are used to establish a limit on the results applicability. The response to RAI FM 01 f, Attachment 2 of Report 0027-0009-014-001, Rev. 0, and Attachment 2 of Report 0027-0009-014-003, Rev. 0 provide specific details on the results of the parameter sensitivity analyses for the Unit I and Unit 2 MCR abandonment calculations. Based on Attachment 2 of Report 0027-0009-014-001, Rev. 0, and Attachment 2 of Report 0027-0009-014-003, Rev. 0, the baseline results presented in the control room abandonment calculations are considered conservative with respect to uncertainty in the parameter values.

Hot Gas Layer Tabulations.The Generic Fire Modeling Treatments report (1 SPH02902.030, Rev.

0), Report 0027-0009-014-005, Rev. 0, and 0027-0053-000-002, Rev. 0 provide times at which the hot gas layer in a generic enclosure will exceed specified temperature thresholds. The computations are performed using the zone computer model CFAST, version 6.0.10 and Version 6.1.1. The methodology for computing the hot gas layer tables is described in detail in Section 6.3 and Appendix B of the Generic Fire Modeling Treatments report (1 SPH02902.030, Rev. 0).

Enclosure I to L-2014-109 Page 33 of 106 Essentially, CFAST is used to balance energy and mass flow through openings and the time at which the hot gas layer temperature reaches a threshold value is reported regardless of the hot gas layer height. The primary input parameters include the fire size, the enclosure geometry, the fuel properties, the opening characteristics, the boundary material properties, and the initial ambient temperature.

The fire size is a prescribed input per NUREG/CR-6850 or is specified with a particular set of input parameters and subject to the parameter constraints (ignition source - cable tray fire scenarios). The room geometry is selected in such a way as to minimize the heat losses to the boundaries and thus varies from volune to volume. Under this assumption, the height of the enclosure necessarily varies with the volume. However, a sensitivity analysis is conducted on the room enclosure shape (Section B.4.4 of the Generic Fire Modeling Treatments report (1SPH02902.030, Rev. 0)) and it is shown that minimizing the enclosure boundary surface area provides a bounding or nearly bounding result for a given enclosure volume when the length to width aspect ratio is varied firom 1: 1 to 1:5 in the cases considered. As the aspect ratio increases, a significant reduction in the temperature is observed indicating that spaces that deviate from a 1: 1 aspect ratio have an increasing safety margin embedded in the hot gas layer temperature results.

The selection of the fuel properties is evaluated in Sections B.4.1 and B.4.2 of the Generic Fire Modeling Treatments report (1SPH02902.030, Rev. 0). Fuel properties are varied over a large range of potential values and the most adverse combination is selected to represent all fuels. In this case a relatively low soot yield material is used because it reduces the radiant heat losses from the hot gas layer to the enclosure boundaries and maximizes the hot gas layer temperature.

The opening characteristics are described in terms of a boundary fraction and are varied over a range of 0.001 - 10 percent in the baseline cases. The hot gas layer associated with the most adverse ventilation case is selected in the fire PRA among the reported ventilation conditions for a given fire size and enclosure volume. The key input parameter that is set is the ventilation geometry (length, width, and base height) given a vent fraction. Section B.4.5 of the Generic Fire Modeling Treatments report (1 SPH02902.030, Rev. 0) provides a sensitivity analysis on the effects of various vent orientations and placements on the predicted temperature. A total of fifty-four vent configurations were examined for the baseline enclosures. It is found that the bounding case can be one of three orientations: one in which the vent width is equal to the enclosure width, located either at the ceiling or at the floor and one in which the vent height is equal to the enclosure height.

All hot gas layer tables reported in the Generic Fire Modeling Treatments report (1SPH02902.030, Rev. 0), Report 0027-0009-014-005, Rev. 0, and 0027-0053-000-002, Rev. 0 are based on the most adverse hot gas layer condition among the three vent orientations and thus represent the bounding configuration for the vent geometry.

The boundary material properties are defined as concrete having the lowest thermal diffusivity reported among available data as described in Section B.4.3 of the Generic Fire Modeling treatments report. The thermal diffusivity of the selected concrete, defined as the thermal conductivity divided by the heat capacity and density, is 5.9 X 10-7 m2/s (6.3 X 10-6 ft2/s) and is about thirty percent lower than the value of 8.9 X 10-7 m 2/s (9.6 X 10-6 ft2/s) recommended in NUREG-1 805. This conservatively biases the results for the boundary materials, though it is shown in Section B.4.3 of the Generic Fire modeling Treatments report (1 SPH02902.030, Rev. 0) that the results are not conservative if they are applied to spaces bound with thermal insulation, lightweight concrete, or gypsum wallboard.

Enclosure I to L-2014-109 Page 34 of 106 The initial ambient temperature is assumed to be 20'C (68'F) in the Generic Fire Modeling Treatments report (1SPH02920.030, Rev. 0), Report 0027-0009-014-005, Rev. 0, and 0027-0053-000-002, Rev. 0. Although an ambient temperature of 20'C (68°F) is not a conservative and bounding assumption, the effect is readily bound by other conservative aspects of the model approach such as the enclosure geometry, ventilation effects, fuel properties, and hot gas layer position.

Finally, a significant conservatism embedded in the CFAST model results is the specification of an adiabatic floor. Radiant heat losses from both the fire and the hot gas layer to the floor are not credited with reducing the hot gas layer temperature. This assumption is expected to conservatively bias the temperature predictions.

Based on the overall conservative bias associated with the CFAST model parameters (collectively), the hot gas layer tables reported in the Generic Fire Modeling Treatments report (1 SPH02902.030, Rev. 0), Report 0027-0009-014-005, Rev. 0, and 0027-0053-000-002, Rev. 0 are considered conservative with respect to uncertainty in the parameter values.

ZOI Calculations.The Generic Fire Modeling Treatments report (1 SPH2902.030, Rev. 0), Report 0027-0009-014-004, Rev. 0, and Report 0027-0053-000-003, Rev. 0 provide ZOI dimensions for various ignition sources and combination ignition sources - cable tray configurations for which fire PRA fire scenarios are developed. The tabulated ZOI dimensions are all based on the methodologies described in Generic Fire Modeling Treatments report (ISPH2902.030, Rev. 0),

except the ZOI dimensions for the ignition source - secondary combustible configurations include the physical offset associated with both the cable tray arrangement and the fire spread in the cable trays. The ZOI dimensions essentially consist of a vertical component derived from a plume exposure correlation and one or more horizontal components, each derived from a radiant heat flux calculation.

The vertical plume calculation uses an empirical model that requires as inputs the fire heat release rate, the ambient temperature, and fire diameter. The fire size is an input parameter specified by NUREG/CR-6850. The fire diameter and ambient temperature are the primary parameters subject to uncertainty. In this case, the fire diameter in the original Generic Fire Modeling Treatments report (1SPH2902.030, Rev. 0) provides ZOI dimensions assuming a variable diameter (as characterized using the heat release rate and heat release rate per unit area). A heat release rate per unit area range between 200 kW/m 2 (17.6 Btu/s-ft2) and 1,000 kW/m 2 (88.1 Btu/s-ft2) is used for transient combustible materials and range Lip to 3,000 kW/m 2 (264 Btu/s-ft2) for electronic panels.

The baseline ambient temperature assumed in the original Generic Fire Modeling Treatments report (1 SPH2902.030, Rev. 0) is 20'C (68°F) with a maximum application limit of 80'C (176°F).

The baseline ambient temperature selection in Report 0027-0009-014-004, Rev. 0, and Report 0027-0053-000-003, Rev. 0 is varied from 20'C (68°F) to 80'C (176°F) and is thus not subject to assumption or uncertainty, at least within the limits of applicability.

The maximum effect of an elevated initial ambient temperature on the ZOI dimensions for transient fuel package fires is provided in Report 0027-0053-000-003, Rev. 0 and in Report 0027-0009-014-004, Rev. 0 for various ignition source - cable tray configurations. The ZOI dimension may change by about two to five percent when the ambient temperature is 40'C (I 04'F) and ten to twenty percent if the ambient temperature is 80'C (176°F) based on various ignition sources and cable tray configurations evaluated in open, wall, and comer locations. This differential is expected to be readily bound by the conservatisms that are embedded in the ZOI development.

Enclosure I to L-2014-109 Page 35 of 106 These conservatisms relative to a transient fuel package fire include the use of steady-state target damage thresholds, a fire diameter that maximizes the ZOI dimension, the use of a ZOI box rather than a cone, and the selection of the most adverse result among a range of methods. Report 0027-0009-014-004, Rev. 0 also provides an assessment of the calculation results to uncertainty in the assumed fire diameter and it is shown that the variation is less than the ZOI resolution implemented in the field.

An additional offsetting conservative factor for the panel fires relative to an elevated ambient temperature environment is the assumed heat release rate per unit area for the electronic panel fires for the vertical ZOI dimension is effectively 3,000 kW/m2 (264 Btu/s-ft2). This means that the characteristic fire dimension for the 9 8 th percentile panel fires is on the order of 0.26 - 0.48 mn (0.9

- 1.6 ft.). The characteristic dimension for the electronic panels as evaluated using the NUREG/CR-6850 guidance would be based on the panel top surface area and will typically be on the order of 0.6 - 1.2 mn (2 - 4 ft.). This indicates a significant bias is introduced by assuming the fire plan area can occupy only a fraction of the panel top. An additional conservative bias is introduced in setting the base location of the vertical ZOI dimension. Per NUREG/CR-6850, Supplement 1, the fire base height may be set 0.3 mn(1 ft.) below the panel top (if the panel does not have significant openings in the top). The vertical ZOI dimensions for the electronic panels reported in the Generic Fire Modeling Treatments report (1SPH2902.030, Rev. 0) use the panel top as the base height reference for the vertical ZOI dimension. This introduces a uniform 0.3 in (1 ft.)

bias in all vertical ZOI dimensions for the electronic panels. As such, the vertical ZOI dimension is calculated using bounding input parameters when viewed collectively.

In the case of the horizontal ZOI dimension for both the fixed ignition sources and the transient ignition sources, the maximum distance as obtained using the more severe prediction among both a solid flame model and the Point Source Model (PSM). The ZOI dimensions for the ignition source

- cable tray configurations are obtained in a similar manner, but include heat flux calculations for using the total heat release rate of all heat sources and the heat flux calculations using the sum of the ignition source and cable trays contributions. The PSM requires as input the fire size and the fire radiant fraction, which is assumed to be 0.4. The fire size is a prescribed input per NUREG/CR-6850. Based on SFPEHandbook of FireProtection Engineering,Section 3-10, the effective radiant fraction for conservative (but not bounding) results is 0.21. A bounding result is obtained when a safety factor of two is used. By assuming a radiant fraction of 0.4, an effectively bounding result is therefore obtained. The solid flame heat flux model requires the fire size and fire diameter as input parameters. The fire size is a prescribed input per NUREG/CR-6850. The fire diameter is varied via the heat release rate per unit area parameter. In this case, the most adverse 2

fire diameter is intermediate with a heat release rate per unit area of about 350 - 400 kW/1i (30.8

- 35.2 Btu/s-ft2 ), depending on the specific case. The value that yields the maximum ZOI dimension is the value used in the analysis.

The electrical panel ignition source ZOIs have additional conservative margins by including an additional calculation that is more conservative than the approach suggested in NUREG/CR-6850, Supplement 1. Per NUREG/CR-6850, Supplement 1, the fire base is located 0.3 m (1 ft.) below the top of the panel and is typically modeled assuming the panel boundaries do not exist (open fire).

The horizontal ZOI dimensions developed in the Generic Fire Modeling Treatments report (1 SPH02902.030, Rev. 0) include an upper horizontal ZOI dimension that is computed in this manner and a lower ZOI dimension that assumes internal flame impingement on the panel boundary. This flame impingement imposes a 120 kW/m 2 (10.6 Btu/s-ft2 ) heat flux on any internal

Enclosure 1 to L-2014-109 Page 36 of 106 boundary that radiates outward from a single side. The lower horizontal ZOI dimension is significantly larger than the upper fire plume base horizontal dimension, typically by a factor of two (compare Tables 5-16 and 5-17 in the Generic Fire Modeling Treatments report (1 SPH02902.030, Rev. 0), for example). The fire PRA selects the most adverse horizontal ZOI dimension and thus incorporates this bias directly.

Based on the overall conservative bias associated with the input parameter, both the horizontal and vertical ZOI dimensions reported in the Generic Fire Modeling Treatments report (1SPH02902.030, Rev. 0), Report 0027-0009-014-004, Rev. 0, and Report 0027-0053-000-003, Rev. 0 are considered conservative with respect to parameter uncertainty.

Containment Cable Tray Fire Calculation.The focus of Report 6372 is to determine the heat flux exposure from a burning cable tray stack to an adjacent cable tray stack in order to assess the potential for raceway damage. The baseline scenarios indicate that the maximum heat flux to an adjacent cable tray stack is less than the threshold value for damage by a factor of two, indicating there is an inherent safety margin of two in the analysis results. A comprehensive parameter sensitivity analysis is provided in Section 9 of Report 6372 in which the parameter input values are varied significantly. Per Table 11 d in Report 6372, the sensitivity analysis indicates that the baseline results are conservative relative to a parameter selection that is consistent with the values recommended in NUREG/CR-7010, Volume 1. As such, it is concluded that the baseline scenarios are considered conservative with respect to parameter uncertainty.

ClosedPanel Heat Release Rates. The purpose of the closed panel heat release rate analysis

("Supplemental Generic Fire Modeling Treatments: Closed Electrical Panels," Rev. B) is to provide heat release rate estimates for panels that are closed and have limited ventilation. A parameter sensitivity analysis is provided in Section 5.4.3 and indicates that the baseline scenarios are considered conservative with respect to parameter uncertainty.

PSL RAI FM 06b NFPA 805, Section 2.7.3.5, "Uncertainty Analysis," states: "An uncertainty analysis shall be performed to provide reasonable assurance that the performance criteria have been met."

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."

Regarding the uncertainty analysis for fire modeling:

b. Describe how the "model" and "completeness" uncertainties were accounted for in the fire modeling analyses.

RESPONSE

Fire model "model" and "completeness" uncertainty was not explicitly accounted for in all fire modeling evaluations or incorporated into the fire PRA at PSL. However, the uncertainty associated with fire modeling "model" and "completeness" uncertainty is addressed through the use of a conservative and bounding analysis. There are five primary areas at PSL in which fire modeling "model" and "completeness" uncertainty is applicable:

Enclosure Ito L-2014-109 Page 37 of 106

• The control room abandonment analyses (Report 0027-0009-014-001, Rev. 0 and Report 0027-0009-014-003, Rev. 0);

• The hot gas layer (HGL) tabulations as contained in 1SPH2902.030, Rev. 0, Report 0027-0009-014-005, Rev. 0, and 0027-0053-000-002, Rev. 0;

" The ZOI tabulations as contained in 1SPH2902.030, Rev. 0, Report 0027-0009-014-004, Rev. 0, and Report 0027-0053-000-003, Rev. 0;

" The closed panel heat release rate estimates as contained in "Supplemental Generic Fire Modeling Treatments: Closed Electrical Panels", Rev. B; and

" The calculation of the cable tray heat flux in the containment building as contained in Report 6372.

MCR Abandonment Calculation.The MCR abandonment calculations (Report 0027-0009-014-001, Rev. 0 and Report 0027-0009-014-003, Rev. 0) provide an assessment of the model uncertainty in Section A2.3 of Attachment 2 of each document using the methods described in NUREG-1934. The uncertainty is assessed for a range of heat release rate bins associated with the primary baseline fire scenario ignition sources in control room area. Table A2-23 of Report 0027-0009-014-001, Rev. 0 indicates that the maximum probability the actual Unit 1 control room abandonment time is fifteen percent shorter than the model predicted abandonment time is 0.0798 percent for the closed electrical panel fire scenarios and is less than 0.0388 percent for the transient and open panel electrical panel fire scenarios. Similarly, Table A2-21 of Report 0027-0009-014-003, Rev. 0 indicates that the maximum probability the actual Unit 2 control room abandonment time is fifteen percent shorter than the model predicted abandonment time is 13.0 percent for the closed electrical panel fire scenarios in the MCR and is less than 0.025 percent for the transient fire scenarios and the closed electrical panel fire scenarios in the support area. Based on the sensitivity analysis presented in Section A2.2 of each document, it is concluded that there is a significant conservative bias in the baseline fire model results that bounds the "model" and "completeness" uncertainty.

Hot Gas Layer Tabulations.The hot gas layer tables are computed using the zone computer model CFAST, Version 6.0.10 and 6.1.1 in the Generic Fire Modeling Treatments report (1 SPH2902.030, Rev. 0) and in the detailed evaluations of secondary combustible configurations (Report 0027-0009-014-005, Rev. 0, and 0027-0053-000-002, Rev. 0). As described in the response to RAI FM-06a, there are a significant number of parameters that are conservatively biased in the model, including the fuel properties, the combustion properties, the boundary properties, the adiabatic floor surface, and the vent configuration. In addition, for a given CFAST geometry, the fire PRA selects the most adverse scenario among a ventilation range between 0.001 and 10 percent of the enclosure boundary area, each of which in turn is based on the most adverse result among three potential vent locations. The approximate effect of each of these parameters (except for the adiabatic floor surface) on the temperature results are provided in Appendix B of the Generic Fire Modeling Treatments report (I SPI-102902.030, Rev. 0). For example, Figures B4-7a through B4-7c and B4-8a through B4-8b in the Generic Fire Modeling Treatments report (1SPH02902.030, Rev.

0) demonstrates that the effect of changing the thermal diffusivity of the boundary materials on the steady state temperature is roughly proportional to the change in the thermal difflisivity at least when centered on a value of 5.9 X 10-7 m 2/s (6.3 X 10-6 ft2/s). Given that this value is about thirty percent lower than the thermal diffusivity recommended in NUREG-1805 for normal weight concrete, a comparable test case would have a temperature reduction of about 93°C (1671F). This

Enclosure I to L-2014-109 Page 38 of 106 sensitivity alone is comparable to the temperature change necessary to reduce the "model" and "completeness" uncertainty to less than two percent as determined using the methods described in NUREG-1934 (i.e., the probability of exceeding the critical value due to model uncertainty is less than two percent when the hot gas layer temperature is about 60°C (108'F) lower than the critical value). When all conservatively biased input parameters are considered together, it is expected that the collective effect on the predicted temperature will result in a low probability of exceeding a threshold value at a tabulated time.

Consequently, it is concluded that fire model "model" and "completeness" uncertainty would not contribute significantly to the risk uncertainty because it is sufficiently bound by the conservatisms in the CFAST hot gas layer analyses.

ZOI Calculations.The ZOI computations provided in the Generic Fire Modeling Treatments report (1SPH02902.030, Rev. 0) and in the analysis of transient fuel packages and secondary combustible configurations (Report 0027-0009-014-004, Rev. 0, and Report 0027-0053-000-003, Rev. 0) rely on a plume centerline temperature, an open source fire radiant heat flux computation, and a radiant heat flux computation from a heated panel or burning array of cables. The plume centerline temperature computation is shown in NUREG-1934 and NUREG-1824, Volume 3 to have a non-conservative bias and a relatively large standard deviation. However, the application considered did not explicitly account for the hot gas layer temperature changes, which are the expected source of the bias and variation. Similar plume correlations used by CFAST and MAGIC show a conservative bias and smaller variation. The application of the plume correlations is limited in the Generic Fire Modeling Treatments report (1SPH02902.030, Rev. 0), Report 0027-0009-014-004, Rev. 0, and Report 0027-0053-000-003, Rev. 0 to 80'C (176°F) or less through the use of the modified critical heat flux, which is intended to adapt the models for elevated internal temperatures. Further, as discussed in the response to RAI FM-06a, the vertical plume ZOI dimension may have as much as a 0.3 m (1 ft.) conservative bias embedded based on the assumed diameters and the base elevations relative to NUREG/CR-6850 and NUREG/CR-6850, Supplement I guidelines.

The horizontal ZOI dimensions are computed using a radiant heat flux model with the radiant fraction set to about two times the value recommended in the SFPEHandbook of Fire Protection Engineering, Section 3-10. Effectively, the radiant heat flux has a bias of two explicitly embedded in the calculation. The probability that the heat flux at a fixed location would exceed the IEEE-383 qualified/thermoset cable limit of 5.7 kW/min (0.5 Btu/s-ft2) given a prediction of 2.70 kW/m 2 (0.25 Btu/s-ft2 ) (i.e., removed conservative bias) may be computed using the methods described in NUREG-1934 with the bias and normalized variance for the radiant heat flux models, which are 2.02 and 0.59. The resulting probability is nearly zero. In the case of the electrical panel fires, an additional margin is provided through themuse of a conservative model beyond that required in NUREG/CR-6850 and NUREG/CR-6850, Supplement 1 for portions of the ZOI below the panel.

Because fire PRA uses the most adverse horizontal ZOI dimension above or below the panel, this additional model introduces a second conservative factor.

Consequently, it is concluded that fire model "model" and "completeness" uncertainty either Would not contribute to the risk uncertainty or are bound by the conservatisms in the analysis, depending on the ZOI dimension considered.

Enclosure I to L-2014-109 Page 39 of 106 Containment Cable Tray Fire Calculation.The focus of Report 6372 is to determine the heat flux exposure from a burning cable tray stack to an adjacent cable tray stack in order to assess the potential for raceway damage. The basic model applied is the Point Source Model with a radiant fraction of 0.4 and the overall safety margin of the baseline result relative to the threshold value is approximately two. As described for the ZOI calculations, the probability that the heat flux at a fixed location would exceed the IEEE-383 qualified/thermoset cable limit of 5.7 kW/m2 (0.5 Btu/s-ft2) given a prediction of 2.70 kW/m 2 (0.25 Btu/s-ft2) (i.e., removed conservative bias from the model) may be computed using the methods described in NUREG-1934 with the bias and normalized variance for the radiant heat flux models, which are 2.02 and 0.59. The resulting probability is nearly zero. This does not include the additional conservative bias provided by the safety margin of about two. As such, it is concluded that fire model "model" and "completeness" are bound by the conservatisms in the analysis.

Closed PanelHeat Release Rates. The purpose of the closed panel heat release rate analysis

("Supplemental Generic Fire Modeling Treatments: Closed Electrical Panels," Rev. B) is to provide heat release rate estimates for panels that are closed and have limited ventilation. There is no specific consideration for model uncertainty; however, the parameters are selected to provide a conservative result as described in the response to FM 06a. The parameter selections are expected to bound the "model" and "completeness" uncertainty associated with the calculation. A simple comparison of the predicted peak heat release rate with the observed peak heat release rate data for full scale closed panel tests involving cable ftiels supports this assertion. Figure FM06b-1 depicts the predicted and observed peak heat release rate for electrical panel fire tests conducted by VTT and Figure FM 06b-2 depicts the predicted and observed peak heat release rate for fire tests conducted by Sandia National Laboratories. The results show the calculation has a significant conservative bias and indicate the "model" and "completeness" uncertainty are bound by the conservatisms input parameter selections.

-* 600 Door opens Conservative 2 inche prediction E 400 W00 4120 Non-conservative a- X prediction 0 100 200 300 400 500 Measured Peak Heat Release Rate (kW)

Figure FM06b Comparison of Predicted and Observed Peak Heat Release Rates in Closed Panels - VTT Tests.

Enclosure I to L-2014-109 Page 40 of 106 1800 1600

• 1400 "V. 1Conservative 1200 prediction R I 1000 800

• 600 S400 Non-conservative

a. X 200 prediction 0 200 400 600 800 1000 Measured Peak Heat Release Rate (kW)

Figure FM06b Comparison of Predicted and Observed Peak Heat Release Rates in Closed Panels - SNL Tests.

