View source for 3F0907-03, License Amendment Request 295, Revision 0: Extension of Allowed Outage Time to Seven Days and Elimination of Second Completion Times Response to Request for Additional Information

{{Adams
| number = ML072490420
| issue date = 09/04/2007
| title = License Amendment Request #295, Revision 0: Extension of Allowed Outage Time to Seven Days and Elimination of Second Completion Times Response to Request for Additional Information
| author name = Franke J
| author affiliation = Progress Energy Florida, Inc
| addressee name = 
| addressee affiliation = NRC/Document Control Desk, NRC/NRR
| docket = 05000302
| license number = DPR-072
| contact person = 
| case reference number = 3F0907-03, TAC MD5241
| document type = Letter type:
| page count = 21
| project = TAC:MD5241
| stage = Response to RAI
}}

=Text=
{{#Wiki_filter:Progress Energy Crystal River Nuclear Plant Docket No. 50-302 Operating License No. DPR-72 Ref: 10 CFR 50.90 September 4, 2007 3F0907-03 U.S. Nuclear Regulatory Commission Attn: Document Control Desk Washington, DC 20555-0001

==Subject:==
Crystal River Unit 3 - License Amendment Request #295, Revision 0: Extension of Allowed Outage Time to Seven Days and Elimination of Second Completion Times Response to Request for Additional Information (TAC No. M1D5241)

==References:==
: 1. NRC to Crystal River Unit 3 Letter dated July 18, 2007, "Crystal River Unit No. 3, Request for Additional Information Regarding License Amendment Request No. 295 Rev. 0, Extension of Allowed Outage Time to Seven Days and Elimination of Second Completion Times (TAC No. MD5241)
: 2. Crystal River Unit 3 to NRC Letter dated April 13, 2007, "Crystal River Unit License Amendment Request #295, Revision 0, Extension of Allowed Outage Time to Seven Days and Elimination of Second Completion Times Limiting Time"

==Dear Sir:==

On July 18, 2007, the Nuclear Regulatory Commission (NRC) issued a Request for Additional Information (RAI, Reference 1) regarding License Amendment Request (LAR) #295 (Reference 2). In accordance with the provisions of 10 CFR 50.90, Florida Power Corporation (FPC), doing business as Progress Energy Florida, Inc., hereby provides the response to the RAI.
This letter establishes no new regulatory commitments.
If you have any questions regarding this submittal, please contact Mr. Paul Infanger, Supervisor, Licensing and Regulatory Programs at (352) 563-4796.
,Y.iA. Franke Director Site Operations Crystal River Nuclear Plant JAF/seb/dar Attachments: A. Response to Request for Additional Information B. List of Regulatory Commitments xc:        NRR Project Manager Regional Administrator, Region II Senior Resident Inspector A, 00 1 Progress Energy Florida, Inc.
Crystal River Nuclear Plant 15760 W. Powerline Street Crystal River, FL 34428

U.S. Nuclear Regulatory Commission 3F0907-03                                                                                   Page 2 of 2 STATE OF FLORIDA COUNTY OF CITRUS Jon A. Franke states that he is the Director Site Operations, Crystal River Nuclear Plant for Florida Power Corporation, doing business as Progress Energy Florida, Inc.; that he is authorized on the part of said company to sign and file with the Nuclear Regulatory Commission the information attached hereto; and that all such statements made and matters set forth therein are true and correct to the best of his knowledge, information, and belief.
                                               /0oA.       Franke Director Site Operations Crystal River Nuclear Plant The foregoing document was acknowledged before me this -I "t               day of 1                  ,'2007, by Jon A. Franke.
Signature of Notary Publi        .. .MY        .DPPOW# DD 408.539 COMMISSION State of Florida                           EXPIRES:Jy8,29 I-     .. 5AW.
dedThNotaiyPubric L w (Print, type, or stamp Commissioned Name of Notary Public)
Personally    7            Produced Known               -OR-  Identification

PROGRESS ENERGY FLORIDA, INC.
CRYSTAL RIVER UNIT 3 DOCKET NUMBER 50-302 / LICENSE NUMBER DPR-72 LICENSE AMENDMENT REQUEST #295, REVISION 0 EXTENSION OF ALLOWED OUTAGE TIME TO SEVEN DAYS AND ELIMINATION OF SECOND COMPLETION TIMES ATTACHMENT A RESPONSE TO REQUEST FOR ADDITIONAL INFORMATION

U.S. Nuclear Regulatory Commission                                                    Attachment A 3F0907-03                                                                              Page 1 of 16 Request for Additional Information Response On July 18, 2007, the Nuclear Regulatory Commission (NRC) issued a Request for Additional Information (RAI, Reference 1) concerning License Amendment Request (LAR) #295, Revision 0 (Reference 2). Florida Power Corporation (FPC), doing business as Progress Energy Florida, Inc., hereby provides the following response to this RAI.
NRC Request
: 1.      In the submittal with regards to the quality of the CR-3 probabilistic risk assessment (PRA), the licensee does not state the scope of their model (i.e., internal events, external events, fires, seismic, etc., as well as at-power, transition, shutdown).
The quantitative results presented are compared to the acceptance guidance of Regulatory Guides (RGs) 1.174 and 1.177, which are based on a full scope risk analysis. The submittal later states that the licensee does not have a fire PRA model.
The licensee is requested to clarify the scope of their risk analysis used for this application.
FPCResponse
: 1.      The ProbabilisticSafety Assessment (PSA) model used in this calculation is the Crystal River Unit 3 (CR-3) PSA Model of Record. This model is an at power (Mode 1) internal events model that includes internal flooding events.      This model does not include fire events, seismic events, shutdown risk, and transition operationfrom power to shutdown states. All Nuclear Energy Institute peer certification Facts and Observations have been addressedin this model.
A gap assessment for the CR-3 PSA Model of Record and the American Society of Mechanical Engineers PRA Standard was completed in May 2007. Items identified were generallyfocused on assumptions and uncertainty requirements of the base model.
The events not included in the model, fire, seismic, shutdown risk and transition operation from power to shutdown states, have been qualitatively or quantitatively assessed as well. These are discussed in the responses to Questions 2, 6 and 7.
NRC Request
: 2.      In the submittal (Attachment A, page 15 of 27), the licensee states that the only potentially significant external events that could influence the risk are fires and severe weather. This conflicts with the conclusions of the topical report BAW-2295-A Rev. 1, which was the basis for the staff approval of TSTF-430, on which this amendment request is based. BAW-2295-A concluded that the impact of the proposed changes had a negligible impact on external events risk. The basis for the licensee's conclusion identifies a review of the CR-3 Individual Plant Examination (IPE) and supporting data. Since external events, including fires, were not in the scope of the IPE, the basis for this conclusion is not clear. The licensee is requested to provide its disposition of external events (other than fires and severe weather) not included in the risk assessment results (such as seismic events, external flooding, high winds, transportation accidents), and a basis for its conclusions regarding the low-risk significance of each of these events for this application.

U.S. Nuclear Regulatory Commission                                                                             Attachment A 3F0907-03                                                                                                       Page 2 of 16 FPCResponse
: 2. The conclusion in the technical report BAW-2295-A that external events are negligible is still valid. The Individual Plant Examination of External Events (IPEEE) can provide insight into external events. To ensure these results are still valid, the following additionalevaluationhas been performed.
Based on the IPEEE, seismic events, transportationaccidents, and external floods all have a frequency below 1E-06 per year. At thisfrequency, with an assumed Conditional Core Damage Probability(CCDP) of 1, and a minimal exposure time, there is negligible impact.
(1E-06 yr       ')  . (7 days) * (1/365 yr/days) = -]E-08 Due to the higher frequency of tornados (greater than 1E-06/yr), the high wind evaluation warrants additional analysis using the IPEEE information and the current PRA model. Table 2-1 provides a summary of high wind events quantified in the IPEEE updated with currentmodel runs that considered the Allowed Outage Time (AOT):
Table 2 Tornado and Unit 1 Stack Fall Impact Total 7        4 Day IPEEE            CDF with        Train A           Train B           Day        Increase Event             Frequency            No AOT          AOT in            AOT in       ICCDP for     ICCDP for 1 /yr)         in Effect            C 2,4               2,5     Limiting      Limiting Effe              Effect          Trains         Train0 Tornado frequency for            2.1 E-05           3.4E-09        6.7E-09           5.3E-09        6.3E-1 1       3.6E-1 1 F1 case and greater Loss of Category 1             6.3E-08            6.3E-08        6.3E-08           6.3E-08                  0 structures due 3 to tornado Unit 1 stack falling on EFW tank enclosure            6.5E-06            2.6E-08        1.1 E-07          1.OE-07        1.6E-09        9.2E-10 due to high winds Table Notes:
: 1. The initiating frequencies for tornado events are taken from section 5.1.1 of the IPEEE. The initiating event frequency for Unit 1 stack falling on Emergency Feedwater (EFW) tank enclosure is from section 5.1.2 of the IPEEE.
: 2. Base Core Damage Frequency (CDF) is calculated by taking the CCDP of the Loss of Offsite Power initiating event without credit given for recovery of offsite power.
: 3. The CCDP for loss of Category 1 structures due to a tornado is assumed to be 1. Due to the low initiating event probability, no further analysis has been performed.
4     Train A Low Pressure Injection (LPI) CDF runs have the Train A Decay Heat Removal (DH), Building Spray (BS),
Decay Heat Closed Cycle Cooling (DC), and associated Raw Water (RW-DC) pumps failed.
: 5. Train B LPI CDF runs have the Train B Decay Heat Removal, Building Spray, Decay Heat Closed Cycle Cooling, and associated raw water pumps failed.
: 6. The Incremental Conditional Core Damage Probability (ICCDP) increase is calculated using the following equation:              ICCDP = (4 days) * (1 yr / 365 days) * ("Limiting Train CDF" - "Base CDF")
: 7. Values in this table were generated using a truncation of 1 E-1 2 for quantified results. Some cutsets may not have been recovered due to the large number of cutsets generated. If recoveries are not applied, the results will be conservative.
: 8. The ICCDP increase is calculated using the following equation:
ICCDP = (7 days) * (1 yr / 365 days) * ("Limiting Train CDF" - "Base CDF")