PSL RAI PRA 01a 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. 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 an 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 using 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 result of a peer review are the facts and observations (F&Os) recorded by the peer review and the subsequent resolution of these F&Os.

Clarify the following dispositions to fire F&Os and Supporting Requirement (SR) assessment 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 PP-AI-01 against SR PP-Al The disposition to F&O PP-Al does not address the statement that "[e]vidence was presented to the [peer] reviewer that raceways supporting PRA equipment exist in the 'no man's land' area between unit I and unit 2". Explain whether PRA-credited equipment exists in this area, which, according to the F&O, is not included as part of a PRA-analyzed compartment.

Enclosure Ito L-2014-109 Page 41 of 106

RESPONSE

Walkdowns were performed of the "no man's land" area between Unit 1 and Unit 2 at PSL. The walkdowns confirmed that that no equipment or raceways impacting the Fire PRA exist above ground in the area. Therefore, no further fire scenarios need to be analyzed and there is no impact on the plant risk.

PSL RAI PRA 01b 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. 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 an 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 using 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 result of a peer review are the facts and observations (F&Os) recorded by the peer review and the subsequent resolution of these F&Os.

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

b. F&O PP-C2-01 against SR PP-C2 The disposition to F&O PP-C2 regarding the exclusion of locations within the licensee-controlled area appears to only address whether locations may contain PRA-credited components. Clarify the treatment of those locations that may contain other than PRA-credited fire sources that could threaten credited equipment or cable items by virtue of a nmulti-compartment fire scenario.

RESPONSE

A walkdown has been perfomled to determine if any fire sources outside of the evaluated fire zones could pose a threat to the evaluated zones. The results of this walkdown have confirmed that no external fire sources exist which pose a threat to the evaluated fire zones.

PSL RAI PRA 01c 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. 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 an 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 using an associated ASME/ANS standard (currently ASME/ANS-RA-Sa-2009) as one acceptable approach for determining the technical adequacy of the PRA once

Enclosure I to L-2014-109 Page 42 of 106 acceptable consensus approaches or models have been established for evaluations that could influence the regulatory decision. The primary result of a peer review are the facts and observations (F&Os) recorded by the peer review and the subsequent resolution of these F&Os.

Clarify the following dispositions to fire F&Os and Supporting Requirement (SR) assessment 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 PP-87-01 against SR PP-B7 and F&O PP-C3-01 against SR PP-C3 The disposition identifies that a draft version of the NISYS-1251-0001, "Fire Hazards Analysis (FHA) Review/Update," was included as a reference in the Fire Ignition Frequency Development report; however, as identified in the F&O, the report appears to still require finalization. Describe the status of the draft report, and clarify what changes have been made after the peer review. Additionally, provide a basis for partitioning performed in the Unit 2 turbine building in light of the fact that Section 2.2 of the Fire Ignition Frequency Development report notes that the FHA does not provide for any separation of the Unit 2 turbine building.

RESPONSE

In lieu of incorporation of fire zone boundary data into the Fire Hazards Analysis, a walkdown of all fire barrier boundaries was performed, supplemented by viewing of available pictures of inaccessible barriers. This walkdown confirmed that the credited fire zone boundaries were substantial boundaries (no significant size openings) or that any significant openings were addressed in the hot gas layer (HGL) and multi-compartment analysis (MCA). Where discrepancies were found, the HGL and MCA evaluations were updated to reflect this configuration change. The Fire PRA has been re-quantified, incorporating the impact of all RAIs, with the associated results (Attachment W) provided in Enclosure 2 to this RAI transmittal letter.

The Unit 2 Turbine Building analyses were divided into sub-zones (2_47 (1) through 2_47 (10)) as a means of more clearly specifying the location of a particular fire scenario (allows identification of scenario location within the sub-zone rather than specifying the location as 2_47, somewhere in the turbine building). The identification of targets for each fire scenario was based on location of targets with respect to the ignition source and its associated ZOI. Separation of a target based on location in a different sub-zone was not a basis for excluding the target for the impacted targets.

Only targets separated from the ignition source by substantial barriers, barriers without significant size openings, were excluded from the target list (same criteria used for inclusion of targets within a ZOI but located in an adjacent fire zone). Due to the configuration of the turbine building being of open construction, no hot gas layer formation is postulated and the use of the sub-zones does not impact the HGL and MCA evaluations (no HGL or MCA scenarios are postulated in the turbine building, for either unit).

PSL RAI PRA Old 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. 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

Enclosure I to L-2014-109 Page 43 of 106 staff for adopting an 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 using 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 result of a peer review are the facts and observations (F&Os) recorded by the peer review and the subsequent resolution of these F&Os.

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

d. F&O ES-CI-01 against SR ES-C l Fire-induced instrument failure should be addressed in the fire human reliability analysis (HRA) per NUREG/CR-6850 and NUREG-1921, "EPRI/NRC-RES Fire Human Reliability Analysis Guidelines." The Human Failure Evaluation report and the disposition to F&O ES-Cl indicate that the availability of instrumentation is predicated on the Appendix R SSA; however, the FPRA credits systems and/or functions not previously analyzed by the Appendix R analysis. Additionally, Appendix R instrumentation, which is specifically indicated by bold formatting in Tables A-I through A-4 of Human Failure Evaluation report, is not identified for all human failure events (HFEs). Current fire response procedures (e.g., 2-AOP-1 00.11) also indicate the availability of essential instrumentation on a fire-area basis; however, documentation provided by Tables A-I through A-4 of Human Failure Evaluation report does not. A review of Human Failure Evaluation report also indicates that the availability of instrumentation associated with fire-specific HFEs (i.e., those not carried over from the internal events (IE) PRA) does not appear to have been evaluated. Moreover, the NFPA 805 Recovery Action Feasibility Evaluations (PSL-FPER-1 1-002) appear to solely list local or remote position and/or indication for the "Systems and Indications" criteria. In light of these deficiencies:

i. Describe how instrumentation that is relied on for credited operator actions was identified and verified as available to a level of detail commensurate with the risk importance and quantification of all human error probabilities (HEPs).

ii. Discuss the extent to which cues and indications have been addressed for fire-specific HFEs.

iii. Discuss the degree to which the post-transition fire response procedures are credited to support the FPRA HRA (e.g., to provide procedure-based cues for actions, to identify available or impacted instrumentation).

iv. Describe how fire-induced instrument failures (including no readings, off-scale readings, and incorrect/misleading readings) are addressed by the FPRA model and HRA. Include discussion of the success criteria assumed for this modeling.

RESPONSE

i. As part of the task of replacing the screening HEP values with detailed FPRA HEPs, FPRA-specific HEPs are being added to the quantification fault tree including instrumentation cues. The required cues were correlated to SSD analysis instrumentation

Enclosure I to L-2014-109 Page 44 of 106 which is identified as available instrumentation in the post-fire shutdown procedures. This imposes a failure of the HEP in any scenario where all associated cues are lost due to fire damage. The treatment of cues is consistent with NUREG-1921, specifically discussion regarding failure of cues due to fire in accordance with NUREG-1921 section 4.5.5.

ii. The fire specific HEPs have also been correlated to the SSA instrumentation. The cues for these HEPs have also been incorporated into the fire PRA fault tree.

iii. The development of the fire specific HEPs included the review of post fire shutdown procedures. Revisions to these procedures to ensure that operators are focused on non-fire impacted cues and to update the required actions in a manner that prioritizes fire PRA required actions while identifying, for implementation as time allows, the defense-in-depth operator actions. This update of the procedures is to be performed in support of the final transition to the new NFPA 805 regulatory framework.

iv. Post fire shutdown procedures will identify instrumentation that is available for each fire area to focus the operators on those instruments that are ensured to be available. By reliance of operator actions on the available instrumentation and the correlation of these instruments to the credited HEPs to ensure that availability of the actions only when the associated cues are available.

PSL RAI PRA 01e 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. 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 an 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 using 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 result of a peer review are the facts and observations (F&Os) recorded by the peer review and the subsequent resolution of these F&Os.

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

e. F&O ES-C2-01 against SR ES-C2 There appears to be no documented process for systematically identifying and defining HFEs that may result in an undesired operator response (i.e., error of omission or commission) to spurious or absent cues and indication as recommended in Section 3.4.1 of NUREG-1921. Describe and further justify the process utilized to identify and model such actions on a fire scenario basis.

Enclosure I to L-2014-109 Page 45 of 106

RESPONSE

An operator does not take immediate action on a single instrument indicator or annunciator. The operator will confirm the signal using another instrument/indicator to verify that the signal is valid prior to taking action. Per ASME/ANS Standard RA-S-2008, SR ES-C2, Capability Category II; only one spurious instrument operation is assumed. Therefore, no adverse impact of errors of commission require evaluation given operator confirmation of instrumentation with other instruments prior to initiation of an action.

The HRA update, prepared to support the responses to RAIs PRA 01 .d, 01 .h, 01.1 and 01 .o, addresses errors of omission due to loss of instrument signals. This is accounted for by incorporation of instrument cues into the Fire PRA model fault tree, thereby, failing the operator action if the associated cues are failed by fire.

PSL RAI PRA 01f 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. RG 1.205 identifies NIUREG/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 an 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 using 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 result of a peer review are the facts and observations (F&Os) recorded by the peer review and the subsequent resolution of these F&Os.

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

f. F&O CS-AI 1-01 against SRs CS-C3 and CS-All as well as F&O FSS-E4-01 against SR FSS-E4 The disposition to this F&O indicates that exclusions based on assumptions of routing were eliminated; however, the disposition does not address how the specific scenarios referenced by the F&O were resolved or provide reference to an updated basis for exclusions that were maintained. Address the following:

Clarify that all cables are either fully routed to a level of detail commensurate with the fire scenario analysis or assumed to impact associated components.

ii. If any exclusion originally based on assumed cable routing is maintained in the FPRA, describe the scope of additional analysis performed to support the continued exclusion, and identify where this effort is documented.

iii. Clarify the reconciliation performed between the Appendix R methodology and that used in NFPA 805, including verification that all exclusions based on "Appendix R component disposition codes", as discussed in Section 12.0 of the Fire Scenario Report, remain valid.

Enclosure I to L-2014-109 Page 46 of 106

RESPONSE

i. No exclusions were credited unless supported by traced, detailed and complete, cable routings. No assumed cable routing is applied to the Fire PRA.

ii. Not applicable as all exclusions are supported by traced cable routings.

iii. Applied Safe shutdown analysis (SSD) exclusions are based on circuit analysis applicable to both the SSD analysis and the Fire PRA.

PSL RAI PRA 01g 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. 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 an 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 using 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 result of a peer review are the facts and observations (F&Os) recorded by the peer review and the subsequent resolution of these F&Os.

Clarify the following dispositions to fire F&Os and Supporting Requirement (SR) assessment 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 FSS-Al-01 against SR FSS-AI The disposition to this F&O does not appear sufficient to fully address the deficiencies identified by the peer review. As a result,

i. Identify the locations in which hydrogen piping is routed, and provide technical justification for excluding miscellaneous hydrogen fires in each location.

ii. Provide technical justification for excluding any flammable liquid fire scenarios for fixed fire sources (e.g., pumps, diesel generators, main station transformers, and turbine generators) that may contain combustible liquids (e.g., as fuel or for lubrication).

RESPONSE

i. The hydrogen system at St. Lucie provides hydrogen to the chemistry lab and the volume control tank in the reactor auxiliary building. These systems are provided with pressure monitoring, guard piping and excess flow check valves which preclude the release of a significant quantity of hydrogen which could cause a challenging fire. The design basis for these lines in the event of a complete line break is for the hydrogen concentration to be limited to no more than 2% hydrogen (Unit 1 UFSAR Appendix 9.5A Section 3.15.2 and Unit 2 UFSAR Appendix 9.5A Section 3.15.1). This is a safety factor of 2 to the flammable limit for hydrogen in air. Therefore, based on these design features no specific scenarios associated with a miscellaneous hydrogen fire are postulated.

Enclosure Ito L-2014-109 Page 47 of 106 ii. The current Fire PRA does not exclude any flammable liquid fire scenarios for fixed fire sources that may contain combustible liquids. A review of the fire scenarios for pumps containing significant quantities of oil identified existing scenarios for the Circulating Water Pumps, Main Feedwater Pumps, Heater Drain Pumps and Condensate pumps. A comparison between Unit 1 and Unit 2 identified several missing scenarios for Unit 2 which were walked down and added to the Fire PRA. Diesel generator fire scenarios are base scenarios which include the impact of a fire given that all targets in the room are impacted. Main, Auxiliary and Startup Transformer scenarios are included in the current Fire PRA, as is a turbine generator related fire.

PSL RAI PRA 01h F&O HRA-A2-01 against SR HRA-A2 As noted by the peer review report, there appears to be no documented process for the identification of fire-specific manual actions. Describe the process utilized to identify fire-related human actions relevant to each fire scenario in the FPRA plant response model (including but not limited to NFPA 805 recovery actions).

RESPONSE

The identification of fire specific manual actions to support the fire PRA was a result of a review of high CDF/LERF scenario quantification and evaluation of the potential for actions to mitigate the associated risk. The initial fire PRA quantification was based on no credit for fire specific operator actions. A review of these preliminary results led to the identification of scenario refinements, more detailed circuit analysis, modifications and fire specific manual actions as a means of refining the risk profile to allow for mitigation of the high risk sequences associated with each high risk scenarios.

PSL RAI PRA 01j F&Os UNC-A 1-03 against SR UNC-A I A review of the FPRA Summary report indicates that the parametric uncertainties were propagated to obtain a distribution for fire-induced core damage frequency (CDF) and large early release frequency (LERF); however, the state-of-knowledge correlation (SOKC) does not appear to have been taken into account. Discuss how the SOKC was evaluated for fire CDF and LERF, identifying the fire-PRA-specific parameters (e.g., hot short probabilities, fire frequencies) that can appear in FPRA cutsets and how they were correlated.

RESPONSE

The Fire PRA Quantification has been updated to incorporate the state-of-knowledge correlation (SOKC) for the following fire-specific parameters in the parametric uncertainty analysis:

1. Ignition Frequencies

2. Non-Suppression Probabilities

3. Severity Factors

Enclosure I to L-2014-109 Page 48 of 106

4. Hot-Short-Probability-Related Altered Events The results of the final quantification are documented in the revised St. Lucie Fire PRA Summary Report.

PSL RAI PRA 01k 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. 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 an 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 using 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 result of a peer review are the facts and observations (F&Os) recorded by the peer review and the subsequent resolution of these F&Os.

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

k. F&O FSS-09-01 against SR FSS-09 This F&O indicates that the FPRA did not postulate failures due to smoke damage.

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

NUREG/CR-6850, Appendix T, Section T.2 identifies four modes (circuit bridging, contact fouling, binding of mechanical movement, and direct chemical/corrosive attack) of smoke damage.

Of these modes of smoke damage, NUREG/CR-6850, Appendix T, Sections T.2.1.2 and T.2.1.4 state that contact fouling and direct chemical/corrosive attack were found to have minimal or no risk significance as testing disproved the potential for such failure modes to cause long-term component failures during the times being considered for the postulated fire and the concurrent shutdown of the plant. As such, these modes of smoke-induced component failure were not considered in the PSL Fire PRA.

NUREG/CR-6850, Appendix T, Section T.2.1.3 states that the binding of mechanical movement mode of smoke damage has no impact on relays, breakers, switchgear, MCCs, and similar devices where the motive force is substantial. The section states that only components dependent on fine mechanical movement involving small driving forces (e.g., strip chart recorders, dial meters, or hard disk drive units) which are not encapsulated (which would prevent smoke penetrating to the moving parts and thereby mitigate the potential for damage) are susceptible to this mode of smoke damage. The section states that this mode of failure is found to have little or no risk significance.

This conclusion of little or no risk significance due to smoke damage causing binding of mechanical movement was further evaluated and verified in the PSL Fire PRA. As discussed in Section 6.2 of the Fire Scenario Report, the only scenarios where smoke damage to indications

Enclosure I to L-2014-109 Page 49 of 106 used in the Fire PRA could occur are for scenarios involving fire in the Main Control Room (MCR) which do not result in evacuation of the MCR. Strip chart recorders are not credited in the Fire PRA model. No credit is taken for dial meters surviving a postulated fire in the same cabinet in which the meter itself is located. As indicated in Section T.1.2.3 of Appendix T of NUREG/CR-6850, only meters which were mounted directly above the fire panel and were destroyed by heat failed, although the test did not have a "not reasonably sealed" mieter as part of the test to determine if they may be susceptible to smoke intrusion. Based on the limitations imposed in the Fire PRA (i.e., not crediting meters located in the same panel as the postulated fire, not using meters exposed to smoke except in non-abandonment scenarios in the MCR, not crediting strip chart recorders for indication) and on the results of the testing as stated in NUREG/CR-6850, Appendix T, Section T.1.2.3, the potential for mechanical binding because of smoke has been adequately considered.

As noted in NUREG/CR-6850, Appendix T, Section T.2.2, "Only one mode of component failure was found in this review to be of potential risk significance; namely, circuit bridging". The components of concern with respect to potential circuit bridging due to the presence of smoke are High Voltage Components and Lower-Voltage Instrumentation and Control Devices. These components are located within well-defined panel enclosures with limited ventilation. The concentration of smoke within these panel enclosures is expected to be significantly lower than that in the surrounding fire zone during a fire. Manual suppression activities, including smoke removal, will tend to disperse the smoke and reduce its concentration. The likelihood of smoke damage to components outside the threshold damage ZOI for non-IEEE-383 rated cables given the enclosure of susceptible components within panel enclosures and the early actions of a fire brigade to disperse and remove smoke is considered extremely small. Smoke damage is expected to be enveloped by the non-IEEE-383 rated cable damage threshold criteria applied for plume and radiant heating as well as the criteria applied for evaluation of hot gas layer effects for all fire scenarios.

PSL RAI PRA 011 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. 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 an 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 using 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 result of a peer review are the facts and observations (F&Os) recorded by the peer review and the subsequent resolution of these F&Os.

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

Enclosure Ito L-2014-109 Page 50 of 106

1. F&O HRA-A2-01 against SR HRA-A2 According to this F&O, the fire-related manual actions were not included as basic events in the FPRA model but rather "were incorporated into the model by altering the failure probability of a related equipment failure basic event" through use of the FRANC (a fire PSA model) Altered Events file. Section 4.1 of the Human Failure Evaluation report further states that "for logic in which multiple basic events (BEs) are correlated to a fire failure and in which these events are input to an "OR" gate (directly or indirectly), [one of] the events is set to the screening [HEP] value while the others are set to their random failure probability to eliminate the potential for non-minimal cutsets." Clarify this statement, and describe the process for applying screening HEPs, including any modeling manipulations made to the PRA logic (e.g., setting other basic events to their nominal random failure probability). Additionally, qualitatively discuss the impact of any logic manipulations on total fire risk and delta risk.

RESPONSE

To address this RAI as well as other HRA related RAIs, the use of screening HEPs incorporated via altered events for associated failure modes, has been eliminated. Detailed HRA has been performed for these fire specific manual actions following the guidance ofNUREG-1921. The logic changes that were made to address the impact of altered events are no longer required with specific HEPs incorporated into the HRA model in lieu of the altered events.

PSL RAI PRA 01m 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. 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 an 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 using 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 result of a peer review are the facts and observations (F&Os) recorded by the peer review and the subsequent resolution of these F&Os.

Clarify the following dispositions to fire F&Os and Supporting Requirement (SR) assessment 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 FSS-H1-01 against SR FSS-HI and F&O FSS-H8-01 against SR FSS-H8 As noted in the disposition to this F&O, the electrical panel factor approach was eliminated from the FPRA; however, in its place, a target-distance-based methodology that is documented in Appendix E of the Fire Scenario Report is applied. Additionally, as noted in the disposition to F&O FSS-H8-01, HGL analysis and MCA has been revised. A review of the HGL and MCA report indicates significant enhancements to the treatment of HGL since the peer review, particularly in the development of HGL scenarios and the fire

Enclosure I to L-2014-109 Page 51 of 106 modeling used to support the calculation of the time to HGL, and the removal of MCA screening criteria. In light of these modeling updates, address the following:

i. In Appendices A, E and J of the HGL and MCA report, non-suppression probabilities (NSPs) associated with the target-distance-based methodology split fractions and those associated with suppression of the fire prior to HGL formation appear to be multiplied together. Given that these NSPs do not appear to be independent, the resulting NSP values applied to scenarios associated with a particular ignition source would appear to be underestimated. In addition, given the manner in which NSP values are being applied, it is unclear if the resultant NSP value for each scenario is below 0.001, the NUREG/CR-6850-recommended lower bound. As a result, describe and justify how NSP values for scenarios associated with a particular ignition are calculated such that any dependencies between fire scenario development branch points are appropriately taken into account, and confirm that for any scenario, the resulting NSP applied is no less than 0.001.

ii. Section 4.2.1.1 of the HGL and MCA report states that if automatic detection is present, the time for suppression begins at time = 0, and if not, 15 minutes is assumed. However, there appears to be no discussion of detector reliability and unavailability. A review of Appendix E of the Fire Scenario Report and Appendix L of the HGL and MCA report appears to indicate that the target-distance-based methodology used to develop non-suppression probabilities does not account for detector reliability. Describe how detection system reliability is addressed by the FPRA. If detection system reliability is not included in the analysis, provide justification of this exclusion, and evaluate its impact on risk results.

iii. Section 4.2.2.1 of the HGL and MCA report states that the electrical fire suppression curve is used for all non-transient fires and is considered bounding; however, per guidance in Supplement 1 to NUREG/CR-6850, this is not always the case (e.g.,

HEAFs, oil fires, etc.). For instance, the more challenging nature of a fire that follows a HEAF should be represented by the HEAF suppression curve from NUREG/CR-6850, Appendix P or FAQ 08-0050 in lieu of the electrical fire suppression curve. Provide further justification for this assumption.

RESPONSE

i. The St. Lucie Fire PRA model implements two types of scenario manual suppression factors for an ignition source: time to target damage and time to hot gas layer (HGL). The time to target damage evaluates the direct heat flux incident on a target due to the fire. The time to HGL evaluates the volume temperature effects due to the fire.

The LAR-submitted model used the approach that these two analyses were independent of each other and therefore one was not conditioned on. the other. The updated approach, generated to support the RAI responses considers the two analyses as dependent. Since the time to target damage in most cases is less than the time to HGL, the time to HGL is conditioned on the time to target damage. For example, consider a time to target damage of 5 minutes and a time to HGL of 30 minutes, in the context of the event tree in Figure 1.

The first node, event MSI, represents the time to target damage. Using the manual non-suppression (MS) distribution from NUREG/CR-6850, Supplement 1, Chapter 14 with a lambda value of 0.102 (electrical fires), the MSI probability is 0.602. The second node,

Enclosure I to L-2014-109 Page 52 of 106 event MS2, represents the time to HGL and is conditioned on the first node. In order to condition MS2 on MS 1, the time credited for MS 1 is subtracted from the time available for MS2. In this example that would leave 25 minutes available for MS2, which, using NUREG/CR-6850, Supplement 1, Chapter 14, has an MS value of 0.079. Figure 1 shows the resulting fire scenario MS values for the three respective fire scenarios applying the node MSI and MS2 values. The HGL fire scenario gets a 0.0474 MS value, which corresponds to a 30-minute non-suppression probability.

__ __ __ USI 1J82 F I bln frewuecy TUmeto bget dwame Time to hot gas M r 0.4 Now-Sevem G 0.4 I SPSP 4 14tS2 1 0.52 Waevere 0.0 1S2 4.74E-02 -A 7.*6E-02 Figure 1. Event tree with no detection or suppression reliability modeled.