U.S. Nuclear Regulatory Commission                                                  Attachment A 3F0907-03                                                                             Page 3 of 16 With the exception of fire and severe weather, external events contribute less than 1E-08 ICCDP for an increase of 4 days per AOT entry for each train of DH, DC, BS, and associated RW-DC trains. Based on these results, it is reasonable to state that these external events have a negligible impact on risk for this application.
NRC Request
: 3. In the submittal, the licensee identified general assumptions relevant to their risk analysis, including that "During performance of the AOT, the corresponding opposite train equipment and diesel are considered to be protected." The term "protected" should be more clearly defined as to what specifically is prohibited while the equipment is protected. The specific scope of equipment should be stated, and appropriate regulatory commitments identified.
FPCResponse
: 3. When a corresponding opposite train is deemed "protected," administrativecontrols are implemented to minimize the possibility of the loss of use of this equipment as described in Administrative Instruction 0500, "Conduct of Operations Department Organization and Administration." There is no planned maintenance or discretionary equipment manipulation on protected equipment or its power supplies. The scope of affected equipment is equipment within the protected train whose loss of function would result in the inability to ensure safe shutdown.          This includes pumps, valves, required instrumentation and power supplies of the systems required for decay heat removal capability, inventory control, power availability, reactivity control, pressure control and containment.
To ensure safe shutdown, the following commitment is made for extended (greaterthan 72 hours)preplannedoutages on the LPI, BS, DC or RW-DC System (see Attachment B):
            "The Remote Shutdown Panel, the Appendix R Cooler and the opposite train of LPI, BS, DC, RW-DC, EFW, Auxiliary FeedwaterSystem, Emergency FeedwaterInitiation and Control System, HPI, and their power supplies (AC and DC) will be administratively designated as "protected" (i.e., no planned maintenance or discretionaryequipment manipulation).
NRC Request
: 4. In the submittal (Attachment E, Section 5.3), the licensee states that 96 additional hours of unavailability are assumed for each train of the affected systems, and that the unavailability would be incurred simultaneously for each train of the systems.
No basis has been provided to justify either of these assumptions based on maintenance practices and past component maintenance history. This assumption is also inconsistent with the topical report BAW-2295-A Rev. 1. BAW-2295-A identified an increase from 80 to 93 hours of unavailability per train per year.
Further, the licensee identified a compensatory measure (Attachment A, page 17) to schedule concurrent outages of raw water and decay heat systems, "if possible."
This appears to conflict with an assumption of concurrent outages, since it identifies that such concurrent scheduling may not be possible in some circumstances. The licensee should justify these assumptions and clarify the intent of the compensatory