The minimum manual non-suppression probability used is 0.001.

ii. Reliability and unavailability of automatic detection systems were assumed in the LAR-submitted model to be incorporated in the manual non-suppression probabilities specified in NUREG/CR-6850, Appendix P, as revised in NUREG/CR-6850, Supplement 1.

Reliability of automatic suppression systems was based on values specified in NUREG/CR-6850, while availability was not considered to impact the reliability data given that plant procedures specify compensatory actions to be implemented when the systems are not available. In order to address the concern that the information inherent in the NUREG/CR-6850 data may not be bounding, the model has been updated to incorporate this additional failure potential. The scenario development event tree incorporates an additional node, before any suppression (manual or automatic) is credited.

The event tree detection failure path includes a 15-minute time delay before manual suppression is allowed to be credited (using SDP guidance for detection time for locations without detection systems). Figure 1 shows an event tree without consideration of detection failure. Figure 2 shows the Updated approach which incorporates detection failure. Note that the MS1/MS 15 and MS2/MS_15 values are bounded to a maximum value of 1. This results in zero ignition frequency being applied to the success branch for instances where the time to target damage or time to hot gas layer is less than 15 minutes.

NUREG/CR-6850 Appendix P suggests a bounding failure probability for smoke detection based on the Halon suppression failure probability. The data used to develop the Halon suppression failure probability included detection failure (smoke detection), so the detection failure probability by itself is bounded by the Halon failure probability.

Enclosure Ito L-2014-109 Page 53 of 106 NUREG/CR-6850 Appendix P does not specify guidance on thermal detection failure probability, therefore the use of the associated suppression system failure probability is applied to the corresponding detection system. The failure probability specified in NUREG/CR-6850 Appendix P for deluge or pre-action sprinkler systems is conservatively applied as the failure probability for thermal detectors associated with actuation of a pre-action system.

Ptob Fire Scoti I-MS1 0.4 No&

Seref 0.552 Sew 0.8 4.74E-02 Figure 1. Event tree with no detection or suppression reliability modeled.

.a --- i pr I Narne I I lMIION FRlUUEtT I F" D* I Mai I U I Deis am I I I FAjue of detection Firsttarget ppession. IASJ15 Hotgasby&T&&Wesak~.

I Re scenanto Owtioenfrequiency I

T '1 I is a 15 minuteOirmdeltay.lAr-I S1isis aISm~,etoe~

A -

14ASI A

l-Fý.DET I-MS2 sl~s 95thPercentilesevaamrn fire B M52 IGF Not PS layerfire C; 1-QAIJIAJS)Non-severe fire A FDEr 1-OAS2AAS.15)

MS1I)4Sj59Mt percentile sev'erefire B MS2JMSj5 C

Hotgastayerfire Figure 2. Event tree with detection reliability modeled.

iii. As discussed in the response to RAI FM 01 p, no credit for an NSP is used in the analysis of HEAF or oil fire HGL impact (all HEAF and oil fire scenarios will be assumed to result in a hot gas layer in the associated fire zone).

Enclosure I to L-2014-109 Page 54 of 106 PSL RAI PRA 01n 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. 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 an FPP consistent with NFPA-805. RG 1.200, "Anl Approach for Determining the Technical Adequacy of Probabilistic Risk Assessment Results for Risk-Informed Activities,"

describes a peer review process using 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 result of a peer review are the facts and observations (F&Os) recorded by the peer review and the subsequent resolution of these F&Os.

Clarify the following dispositions to fire F&Os and Supporting Requirement (SR) assessment 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 FSS-A1-02 against SR FSS-AI As indicated by this F&O, only manual suppression was credited in the FPRA at the time of the peer review; however, following the peer review, credit has been taken for automatic suppression systems. As a result, address the following:

i. Explain how automatic suppression is credited in the analysis.

ii. Provide additional justification that generic estimates reflect actual system unreliability and unavailability. Section 4.2.1.2 of the HGL and MCA report indicates that generic estimates of suppression system unreliability from NUREG/CR-6850 were utilized; however, it appears that no review of plant records was performed to determine if the generic unavailability credit is consistent with actual system unavailability.

Additionally, F&O FSS-D7-01 indicates that such a review for outlier behavior was not performed.

iii. Describe the dependencies that exist amongst credited suppression paths, and discuss how they were evaluated and modeled.

iv. Provide a technical justification for the methodology used to determine or calculate sprinkler activation time (in any fire area). If sprinkler activation time was not calculated, provide a technical justification for not having to determine this time quantitatively. Section 4.2.1.2 states that automatic suppression is credited to prevent the formation of a HGL; however, there appears to be no fire detection analysis conducted in support of the activation of fixed suppression systems. Section 4.2.1.1 of the HGL and MCA report also appears to indicate that automatic suppression is assumed to occur at time = 0.

Enclosure Ito L-2014-109 Page 55 of 106

RESPONSE

i. Automatic suppression is credited in the Hot Gas Layer (HGL) and Multi-Compartment Analysis (MCA). Only ionization detection actuated automatic suppression systems, where available, are credited to actuate prior to a HGL temperature of 80' C. For specific fire zones, such as the cable spreading room, Halon and preaction systems actuated by smoke or heat detection are credited for actuation prior to a HGL temperature of 131' C (Halon system for Unit I and Pre-Action water system for Unit 2). Area wide automatic suppression systems in other areas are credited prior to a HGL temperature of 205' C for the MCA.

ii. A review of plant unavailability/unreliability data, for the period 2011 through 2013, for fire protection systems confirmed that unavailability/unreliability values for the systems credited are bounded by the NUREG/CR-6850 values specified on p. P-6. This review confirmed that none of the systems credited have experienced outlier behavior relative to the unavailability/unreliability data credited from NUREG/CR-6850.

iii. The analysis has been revised to address the dependency between detection systems and the manual non-suppression system probabilities credited. Further discussion is provided in the response to RAI PRA 01 .m.

iv. Credit for automatic suppression systems is taken for those systems actuated by ionization smoke detection systems and for systems which will actuate prior to the credited hot gas layer impact timing. For the smoke detection actuated systems the actuation time is very short and is expected during the incipient stage of the fire (effectively setting the time of detection to zero relative to the heat release rate versus time assumed beyond the incipient stage of the fire which is the primary driver for the hot gas layer analysis in which in this credit is taken). For deluge/preaction systems, the actuation is based on heat detection systems and sprinkler head fusible link actuation. The credit taken is for temperatures 131 'C, and 205 *C. These temperatures have been confirmed to be well above the settings for the detection systems and the sprinkler head actuation temperatures (typically in the 200 TF to 220 TF range, 93 VC TO 104 'C, providing a minimum margin of over 25 'C). Given that credit is taken at temperatures well above the setpoints, the use of a time = 0 criteria is conservative.

PSL RAI PRA 01o 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. 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 an 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 using 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 result of a peer review are the facts and observations (F&Os) recorded by the peer review and the subsequent resolution of these F&Os.

Enclosure I to L-2014-109 Page 56 of 106 Clarify the following dispositions to fire F&Os and Supporting Requirement (SR) assessment identified in LAR Attachment V that have the potential to impact the FPRA results and do not appear to be fully resolved:

o. F&O HRA-83-01 against SR HRA-83 and F&O HRA-A2-01 against SR HRA-A2 HRA modeling performed to support the FPRA does not appear to have been sufficiently well documented to support peer review. For example,

• F&O HRA-B3-01 states that fire-specific HFEs are in the model "with limited documentation and no characterization," further noting that "the definition of these HFEs is not complete and provides no scenario specific information beyond the fire scenario ID in the [FRANC] Altered Events file."

" F&O HRA-A2-01 states that "[the] documentation is not sufficient to support FPRA peer review and future use."

Furthermore, the HRA screening analysis does not follow the guidance in NUREG/CR-6850 or NUREG-1921.

i. Evaluate the impact on risk results of using approaches in NUREG/CR-6850 and/or NUREG-1921, particularly where this would employ higher HEPs than in the HRA method.

ii. Summarize the impact of applying the criteria discussed in Section 4.3 of NUREG-1921 in assessing the feasibility of operator actions in lieu of making the assumption that all actions are feasible.

iii. Discuss the steps that will be taken to peer review this HRA method to support post-transition change evaluations.

RESPONSE

The Fire PRA HRA documentation has been updated to adjust the Full Power Internal Events (FPIE) Human Reliability Analysis (HRA) to address the impact of fire on the associated Human Failure Events (HFEs) in lieu of the multiplier approach originally used to address fire impacts on these HFEs. This update was performed in accordance with the guidelines ofNUREG-1921. For the fire specific HFEs, a detailed 14RA has been developed using the same methodology used to develop the FPIE HRA with appropriate adjustments to reflect the impact of fire on the associated Human Error Probabilities (HEPs). The fire impact on these HEPs was evaluated in accordance with the guidelines ofNUREG-1921.

i. The updated fire PRA quantification addressed in the submittal of the 120 day RAI responses includes the effect of the HRA revision as described above.

ii. The impact of fire on the credited HEPs was addressed by failing HEPs associated with required pathways for credit of the associated operator actions. For fire specific HEPs credited, a feasibility analysis was performed, using the same methods used to demonstrate feasibility for the deterministic analysis credited operator actions, to ensure the feasibility of these actions. The feasibility evaluation addressed the assessment factors specified in NUREG 1921, Section 4.3.

iii. Given that the HRA analysis revisions utilize the same BRA methodology used for the FPIE model, these changes are not considered to represent an Upgrade but a revision based

Enclosure I to L-2014-109 Page 57 of 106 on clearly defined adjustments of the HEPs to reflect the impact of fire, a peer review is not considered necessary.

The Fire PRA will be quantified incorporating the impact of all RAIs with the associated results (Attachment W) provided in Enclosure 2 to this RAI transmittal letter.

PSL RAI PRA 01p 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. 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 an 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 using 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 result of a peer review are the facts and observations (F&Os) recorded by the peer review and the subsequent resolution of these F&Os.

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

p. F&O FSS-H 1-01 against SR FSS-HI This F&O notes that in several cases, the FPRA implemented methods beyond those available in accepted guidance documents (e.g., NUREG/CR-6850 and its supplement).

Identify and describe all deviations from accepted methods and approaches. In addition, clarify whether guidance from the letter from NRC to NEI, "Recent FPRA Methods Review Panel Decisions and EPRI 1022993, 'Evaluation of Peak Heat Release Rates in Electrical Cabinets Fires"' dated June 21, 2012 (ADAMS Accession No. ML12171A583) was used in applying related methods. For any identified deviation from accepted methods and approaches, evaluate the impact on risk results as part of the aggregate change-in-risk analysis.

RESPONSE

Unreviewed Analysis Methods were eliminated (panel factors methodology was eliminated, wall and corner factors were applied where appropriate) or revised (revised use of 69 kW HRR for transient fires to limit its use to locations specific locations) by application of the guidance provided in the June 21, 2012 Joseph Giitter to Biff Bradley memo. The items addressed in this memo and their disposition with respect to the PSL Fire PRA is addressed below:

1. Frequencies for Cable Fires Initiated by Welding and Cutting - not used

2. Clarification for Transient Fires - methodology applied is consistent with the approach accepted by the methods review panel and the NRC. See RAI PRA 4 for further details.

3. Alignment Factor for Pump Oil Fires - not used

Enclosure I to L-2014-109 Page 58 of 106

4. Electrical Cabinet Fire Treatment Refinement Details - eliminated from Fire PRA supporting PSL LAR submittal as stated in PSL NFPA 805 LAR Section V.2 and in RAI PRA 01.m.

5. EPRI 1022993 - "Evaluation of Peak Heat Release Rates (HRRs) in Electrical Cabinet Fires" -

not used No other methods used in support of the PSL Fire PRA are considered deviations from accepted methods and approaches.

PSL RAI PRA 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 FPRA and endorses, with exceptions and clarifications, NEI 04-02, Revision 2, as providing methods acceptable to the staff for adopting an FPP 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.

Discuss how sensitive electronics are identified and treated. Include in this discussion a description of the damage criteria that are used for sensitive electronics inside and outside (i.e., exposed) of cabinets, and provide justification for these thresholds. Additionally, indicate whether this treatment is consistent with guidance in NUREG/CR-6850. Note that draft FAQ 13-0004 discusses the treatment of sensitive electronics only within an enclosure.

RESPONSE

Sensitive electronics (i.e., computers, digital converters, digital amplifiers, digital communications equipment, electrical devices that contain a semiconductor or an integrated circuit board as a key element) that were part of the internal events model and were required for evaluation in the fire PRA and that could be damaged by heat from a fire were assessed. This assessment was done using an approach consistent to that described in Fire PRA FAQ 13-0004. In other words, the shielding of components from direct radiant exposure by the robust enclosure of the panel within which the heat sensitive components are located can be conservatively substituted by modeling the component as being a non IEEE-383 thermoplastic cable - which Would not be as sensitive to heat but would also be modeled as receiving direct radiant exposure from the postulated fire. The use of a non IEEE-383 thermoplastic cable criteria for defining a ZOI for damage of cables or panels provides a more conservative criteria than the IEEE-383 thermoset cable criteria specified for use in Fire PRA FAQ 13-0004.

Walkdowns have been performed to identify components required to support the fire PRA which are located outside of panel enclosures. No components required to support the fire PRA were identified which are mounted outside of a panel enclosure. Therefore, the conservative use of non-IEEE thermoplastic cable damage criteria as a basis for a fire ZOI provides a bounding analysis consistent with that specified in Fire PRA FAQ 13-0004.

Enclosure I to L-2014-109 Page 59 of 106 PSL RAI PRA 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 FPRA and endorses, with exceptions and clarifications, NEI 04-02, Revision 2, as providing methods acceptable to the staff for adopting an FPP 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.

In regard to the treatment of MCR quantification, address the following:

a. According to Supplement I to NUREG/CR-6850, back panels housing items such as balance-of-plant and off-site power controls and indicators should be excluded from the main control board (MCB) and treated as general electrical panels. In light of the audit walkdown observation that the MCB forms the front face of a sub-enclosure that is shared with a row of back panels separated by a distance of greater than about 3 feet, provide justification that the back panels are part of the MCB in terms of frequency apportionment.

Additionally, evaluate the impact on risk results of treating these back panels as general electrical panels in terms of frequency apportionment and fire scenario development, including propagation to the MCB panels.

b. Clarify whether NSP values lower than 0.001, the NUREG/CR-6850-recommended lower bound, were utilized in the MCR abandonment analysis.

RESPONSE

a. The walk-through vertical main control board used in the St. Lucie plant control room includes key functions used for plant operations. The PSL control room consists of a series of six panels making up this main control board. No other control panels exist in front of these boards other than operator consoles which are work areas and not plant controls.

Therefore, these six panels, which are designed as walk-through panels, comprise the main control board. Separating the front and rear sections of these panels is not appropriate given that cables associated with functions on the panels are routed in the front and rear sections with cable routed over the walk-through section to terminate at the front of the panel. Therefore, these panels constitute the main control board as defined by NUREG/CR-6850, Supplement 1, Chapter 5.

The criteria specified in NUREG/CR-6850, Supplement 1 for defining the panel(s) in the control room to be designated as the main control board (NUREG/CR-6850, Task 6, Bin 4) are provided below. The bases for the walk-through panels at PSL meeting this criteria are provided in brackets after each item:

(1) serving as an integral part of the main plant monitoring and control functions; [these panels are the main plant monitoring panels and they house the main control functions for the plant, no other panels at St. Lucie serve this function]

Enclosure I to L-2014-109 Page 60 of 106 (2) located in the center of the operators' main work area; and [no other panels in the main operator work area house the primary plant monitoring and control functions]

(3) manned on a nearly continuous basis. [manning/monitoring functions are focused on these panels]

Therefore, these panels fulfill the main control board function for the St. Lucie plants as defined in NUREG/CR-6850, including Supplement 1.

b. The use of an NSP less than 0.001 is only applied for control room abandonment scenario heat release rate bins where both the temperature and visibility conditions peak prior to 25 minutes. For these heat release rate bins abandonment conditions are not reached and the associated NSP will be zero indicating that these bins will not contribute to control room abandonment.

PSL RAI PRA 16 As documented in LAR Attachment U, the peer review of the IEPRA was performed using the criteria in NEI 00-02, and as a result, a self-assessment is required according to the guidance within RG 1.200 (Revision 2) to demonstrate that the technical adequacy of the PRA is of sufficient quality to support the application. Additional self-assessment subsequent to the July 2002 NEI 00-02 peer review have been performed; however, they do not appear to be sufficient demonstrate the technical adequacy of the PRA in accordance with the guidance in RG 1.200 (Revision. 2). In particular,

• Focused-scope peer reviews have only been performed on select technical elements of the ASME/ANS PRA standard (i.e., large early release (LERF) evaluation (LE), human reliability analysis (HRA), internal flooding (IF), and data analysis (DA) as well as supporting requirement (SR) related to common cause failure). Also, the July 2009 LERF focused-scope peer review was not performed against the latest version of the standard as endorsed by RG 1.200, Revision 2.

" A December 2005 independent review of the PRA was performed using a dated version of the standard (i.e., ASME RA-Sa-2003 "to guide future PRA enhancement activities"). This "basic assessment" does not appear to qualify as a self-assessment. No formal process or review criteria were cited as being used to judge the adequacy of PRA against SRs. Additionally, the exact scope of SRs and PRA documentation reviewed is unclear and is acknowledged by the review team as being limited.

" An October 2007 assessment of the St. Lucie PRA was performed to addresses deficiencies raised by the December 2005 independent review and F&Os from the July 2002 NEI 00-02 peer review; however, this assessment does not appear to follow the self-assessment process outlined in Appendix B of RG 1.200. Additionally, the results are not documented in such a manner that it is clear why each requirement is considered to have been met.

Although self-identified action items to comply with RG 1.200, Rev. 1 are briefly summarized, they are not tied to specific SRs.

Explain how the 2002 peer review and these subsequent focused-scope peer reviews and assessments are consistent with or equivalent to the peer review and self-assessment process in

Enclosure I to L-2014-109 Page 61 of 106 NEI 00-02, Revision 1, as endorsed by RG 1.200, Rev. 2, with clarifications and qualifications, for demonstrating the technical adequacy of the IEPRA for the NFPA 805 application.

In addition, provide the deficiencies (or "gap") identified by the latest gap assessment in a manner analogous to the format in which F&Os are presented and dispositioned in LAR Attachment U, including an assessment of each identified gap's impact on the FPRA.

RESPONSE

A response to this RAI was given during the 90-day response period by letter L-2014-083 dated March 25, 2014. The purpose of this 120-day response is to address the last paragraph of the question above. Table (1) shows the list of "gaps" identified in the St. Lucie Gap Assessment, as described in the 90-day response. These gaps are associated with F&Os that resulted from the ISLOCA Focused Peer Review conducted in December 2013. An assessment of each gap and its impact on the Fire PRA is identified below.

As part of resolving these findings and assessing their impacts, a sensitivity analysis was developed whereby the ISLOCA Internal Events fault tree model logic was completely revised and updated. F&Os IE-C5-01, IE-C9-01, and SY-A2-01 were all addressed in this update. F&O IE-C6-01 is a documentation issue and does not result in a change in modeling. The cumulative impact of the F&Os resulted in a decrease in Internal Events CDF and LERF for St. Lucie Unit 1 and Unit 2. The following are baseline Internal Events sensitivity results:

PSL1 CDF (in LAR document) = 5.34E-06 /yr.

PSL1 CDF (w/ ISLOCA update) = 4.92E-06 /yr.

-8% decrease PSLI LERF (in LAR document) = 7.79E-07 /yr.

PSLI LERF (w/ ISLOCA update) = 3.57E-07 /yr.

-54% decrease PSL2 CDF (in LAR document) = 6.77E-06 /yr.

PSL2 CDF (w/ ISLOCA update) = 6.74E-06 /yr.

<1% decrease PSL2 LERF (in LAR document) = 2.32E-07 /yr.

PSL2 LERF (w/ ISLOCA update) = 2.03E-07 /yr.

-12% decrease Based on the nature of the ISLOCA update which was limited to logic and fault tree structure changes and elimination of conservatism, as well as the resulting reduction in internal events CDF and LERF for the PSL Internal Events model, the changes, and associated F&O resolution, will not adversely impact the fire PRA model and the resulting quantification.

Enclosure I to L-2014-109 Page 62 of 106 Table (1): Listing of ISOLCA Focused Peer Review Findings and Observations (F&Os).

Other Impact on Fire F&O # Element Affected Issue PRA SRs IE-C5 The approach for generating the initiating event frequency CDF/LERF IE-C5-01 utilizes an adjustment factor that ratios the exposure time. Decrease. No The exposure time is not the mission time for reliability and Impact.

the approach does not produce an appropriate value for frequency of occurrence.

Utilize the idea of initiating event as the first valve failure based on having to remain isolated for the period of one year. Then consider the unavailability of the other valves in the line based on exposure time. Systematically address each valve as if it is the holding valve.

IE-C6 A screening approach is utilized for some lines based on Not a modeling E-C601 low frequency but this is not quantified. The SR indicates a issue.

frequency expectation for screening. Documentation Define the estimate for the lines screened on low frequency Issue only. No and show that the calculated frequency supports screening. Impact.

IE-C9 IE-C10, The fault tree model used for the ISLOCA paths assumes CDF/LERF IE-C9-01 SC-A5 that the status of all valves is known when the plant is Decrease. No brought online and the corresponding exposure time is the Impact.

refueling interval. However, based on discussions with knowledgeable staff, there is no positive means to know that more than one isolation valve is actually holding. Use of status lights is not definitive since there is a +/-5%

margin between light changing and valve seating. The exposure time should be based on a positive flow test which may not occur on a refueling basis but based on other studies could be as much as the life of the plant.

Additional input from the PRA staff identified the presence of a relief valve which could offset some of this uncertainty, but would require operation of the relief valve to annunciate the failure which is not currently modeled.

SY-A2 QU-D2 The current ISLOCA model includes a failure of the SDC CDF/LERF SY-A2-01 isolation valve to fail to close as one manner by which a Decrease. No loss of isolation may occur combined under an "OR" gate Impact.

with a pre-initiator error involving the operator failing to correctly close the valve. Ifthe valve mechanically failed during startup the operators would not enter into power operation so the failure mode is not valid. It could be postulated that if it failed the operators could fail to take appropriate actions which would be a pre-initiator action, but this would require the two events to be "ANDed" which would substantially decrease the likelihood of occurrence.

Closing the valve at power is not plausible due to the high RCS pressure so the closure would not be valid with regard to isolation.

Enclosure I to L-2014-109 Page 63 of 106 Att. J Markup

Enclosure I to L-2014-109 Page 64 of 106 Florida Power & Light Attachment J - Fire Modeling V&V This attachment documentsthe Verification and Validation (V&V) basis for the St. Lucie Nuclear Power Plant (PSL) Fire Probabilistic Risk Assessment (FPRA) fire modeling applications. Plant specific ire modeling usedto support the PSL FPRA consists of the following:

" The calculation of the Main Control Room (MCR) operator abandonment times in both the Unit 1 Unit 2 control rooms (Refs. 112;J.

" The use of generic fire modeling treatments and its associated supplements as applicable to develop Zones of Influence (ZOls) (Refs. J43. 4.15));

" A-Ddetailed assessments of plant specific ire scenarios involving secondary cable tray combustibles (Ref. J26i. 17 8); and

" A detailed analysis of the cable tray separation within the Unit 1 Containment Building in support of the K1 Exemption (Ref. J39).

Main Control Room Abandonment Calculations A separate MCR calculation is provided forthe PSL Unit 1 and Unit 2 control rooms (Ref.-j). The report titled'Evaluation of Unit 1 Control Room Abandonment Times at the St. Lucie Plant" (Ref. J41) addresses the Unit 1 MCR abandonment times and the report titled "Evaluation of Unit 2 Control Room Abandonment'Times at the St. Lucie Plant" (Ref. J52) addresses the Unit 2 MCR abandonment times.

The goal of the MCR abandonment reports is to compute the time operators would abandon the PSL Unit 1 or Unit 2 MCR given a fire in either the Unit 1 or Unit 2 MCR, respectively. The abandonment times are assessed for various electronic equipment fires and for ordinary combustible fires as defined by the discretized heat release rate conditional probability distributions presented in Nuclear Regulatory Guide I (NUREG)/Contractor Report (CR)-6850 (Ref. J1_1O). The abandonment time in the main control room is estimated by calculating the time to reach threshold values for temperature and visibility as identified by NUREGICR-6850 (Ref. J610). Transient fire scenarios that may be more challenging than NUREGICR-6850 (Ref. J61__) Appendix E Case 8 are assessed using a workstation fire scenario. This fire scenario is characterized using the data provided by National Institute of Standards and I Technology (NIST) Interim Report (IR)4833 (Ref. J-711) and the Societyof Fire ProtectionEngineers(SFPE)Handbook of Fire ProtecionEngineering,Section 3-1 (Ref. J812).

The focus of the MCR abandonment evaluation is on the first twenty-five minutes after ignition because the non-suppression probability (NSP) decreases to 0.001 attwenty minutes (Refs. J610, 19L3). The abandonment calculations are performed using the zone fire model Consolidated Fire and Smoke Transport (CFAST). Version 6.0.10 as described in NIST Special Publication (SP) 1026 (Ref. J104) and NIST SP 1041 (Ref.

J 1-45).

The Unit 1 and Unit 2 control rooms are geometrically similar but have different internal configurations, boundary materials, and ventilation flow rates, and cable types. The control rooms are completely separated from one another so that two independent MCR abandonment calculations were developed, one for each unit.

Revision O Pag~eJ-2

Enclosure I to L-2014-109 Page 65 of 106 PAGE J-3 INTENTIONALLY LEFT BLANK.

Enclosure I to L-2014-109 Page 66 of 106 Florida Power & Light Attachmentt J - Fire Modeling V&V Unit 1 MCR Abandonment Calculation The Unit 1 MCR abandonment calculation is provided in the report titled t Evaluation of Unit 1 Control Room Abandonmentlimes at the St. Lucie Plant' (Ref. J4_). The Unit 1 MCR area is evaluated using &4ýt room geometry (Computer Room, the MCR Ao* and the Computer Room - Staff SupportArea). The geometry and fire parameters forthe simulations fall within the model limits listed in NIST SP 1026 (Ref. J4014) and NIST SP 1041 (Ref. J 145). Specifically, the vent area to enclosure volume ratio is less than two and the aspect ratios of the enclosures that define the control room sub-areas in the CFAST model are less than five. (forthe true geometry). The only

,excption to this is the Main Control Board (MC B)sub-space@, which has. an aspect rati ofabou t 6.5.

su-b-area is effectivela volume obstuction with leakage bound*ay cond9itins. NIST SP 1041 (Ref. Ji !) recommends using the CFAST codor flow sub-modeal w."hen the aspec.t rtio exceeds, f.e; however, sinc. the timing ofthe ga layer flows within the ACD itself are not important to-the c-ond-itioans that develop insides the control room prope9r, t he largeaa specGt rat iofor this su b-e_ nclosurea is conPsi dere dacce ptablea fcr th isr applicap.n The physical input dimensions forthe MCR sub-endosures are adjusted to account for obstructions and boundary heat losses and the resulting model geometry has a length to width aspect ratio gat.-Iess than five t,-in allsgml spaces. 4wevTi-the input geometry conserves the bcundaww-arnendosure aspect ratio, the enclosure volume, and the enclosure height Therefgor, a corridor flw mo.de li ;i to*ntenti0nAly av-oid.ed because the true geomety has an aspect ratio that is within the modal lumotafon&r The verification for the CFAST model (Version 6.0.5) is provided in NUREG-1824, Volume 5 (Ref. J 12D_). Supplemental verification for CFAST, Version 6.0.10 is provided as an appendix to the Unit 1 MCR abandonment calculation (Ref. J41) as well as in NIST SP 1086 (Ref. J 13Z).

The non-dimensional parameters that affect the model results, as documented in NUREG-1824, Volumes 1 and 5 (Refs. J126, J148) and NUREG-1934 (Ref. J159),

include the model geometry, the global equivalence ratio, the fire Froude Number, and the flame length ratio. Non-dimensional parameters that relate to target exposure conditions (heat flux) and sprinkler actuation (ceiling jet) are not applicable tothis calculation because these output parameters are not used. Section A4.5.1 of Ref J 1 provides a detailed assessment of the non-dimensional parameters described in NUREG-1824. Volumes 1 and5 (Ref. J16, Ji18 and NUREG-1934 (Ref. J19) as they apply to the Unit 1 MCR analysis. Overall. the application of CFAST. Version 6.1.1 in the PSL Unit 1 control room falls entirely within the NUREG-1824. Volume 1 (Ref. J 18)

V&V parameter space for applicable parameters for all times upto the predicted abandonmenttime orwould generate a conservative result relative to a case that fell within the NUREG-1824. Volume 1 (Ref J1i)V&V parameter space. The model results after abandonment is predicted may be based on an application outside the NUREG-1824, Volume 1 (Ref. J 1i) V&V parameter space, but the results are not used in the FPRA.

Revision 0 Page J-4

Enclosure I to L-2014-109 Page 67 of 106 Florida Power & Light Attachment J - Fire Modeling V&V The nondimansional geometry parameters (length to hemiht and width to height) rapng from about 1.5 - 65 for the twoe geometry for each sub-area. Except forthe MCB sub area, the geome@' parameters- h! within the NUREG 1824, Volume 1 (Ref. i ll) validatio range (0.6 5_7) As deribedabove, the MOB sub areais effecely an obsthnru-ton with leakage boundary codtopns. The intarmal cognrditions ofthae MOBIw;ilI not affeathe aba*donmenlt tm inthe control room proper,therefore, the large geometry patio forthis sube*c-losureis consideed aceptable forthis appliat;on.

CEAST, Varsion 6.0.10 does not use a fire diameter (Reft. 110, ill); thus, the detArmiPnaftion fthe appropriate fire Frouwd Numb*r is basedhon the application raher than the fire model inputs. The fire scenarios considered in"the abandonment Calcultion inde electrical panals and transiant ignition sourcesr that are typical of clUear pQwIer plants and comparable to the types of fir scenrios envsoed in the NUREG 1824,Volume . (Ref. il4)IV&V effor The application ofthe .fire modelin resultsa towrard, ignition sour*lsrthat fallkwithin tha N1 GR 6850 (R f 1ar6)

EIO conona-;ýnl probability distibution fortransient an-d eletrical Panel in*ition sourer, and are thlu consideared to typical of thos Iused in NUREG 1821, Volumes 1 and 5 (eft.

12, i 4) used to v2lidate the CFAST fire model. The exceptions to this are the wo,. rkstation fueiol package fire and the cable t*ray fire. The workstationfi-,- Airae in.volves a.

ralatvely large fire over a desk footprn;t The fire Froude Nmber as computed using the mathods desrihbed in INREG1 931 (Retf A15) is abo*ut 1.32 assuming a 1.2 x 0 76 mr1 x 2.5 ft) desk plan, which iswithin the Nil.REG 182, Volume 1(Ref i4) aliation range of 0.1 2.- T4 fir Frod N*u*mberr frthe cable tray fire scenario is not readily computed using the methods descr3ibe* in NUREG 1931 (Ref ( 15t due to the geometry and fire aspect ratio (i.e., the cable tray fire is a line type fire). Beaurasea line type fire has more entrainment per unit length than an equivalent area sour. e firm the use@ Of 2An axiymmeric plume model for a type fire is expected to be ne conseroaative wihen computing the temperature and smoke density. The fire Frou de Number computed on a per unit length basis provides a rough indic;ion ofthe fire strength as comparedtothe mres assessed in NUREG :noums 1 and 5 (Refs 112, ill). UndeArth . tWin a..stack of two 0.6 m (2 ft\wide cabletrayswould

.coni havea* fmi*FrdN ,,Numbar of 1.22 pna unit length basis, which fallsw,^ithinthe NUREG 1821, Voum ~ 1Re.4ll 4W validation ranige of 0.1 2.1.

The global equiv*aenca rafti app-rcable to the aentire PSI UI It1 *MR oImain* (MR Area, Computer Room Paltion, and StaffSuppor. Area) for normal Heating, Ventilation, a'nd AirCon4ditioning Ill(HVAC) copdiions, may be assessed using the ratmoof t.hea maximu.m suppcted fir size to the fire size postulated. Based on the fresh air supply flg. ofO .35v mt's(750 c p(Refjn16), the omaximm fire size that could be supportedas about 1.27 MW (1,210 Btus) *(Rof 4**. The maxim.um fire size postulated is about 1.70 !'AW (1,150 Btu)* for the o electrial panel fire that propagates, to ern adjacent panels. Thus, the maximum global equivalepce ratio asexpected to be abouit 3.7and therefoe xcee*ds the NI IUREG1-2 Volume 1 (Ref il4) validotion range of 0.01 0*.6 Tha max;mum average heat release rate, whirh baeler reflects tha oxygna consumpton that wmou ld be expected over the twenity-fiv minute interwa, isabouta 21 MX"! (1,990 Mtu*s). This means the maximum global equivalence ratio is, expetoe to be onnhthAorder which stilllexced the !NUREG-1824,Volumo

• of1.65, l(Refill) valid;ation range of 0.01 0 O.6. Nhen the initial oxygen reservoir in the gross; PSI IUnit 1 Revision 0 Page J-5

Enclosure I to L-2014-109 Page 68 of 106 Florida Power & Light Attachment J - Fire Modeling V&V MACR volume of 2,371 m3' (83,670 q as tabulated inpRot. 14 forthe CFAST mpodeis, conAsidered in comrbina2tion w.ith the-freSh air supply flow, it can -be-shmown that this oxygen reservoir is capable of supporting a 1. 18 MW (3,960 Buls) source fire for twenty -five minutes at a global equivalence ratio of 0.*6 This means thatthe maxmum global equivalenice ratio- decreases to-abhout 0.67 (for maximumn fire size) or 0.30 (for max.imum average fire size) and ths ,falls .w hin the range considerab*d b I 'yREG 1821, Volume 1 (Ref. il4) There is no designaed smoke purge. mode per Degnký Basis Document (DBD}-HVAC*1, Re,Asion 2 (Rot J16), Dmrwing 8770-G-862, Revision 34 (Ref- i), apnd AR-1-18003023, Revsioin 29A (Ref. J19} Therefore, the maxmum eqialence ratio for any fire scenari cnsdrdi on the orde "r of 0-20 inthe UIt

" 1 MCR,.when there is forcedventiation.

In theca-se of no foce ventilati, the maximum. global equivalents ratio may also be determined using the inital mass of ygen available while conservatively ignoing the boundary leakage flows. The bounding casewith respect tothe NUI-RE=G482H', Volume 1 (Ref. Ji 4) validation spats is when the bounda.' doors remain closed. The initial oxygen, reservoir ran support a 3.11 MW (3,20 Btul,,) fire fortwet, five minutes at an; equialence ratio of 0.6. This mean. that the maximum global equv.alents ratio w.ould be about 0.83 when considering the peak fire size only or 0.37 when considering the largest average- fire s"e over the twenty fve mpinute inpterval. This e-quivalents ratio 0onservatively ignores the bactthatthe large firs res, it in abando*nment tirmes on; the order nf sixto seven minutes, .hich would decrease the maximum equivalence ratio by a factor of at least three to about 0.12. It is thus conduded that even when the doom am closed, the m~aximump equivalence ratio remains woithin orbelow the NOUREG-1991, Volume 1 (Ref.il4) validation range.

Ma*n fire sc*naos will act-ally have an equivalents ratio thatis below the minimu.m va'lue of 0.01 assessd in NUREG 1824, Volumes ! and 5 (Refs. J12, Jl4).This means that thera is a suffident supply of xygeAn awailable for the fire up to the time at which abandonmen is prdicted. Fu all aos begin with an equivalence ratio of 0.0, including those that form the NUREG 1821, V*ou 1 (Retf ill) validatvon basis; thus, the scenario evaluated in the PSL Unit I MCR are not inconsiatent w.mith the validation

.cenarios conside;4red in UREG1*821, Volume 1 (Ref. il.4) simply because the rati is ow. iven that this is-the most aadverse electrial panel ire scenario equivlenc postulated, the global equivalence ratio at the prediced abandonment time is experted to be comparable- or lower f~rthe less severe eletrical panel fire scearios0 anpdthe transient fi scnro. Conm.sequenly. even when the HVAC is inopemrtve and the bounda' doorm are closed, the maximum global equivalents ratiow.ithin the MCR do'nma.in is expeded to remain within th *N-.UREG 1821,Volume 1 (Ref. J1I)v.alida-tion range Up upntil the time atwhich abandonment is predicted.

rinally, the flame length ratio is normally met but inthe case of the largest fire sbes.

postulated, the flame height may reach or *ex*cd the heffight. Because sprinkler actuation and thermal radiation to targets are not computed with the CFAST model, this parameter is not an applicablea metrc. Rather, the plume entrainment hblowIm the hot gar, layer c layer descept ime anr, the concenrtion ofsoot pordu*s* in the layer hntrolethe This naset oftha mrodel is nOtaffeted byhthe flame heightto eilinmg height ratio.

Revision 0 Page J-6

Enclosure I to L-2014-109 Page 69 of 106 Florida Power & Light Attachment J - Fire Modeling V&V Consequently, the applcton of CFAST to model fims inthe PSL Unit control room fal*s entiralywithin the N'UREG-1824, Volume 1 (Ref. A4l) validation spa. .

Additional V&V studies are contained in NIST SP 1086 (Ref. J13Z) and Naval Research Laboratory (NRL)Memorandum Report (MR)16180--04-8746 (Ref. J202_). These studies have a broader parameter validation space than NUREG-1824, Volume 1 (Ref.

J 148_). NIST SP 1086 (Ref. J 13Z) is based in part on the methods of American Society forTesting and Materials (ASTM) E1355 (Ref. J241J). NRL/MR/6180-04-8746 (Ref.

J2--2_) provides a Navy specificV&V study, which includes an assessment of CFAST, Version 3.1.7 predictions in multiple enclosures and multiple elevation configurations.

These additional studies extend the range of the validation space to include configurations and conditions applicable to the MCR abandonment sensitivity analysis (Appedixttachment 2B of Ref. J41$.j)

Table J-1 provides a summary of the validation and verification basis for CFAST, Version 6.0.10 as applied inthe Unit 1 main control room abandonment report.

Unit 2 MCR Abandonment Calculation The Unit 2 MCR abandonment calculation is provided in the report titled "Evaluation of Unit 2 Control Room AbandonmentTimes at the St. Lucie Plant" (Ref. J52). The Unit 2 MCR area is evaluated using aj*t mroom geometry (the MCR Area-andthe-Staff Support Area,-.d=MCBs). The geometry and fire parameters forthe simulations fall within the model limits listed in NIST SP 1026 (Ref. J 104) and NIST SP 1041 (Ref.

J 14-). Specifically, the vent area to enclosure volume ratio is less than two and the aspect ratios of the enclosures that define the control room sub-areas in the CFAST model are lessthan five. (forthe true geometr,. The only exc.. on to thisis the MCD sub space, whi c has a .a ratioof abou....it 6..5. oweve r, t.he condition wihn t.hi.

space are nAt se iclly u to aess the abandonment ratherthe flro^ws bet.Afeen this ..... p. r and. the......MCR area are phenomena I.. J...:. X..L. . ftAll * ... of f fD interest

.....L.. Further,

. W.. . . ..in cases

. . whera

... . the obtPct.-in.,AWith lea-ka bou-ndar" mnditions NIST SP 1041 (Reaf. Jill) rcommends us.ig the CFAST co-dor flo,0 sub- model when the as ped rat io exceeds five; how.. ver; sincea the timingof'the gas layer flo-wAs*within the MCD itself iap not importan Rto the conditons that develop insidethe conptprol Mom p..per, the large spect-ratio forthis enclsure

-sub ' is consiered accptabe fopths .. appli*caion.

The physical input dimensions forthe MCR sub-enclosures are adjusted to account for obstructions and boundary heat losses and the resulting model geometry has a length to width aspect ratio greate-that is less than five in all fop some spaces. HowmveiThe-the input geometry conserves the boundaw-area enclosure aspect ratio the room enclosure volume, and the enclosure height. Therefore, a c.oridor fl-ow. model is intentionally avoided b the true geometry w

has an aspec ratio that i-s wih"i the model limitMtions.

The verification forthe CFAST model (Version 6.0.5) is provided in NUREG-1824, Volume 5 (Ref. J12_). Supplemental verification tor CFAST, Version 6.0.10 is provided as an appendix totheUnit 2 MCR abandonment calculation (Ref. J52) as well as in NIST SP 1086 (Ref. J 13Z).

Revision P Page J-7

Enclosure I to L-2014-109 Page 70 of 106 Florida Power & Light Attachment J - Fire Modeling V&V The non-dimensional parameters that affect the model results, as documented in NUREG-1824, Volumes 1 and 5 (Refs. J126, J148) and NUREG-1934 (Reft J152),

include the model geometry, the global equivalence ratio, the fire Froude Number, and the flame length ratio. Non-dimensional parameters that relate to target exposure conditions (heat flux) and sprinkler actuation (ceiling jet) are not applicable tothis calculation because these output parameters are not used. Section A4.5.1 of Ref- J2 provides a detailed assessment of the non-dimensional parameters described in NUREG-1824. Volumes 1 and 5 (Ref. J 16, J 1) and NUREG-1 934 (Ref. J19) as they apply to the Unit 1 MCR analysis. Overall, the application of CFAST. Version 6.1.1 in the PSL Unit 1 control room falls entirely within the NUREG-1 824, Volume 1 (Ref. Ji18 V&V parameter space for applicable parameters for all times up to the predicted abandonment time orwould generate a conservative result relative to a case that fell within the NUREG-1 824, Volume 1 (Ref- J18) V&V parameter space. The model results after abandonment is predicted may be based on an application outside the NUREG-1824, Volume 1 (Ref. J 18) V&V parameter space, but the results are not used in the FPRA.

The no iesoalgemetry parameters (length to heigh~t and width to height) range from 5 forthe true geometry for each sub- .ra.. AExcep f,,;orthe, MCDA sub*,

area, the geometry parameters fall within the, NUREG 1824. Volume 1 (Ref. Ji,)

va1IUllIu[li Lrf19U \U.U86WA t UtIU IU 209M , Lilt PA%.LJ IILLJuj 18U 0U+4I obstruction with leakage boundary cnditions. The intemal conditions of the MCD wii-ll not-2 affhe abandonmenttimes in the control room pwrper, therefore, the large geometry ratio for this s.b-_eclosurae is considered acceptable forthis applic.tio*n CFAST, Version 6.0.10 does no use 2 fire diameter (Refs- Jig, ill); thus, the detewr*mintion ofthe approproiate fire Fu*da Numberis bas -Rodon the application rather thnhefie odel inputs. The fiescnros osidered inthe abandonment c-alculation in-dude elerical pa'nelss a-nd transient ignition sourceas that are typ*a*l f nuclear power plants and comparable to the types of fire sc.nario*s envisioned inthe NUREG 1821,Volume 1 (Ref. i- V14) V effort The applicat*io'nothe fire mo..ling resulits arottoward ignitio sources that ISOliti th NREG!CR-6850 (Ref JIM conurnOnASI pPWMAIOYISIayuMIDuuuur It;r LIMTIUGfAIU 2 uPua 0861GI. prauil I9RIII tu ~

RAu QU[IEG are thu-s considered to typical of those u'sed in NUREG 1824, Volu'mes 1 and 5 (Refs1 J 12, J!4) used to,- vidate the CFAST fire model. The exception to this is the wo.0rkstation &-el package fire. The wA-orkstationA fire involves; A.relatively large fire9 over a.

deskf4 fotpin The9 fire F roude Number as com~puted using the methods. described in NUREG 1931 (Ref. J5u) is about 1.32 assuming a 1.2 x 0.76 m (, x2-5 ft) desk plan, which is within the NUREG 1824. Vol-ume 1 (Ref. 114) validation rapne of 0.4 2 .1 The global equivalence ratio applicable to the entire PSL Unit 2 MCR domain (MCR

.Area an-d Staff Suppoll Area) fornormal HVAC codtinpay be assessed using the ratio of.the maximum supportedfire sizo.to.the fire size postulated.-. B..adon.the frsh air supply fiNofn.7m *0.7m! (1 J22), the maxmum fire sizoethat could be

),000r,(Ref supported is about 1.69 MW (1,600 Btkuls) (Ref J17). The ma-ximum fire size postulated isabou-t 3. MW (2,910 Bltuls) for the multiple bahund-le e-lectrical panel fire that p to adjacent panels..Thus, the axi.mum global equivalene rtio is expected

.ro.agates to beA .,bout1.83, which ex. ,ees the INIUREG-1 821, Volum.e 1 (Ref i1l4) validation Revision 0 PageJ-8

Enclosure I to L-2014-109 Page 71 of 106 Florida Power & Light Atrhrnment J - Fire Modeling V&V range of 0-04 0.-. The maximum average heat release rate. which boter rmfleds th oxygen...P..csu. On that would be ex pareddover the twenW-five mnute inteOrval, is about 1.3 MW (1,230 tis). This meansthe ma.imum global .. ui aenc ra..o is expectedto be on the orderof 0_7, whhic still exceedsthe NUREG-1821, Volume 1 (Ref. ill4) validaton range of0. 0. Q6. When the initial oxgnreevi inthe gross PSL Unit 2 MCR volume of2,22 m8 (85,170 f as tabul2ated in Ref 15 forthe CFAST modcil conse*nred iowith the **esh aur supply flow, it capn be shOAn that this oxygen reservoir is capable of supportinig a 415 MW (4,270 Btuls) source ire for twenty-five minuters at a global equivalence ratio af 0.6. This means. that the adu global equivalenc~e ratio-decreases to abou t 0.11 (for maximum fire Size) or0.7(o m*x imm ave..rage fire Siz) aPd thus falls within the range considered by NUREG 18.2...1 Volu (Refill).. Thm. no. designated

.. smoke purge mode per DBDHA 2, Revision 2 (Ref j22) and Drawing 2998 - 862, Re'sio; 30 (Ref. 2) Thus, the maximum equivalenc ratio for any fire scoriano considered is on the order of 0.17 in the Unit 2 MCR w.-hen there is fomd v'entiQa 1onp In the r-a-se of non forrcead- ventilation, the maximum global equivgalenPC rgati mrpay also be dete-iind using the inital mass of oxygen availablemwhile conservatively ignoring the boundary leakage flows. The bounding casewith respect tothe NUl REGG 1821, Volume

! (Ref. i 4) validation space is when the bounda.y doom, remai closed. The initial oxyge.n m.sepv.ir can support a 3-19 MW (3,310 Btls) fire fbrtwe ... five moipn ute at an equivaleoce ratio of 0. 6 This means thatthe max.imum globl equivalence ra-ti.owould beoabu 0.53 whncnieigO the peak fire size or 0.22 when considering the larged average fire sie ov.,er the twenty five minute inWtersal This equivalenc ratio conservatively ignores the factthatthe large fires raesult in abandionme-nAt times On the order of six to seven minutes, which woul,d deease the maxipmum equivalence raio by a factor ofthree to ab2ot 0. 07. It is thu-s condu,_ded that even w. ;hen the doo2are cloed th mximum equivalence ratio remains within or belaN the NU REG 1821, Volume 1 (Ref- Ji1) validation rang..