U.S. Nuclear Regulatory Commission                                                                        Attachment A 3F0907-03                                                                                                   Page 4 of 16 measures regarding concurrent outages, or provide more realistic risk analyses accounting for non concurrent outage scheduling.
FPCResponse
: 4. The increased average yearly unavailability,for extending this AOT is expected to be less than or equal to the 13 hours presented in BAW-2295-A. The PRA calculations used 96 hours (the difference from 3 to 7 days) to bound results. The risk analysis has been re-performed with the 13 hour increased unavailability assumption. Results are presented in Table 4-2.
The Reference 2 submittal assumed DH, BS, and RW-DC would all be removed from service at the same time for maintenance associated with the extended AOT. This concurrent maintenance assumption has been removed in the results presented in Table 4-2. Table 4-1 provides modified test and maintenance terms calculatedfor the fidl AOT (7 days) and the expected average yearly value (current plus 13 hours). The expected average yearly value is based on the 13 hour increaseprovided in BAW-2295-A.
Table 4 Unavailability Values for Non-Concurrent Maintenance Evaluation Description                            Nominal               Full AOT           Expected Model Value              Values              Values DH P-1A Train in Maintenance                       1.03E-02             2.13E-02 1          1.18E-02 3 RB Spray Train A in Maintenance                       9.4E-03             2.04E-02 1          1.09E-02 3 RWP-3A in Maintenance                            8.6E-03              1.96E-02 1         1.01 E-02 3 2         5.48E-03 3 DC Train A in Maintenance                          4E-03              1.92E-02 DHP-1 B Train in Maintenance                       1.03E-02             2.13E-02 1          1.18E-02 3 RB Spray Train B in Maintenance                       9.4E-03             2.04E-02 1          1.09E-02 3 RWP-3B in Maintenance                            8.6E-03             1.96E-02 1          1.01 E-02 3 2         5.48E-03 3 DC Train B in Maintenance                          4E-03              1.92E-02 Table Notes:
: 1. These Full AOT values are calculated by adding 4 days of unavailability per year to the current PRA model unavailability value. This represents the 4 day increase from 3 days to 7 days.
: 2. These Full AOT values are calculated by using 7 days of unavailability per year. The current value is less than 3 days per year, so the addition of 4 days would not have encompassed the full 7 day period.
: 3. The expected values for mean duration are calculated by adding 13 hours of unavailability per year to the current PRA model unavailability value.
Table 4-2 compares BAW-2295-A results to the current CR-3 PRA model results.
Current results are provided with and without the concurrent maintenance assumption.
A mnean duration increase of 25 hours is used. BAW-2295-A evaluation used 13 hours.
Results shown in Table 4-2 show the removal of the concurrent maintenance assumption does not significantly impact the results. The PRA calculations result in lower risk values than those presented in BAW-2295-A. Concurrent maintenance will still be performed whenever practical.

U.S. Nuclear Regulatory Commission                                                                             Attachment A 3F0907-03                                                                                                       Page 5 of 16 Table 4 Comparison of BAW-2295-A to Current Model Results Current Model Results, Risk Measure                       B W 9                                     without Concurrent Maintenance      Assumption Base CDF (/yr)                                8.18E-06                              4.70E-06 4 4 Day Increase ICCDP                              4.6E-07                               3.E-07 1 A (Single AOT Risk)
Total 7 Day ICCDP                               8.OE-07                               6.OE-07 4 5, (Single AOT Risk)
Incremental Yearly CDF (Mean expected duration)                             4.4E-07                               1.9E-07 2.4
(/yr)
Incremental Yearly CDF (full                           2.3E-06                              1.67E-06   2,4 AOT) (/yr)
Base LERF (/yr)                                1.84E-07                               3.69E-07 Incremental CLERP (Single                              1.5E-08                              4.03E-10 6 AOT Risk)
Incremental Yearly LERF                                9E-09                                    3 (Mean duration) (/yr)
Incremental Yearly LERF (full                           4.5E-08                                    3 AOT) (/yr)
Table Notes:
: 1. Based on Train B Out Of Service (OOS) which gives results that are more conservative. These results use a 4 day period which is equal to the change from 3 to 7 days. All maintenance events were at nominal values. If compensatory measures limit maintenance on Emergency Feedwater System (EFW), Auxiliary Feedwater System (AFW), Emergency Feedwater Initiation and Control System (EFIC), High Pressure Injection (HPI), Appendix R Cooler, and their power supplies, the resulting value would be 2.5E-07.
: 2. Quantified using values from Table 4-1.
: 3. The LERF values have not been re-quantified based on the Conditional Large Early Release Possibilities (CLERP) value. These values are the Incremental LERF values and will not provide additional insight.
: 4. Model truncation of 1E-1 1 has been used to quantify level 1 results.
: 5. Based on Train B OOS which gives results that are more conservative. These results use a 7 day period. All maintenance events were at nominal values. If compensatory measures limit maintenance on EFW, AFW, EFIC, HPI, Appendix R Cooler, and their power supplies, the resulting value would be 4.4E-07.
: 6. Based on Train B OOS which gives results that are more conservative. These results use a 7 day period. All maintenance events were at nominal values.
NRC Request
: 5. The submittal did not analyze emergent component failures (i.e., unplanned maintenance) and the potential impact on risk from increased probability of common cause failure. Since the applicability of the extended AOT is not limited to planned maintenance activities, the risk analysis should address emergent repairs per RG 1.177 Section 2.3.3.1 and Appendix A Section A.1.3.2.
FPCResponse
: 5. Random failure rates were left at their nominal value for risk metrics evaluations.
Unplanned testing and maintenance contributions were included in the delta CDF and delta Large Early Release Frequency (LERF) calculation. The expected increase in unavailabilit-, was added to the existing tenr to ensure higher levels of planned maintenance and unplannedportions were not lost.