Many fime scnari-os-will actally have an equivalencep ratio that is; belmwthe minimump valu of 0..0 ;assessed . in NU REG1821, Volumes 1 and 5 (Refs. e12,1i1)** This means that ther is a. suffident supply of oxygen available forthe fire up to the time at which abhnadonment is predited. Further, a2l scenarios begin With an equivalence ratio of 0.0,

r.Inc ,igtho*e th*t fopmthA NUREG 1821, Volume! (Ref 1A1)-validaIon basis; thus, the scenario evaluatd inthe PSL Unit 2 MCR am not inconsiwent ,ith the validation senarios' co.nsidred in NIREG 1824, Volume I (Ref. i14l) sim because the eq'-iv.aa!ence ratio is .,. Givoen that this is the mos ad-eme eledrical panel fre sc*nario postulated, the global equivalence ratio at the predided aband-o*Amen time is expected to be comparable orp l-Aer for the les s seve-re eledriical panel fire sceara-ios. and the tran.ient fire scena.ri. Consequently, even whe. theHVAC is inop-nerive and the boundary doom are closed, the maximum global equivalence- ratio w.it.hin the M4CR domain is expectedto emarin w*ithin the NIUREG-182*,*Volume 1 (Rfi 14) valida*t*i ranAg Up uIntil the time atw4Ahic~h abandonmepnt is prediced.

Finally, the flame IenAgth ratio is.noally met but in the cars ofhe largest fire sies postulated, the flame height may re*ah or exce-ed the coiling height. Because sprnlder actuation and the.al radiation to targets are not computed with the CFAST model, this Revision 0 Page J-9

Enclosure I to L-2014-109 Page 72 of 106 Florida Power & Light Attachment J. - Fire Modeling V&V parameter isnot an applicable metrici.- Rather, the plume entrainmenAt belowothe hot gas

!ayer controls the layer d*escet time and the concn.ttion ofsoc produds inthe layer-This aspect of the model is not affe-ded by the flamne height to ceiling height ratio.

Consequently, the appliation of CF-RAST to mode!Ifires in the PIS l UInit ontro0 l mom falls entirely within the NUREG- 1824, Volume 1 (Ref. l4)V'..lida..on spac.

Additional V&V studies are contained in NIST SP 1086 (Ref. J137) and NRLIMR/6180-04-8746 (Ref. J202_). These studies have a broader parameter validation space than NUREG-1824, Volume 1 (Ref. J 14D). NIST SP 1086 (Ref. J 13Z) is based in part on the methods of ASTM E1355 (Ref. J2-1). NRIJMR/6180-04-8746 (Ref. J2-0__) provides a Navy specific V&Vstudy, which includes an assessment of CFAST, Version 3.1.7 predictions in multiple enclosures and multiple elevation configurations. These additional studies extend the range ofthevalidation space to include configurations and conditions applicable to the MCR abandonment sensitivity analysis (Appendi 13 of Ref- J52).

Table J-1 provides a summary ofthe validation and verification basis for CFAST, Version 6.0.10 as applied inthe Unit 2 main control room abandonment report.

Summary It is concluded thatthe CFAST applications inthe "Evaluation of Unit 1 Control Room AbandonmentlTimes at the St Lucie Plant" (Ref. J41) and "Evaluation of Unit2 Control Room Abandonment Times at the St. Lucie Plant" (Ref. J52) has a validation and verification basis that meets the requirements of National Fire Protection Association (NFPA) Standard 805, Section 2.4.12.3 (Ref. J242).

Page J.10 Revision 0 Page 3-10

Enclosure I to L-20 14-109 Page 73 of 106 Florida Power & Light Attachment J - Fire Modeling V&V Generic Fire Modeling Treatments The =Generic Fire Modeling Treatments," Revision 0 (Ref. J253) document is usedto establish zones of influence for specific dasses of ignition sources and primarily serves as a screening calculation inthe PSL FPRA under NUREGICR-6850 (Ref. J610_,

Sections 8 and 11. The "Generic Fire Modeling Treatments", Revision 0 (Ref J252) document has two fundamental useswithin the PSL FPRA:

" Determine the ZOI inside which a particular ignition source is postulatedto damage targets or ignite secondary combustible materials; and

• Determine the potential of the ignition sourcesto generate a hot gas layerwithin an enclosure that can either lead to full room burnout or invalidate the generic treatment ZOls for a particular class of combustible materials.

TheZOI is determined using a collection of empirical and algebraic models and correlations. The potential for a hot gas layer having a specified temperature to form within an enclosure is determined using the zone model CFAST, Version 6.0.10 (Refs.

Verification The calculation development and review process in place at the time the "Generic Fire I ModelingTreatments', Revision 0 (Ref. J2§ijwas prepared included.

contributions from a calculation preparer, a calculation reviewer, and a calculation approver. The responsibilities for each are as follows:

" The calculation preparer develops and prepares the calculation using appropriate methods.

• The calculation reviewer provides a detailed review ofthe report and supporting calculations, induding spreadsheets and fire model input files. The reviewer provides comments to the preparer for resolution.

" Calculation approver provides a reasonableness review of the report and approves the document fur release.

The calculation preparation occurred over a two year period ending in 2007. The review stage was conducted in 2007 at the completion of the preparation stage. The calculation was approved January 23, 2008. The approved document, the signature page, and an affidavit were transmitted to the Document Control Desk at the Nuclear Regulatory Commission in Washington, D. C. on January 23, 2008.

In the case of the empirical equations/correlations that form part of the basis of the "Generic Fire Modeling Treatments". Revision 0 (Ref. J253) a considerable amount of verification was performed during the preparation stage bythe preparer. The empirical equations/correlations were solved with Microsoft Excel I spreadsheets using either direct cell solutions (algebraic manipulation) or Visual Basic macros. All direct cell solutions were validated by the preparer through the use of alternate calculation. For simple equations, this entailed matching spreadsheet solution to the solution obtained Revision 0 Page J-1 1

Enclosure 1 to L-2014-109 Page 74 of 106 Florida Power & Light Atftahment J - Fire Modeling V&V using a hand calculator. For more complex calculation schemes, the alternate calculation verification entailed "Ahf yidigh "ji "i r

• gM ant matching the solution using a hand calculator or matching the solution I to a verified solution (i.e., the NUREG-11805 (Ref. J263) solid flame heat flux models).

The verification of the Visual Basic macros also depended on the type of macro. In situations where the macro is used to perform multiple direct computations, the macro results were verified against the verified spreadsheet solutions that were verified through alternate calculation. In caseswhere the macro is used to find a root, the root was verifiedto be a zero by direct substitution into an alternate form of the solved equation.

The empirical equations/correlations were further verified bythe reviewer using a Design Review method as indicated in the signature sheet. An independent reviewer was provided access to the draft report and all supporting calculation materials in late 2007. The reviewer conducted a detailed review of the implementation of the equations within the spreadsheets and the reporting of the equation result inthe draft report.

Comments and insights were provided tothe preparer over the review period and were addressed to the satisfaction of the reviewer. Upon the completion of the review, a revised draft was prepared for review by the approver. The approver provided a higher level reasonableness check of the methods, approach, and the results. Comments and insights that were provided by the approver were addressed to the satisfaction of the reviewer and Revision 0 of the reportwas prepared and approved on January 23, 2008.

The verification forthe CFAST model (Version 6.0.5) is provided in NUREG-1824, Volume 5 (Ref. J 12fi). Supplemental verification for CFAST, Version 6.0.10 is provided as an appendix totheMCR abandonment calculations (Refs. J41, J52) rpport aswell as in NIST SP 1086 (Ref. J13Z).

Validation The empirical equations and correlations are drawn from a variety of sources that are documented invarious chapters of the SFPE Handbookof Fire ProtecionEngineering, peer reviewed journals (e.g_, the Fire SafetyJouma/), or engineering textbooks. The empirical models primarily fall into three groups:

" Flame height;

" Plume temperatures; and

• Heat fluxes (at a target location).

Table J-2 of this attachment identifies the empirical models that are used either directly or indirectly in the "Generic Fire Modeling Treatments", Revision 0 (Ref. 2.

The table also identifiesthe original correlation source documentation andthe correlation range in terms of non-dimensional parameters. The table also provides where applicable supplemental validation work that may have been performed on the correlations and provides limits applied inthe "Generic Fire Modeling Treatments",

Revision 0 (Ref. J25_110,= as applicable.

Except for the cable tray ZOI calculation, the flame height calculation is used only as a means of placing a limit on the applicability of the ZOI tables which are based on the Revision 0 Page J-12

Enclosure I to L-2014-109 Page 75 of 106 Florida Power & Light Attachment J - Fire Modeling V&V plume temperature and thermal radiation heat flux. The flame height calculation for axisymmetric source fires is robust and has considerable pedigree. The original documentation and basis of the flame height correlation is Heskestad (Ref. J2.7-4) as noted in Table J-2 of this attachment. Although there are earlier forms of the flame height equation, Heskestad provides a link between the flame height and plume centedine temperature calculation and identifies the range overwhich the plume equations are applicable. Because the flame height and plume centedine temperature I equations are linked, the plume centerfine range cited by Heskestad (Ref. J274) applies to the flame height calculation as well. The plume centedine temperature equations, and thus the flame height correlation, are applicable overthe following range as noted in Table J-2 (Refs. J2_74, J28_:

-5Z I* logDZ where c. is the heat capacity of ambient air (k/lkg-K [BtuJIb-°R]), 7T.is the ambient temperature (K['R]). 9 is the acceleration of gravity (mis' [ft/sJ), p. is the ambient air density (kglm2 [lb/ftlj), Q is the fire heat release rate (kW [Btuls]), r is the stoichiometric fuel to air mass ratio, D is the fire diameter (m [11]), and AH, is the heat of combustion of the fuel (kJfkg [Btu/lbD. Application of Equation (3-1) depends on the fuel as well as a non-dimensional form of the fire heat release rate (fire Froude Number). In practice, the heat of combustion to air fuel ratio for most fuels will fall between 2,900 - 3,200 kJ/kg (1,250 - 1,380 BtulIb), and fortypical ambient conditions the L- ratio forwhich the plume equations have validation basis is between 7-700 kW*Sm (2.1 -210 Btu2 */s 2l-ft) (Refs. J25, For fire sizes on the order of 25 kW (24 Btuls) or greater, this 2 means that the plume centerline equation is valid for heat release rates of 100 kWIm 2

(8.81 Btuls-ft=) to well over 3,000 kW/m- (264 Btuls-ft ). For weaker fires (e.g., unit heat release rates less than 100 kWemn [8.81 Btu/s-ft21], the tendency of the model is clearly to over-predict the temperature and flame height thus for applications outside the range I but below the lower limitthe resultwill be conservative. The concern is, therefore, entirely on the upper range of the empirical model. The tables in the °Generic Fire ModelingTreatments", Revision 0 (Ref. J25,3) are specifically developed with transient, lubricant spill fires, and electrical panel fireswith a heat release rate per unit area within the validation range. When the heat release rate per unit area falls outside the applicable range, the table entry is not provided and it is noted thatthe source heat release rate per unit area is greaterthan the applicable range forthe correlations. This applies to the flame height andthe plumetemperature for axisymmetric source fires.

The flame height and plume centedinetemperature for line type fires (fires having a large aspect ratio) are applied only to cable tray fires. The correlation used has pedigree and has existed in its general form since at least Yokoi (Ref. J3302_). Most recently, Yuan et al. (Ref. J34M28 provide a basis forthe empirical constant using experimental data with source fires having a width of 0.015 m - 0.05 m (0.05 - 0.16 ft) and a length of S0.2 - 0.5 m (0.7- 1.64 ft) (Ref. J342_). When normalized, the applicable heightto heat release rate per unit length range C) forthe correlations based on the experiments of Yuan et al. (Ref. J3428) is between 0.002 and 0.6. This range indudes the flame height Revision 0 Pa ge J -13

Enclosure I to L-2014-109 Page 76 of 106 Florida Power & Light Attachment J - Fire Modeling V&V as well as the elevation at which the temperature is between 204- 329°C (400 -

625'F). the temperature at which cable targets are considered to be damaged under steady state exposure conditions. Yuan et al. (Ref. J3428) also provide a tabular comparison ofthe empirical constant against seven preceding line fire test series, which include a broader range of physical fire sizes and dimensions. The Yuan et al. (Ref.

J3428) constant is greaterthan the other seven and thusthe temperatures and flame heights are more conservatively predicted using the Yuan et al. (Ref. J3428_ data. The application of the Yuan et al. (Ref. J3428) correlation inthe "Generic Fire Modeling Treatments", Revision 0 (Ref. J2-5) document falls within the normalized applicability range reported by Yuan et al. (Ref. J34Z-2)

Four flame heat flux models are used inthe "Generic Fire Modeling Treatments",

Revision 0 (Ref. "' u as described in Table J-2 of this attachment: the Point Source Model (PSM), the imple Method of ,ýW and Beyler, the Method of MUft and Croce, and the Detailed Method of %ot and Beyler. The formertwo are simple algebraic models using the heat release rate, separation distance, and the fire diameter.

The latter two are considered detailed radiant models that account for the emissivity of the fire and the shape of the flame. Dueto limitations inthe target placement, the (Simple) Method of, hand Beyler are shown to be inapplicable for calculating the ZOI dimensions. Similarly, forthe fuels considered, it is shown thatthe Method of M and Croce produces a net heat flux that exceedsthe fire size. The ZOIs are, therefore, determined using the Point Source Model and the Detailed Method of gti],

and Beyler. The method that produces the largestZOI dimension is used for each fuel and fire size bin.

The Point Source Model and the Method of,*hol and Beyler have been shown inthe I NUREG-1824, Volume 3 (Ref. J3229) verification and validation study to provide reasonably accurate predictions when thetarget separation to fire diameter R,) ratio is between 2.2 and 5.7 per NUREG-1824, Volume 1 (Ref. J14_). Furthermore, the fire size ranges considered inthe Generic Fire Modeling Treatments", Revision 0 (Ref. J2-53) report are between about 25 - 12,000 kW (24-11,400 Btuls) and the heat release rates per unit area range between about 100 - 3,000 kW/m' (8.1 - 264 Btu/s-ft=) for all fuels and fire size bins.

Using this information, the following table may be assembled forthe applicable target heat flux range, based on the NUREG-1824, Volume 1 (Ref. J148) validation range:

Revision 0 Page J-14

Enclosure I to L-2014-109 Page 77 of 106 Florida Power & Light Attachment J - Fire Modeling V&V Heat Release Point Source .5b0ft and Fire Size Rate Per Unit Fire Diameter, Model Heat Flux Beyler HeatFlux 2

kW (Bhuds) Area, kWlm m (it) Range, kWImr Range, kWim2 (Btuls-fF) (Bufrus--W) (Btuts-W-)

25 008.8 0.07-0.45 0.36-3.8 (0.005 - 0.04) (0.03 -0.4) 25 (24) 3.000 (264) 0.1 (0.3) 2-13.) 2.84-10 10.07-0.45 0.553-05 100 (8.8) 12.4 (41) (0.007-0.04) (0.05 -0.4) 12,000 (11,400) 2-13.6 0.45-04.4 12,000 (11,400) 3,000(264) 2.3(7.4) 2-1.2 0.4-0.4 11 (0.2 -1.2) (0.04 -0.4)

The threshold heat fluxes that define the steady state ZOI dimensions range from 5.7 -

11.4 kWlma (0.5- 1 Btu/s-ftl). Transient ZOI dimensions, addressed in the "Supplemental Generic Fire Modeling Treatments: Transient Fuel Package Ignition I Source Characteristics', Revision 0 (Ref. J334) mayapproach 16-18 kW/m2(1.4-1.6 2

Btuls-ft ). Clearly, the steady state ZOI dimensions based on critical heat fluxes of 5.7 -

11.4 kW/ma (0.5 - 1 Btu/s-ft2 ) overlaywith the range of valid predicted heat fluxes I identified in NUREG-1824, Volume 1 (Ref. J 148). Fuels that identify the most conservative value over a range of heat release rates per unit area (transient and electrical panels) will thus include at least one point within the validation range (i.e., 5.7 kWIm 2 [0.5 Btu/s-ftl or 11.4 kW/m- [1.0 Btuls-ftq, depending on the unit). Since the algorithm searches forthe most adverse value, the result will be at least as conservative as the value obtained within the model validation range.

There are combinations of fuels and source strength ranges that do not produce heat fluxes that fallwithin the validation range. This is especially true for the higher target heat flux values (11.4 kW/m 2 [1 Btuls-ft2 ] and higher) combined with the lowertransient fuel package heat release per unit area range (200 - 1,000 kW/m2 [17.6 - 88.1 Btuls-ft]). This is addressed through an extended validation range of the heat flux models I provide by the SFPE (Ref. J430). As noted inTable J-2 of this attachment, the SFPE assessed the predictive capabilities of the Point Source Model and the Detailed Method of *SWi and Beyler against available pool fire data. The pool diameters ranged from 1

- 80 m (3.3 -262 ft). The conclusion was that the Point Source Model was conservative but not necessarily bounding when the predicted heat flux is less than 5 kWIm 2 (0.44 Btu/s-ftt ) and the empirical constant (radiant fraction) is 0.21. The method is bounding when a safety factor of two is applied to the predicted heat flux. The application inthe I "Generic Fire ModelingTreatments', Revision 0 (Ref. 125_) uses an empirical constant (radiant fraction) of 0.35, indicating the application is essentially bounding. Similarly, it was concluded that that Method of $&Ui and Beyler is conservative when the predicted heat flux is greater than 5 kW/m2 (0.44 Btuls) and the method is bounding when a safety factor of two is appliedtothe predicted heat flux. The implementation in the "Generic Fire Modeling Treatments", Revision 0 (Ref. J253) is conservative, though not bounding. Although the SFPE considered fire diameters greaterthan about 1 m (3.3 Revision 0 Pag~eJ-15

Enclosure Ito L-2014-109 Page 78 of 106 Florida Power & Light Attachment J - Fire Modeling V&V ft), smaller diameter pool fires are not optically thick and have a lower emissive power (Ref. J812). Thus, the use of the methods fbr smaller fires is conservativethough outside the SFPE validation range.

The use of the heat flux models largely falls within the NUREG-1824, Volume 1 (Ref.

J 148) validation parameter space range; however there are caseswhere this is not so.

For larger diameter fires, the SFPE provides comprehensive validation against full scale test data of the methods applied. The application in the "Generic Fire Modeling Treatments", Revision 0 (Ref. J253) report and the applicable supplements necessarily fall within the validation range or are more conservative because the solution algorithm identifies the most adverse solution among the methods. Smaller fires may fall outside the validation range of both studies, but such fires have a lower emissive power and are conservatively treated using the methods designed for high emissive power source fires.

A number of other empirical models that appear inthe generic fire modeling treatments are applied within the stated range of the models orthe data forwhich the models were developed. For example, the cable heat release rate per unit area model is based on cables that have a small scale heat release rate that ranges between 100- 1,000 kW/m2 (8.8- 88.1 Btuls-ft 2

). The solution tables are provided forthis range. The unconfined spill fire model (heat release rate reduction factor) is based on observations of pool fires having a diameter between 1 - 10 m (3.3 -33 ft). The diameter range for which ZOI data is provided is 0.7- 5 m (2.2- 17 ft). The lower rangevalue is less of a concern due the reduction in the optical thickness ofthe fire when the diameter falls below 1 m (3.3 ft). The upper range is maintained in the ZOI solutions. The offset distance for flame extensions outside a burning panel have an upper observational limit of about 1,000 kW (950 Btuls), though it is applied in a normalized form (extension to panel height ratio). The ratio is applied as determined from the test data.

The CFAST applications in the "Generic Fire Modeling Treatments', Revision 0 (Ref.

JRApjB consist of simple geometries with a single natural vent path connected to an ambient boundary condition. The simulations are used to determine the time after the start of the fire thatthe hot gas layer temperature reaches a predetermined critical temperature. No consideration forthe hot gas layer depth is made; if the hot gas layer temperature reaches the critical temperature at anytime, then this time is the sole output parameter used inthe "Generic Fire ModelingTreatments' report. The enclosure geometry is specified as a function of the volume in such a way as to minimize the heat losses to the boundary. Three vent configurations are evaluated for each volume-room geometry-vent fraction; the most adverse result among the three vent configurations is used.

The room geometry and fire parameters for the 'Generic Fire Modeling Treatments",

Revision 0 (Ref. J2-5) simulations fall within the model limits listed in NIST SP 1026 (Ref. J 104) and NIST SP 1041 (Ref. J 145). Specifically, the vent area to endosure volume ratio is less than two and the aspect ratios of the enclosures are less than five.

The non-dimensional parameters that affect the model results as documented in NUREG-1824, Volumes 1 and 5 (Refs. J126, J148) and NUREG-1934 (Ref. J159) include the model geometry, the global equivalence ratio, the fire Froude Number, and Revision 0 Page J -16

Enclosure I to L-2014-109 Page 79 of 106 Florida Power & Ligght Attachment J - Fire Modeling V&V the flame length ratio. The non-dimensional parametersthat relate to target exposure conditions (heat flux) and sprinkler actuation (ceiling jet) are not applicable to this calculation because these output parameters are not used. The non-dimensional geometry parameters (length to height and width to height, which range from 3.3 -4.3)

I fall within the NUREG-1824, Volume 1 (Ref. J 14_) validation range (0.6- 5.7). As previously noted, CFAST does not use a fire diameter, therefore, it is possible to specify a fire that fallswithin the range of fire Froude numbers considered in the NUREG-1824, Volumes 1 and 5 (Refs. 1 i i Thesource fires considered are consistentwith those described in NUREGCR-6850 (Ref. J61_0) and thus those that are the subject of the NUREG-1824, Volume 1 (Ref. J 148) validation effort- The global equivalence ratio does exceed the ratio validated in NUREG-1824, I Volume 1 (Ref. J148), in some cases by a significant margin. Large fires in very small volumes with law ventilation could effectively result in equivalence ratios that even exceed the maximum values observed in fully developed fires (3 - 5) (Refs. N 2 However, the limiting oxygen index used inthe model is zero, which forcesthe combustion process to use all available oxygen within the enclosure and the heat release rate to decrease to a value setby the natural ventilation oxygen inflow. The maximum temperature over the course ofthe fire occurs at some time priortothe oxygen being consumed inthe enclosure, thus the global equivalence ratio for the data reported is based on a condition where it is less than unity and within the validation I basis of NUREG-1824, Volume 1 (Ref. J 148). Further, for a given volume and fire size, an optimum ventilation condition will occur overthe vent range considered. Because of potential variations in a ventilation condition, the FPRA uses the most adverse time over the reported range and effectively performs an optimization on this parameter.

Finally, the flame length ratio is not always met, especially for large fires postulated in small enclosures. Because sprinkler actuation and thermal radiation totargets are not computed with the CFAST model, this parameter is not an applicable metric. Rather, the plume entrainment below the hot gas layer controls the layer decenttime and the concentration of soot products in the layer. This aspect of the model is not affected by the flame height to ceiling height ratio.ii Additional V&V studies, which are useful for extending the range of applicability of the model, are contained in NIST SP 1086 (Ref. J13_) and NRLUIvR/618O-04-8746 (Ref.

J2-020. These studies have a broader parameter validation space than NUREG-1824, Volume 1 (Ref. J 14B). NIST SP 1086 (Ref. J 13_) is based in part on the methods of ASTM E1355 (Ref. J241). NRUMR/6180-04-8746 (Ref. J2-02) provides a Navy specific V&V study, which includes an assessment of CFAST, Version 3.1.7 predictions in multiple enclosures and multiple elevation configurations. These addtional studies extend the range of the validation parameter space to include configurations and conditions presented inAppendix B of the "Generic Fire Modeling Treatments', Revision 0 (Ref. J25,3) report Appendix B of the "Generic Fire Modeling Treatmentse, Revision 0 (Ref. J253) report provides an in depth analysis of the parameters used as input and Table B-2 indicates the basis forthe input parameter selection. The parameters are either selected as Page .1.17 Revision 0 Rejision Page J-17

Enclosure I to L-2014-109 Page 80 of 106 Florida Power & Light Attachment J - Fire Modeling V&V absolutely bounding over the credible range or establish an application limit (e.g.,

elevated temperature environment and boundary thermal properties).

A summary of the validation basis for both the CFAST and the empirical models is provided in Tables J-1 and J-2 of this attachment Based on the information inthe tables and the preceding discussion, it is shown thatthat the empirical fire model applications in the "Generic Fire Modeling Treatments' either fail within the original correlation bounds orthey are outside the bounds but used in a way that is demonstrably conservative. Likewise, CFAST is usedwithin the model limitations described inthe CFAST Users Guide (Ref. J 1.45) andthe CFASTTechnical Reference Guide (Ref.

J 104). The results as reported in the "GenericFire ModelaigTreatments", Revision 0 (Ref. J25_) document are based on conditions that fall withn the parameter space considered in NUREG-1824, Volumes 1 and 5 (Refs.

• MW Vjj*&

Sg UW The use of the"Generic Fire Modeling reatments", Revision 0 (Ref. J25_3) inthe FPRA performs an optimization over the ventilation fraction and is necessarily based on a condition that falls within the I NUREG-1824, Volumes 1 and 5 (Refs. 1 i ti

&g ,q&_%*,, Given these considerations, it is concluded thatthe CFAST I application in the "Generic Fire Modeling Treatments", Revision 0 (Reft J2-53) document has a validation and verification basis that meets the requirements of NFPA 805, Section 2.4.12.3 (Ref J242).

Generic Fire Modeling Treatments Supplements There are five generic supplements to the "Generic Fire Modeling Treatments'. two of which are used by the PSL FPRA4R4f.4J4):

" Supplement 1: "Supplemental GenericFire Modeling Treatments: Closed Electrical Panels," Revision B (Ref. J375); and

" Supplement 2: "Supplement'al Generic Fire Modeling Treatmen-ts: Hot Gas Layer Tables," Revision H (R4f J38)j-and

" Supplement 3: "Supplemental Generic Fire Modeling Treatments: Transient Fuel Package Ignition Source Characteristics," Revision 0 (Ref J334).

Suppalement 4, "Supp lemena tal GeQ-nerPF*re Kodel1in TreatmePtAs Transient Targe Response to Transient Ignition S-ource Fire Exposures;" Rsvision A (Ref J39), and Supplement 5 "Supplempental GeQ-ne-ric Fire Modeling TreatmenAts: Solid S-tate COntrLA Component ZO7 And Hot G as LayerTables", Re;ision 0 (Ref. J40) are not Used inthe PSL FPRA (Re&f i".

Supplement 1 Supplement 1, "Supplemental Generic Fire Modeling Treatments: Closed Electrical Panels," Revision B (Ref. J37) considersthe maximum potential fire size thatcan develop in electrical panels that are vented. The supplement consists of a theoretical model development (energy and mass balance on a control volume) and a parametric sensitivity analysis. This consists of two parts: the first part provides the maximum heat Revision 0 Page J-118

Enclosure I to L-2014-109 Page 81 of 106 Florida Power & Light Attachmert J - Fire Modeling V&V release rate that can be supported by a vented metal box and the second part uses this heat release rate information as input into the same calculation process used by the

'Generic Fire Modeling Treatments', Revision 0 (Ref. 43.*j_ N to generate the ZOI dimensions. The numerical solution for the heat release rate calculation was independently reviewed inthe same manner as was the original "Generic Fire Modeling Treatments" calculations. The ZOI dimension calculation is the same procedure used in the original "Generic Fire Modeling Treatments", Revision 0 (Ref. a and was verified and approved as previously described.

An alternate method verification is also provided in Section 6 of"Supplemental Generic Fire Modeling Treatments: Closed Electrical Panels," Revision B (Ref. J-V5_) The alternate method forwhich the model is compared was developed by Electric Power and Research Institute (EPRI) as described in Ref. J375. The alternate calculation comparison only considers the ventilation limited aspect of the EPRI model. No consideration of the heat release rate conditional probability distribution is provided in "Supplemental Generic Fire ModelingTreatments: Closed Electrical Panels," Revision B The model is validated against available test data for either electrical panel fire tests or tests in metal enclosures having dimensions comparable to electrical panels. These tests are documented in the following references:

• Mangs et al. (Refs. J4433-J4335);

• Mangs (Ref. J443_);

• }ietal. (Ref.J453__-;

• NUREGICR-4527, Volume 1 (Ref. J463_); and

• NUREGICR-4527, Volume 2 (Ref. J47_9).

Approximately thirty fire tests are represented inthe references cited. A common attribute among the tests was the desire to drive the intemal conditions to a fully-developed (i.e., ventilation controlled) burning regime. Not all tests were successful, and consequently a fully developed modelwill over-predict the heat release rate. It is shown by Hunt (Ref. J499) that the fully developed model predicts or over-predictsthe heat release rate from all fire tests except for five tests. Four of the tests involved a p;pjyMftdJ3*I * (PMMA) fuel that exhibited combustion external tothe panel at the vents (Ref. J45&3). The recommended procedure for addressing external burning at the vents is to assume a global equivalence ratio within the panel of two, with half of the combustion occurring inside the panel and half outside the panel. This applies only to fuels that are shown to re-ignite outside the panel and would not typically be assumed for cable materials. The fifth test for which the heat release rate was not conservatively predicted involved a change inthe status of the opening (i.e., the door latch mechanism failed and the door opened). The recommended practice to account forthis situation is to use the Frequently Answered Question (FAQ) 08-0042 (Ref. J913) guidance for determining whether a panel is robustly secured.

The mass and energy balance model used to calculate the maximum fire size that can be supported within an electrical panel is conservative and is shown to predict the heat Page J.19 Revision Reviciori 00 Page J-I9

Enclosure I to L-2014-109 Page 82 of 106 Florida Power & Light Attachment J - Fire Modeling V&V release rate among a large number of fire tests. The tests consist of a large range of ventilation fractions and panel sizes. In addition, the input parameters are conservatively biased as described in Section 5 of"Supplemental Generic Fire Modeling Treatments: Closed Electrical Panels,' Revision B (Ref. J375). As such, the model is considered to be validated over the range for which the ZOI dimensions are tabulated in the supplement Supplement 2, "Supplemental Generico Fire Modsling Treatments: Hot Gas Layer Tables, t Revision H (Ref- i3B) prevides hot gas, layertables for additional crGia temperatures., and ignition sourc. heat release rates, including some ignition ourco_

secondar' fuel packaga combipatiops. ln addhtiop ntheZ=l dimnsiopns for saesitive

~omnponent targets ame provided. The hot gas layer tables and Z01 dimensions a calculated --sing the Same calculation procedureas w,...re. u1ed forthe orgin.s! Ge-e.c Fire ModelingTreatments report butwith diffar-ept input pa.rpeamrs These procedures we0re verified and approved as previously described. The validation.~basis for the ZOI dimeansions is identiOcal to that ofthe original Goneric Fire Modeling Treatmentst ,

Revision 0 (Ref J25) report Hot gas layers are provided fora single generic se.onda combustible ,- configuration.

two horizontaGcable trays located 0 I above

. an eletcal

.f.) panel ig.tin source that may.be located in the ope,* near a wall, opr in a coamer. The cable tra*ls are simultaneously ignited five min,utes f ter ignition at a sinepn point and the fire is all..e to propagate laterally in each direcio-n. The hoizontal flame propagat.io rate used in the a naly siS forh ormp*stic Pad theoseat cables is as raconmm-eaod in INUR IEIF G!CR_

6850 (Ref. J6), or 0.9 mrr.s (0.3 is)ad03 mins (0.12 mIps), respectvely. This proPa*gtion rate h*a been shaan to be broaly applicable to the cable class n Nl'IREG!CR_7010, Volume 1 (Ref J49). The assumad heat release rate per unit area for the abe rzisa censtant and assumed to be 25 kWpm 2 (19.8 BfowIs Wf1, whichi slightly less than the value recommanded in IUREG!CR_7010, Volume 1 (Ref JI4 for t

the..oplaic. rcables (250 kW.m. [22 Btu. s .ft.. applicable value for.the..

The . o.

cables is 150 k.WP.1 (13.2 Btu!s-.f per NUREG-7010, Volume 1 (Ref. J,9). Wall and cnreo r configura-tio-ns are add-ressed using the 'Image' Meffthiod in which the sourer heat release rate and area are doubled for a wAll1 configuration an~d quadrupled for acomer configuration. The enclo-sure bounda.* su.farce 2rea and ventilation are also doubled and quadnipled-forwa0ll a nd comer copnfiguratios* respecively. This treatmenttakes advantage of the proprionality of the entrainmenttothe f4ie per:imater and the constard plume angle (Refs- j29, Ij*OJ52)l an results in more adverse copnditions wnhen the antrainme*nt*fire pr*imaer ratioi is redu-cm-d The secondary combustible hot gas layertablos are prmarIly used as a tool fr addressing sceanaros with the potetin W invoke one or tpo cable trays as secoda*

to y comustble. Scenarios.. withMore adverse cable tray aUangemen. . ar. addr esed in the plant spec-ific; destailed fire modeling report (Ref 12).

Revision 0 Page J -2 0

Enclosure I to L-2014-109 Page 83 of 106 Florida Power & Light Athcsbment J - Fire Modeling V&V Suoplement 3 The focus of Supplement 3 tothe "Generic Fire ModelingTreatments", Revision 0 (Ref.

J253) report, "Supplemental Generic Fire Modeling Treatments: Transient Fuel Package Ignition Source Characteristics", Revision 0 (Ref. J334) is to provide an analysis of and basis for the transient ignition source heat release rate per unit area, the fire duration, and flame height. The analysis uses the original transient fire test data referenced in NUREGJCR-6850 (Ref. JI10)to estimate the transient ignition source characteristics of interest in orderto provide a narrower range of input parameters forthe ZOI calculations addressed in the "Generic Fire Modeling Treatments", Revision 0 (Ref. J253) report.

Supplement 3 is primarily an analysis of test data; however, several revised ZOI tables using the results of the analysis are provided. The ZOI tables determined using the same processes and fire models used to generate the original ZOI tables inthe "Generic Fire Modeling Treatments", Revision 0 (Ref.- The validation and verification developed for the "Generic Fire Modeling Treatments", Revision 0 (Ref.

4&5j~a= forthe model is thus applicable to this supplement.

Detailed Fire Modeling Calculations The detailed fire modeling calculations as documented in Report 0027-0053-000-002, Revision 0 (Ref. J2*. Report 0027-0009-014-004. Revision 0 (Ref. J7). and Report 0027-0009-014-005. Revision 0 (Ref. JB) assess the potential for hot gas layers to exceed certain critical temperature thresholds when secondary combustibles are involved. The calculations providethe hot gas layer and ZOI data for ignition source -

cable tray fuel packages in generic PSL volumes provides detaile d calculaons of the hot gas layerte.p.rat...... gInei-'-umsRPSL for app....roxatel te, igniio so.urc. -cable tray configurations, pri..mIaly thnse that invoak' more than twoJo cable trays.

-The calculations use&two different fire models or Gala 2tj methods:

• The Flame Spread over Horizontal Cable Trays (FLASH-CAT) calculation method, as incorporated in plant specific Microsoft Exceltm Spreadsheets (Refs.

J26 J7 JB, J4S91_; and

• CFAST, Version 6.1.1 (Refs.J104,J-114).

The FLASH-CAT method (Ref. J491) essentially involves a group of recommended heat release rate and flame spread parameters for cables in cable trays and is used in Report 0027-0053-000-002, Revision 0 (Ref- J2_6). Report 0027-0009-014-004, Revision 0 (Ref. J7). and Report 0027-0009-014-005, Revision 0 (Ref. J8) to generate the heat release rate contribution from secondary combustibles. Finally, CFAST Version 6.1.1 (Refs. J 104, J 145) is used to generate hot gas layertables for generic plant spaces with specific secondary combustible configurations. The CFAST results forthe hot gas layer temperature are evaluated over a range of natural ventilation conditions (0.001 - 10 percent of the boundary). The large natural ventilation range considered inthe analysis readily encompasses the ability of a forced ventilation system to provide oxygen while conservatively ignoring the mixing or diluting aspects of such systems. In otherwords, a Revision 0 Page J-21

Enclosure I to L-2014-109 Page 84 of 106 Florida Power &Light Attachment J - Fire Modeling V&V forced ventilation system is not postulated to provide more oxygen than is already assumed over the range of natural ventilation conditions andthe system wouldtendto improve the resultwhen dilution of the hot gas layer is considered.

FLASH-CAT Calculation Method The FLASH-CAT applications in Report 0027-0053-000-002, Revision 0 (Ref. J2,6).

Report 0027-0009-014-004. Revision 0 (Ref. J7), and Report 0027-0009-014-005, Revision 0 (Ref. J 8) iare used to generate the temporal heat release rate for specific cable tray arrangements. The input parameters that are used are those recommended in NUREG/CR-7010, Volume 1 (Ref. J49_) and the initial conditions (initial area and ignition criteria) are those recommended in NUREG/CR-6850 (Ref. J.1_0). The calculations themselves -it-lfi are performed using a-Microsoft Excel'm spreadsheets.

The verification basis for the FLASH-CAT model as incorporated in the Microsoft ExcelmI Spreadsheet involves numerical comparisons against results presented in NUREG/CR-7010, Volume 1 (Ref. J491). These comparisons are providedwith the I detailed fire modeling reports (Refs. J26. J7 and JB) and serve asthe verification that the model is correctly implemented as a Microsoft Excel' spreadsheet The validation I forthe FLASH-CAT method is provided in NUREGICR-7010, Volume 1 (Ref. J491) using about thirty different cable samples. The samples include cables having the same or similar materials as the predominant cable types used at PSL (e.g., polyvinyl chloride

[PVC], polyethylene [PE], and cross-linked polyethylene [XLPE] per Ref. J41 and Ref.

J52) such that the results and conclusions are applicable. An added measure of conservatism is provided inthe FLASH-CAT analysis by assuming thermoplastic cable flame spread and propagation properties for all but one configuration (Ref. J26 _J7 J8).

There is no validation range per se specified forthe FLASH-CAT method (Ref. J491).

Rather, it may be inferred that if the configuration is similar (i.e., horizontal cable tray stacks) and the cable composition is similar, the results are applicable and NUREGfCR-7010. Volume 1 (Ref. J491) serves as the validation basis. The FLASH-CAT applications described in Report 0027-0053-000-002, Revision 0 (Ref. J2) Report 0027-0009-014-004. Revision 0 (Ref. J7). and Report 0027-0009-014-005, Revision 0 (Ref. J) -involve horizontal cable tray stacks with some vertical or vertically sloped segments involving materials that are among those tested. The horizontal segments I conform to the NUREGICR-70 10, Volume 1 (Ref. J491) test configuration butthe vertical segments do not However, the vertical segments are conservatively assumed to propagate at a faster rate as recommended in NUREGICR-6850 (Ref. J1__.

Therefore, the FLASH-CAT application has a validation and verification basis that meets the requirements of NFPA 805, Section 2.4.1.2.3 (Ref. J242).

CFAST CFAST, Version 6.1.1 (Refs. J 104, J 1-_) is used to generate hot gas layer tables that provide the time various temperature thresholds are reached in the generic PSL volumes for spedfic secondary combustible arrangements spaces using the FLASH-CAT temporal heat release rates. The CFAST hot gas layer temperature applications Revision 0 Page J-22

Enclosure I to L-2014-109 Page 85 of 106 Florida Power & Light Attachment J - Fire Modeling V&V are identical tothe approach adopted inthe =Generic Fire Modeling Treatments",

Revision 0 (Ref. J2-53).

The CFAST analysis assesses the time the hot gas layer temperature reaches threshold values over a range ofventilation conditions (0.001 - 10 percent of the boundary area). The ventilation condition that results inthe most adverse time for a given scenario is used inthe FPRA#Z(.Q.4-The verification forthe CFAST model (Version 6.0.5) is provided in NUREG-1824, Volume 5 (Ref. J 12§). Supplemental verification for CFAST, Version 6.1.1 is provided as an appendix to Report 0027-0053-000-002, Revision 0 (Ref. J2) and Report 0027-0009-014-005. Revision 0 (Ref. J8) rp-rtas well as in NIST SP 1086 (Ref. J 13_).

The validation for CFAST described forthe original Generic Fire Modeling Treatments",

Revision 0 (Ref. J25J3 applies, includingthe following:

The equivalence ratio for some ventilation cases will fall outside the NUREG-1824, Volume 1 (Ref. J 14R) validation parameter space. However, at least one ventilation condition will be within this range, and the results are thus no less conservative than a case that fallswithin the NUREG-1824, Volume 1 (Ref.

J 14D)validation parameter space. In general, the most adverse resultswill be predicted when the equivalence ratio is near unity (optimum burning conditions).

Validation work has been performed for CFAST at these equivalence ratios (e.g.,

Refs. J 137, J20) and applies to the PSL calculation.

Based on these considerations and those provided for the "Generic Fire Modeling Treatments". Revision 0 (Ref. J253), it is conduded thatthe V&V basis for the CFAST application analysis meets the NFPA805, Section 2.4.12.3 (Ref. J242) requirements.

KI Exemption A detailed analysis of the cable tray separation inthe Unit 1 Containment Building was performed as part of Exemption K1 (Ref. J39) and documented in Report 6372 (Ref.

J5342). This cable tray analysis is credited in the FPRA as a basis forthe adequacy of the cable tray separation in the Unit 1 and Unit 2 Containment Buildings-(Rot i54). The analysis evaluates the effect of a cable tray stack fire on an adjacent cable tray stack in the Unit 1 Containment Building that may be as close as 2.1 m (7 fi). The cables are thermoplastic, but are coated with [Ejjiantiý The analysis is performed both with and without crediting the ULj astij for increasing the damage temperature. The analysis develops an exposure heat release rate within the initiator stack using avaiable test data on cable trays, applies empirical calculations to quantify the heat flux, and conducts a sensitivity analysis on a number of factors including forced ventilation and enhanced fire spread rates. The heat transfer model HEATING, Version 7.3 (Ref.

I J5543) is used to further compute the relative safety margin when the cable damage threshold is assumed to be that ofthermoplastic cables.

I An alternate calculation is also provided in Report 6372 (Ref. J5342) using the Fire Induced Vulnerability Evaluation (FIVE) methodology, though the FIVE model itself is Revision 0 Page J-23

Enclosure I to L-2014-109 Page 86 of 106 Florida Power & Light Attachment J - Fire Modeling V&V not used; rather, the Point Source heat flux model is applied using parameter assumptions recommended by EPRI (Ref. J5644).

The empirical calculations and input files were verified by independent review as indicated on the signature page of the fire hazards analysis report (Refs. J39, J5342).

Report 0027-0009-014-002. Rev. 0 (Ref. J5-45) provides additional verification of the fire model calculation components using an alternate calculation as well as a validation basis for each calculation component associated with the baseline fire scenarios.

Report 0027-0009-014-002, Rev. 0 (Ref. J5745) also provides a verification and validation basis for the thermal evaluation performed usingthe heat transfer model HEATING. The key empirica ca-lcul-aons that require validation inv.. oe the cable t*a*

fire rel~eas rte÷ deeop~~lrment hich sein effectaen aJltemaetotthe FlAS C*-_AT metho hetherth analysis his ...sesitie tov...ous assumed conditions (air fiow effect) The u-se of these comQatio÷ns is notcentraltothe analysis orits condusicns andas such do not rcwuire validatiion in this attachment !n addition, the alternate mpethod cal ,lclion (i~e, Pontte oure het fux odelq is Pot direcotly used in the analysis,and does not require vaidation in this attachment. The altermate cal*olation does provide a degree of verilfication of the calculation r~euts, ha*^vr.

The detailed cable tray separation analysis forthe 'Jnit 1 Conptainment Building (Ref.

J53) is also used tonc*redt the cable t.ay separation n the Unit 2 Containment Building (Ref. J54). The basis for the Unit 2 croef is thaqt the IUnit 1 configuration is Similar.to or boun-llds the Unit 2 configur9"atini terms- of the fuel load, the cable tray separation, the cable flame spread rate, and the c abhle hat release rate per unit aPra (a. ., Unit 1 har thermoplas c cbLes .. and Unit? has-thermoset cables) (see Ret j57)..

Revision 0 Page J-24

Enclosure I to L-2014-109 Page 87 of 106 Florida Power & Ligaht Aftachment J - Fire Modelina V&V The empidrial c n the cmputation oftth fire size, fire dimenswons-and maianAt heat flux. The-ire spreaad and propagdiopn through the cable!tra stac as daesribed in'A Repo 6372 (Refs. M, j53) is readily shg.. to he mor..consr vatv than the FLSHAC-AT method (Ref 1j.9). Spedfically-

- The baseline flame spread rate is 2 mm!s (0.078 in!s). The flame spread rate re o ed d i* NWREGCR-46850 ied (Rot 26) and u.sed by the h LASH CAT method (Ref. j49) is 0.9 rmm~s (0.035 m~)forthernopjastic cables. This aconervt~' faer f a lest wo intrducs wth respectto fire grptmh along a cable %Fay

_ The propagation tim bet.ween cable trays ,within the same tray stack*is intnt The propagation time recommended in NUREGI!IC-6850 (Ref .A) an;d *&ed by the FLASH- CAT method (Ref. J49) is Thur minutes to the second tray, three add itioanal m inuteas tot Qthe t hirdttray, two additio-nal m-ainutesto-t he. fouth t ray. an One addtoa miut o ah subsequenttray. Trays that are coated with Flaematici~lhave an wpignfio delay of tAweve mwinue betwee thist and sezond cable tray (Ref. 26).

• The ma~ximu-m baseafine heat releasse rate on wAhich the resul-ts are based is450

'M2 (39.6 Btuas whe covfe toa'w clevle m usingthe method describedinp National Bureau1 of StAndrs 396 (Ref-.

-3ei~prtNSR8 258). The valueA recommenndedin WINRUE!CIR701 0,Voluwme 1 (Ref 119) i2S kW'm2 (22 Btuis fiý. This introduces a consr.QpaW' fadenrof aboia twmmýomwth re spect togtheA fire simze at a local ized point

• Temaximump firs size develops inpstantly PAt a fixond location. NUREG!CIR-7010, V oLume@ 1 (Ref. 2149) re commenPds a ramp-u pti me equal to oneP -sixt0h the toal buming duration.

The- thre-shold temperature criterio for thermog0astic cables is21 8 (125EF),

censistentwith the recommendations atthe timpe the reportwas prepared (Ref J56)m However, the results are applicable tothresholdtemperaturos of 201 C(400'F) since they are reporteA.d inte*s- of the maiu cable suFIace temperature. The baselne cases remain bealm boththe Or*iinPal threshold temperature and the curranrtlya copt va,,lue of 2010 C (4O0'F) for a genercthermoplastc cbe Based on these comparsons, it m.ay b cOncuded that the cable trFay fire propagatin Ra alysis prese.ted InARepor 6372 (Ref. 253) bounds the FLASH CAT method (Ref Jh9). The :I LASH-CATalidati*n proosnted in N'REIRF 7 Vlume 1 (Ref. A 9) therefore appklis The fire dimensions..... are base-d on the popagaion distanc* e ad flam*e height. A lins fire flame height coixelation isused that ismore consrvpativethan the flame height correlation described in the af*enern Fire Modeling Treatments, Rvision0 (Ref. 225) report. The applation is 0.Within the range used by the GenericF=ire MOde4ing Treatments', Revision 0 (Ref. 225) report and-1the valmoidation for of the line fire cO-el"on thus applies to this asped of the Ki Exemption (Ref /3) calculation. The radiant heat flu is4; dete4m*A@edby computinthe ge*omti.cvie aor wf* awn dradit*ing fif' perQent of the eatrelaserate energy, abounding vaIlue amongA ThU scale test data (Refs. 231, Revision 0 Page J-25

Enclosure Ito L-2014-109 Page 88 of 106 Florida Poweer & Light Attachment J - Fire Modeling V&V j59). The view factor computation is a geometi.c computation applicable to the tray configuration in *th Ie It1 Cont2inment Building. Inthis context.,the focus ofthe

,aMlIMMOPi ap .. l .. Pi QPh tthA v l' d,*

.-...- n Th ~ ~

.. a "-+'w n .... +g~ig is ,,I ,o,-ribe,,,dip Report 6372 (Ref. j53) anpd shan to biXythe be-nse- with or bouAnd b geometric shape assumed Verificaon ~.and V2lidation ofthe HEA*ING Model The heat transfer computations conducted using HEATING, Version 7.3 (Ref. j55) are provided n the oriQinala Ki Exemption fire hazards analysisto daemon-trate the tras.ien t temperature rsponsrA ofthe cable targets, in particular frthermoplati cables (i.e.,

Flamemastic no~t crediteGD. Aithough this aspectof the analysis, is provided to show.

margin un~derthe moSt ades supinisAn essential aspect to the current analysis credit inthe FPRA_ This is because the Flamemactc cred. taken in Report 6372 (Ref. 253) is not c..nsite"ntwith Appendix 0-of N'UREG!CR-6850 (Ref J),which means the reauts forth& saenarios in which Flamemasticis npot c-rdited are Offctively Scon, te citial emperature used in the or9igial analysis, is the aseinescearis.

gre..ethrhn the critical temperatue currently used forthermoplastic cables, (20

[4252F] vs. 200'G [ 0 'FD. ......... +Gut, the HEATING, Versiop 7.3 (Ref 255) results are necessary forF shawingthatthe effectimve ba-seline scenarios are acceptable when intemreted under~the_ NIUREIG!CR-6850 (Ref 26) guidelines.

HEATING, Vaemion 7.3 (Rtf j 55), was d leloped at the Oak Ridge National Laboratories as a general purpose fiitodifferwnc* heat transfer model f' use8in the commegrdal and goQvemmnFGt nearI-2 induMStries (Ref 255). There are a; number of validaion andverification report and ben@hmark solution cses for general a.plications ofthe HEATING model (e.g., Refs. 640J6*. Averification and validation st4ud for fire related appliciopns, is docum-ent in NI'IMIRIi0 _01-8716 (Ref 20)4.

The model verifica"tion summarized in NRL'MAR6180 -018746 (Ref J20) is based on the mehodoloe developed byWisrfrb-m and Pilsson (Refs. 234_265) for which the sutions of eightfire exposure configurations of increasing complex.t' are provided.

The simple.st involve aG ca2..e..

compariSon hav e.xa..t anal4ic soluAtons whereas the more complex cases against a baselinA heat tra*n*s@r solution genena'ted bythe conduction finite element model TASEF (Ref. A6). The model validation documented by NRL'MPJ6180-4 8746 (Ref 220) is based on eighttst case+ s of inceasing complexi- forw,-hich meas-urd data is available.

Summary Based on Report 0027-0009-014-002. Rev. 0 (Ref J5745)these consderations a*nd those provided for the Fi*reM*ingTretets, Revwisio 0 (Ref 225) report,

,-eneric_

it is concluded that the V&V basis forthe fire modeling tools and calculations contained in the KI Exemption (Ref. J 3) fire hazard analysis report (Refs. J39, J5342) meetthe requirements of NFPA 805, Section 24.1-2.3 (Ref. J242).

Revision 0 Page J-26

Enclosure I to L-2014-109 Page 89 of 106 Florida Power & Lioht Attachment J - Fire Modelingn V&V Table J-i - V & V Basis for Fire Models I Model Correlations Used in the FPRA.

Calculation Application V & V Basis Discussion MainCRAbandonment Calculationofoperator NUREG-1824. Volume I (Ref. Ti7heabandonment timein the Unit l and Unit 2 main control (Units 1 and 2). abandonment times in the Unit J 148) rooms is determined by computing the timne for the visibility 1 and Unit 2 Main Control NUREG-1824, Volume 5 (Ref. and temperature to reach thresholds as specified in Rooms. J126 NUREGICR-6850 (Ref. J610). CFAST, version 6.0.5 has NISTSP 1026 (Ref. J104 been varldated for certain configurations in term s of predicting the temperature increase in an enclosure in NISTSP 1041 (Ref. J 14* accordance with NUREG-1824, Volume 5(Ref. J4123b.

NIST SP 106 (Ref. J 13Z) CFAST, Version 6.0.10 is found to provide a reasonable and NRLJMJR6180--04-8746 (Ref. J20) conserv-tive estimate of both the hot gas layer temperature and visibility as a function of time given the input fire size for a control room like enclosure. This information is documented in Appendix D of the reports entitled

'Evaluation of Unit 1 Control Room Abandonment Times at the St. Lucia Plant (Ref. J44) and 'Evaluation of Unit 2 Control Room Abandonment Times at the St. Lucie Plant (Ref. 4J52.

The MCR abandonment application fals within the non-dimensional parameter space for the NUREG-1824.

Volumes 1 and 5 (Refs. J126, J4148_V&V report as estimated using the methods described in NUREG-1934 (Ref. J 159). The application also falls within the model limits as specified in NIST SP 1026 (Ref. J10) and NIST SP 1041 (Ref. J11-. Additional V&V documentation is provided inNISTSP 1086 (Ref. J13D and NRUMRI61I--04-746 (Ref. J20) that expand the validation parameter space from that included in NUREG-1824. Volumes I and 5 (Refs.

J 126, J414§). including multiple compartment applications.

Generic Fire Modeling Definition of zones of NUREG-1824. Volume I (Ref. Table J-2 provides a summary of the validation basis for the Treatments. Revision0. influence about specific J148_ empirical models used in the Generic Fire Modeling classes of ignition sources. NUREG-1824. Volume 3 (Ref. Treatments', Revision 0 (Ref. 4.

Scenario screening for the J3228 The "Generic Fire Modeling Treatments', Revision 0 (Ref.

multii-compa*rtment analysis. NUREG-1824. Volume 5 (Ref. J253_ report uses CFAST. version 6.0. 10 in a simple 4 12g) geometry that minimizas the boundary heat losses given a volume. Forthe volume postulated, the configuration NISTSP 1026 (Ref. J 10_) produces the most adverse result regardless of the actual NISTSP 1041 (Ref. J 1-1) dimensions used.

NISTSP 106 (Ref. J1373 The application falls within the model limits as specified in Table J-2 NISTSP 1026 (Ref. J1043 and NIST SP 1041 (Ref. J414.

Except forthe global equivalence ratio, the non-dmensional Revision 0 Page J -2 7

Enclosure I to L-2014-109 Page 90 of 106 Florida Power & Li aht Afachment J - Fire Modelinr V&V Table J V & V Basis for Fire Models I Model Correlations Used in the FPRA.

Calculation Application V & V Basis Discussion parameters fall within the V&V s pace of N U REG-1824 Volumes 1 and 5 (Refs. J126. J14g). Although equivalence ratios are considered over a much larger range than addressed by the NUREG-1824. Volume I {Ref. J 14g) validation tests, the results are based on a single time point based on an equivadle*n ratio that is close to unity orlower and thus may falldirectly within the NURE--1824, Volume I (Ref. J 148) validation parameter space.

Additionra validation results that consider the higher predcrtive capability under hi,"er equivalarce ratios are provided inNISTSP 1036 (Ref. J137).

Supplemental Generic Definition of the ZOI about Mangs et al. (Refs. J4133-J435J Table J-2 provides a summary of the validation basis for the Fire Model Closed vented electrical panels. empirica models used to generate the ZOls in the *Generic Electrical Panels Fire Modeling Treatments", Revision 0 (Ref. J253) report.

(Supplement 11). The pred ictive capability of the heatI release rate model is Mangs (Ref. J4420 validated against about thirty fire tests involving a wide NUREGICR-4527, Volume 1 (Ref. range of panel sizes and ventilation configurations. The J46M) model is only capable of predicting the maximuma beat NUREGICR-4527, Volume 2 (Ref. release rate that can be supported within a robustly secured J4.Zn panel as defined in FAQ 08-0042 (Ref. J9.3). Fuels that Iy etal. (Ref. J4537) may exhibit external burning (i.e., internal equivalence ratio greater tlan 1 - 1.3) are addressed by assuming an Table J-2 equivalence rnaio of two. with haOl of the burning occurring inside the panel aid half occurring outside the panel. This applies to PMMA, which is not a common material used in cable construction.

______-au ________ WIEý82 m 5 Ra 12 The - am-mf ,ehdi;divsopd n GiiicFimMoQg R- MR6.M.-.44446 (Rat i2o)

N'~ei rAts.....dsdbyNllCG!CR ... 0 (Rat

______________t "aled fm -w'ide gangs of cabia copoitions 7

hliI~j~

IREGCR 0S.Uot-'o"n I (Rat """Ef3CR 0IO. Wulme 3(Rat '49)

J49)

TkAbleI Revision 0 Page J-28

Enclosure 1 to L-2014-109 Page 91 of 106 Florida Power & Light Att3chment J - Fire Modeling V&V Table J V & V Basis for Fire Models I Model Correlations Used in the FPRA.

Calculation Application V & V Basis Discussion Supplemental Generic Characterization of the heat NUR EG-1824. Volume 1 (Ref. The supplement provides an analysis of the transient fuel Fire Model Treatments: release rate per unit area, fire J 148) package fire tests in order to better characterize the heat Transient Fuel Package duration, and flame height for NUREG-1824, Volume 3 (Ref. release rate per unit zrea, the fire duration, and the flame Ignition Source transient ignition sources. jI height. These parameters are used in the development of CharSpcteristis Provides revised ZOI tables the ZOI in the original "Generic Fire Modeling Treatments',

NUREG-1824, Volume 5 (Ref. Revision 0 (Ref. J252) report and prior to the development based tr-3).on hefel pis of t' of Supplement 3 (Ref. J334) were conservatively bounded.

based on fire ttransient the test arlsis data. of the NISTSP 1026 (Ref. J 10_) Supplement 3 provides the basis fora narrower parameter NISTSP 1041 (Ref. J110 vaue range as determined from the actualfire test reports ISTS 108 (Ref. J132) on which the NUREGCR-6850 (Ref. J6S__ conditional ITabeJ2 .probability distribution was established.

Revised ZOI tables are developed for transient ignition source fuel packages using the results of the fire test data analysis. The ZOls are computed using the same processes as the original 'Generic Fire Modeling Treatments% Revision 0 (Ref. 4j3 and the V&V basis is therefore the same.

Detai.ed Fire Scenario Calculation of the time the hot NUREG-1824, Volume 1 (Ref. Detailed evaluations are provided for specitic ignition Calculations. Revision 0 gas layer reaches critical J 14§) source-secondary combustible configurations involving temperature thresholds for NUREG-1824. Volume 3 (Ref. multiple cable trays.

scenarios (muoplvg secondab jl Two fire modeling tools or calculation methods are used in combustibles (muitiple cabl NUREG-1824, Volume 5 (Ref. this assessment: FLASH-CAT(Ref. J431) and CFAST, trays). Jl126) Version 6.1.1 (Refs. JM._4, JSI). The FLASH-CAT method is used tocompute the temporal heat release rate profiles for NUREGICR-7010, Volume 1 (Ref. specific cable tray arrangements where secondary J491_ combustibles areincluded. CFAST, Version 6.1.1 is used to NISTSP 1026 (Ref. J 104) compute the time the hot gas layer temperature reaches NIST SP 1041 (Ref. J .15 various threshold values given the ignition source and NISTSP 1086 (Ref. J13f) secondary combustible heat release rates.

NRLIMRI61SO--04-746 (Ref. J20) NUREG-1824. Volume 5 (Ref. J 12@6provides the verification and validation basis for CFAST. Supplemental verification is provided in Report 00274053-000-002.

Revision 0 (Ref. J2) and Report 0027-O0-)9-014-005.

Revision 0 fRef. Js) -for the specific CFAST version used.

Verification for the FLASH-CAT method implementation in Microsoft I is provided in Report 0027-0053-000-002.

Revision 0 (Ref. J2) and Report 0027-0009-014-004.

Revisionfi (Ref. J)-.via comparisons with NUREG/CR-Revision 0 Page J-29

Enclosure I to L-2014-109 Page 92 of 106 Florida Power & Licht Attachment J - Fire 10odelina V&V Table J V & V Basis for Fire Models I Model Correlations Used in the FPRA.

Calculation Application V & V Basis Discussion 7010, Volume 1 (Ref. J491) results.

CFAST, Version 6.1.1 is used wthin the NUREG-1824.

Volume I (Ref. J 148_ parameter space for at least one ventilation condition per hot gas layer fire scenario: the most adverseventilation condition is selected for each scenario:

thus, the results are at least as conservative as a case that falls withintheNUREG-1824, Volume I (Ref. J14J8).

vaida-tion space. The FLASH-CAT model uses the recommended input parameters of NUREGICR-7010.

Volume 1 (Ref. J 49_) and is used to calculate the heat release rate in horizontal cable trays containing cables similar to those tested. Therefore, the applcation falls within the valida*Td range for the FLASH-CAT method.

2 Ki Exemption Fire Hazard Basis for cre"ltng raceway IQiRCt-flR-R7jI, _aM.&

p (I The or-elna1 (,IExe*nv*onana'ssa (Ref. J53,42 was Analsis Report separation in the Unit 1 .4 earfcTrmed usýsno mlne-ed t*ha ,redate Vhe ci-*e* NUREG Containment Building. -VRI LUMII ¶'1-46 P-1; 1-1. Wtdance. t*r showin FnRerel G0V27-CVO9-014-C02, Rev. 0 (Ref. J1547U that ihe cffital ass'rsmr:,zcTr rad to a ccnservt!.ve resuts reEntive the current NUREG Qoufdance (Re 4Z0+witcits tire v(-nrrse for va"~datonm. The ca~crilatfn~resu-ls R-QtbS-~Z fcr thebass~ne scenaifes are veuffzndtirr~cuujh aTerale (R.-.l.*-16 cacuafats as doc*n*neted EnRewTe CQr27-C,2.-! t4--(42, 4

c 1R -P= 0 (Ref. J554Rev.

nq"" - _ '-- y ,e" ,'-U."y J62) i9 01 .%aXf E c.= - (Re-13) R -pc 4 6t312 (

IZab4-J-ReLven 03274103-014- '53wx-=q;;" 7W spp cdhý exampgaman Rev. 0 (Ref. J5Al) _.tc.j ,_ mH5_-.3 . "he =-,"_' "m_, a yýog 627- fRef 153 - rcaapazatZZ

-grf-X~a ma&"7)Aý

- -h . . Q. . . ., . .. . . . . . . . . . .

...- ."d- . ... =4a . .

{~f9-n"-m p4~ W~th~h~nn me.

Revision 0 PagfeJ-30

Enclosure I to L-2014-109 Page 93 of 106 Florida Power & Light Attachimernt J - Fire Modelinia V&V Table J V & V Basis for Fire Models I Model Correlations Used in the FPRA.

Calculation Application V &V Basis Discussion 6, -;;ch. fc-hs

-- 1 125) 5a mod-' F-mat Rat j53) am spo the V.=m;mI.:,dsd zw'uss follhai" ASIA-mficogme Q4"I- c",- -xg- ý- Cmp,=d

Ca ;p,,f-I Hr

AllIG 73 155)

=d 7z,',;dia ma-z"I pzcpart7ra %WZ-h

,wmFA--:Uxa -ad"-,' I , . I %-,

-:y zah tým-

-RI rDIR -320--Ciiuilra (Rfij j-,,ý pum,;" ,

aý vz:d-ýQm -rasmw- c4 HEAM-110. Nl"-;Qn 7 3 Mal 155) to E-A 163 165) liza -'Ga-ed ýaAWAZARZW-04 P246 (Rat -ad qQQMfita&&4h=

GA Q'dfiX MA"F-25 jw) zacl,,ZQO Rap= YjC,-DmAzt (RQf 163) -Ad Revision 0 PageJ-31

Enclosure I to L-2014-109 Page 94 of 106 Florida Porter & Light Attachment J - Fire M~todeling V&V Table J V &V Basis for Fire Models I Model Correlations Used: "Generic Fire Modeling Treatments". Revision 0 (Ref. J253) Correlations.

Location in Ref. Original Subsequent Correlation 2atn Reference Application Original Correlation Range Validation and Limits in Ref. 253 Verification Flame Height Page 18 Heskestad Provides a limit Directly (Ref. J2Z7); on the use of the If NUREG-1824, Heskestad Zone of -E< log " Volume3 (Ref. affI (Ref. J2&5) Influence  :~,')J32M~S 30 In practice. wood arid hydrocarbon NUREG-1824.

fuels, momentum or buoyancy Volume 5 (Ref.

dominated, with diameters between J 12b-0.05- 10 m (0.16 - 33 ft). (Correlation used mCFAS')

Point Source Page 1 mes Lateral extent of NREG 2 Prdted heat flux at Model (Ref. JZ40 ZOI- target is Less than 5 omparisonVto Isotropic flame radiation. Vu (Ref. kWm (0.4-4 Btus-ft-) peT Compared J32)j: SFPE (Ref. J34J.

othermethods withdata for 0.37 m (1.2 fi) diameter PMMA pool fire an a target located SFPE (Ref.

ata ratio of 10.

Method of Page 19 § et a]. LateraJ extent of Pool aspect ratio less than 2.5. SFPE (Ref. Ground based vertical

§Wo and (Ref. J6_47.) ZOI- Hydrocarbon fuel in pools with a J340) target.

Beyler comparison to diameter between I - 30 m (3.3 - 98 NUREG-1824.

othermethods ft). Volume 3 (Ref.

Vertical target, ground level. J32pj Method of Page20 M Lateral extert of Round pools; SFPE (Ref. Totalenergy emitted by MM (and (Ref. J634D ZOI- Hydrocarbonfuel in pools witha J340J thermalrad.-tion less Croce) comparison to diameter between 0.5 - 80 m (1.64 - than total heat released.

other methods 262 ft).

Method of Page 20 et a]. Lateral extent of Round pools; SFPE (Ref. Predicted heat flux at gmand (Ref. J6847D ZOI Hydrocarbon fuel in pools with a J340 target is greater than 5 Beyler dmelete between 1 - 50 m (3.3 - kWlm t (0.44 Btns-ft-) per 164 ft). NUREG-1824. SFPE (Ref. J340O.

Volume 3 (Ref. Shown to produce most J3220 conservative heat flux over range of scenarios considered among all methods considered.

Revision 0 Page ,J-32

Enclosure I to L-2014-109 Page 95 of 106 Florida Power & Light Attchment J - Fire Modeling V&V Table J V & V Basis for Fire Models / Model Correlations Used: "Generic Fire Modeling Treatments". Revision 0 (Ref. J25) Correlations.

Subsequent Correlation Location in Ref. Original Application Original Correlation Range Validation and Verification Limits in Ref. 253 -

J2im_ Reference Plume heat Page 22 Wakamatsu Vertical extent of Fires with an aspect ratio of about 1 Wakamatsu et Area source fires with fluxes etal. ZOI randhaving a plan area lessthanI l. forlarge fires aspect ratio- 1. Used (Ref. J.O.) m=(0.09f-'. TI..JZ4IJ with plume centerline SFPE Handbook temperature correlation; of Fire Protection most severe of the two is Engineering, used as basis for the ZOI Section 2-14 dimension. This is not a (Ref. JZtQ*J constraint in the fire model analysis for the cases evaluated.

Plume Page23 Yokoi Verticalextent of Alcohollamp assumed to effectively NUREC-1824, Area source fires with centerline (Ref. ZOI be a fire with a diameter -0.1 m (0.33 Volume 3 (Ref. aspect ratio - 1. Used temperature J32; ft). J322n): with plume flux Beyler SFPE Handbook correlation; most severe (Ref. J5Oj of Fire Protection of the two is used as Engineering. basis forthe ZOI Section 2-1 dImension.

(Ref. J2,j Hydrocarbon Page 51 SFPE Determine heat Hydrocarbon spill fires on concrete None. Based on None. Transition from spill fire size Handbook release rale for surfaces ranging from - to -10 m limited number of unconfined spill fire to of Fire unconfimed (3.3 - 33 ft) in diameter. observations, deep pool burning Protection hydrocarbon spill assumed to be abrupt.

Engineering, fires.

Section 2-15 (Ref.

J*z2 Flame Page 100 SFPE Determine the Comer fires ranging from -10 to None. Eased on None. Offset is assumed extension Handbook fire offset for -11.000 kW (9.5 - 348 Btuls). Fires limited number of equal to the depth of the of Fire open panel fires. included gas burners and observations. ceiling jet from the Protection hydrocarbon pans. experiments.

Engineering, Section 2-14 (Ref.

_Z3)

Revision 0 Pag~e J-33

Enclosure I to L-2014-109 Page 96 of 106 Florida Power & Light Attachment J - Fire Modeling V&V Table J V & V Basis for Fire Models/ Model Correlation s Used: "Generic Fire Modeling Treatments", Revision 0 (Ref. Jl2,) Correlations.

Subsequent Correlation Location in Ref. Original Application Original Correlation Range Validation and Verification Limits in Ref. 253 aJ253 Reference Line source Page 101 RDetermine the Theoretical development. SFPE Handbook None. Transition to area flame height (Ref. J254) vertical extent of of Fire Protection source assumed for the ZOI Engineering, aspect plan ratios less Sectior 2-14 than four. Maximum of (Ref. JZ...D area and fine source predictions used in thtis region.

Corner flame Page 103 SFPE Determine the Comerfires ranging from -10 to None. None.

height Handbook vertical extent of -1.000 kW(9.5- 948 Etads). Fires Correlation form of Fire the ZOI included gas burners and is consistent with Protection hydrocarbon pans. other methods; Engnering.

Section 2-14 (Ref. J53) the bsis.

Air mass flow Page 140 Kawagos Compare Small scale., % scale, and fullscale pjyft(Ref. None. SFPE (Ref. JZhQ through (Ref. J-55 mechanical single rooms with concrete and steel JiZ50; spaces with a wide range opening ventilation and boundaries. Vent sizes and thus SFPE (Ref. of opening factors.

natural opening factor varied. Wood crib .ZS&J) ventilation fuels.

Line fire flame Page 210 Yuan et a]. Provides a limit None. None.

height (Ref. J3128) on the use of the z Correlation form Zoneof 0.002 < <0.6 is consistentwith Influence (ZOI); other methods; Extent of ZOI for comparison to cable tray fires. In practice, from the base to several cl-taset from times the flame height based on Yuan at al. (Ref.

0.015- 0.05 m (0.05 - 0.16 ft) wide J21t0 provides gas burners. basis.

Revision 0 Page J-34

Enclosure I to L-2014-109 Page 97 of 106 Florida Powver & Liguht Attachment J - Fire Modeling V&V Table J V & V Basis for Fire Models/ Model Correlations Used: "'Generic Fire Modeling Treatments". Revision 0 {Ref. J=_) Correlations.

Subsequent Correlation 2in f Original Application Original Correlation Range Validation and Limits in Ref. 253 or n253 Reference Verification Cable heat Page 210 N BSIR 85- Provides Cables with heat release rates per None. Correlation preicts a release rate 3196 assurance that unit area ranging from about 100 - lowerheat.release rate per unit area (Ref. J58) the method used 1,000 kW/mr (8.8 - 8& Bturs-ft). than as sumed in the is bounding Treatments and is based on test data.

Line fire plume Page 212 Yuan et al. Provides a limit None. None.

centerline (Ref. J3428) on the use of the z Correlation form temperature Zone of 0.002 < T- < 0.6 is consistent with Influence (ZOI); other methods; Extent of ZOI for to seral comparison to cable tray fires. In practice, from t. . base dataset from times the flame height based on 0.015- 0.05 rm (0.05 - 0.16 ft) wide Juanet a]. (Ref.

provides gas bbasis.,)

Ventilation Page 283

• Assessing the SFPE (Ref. None. Provides depth in limited fire size (Ref. JUQý) significance of Ventilation factors between 0.06 - JZhý7) the analysis of the vent position on 7.51. selected vent positions.

the hot gas layer The global equivalence temperature Fire sizes between 11 - 2.800 kW ratio provides an alternate (10- 2.654 Btufs) measure of the applicability of the Wood, plastic, and natural gas fuels. analysis and for reported output is within the validation range of CFAST.

Revision 0 Page J-35

Enclosure I to L-2014-109 Page 98 of 106 Florida Power & Light Company Attachment J -Fire Modeling V&V References ii. Report 04'930G0006 10,ilSt. Luci Units 1 and Fire Probabl* istDicisk A.SSesR .smentFirs S ReNpo nR.na.o RE*

IC R- 850 Tas ks an d 11," Revisio n ,

ERIN Engineering, Walnut Creek; CA, 2013.

J1. Report 0027-0009-014-001 'Evaluation of Unit 1 Control Room Abandonment Times at the St. Lucie Plant.' Revision 0. Hughes Associates. Inc., Baltimore, MD.

2014.

J2. Report 0027-0009-014-003, 'Evaluation of Unit 2 Control Room Abandonment Times at the St. Lucie Plant." Revision 0, Hughes Associates, Inc., Baltimore, MD, 2014.

J3. Hughes Associates, Generic Fire ModelingTreatments." Proiect Number 1SPH02902.030, Revision 0. January 15, 2008.

J4. Report 0027-0053-000-003, "Supplemental Generic Fire Modeling Treatments:

Transient Fuel Package Ignition Source Characteristics,' Revision 0. Hughes Associates, Inc., Charlotte, NC, February, 2013.

J5. Hughes Associates. "Supplemental Generic Fire Model Treatments: Closed Electrical Panels." Revision B, Baltimore, MD. October 11. 2011.

JjReport 0027-0053-000-002, 'Evaluation of Development andTiming of Hot Gas Layer Conditions in Selected PSL Fire Zones," Revision 0, Hughes Associates, Inc., Charlotte, NC, February, 2013.

J7. Report 0027-0009-014-004. "Combined Ignition Source - Cable Tray Fire Scenario ZOls for St Lucie Nuclear Power Plant Applications," Revision 0, Hughes Associates. Inc., Baltimore, MD, 2014.

J 8. R.O 02.7-00.09-01.4.0059. 'gvaluationofte.Dev.elopment, andTlirming ofHot Cp.mb*u~stbles." .Reyisi.pn..0,.du~ghefs.As~sociate~. 1nc*.Ealtim.pre, MD..201.!.

J9j ML021410012, "St. Lucie Unit 1 Docket No. 50-335 FPL Response to Request for Additional Information for 10 CFR 50 Appendix R K1 Exemption Clarification/Request," Florida Power and Light Company, Jensen Beach, FL, May 15, 2002.

j4. Hughes Associates, 'Evaluation ofUIt Control Room Abandonment~imes at the St. Lucie Plant" Revision 0, Hughes Assodates, Inc., alti**ore, MD, RevisionS Page J-36

Enclosure Ito L-2014-109 Page 99 of 106 Florida Power & Light Company Affachmeril J -Fire Modelirtg V&V jS. Hughes Associates, "Evaluation of Unit 2 Control Room AbandonmentTmes 2t the StLI wcai IPa*,* R8,i*on 0, Hughes Ass"o'tas, Inc., Baltimoe, M*.

Nov"emb"r 6,_2W09 1" NUREGICR-6850, -EPRIINRC-RES Fire PRA Methodology for Nuclear Power Facilities Volume 2 Detailed Methodology," EPRI 1008239 Final Report, NUREG/CR-68501EPRI 1023259, Nuclear Regulatory Commission, Rockville, MD, September, 2005.

I j NISTIR 4833, "A Sprinkler Fire Suppression Algorithm for the GSA Engineering Fire Assessment System, , Daniel and Y Robert, National Institute of Standards and Technology Interim Report, Department of Commerce, National Institute of Standards and Technology," Gaithersburg, MD, 1992.1 j 1 SFPE Handbook of Fire Protection Engineering, Section 3-1, "Heat Release Rates," §, V., The SFPEHandbookof Fire ProlectionEngineering,4m Edition, P. J. DiNenno, Editor-in-Chief, National Fire Protection Association, Quincy, MA, 2008.

I 1 NUREG/CR-6850, Supplement 1, "Fire Probabilistic Risk Assessment Methods Enhancements," EPRI 1019259, Technical Report, NUREG/CR-6850 Supplement 1, Nuclear Regulatory Commission, Rockville, MD, September, 2010.

I P,. NIST SP 1026, "CFAST- Consolidated Model of Fire Growth and Smoke Transport (Version 6) Technical Reference Guide," Jones, W_ W., Peacock, R. D.,

Fomey, G. P., and e P. A., National Institute of Standards and Technology, Gaithersburg, MD, April, 2009.

I NIST SP 1041, "CFAST- Consolidated Model of Fire Growth and Smoke Transport (Version 6) User's Guide," Peacock, R. D., Jones, W. W., Fe.Df P. A.,

and Fomey. G. P., National Institute of Standards andTechnology, Gaithersburg, MD, December, 2008.

I NUREG-1824, Volume 5, "Verification and Validation of Selected Fire Models for Nuclear Power Plant Applications Volume 5: Consolidated Fire Growth and Transport Model." NUREG-1824/EPR1 1011999, Salley, M. H. and Kassawara, R.

P., NUREG-1824, Final Report, U.S. Nuclear Regulatory Commission, Office of Nuclear Reactor Regulation, Washington, D. C., May, 2007.

I NIST SP 1086, "CFAST- Consolidated Model of Fire Growth and Smoke Transport (Version 6) Software Development and Model Evaluation Guide,"

Peacock, R_ D., McGrattan, K., Klein, B., Jones, W. W., and fe)1fJe P. A.,

National Institute of Standards and Technology, Gaithersburg, MD, December, 2008.

Page J.37 Revision 0 Page J-37

Enclosure 1 to L-2014-109 Page 100 of 106 Florida Power & Light Company Attachment J -Fire Modeling V&V J41 L NUREG-1824, Volume 1, "Verification and Validation of Selected Fire Models for Nuclear Power Plant Applications Volume 1: Main Report." NUREG-1824IEPRI 1011999, Salley, M. H. and Kassawara, R. P., NUREG-1824, Final Report, U.S.

Nuclear Regulatory Commission, Office of Nuclear Reactor Regulation, Washington, D.C., May. 2007.

I NUREG-1934, "NuclearPower Plant Fire Modeling Application Guide," Salley, M.

H. and Kassawara, R. P., NUREG-1934dEPRI-1019195, U.S. Nuclear Regulatory Commission, Office of Nuclear Reactor Research, Washington, D. C., November, 2012.

J1. DBRD HV.AC 1, "Safety Related .4VAC y'Wems;" Design Basis Doc-umen Revision 2, Florda Paxer & Light Co~mpany, St- Lucie Plant, Jeansen Beach, Florida, December, 2005.

j17. Drysdale, D. D.,FiFlynargng, 21 Edition, John XMay and Sons, New York, N*w York, 1999.l JIB. Drawing 87700G862, 'HVAC- Air Flc~itDiagr:am,"Revision 34, Flor1idaPRgNeir&

Light Cormpan, SLc Plant Jensen Beachr.h Florida.

J119-. AP 11800023, 'Unit 1 Fire Fighting Strategies,"- Revisioni 29A, Florida Pc~er&

Light Company, St. Lucia@ Plant, Jensen; Beach, FloridaApril 3,2008.

J20. NRLJMR/6180-04-8746, "Verification and Validation Final Report fbr Fire and Smoke Spread Modeling and Simulation Support of T-AKE Test and Evaluation,"

Tatem, P.A., Budnick, E.K., Hunt, S.P.,Trelles, J., Scheffey, J.L., White, D.A.,

Bailey, J., Hoover, J., Williams, F.W., Naval Research Laboratory, Washington, D.C., 2004.

J21. ASTM E1355, "ASTM Standard Guide for Evaluating the Predictive Capability of Deterministic Fire Models," American Society forTesting and Materials, West Conshohocken, PA, 2004.

QJ2Q_.\DAD-_*" -Safety Related HC Sy,*ms," Revision 2J en.sen Bacrh, Filorida, December, 2005.

j23. Drawing 2998-0 862, "HVAC -Ar Fl Diagram," Revision 30, St. LuciaNuc.e.- r Staton, Unit?2, Jensen Beach, Flor:ida, July 31, 2008.

4i4, FP  :"I-12PH02902.030, Revision 0, Januay' 15, 2008.

RevisionP Page J-38

Enclosure I to L-2014-109 Page 101 of 106 Florida Power & Light Company Attachment J - Fire Modeling V&V JZQ NUREG-1805, "Fire DynamicsTools (FDI',"Lqlk8L N. and Salley, MR H_,

NUREG-1805, Final Report, U.S. Nuclear Regulatory Commission, Office of Nuclear Reactor Regulation, Washington, D. C., October, 2004.

J4 Heskestad, G., "Peak Gas Velocities and Flame Heights of Buoyancy-Controlled Turbulent Diffusion Flames," EighteenthSymposium on Combustion, The Combustion Institute, Pittsburg, PA. pp. 951-960, 1981.

I 2o t %A G-, 'Engineering Relations for Fire Plumes," FireSafety Journal,7:25-32, 1984.

I J262-. SFPE Handbook of Fire Protection Engineering, Section 2-1, "Fire Plumes, Flame Height, and Air Entrainment," Heskestad, G., The SFPEHandbook of Fire ProtectionEngineering,4" Edition, P. J_ DiNenno, Editor-in-Chief, National Fire Protection Association, Quincy, MA, 2008.

I JA2 Yokoi, S., "Study on the Prevention of Fire Spread Caused by Hot Upward Current," Report Number 34, Building Research Institute, Tokyo, Japan, 1960.

IJ Yuan, L. and Cox, F.,"An Experimental Study of Some Line Fires," Fire Safety Journal,27, 1996.

I1;2& NUREG-1824, Volume 3, 'Verification &Validation of Selected Fire Models for Nuclear Power Plant Applications, Volume 3: Fire DynamicsTools (FDTS)," NUREG-1824/EPR1 1011999, Salley, M. H. and Kassawara, R. P.,

NUREG-1824, Final Report, U.S. Nuclear Regulatory Commission, Office of Nuclear Reactor Regulation, Washington, D. C., May, 2007.

I33. Report 0027 -0053-000003, "Supplmental Generic Fire Modeling Treatmants-Transi;nt IFuel Package Ignition Sou.rce Characteristics," Revision 0, Hughes A)4Q SFPE, 'The SFPE Engineering Guide forAssessing Flame Radiation to External Targets from Pool Fires," Society ofFire Protection Engineers, National Fire Protection Association, Quincy, MA, June, 1999.

I 3_5 SFPE Handbook of Fire Protection Engineering, Section 2-5, 'Effect of Combustion Conditions on Species Production," Gottuk, D. T. and , B. Y.,

The SFPE Handbook of FireProtectionEngineering,4Th Edition, P. J. DiNenno, Editor-in-Chief, National Fire Protection Association, Quincy, MA, 2008.

J3& SFPE Handbook of Fire Protection Engineering, Section 3-4, "Generation of Heat and Chemical Compounds in Fires," Thkn. A., The SFPEHandbookof Fire ProtectionEngineering,4 1 Edition, P. J. DiNenno, Editor-in-Chief, National Fire Protection Association, Quincy, MA, 2008.

Page J-39 Revision 0U Page J-393

Enclosure Ito L-2014-109 Page 102 of 106 Florida Power & Light Company Attachment J - Fire Modeling V&V j37- Hughes Asociaes, 'Supp@lemental Gepnric Fire Model Treatments: Cloed Electrinal P.anel Revision B, Baltimore, MD, October 11, 2014-1r j38. Hughes Associates, t Supplemental Generic Fire Model Treatments: Hot Gas Laye@

Tables.' Revision H,Baltimore, MD, August 10, 2012.

j39. Hughes Associates, t Supplsernntal Gene ric Fire- MAod-el TFtrea ~tmets:Transient Target Response to-Transient Ignition Source Fire Exposures," Revision A Baltimore. MD, October 13, 2011.

jI4. Report 1JI\A21020.00 02, Revisio;n0, 'Supplemontal Generic Fire M odeling Tratments:- olid S*t*ate Conpol Compo.entZ.70l andI Hot Gas LayerTables,"

Hughes Associate... .leinsorg Group Risk Servis, Baltimore, MID, Sep.tembe I l Mangs, J. and K&eaQý J3. 0., 'Full Scale Fire Experiments on Electronic Cabinets Ilt VTT Publications 269, Technical Research Centre of Finland, Espoo, 1996.

L42& Mangs, J., F J., and gk , 0., "Calorimetric Fire Experiments on Electronic Cabinets,t Fire Safety Journal,38, pp. 165- 186, 2003.

1 JQ, &1S., I L., Such, J.MgjC.3,.Q j Qa.

C.,JMgq of Electrical Cabinet Fires Based on the CARMELA Experimental Program," 4MpjgA Forum, 2004.

A444 NUREGICR-4527, Volume 1, "An Experimental Investigation of Intemally Ignited Fires in Nuclear Power Plant Control Cabinets Part I: Cabinet Effects ests,t NUREGICR-4527/ SAND86-0336, Volume 1, Chavez, J. M., Nuclear Regulatory Commission, Washington, DC, 1987.

I44Q NUREG/CR-4527, Volume 2, "An Experimental Investigation of Intemally Ignited Fires in Nuclear Power Plant Control Cabinets Part IH: Room EffectsTests,:

NUREGICR-4527 / SAND86-0336, Volume 2, Chavez, J. M. and Nowlen, S. P.,

Nuclear Regulatory Commission, Washington, DC, November, 1988.

RevisionP Page J-40

Enclosure I to L-2014-109 Page 103 of 106 Florida Power & Light Companv Affachment J -Fire Modelinr V&V I J4 Hunt, S. P_, 'Maximum Fire Size in Closed Vented Electrical Panels," Fire ProtectionInformation Forum, September, 2009.

I d NUREG/CR-7010, "Cable Heat Release, Ignition, and Spread in Tray Installations Qqopjg Fire (CHRISTIFIRE) Volume 1: Horizontal Trays," Final Report, McGrattan, K., Office of Nuclear Regulatory Research, Nuclear Regulatory Commission, Washington, DC, July, 2012.

J5O. Beyler-, C ., ' Fire Plumnes anpd- CeoilingJets,"-Fife Safelyigumal, Volume 11, n Number 1, Elsevier Sequoia S.A-, Lausanne~, Switzedand, 1986.

J51. Th.mas, P. H., Hinkley, P. L., Theobald, C-R, and Sims, D. L,"Fire Technical Paper No. 7," H. M. Stgonery Ofc, Joint Fire Research Organization, London, JS12NiST.G CR-90-580, "Develop. TOIGA f an Instretiona Program fcPrctdng rngine~ers riazaro users,- ujampeu, J.rX. ano 9Yler, 6. L., General Lontractor Report (GCR), Nainl Istitute of Standards and Techn;olow, GaithersburgMD, huly, 113Cli' I 4J Report 6372, "Fire Hazard Assessment of Exposure to Safe Shutdown Raceways, St. Lucie Unit 1,"Final Report, Hughes Associates, Inc., Baltimore, MD, March 28, 2002.

i5I4 Report 04n920Gn0n1506.102 ,"St. Lucie Units 1 and -2Fire Proabili-st-ic Risk

.Assssment Humn Failr-,ir Evaluation Repor", Revision 1, ERIN Enqge*-n*g,,

.a.lnut Cree, C'A_.., Feb.mar ,', 2013.

J,4Qj Technical Report PSR-199, "HEATING 7: Multidimensional, Finite-Difference Heat Conduction Analysis Code System," Childs, K. W., Oak Ridge National Laboratory (ORNL), Oak Ridge, TN, 1998.

d,540, EPRI, "FIVE Fire Induced Vulnerability Evaluation Methodology," Electric Power Research Institute, Palo Alto, CA, 1991.

J*..5.. Report 0027-0009-014-002, 'Verification and Validation Basis forthe 10 CFR 50 Appendix R K1 Exemption ClarificationlRequest Fire Modeling Report."

Revision 0 HuinhesAssnriates. Inc_ Baltimore_ MD February~ 2014.

2014.

Revision 0 Hunhes Associates Inc Baltimore MD Februarv J 7 PSL-' I. FPER-01_-025, "Response to GL 2006n03 Potentially Nlonconpforring Hemy.c and MT Fira Ba;er Configuraons," Fire Protection Evalution RepPrt (FPER), St.

LuciePlý2antUit 1 and Unit 2, FloiridaPR mvr and Light GCompany, Jensen Beach, FL,eceber14,2006.

Revision U Page J-41

Enclosure I to L-2014-109 Page 104 of 106 Florida Power & Light Company Attachment J -Fire Modeling V&V 158. NRSIR 85-3196, 'Heat Ralea-s Rate Chapracerstics of SnmeP Combustible Fuel Sources in Nulo-*roerRlants" Lee, B.T., Nati**nl Bureau of Standards; (NBS), Gaithersburg, MID, 1985.

J5.SFPE HanAdboo-k -ofFire Protection Engineering, Section 3-10, 'Fire Hazr Calcul-Wations for Large, Open Hydrocarbon Fires," B*y!er-, C L., ThSFE Handbook of Fire Protection Erngineofln, 4' Edition, P. j. DiNenno, Editor-in-Chief, National Fire Proecion Assodaton, Quincy, MA, 2008.

J60. Technical Report. K'CSD/!NF-8911, t HEA.TCHEK: A Computer Pmgr...mto Automate\'Nri6ficatinr ofNewolersionn of HFATING," Chu, II.,Uni;o Carbide Corp, Nuc.lea~r Di. C-aJeou lDiffusion Plant, Oak Ridge, Tennessee, 1989.

161. Trchnicr-al Report K/CSD'T*Mr61, HEA.TNG6 Verification," Bt'a, C. B., Childs, K.

'A.,and G"ies, G.E., Oak Ridge National Laboaraty, Oak Ridge, Tennessee,

_16)2 Tec-hnics! Reoert ORN~b.NLIRE=G/CSD-2./0/D3, 'HEI*ATING";-A Copu....er Program.

for Mu-tidimensional Heat Tra.nsfrrAnasis eVersion 6.1), Sect. FlO of SCARLE: A Modul!ar Code System fcr Pe.do~rming Sta*ndardized Computer Analysis fcr Icensing' Evaluaion," Childs, K.W., Giles, G."E.,Bryan, C. B.,Iand Cobb, C. K, Marinmp Marietta EnerOY Systems, Inc;., Oak Ridge National Labomtor', Oak Ridge, J Wffic*-strbmm, U., "An Evaluation Srche*m*e f Computer Code forCalull-ting Temperature in Fire Exposed Structures," in S. Gpayson (Ed.), !n eR am '99 (pp,.

1033 1011), Londo-n lntIn@rsdenc CommunicaPtions Ltd., 1999.

J1614. Wicstrnm-,I,U., and Pilsson, ., 'A SGcama fo-th- Vericatopn of Computer Codes for CalculatingTemperature in Fir; Exposed Structures," SP Report 1999:36 Bor:is, Sweden: SP Swedish NationalTesting d Research lnst~ute, 1999.

165. PRlsson, 1-1and Wui-kstrbm, U.,"A Sc*heame forVerifica-tion of Computer Codes f*r CalclatIng* Temperature in Fire Exposed... t u....e !nteematkna!i1

"*, *.A.*.p on StR."ct'oras in Fi.res. Cooenhaeon. DenmPaA.". June. 2000.

166. Sterne~r, E., and Xhickstrbm, U. qTA 5FF Temps PR-ture Anqalysis, of StrudtureE6

~~.,.1.

Exposed to Fire," SP Report 1990:05, SP Swedh Institute, Boris, SwedAn, 1990.-

4§, Mp , A. T., 'Thermal Radiation from Pool Fires." Technical Report 22361-5 RC-B-67, Factory Mutual Research, Norwood, MA, August 1976.

.Inrnnl nf Firm PrnilrfinnFnningnpnn V/nI 1 Kn A nn 1AI-1 rfl 1QRQ

Enclosure I to L-2014-109 Page 105 of 106 Florida Power & Light Company Affachment J - Fire Modeling V&V I J489. Mu!J,,K S . Them. ia dOPl E,,M Energy Combustion Sci., Vol. 10, 1984.

1 JI Wakamatsu, T., ag§Xji Y., Kqgiy% K, and Kikv, D., 'Heating Mechanism of Unprotected Steel Beam Installed Beneath Ceiling and Exposed to a Localized Fire: Verification Using the Real-scale Experiment and Effects of the Smoke Layer," Proceedingsof the Seventh InternationalSymposium on Fire Safety Science, Intemational Association for Fire Safety Science, London, UK, 2003.

J4 SFPE Handbook of Fire Protection Engineering, Section 2-14, 'Heat Fluxes from Fires to Surfaces," L.Oep B. Y-, The SFPE HandbookofFire ProtectionEngineering,4 0 Edition, P. J. DiNenno, Editor-in-Chief, National Fire Protection Association, Quincy, MA, 2008.

J 51. Beyler, C. L. 'Fire Plumes and Ceiling Jets," Fire Safel'Joumal, Volume 11.

Number 1, Elsevier Sequoia S. A., Lausanne, Switzerland, 1986.

I Fu SFPE Handbook of Fire Protection Engineering, Section 2-15, "Liquid Fuel Fires," Gottuk, D. and White, D., The SFPE Handbookof Fire Protection Engineering,3Yl Edition, P. J. DiNenno, Editor-in-Chief, National Fire Protection Association, Quincy, MA, 2002.

j4Dj SFPE Handbook of Fire Protection Engineering, Section 2-14, 'Heat Fluxes from Fires to Surfaces," L&Ojro B. Y., The SFPE Handbookof Fire ProtectionEngineering,310 Edition, P. J. DiNenno, Editor-in-Chief, National Fire Protection Association, Quincy, MA, 2002.

5 Kawagoe, K, =Fire P.*f ,in Rooms," Report Number 27, Building Research Institute, Tokyo, Japan, 1958.

J56. DQysd**g.. _P_ FeD-vnmia.2*

...M EZditio*m.

.ph Wiley an~dSons.Nmew..ork.New j SFPE, "Fire Exposures to Structural Elements," Engineering Guide, National Fire Protection Association, Quincy, MA, May, 2004.

Revision 0 Page J-43

Enclosure 1 to L-20 14-109 Page 106 of 106 Florida Power & Light Company Atlachment J - Fire Modeling V&V

_Sqources in.Nuc~. rPvoe~r Pan1ts.* ee.T.- Natiponal Bureau of S§tandards (NBS)..

a ithre4?.ro. MD.1 ...

JP749h'Md. Fjrq TegjnV.'lpg*

RevisionO Page J-44