U.S. Nuclear Regulatory Commission                                                                     Attachment A 3F0907-03                                                                                                Page 6 of 16 The CCDP values were originally calculated with test/maintenance terms set to zero for the opposite trains of LPI, BS, DC, and RW-DC. Additionally, the test/maintenance terms were set to zero for the Appendix R Cooler,for all trains of EFW, AFW, EFIC, and HPI, and for their power supplies. To evaluate the possible impact of these test and maintenance events, the ICCDP change was re-quantified with average maintenance includedfor all components.
Table 5 Impact of Test and Maintenance Cases Nominal Testing                     Reduced Test and Maintenance                          Maintenance Total 7 Day ICCDP                               6.OE-07 3                          4.4E-07 3 (Single AOT Risk) 4 Day Increase ICCDP                              3.4E-07 1                          2.5E-07 1 (Single AOT Risk)
Table Notes:
: 1. Based on Train B OOS which gives results that are more conservative. This result uses a 4 day period which is equal to the change from 3 to 7 days and is described in Table 4-2.
: 2. Model truncation of 1E-1 1 has been used to quantify level 1 results.
: 3. Based on Train B OOS which gives results that are more conservative. This result uses a 7 day period and is described in Table 4-2.
Common Cause Imnpact:
Unplannedfailures and maintenance can result in AOT entries when the status of the complementary equipment of the same type and fiinction may not be known. In these cases, the common cause term is larger than the nominal value. To evaluate common cause contribution, ICCDPs have been calculatedfor the DH, DC, and RW-DC trains.
BS was not evaluatedfor this sensitivity because BS does not affect the level 1 results and does not impact the level 2 results significantly. Therefore, the common cause contributionfor BS is considered to be negligible when compared to the level 1 results of DH, DC, and RW-DC.
Each train is evaluated for common cause impact by setting the pump for the desired train as failed, and increasing the common cause term for the pump to its largest respective common cause beta value. This results in conservative values, but provides some insight into the maximum possible impact.
A sensitivity of equipment recoveries was not performned because equipment recoveries are not credited in the base PRA model.
2 Table 5 Common Cause Total 7     4Day Failed                                  Model          Beta           Base     New                 Increase Component           Description           Value         Value           CDF      CDF      ICCDPa Chne DHP-1A          DHP-1A,1B CCF4         1.02E-04      5.01E-02       4.70E-06 5.94E-05   1.05E-06    5.99E-07 DHP-1A and DHP-1B            DHP-1B fail to       1.02E-04      5.01E-02       4.70E-06 6.16E-05   1.09E-06    6.24E-07 start DCP-1A          DCP-1A,1B     CCF      3.52E-05      1.52E-02       4.70E-06 3.39E-05   5.60E-07    3.20E-07 DCP-1B          DCP-1A,1B     CCF      3.52E-05      1.52E-02       4.70E-06 4.72E-05   8.15E-07    4.66E-07 RWP-3A          RWP-3A,3B     CCF      2.97E-05      3.55E-02       4.70E-06 4.90E-05   8.50E-07    4.85E-07 RWP-3B          RWP-3A,3B     CCF      2.97E-05      3.55E-02       4.70E-06 6.21 E-05  1.10E-06    6.29E-07

U.S. Nuclear Regulatory Commission                                                         Attachment A 3F0907-03                                                                                   Page 7 of 16 Table Notes:
: 1. This result uses a 4 day period which is equal to the change from 3 to 7 days.
: 2. Model truncation of 1E-1 1 has been used to quantify level 1 results.
: 3. This result uses a 7 day period and base model alignments.
: 4. Common Cause Failures.
The large CCF ICCDPs calculated are about twice as large as the point estimate value with nominal common cause values used. The impact of the common cause term is expected to be minimal for several reasons:
       " The unplanned maintenance unavailability is a small fraction of the overall pump unavailability.
* Only a small percentage of unplanned maintenance events last longer than three days, the current Technical Specification window. If the failed component is fixed within three days, there is no increase in risk to the public.
       " An equipmentfailure or deficiency requires an assessment to determine if thisfailure has a common impact on redundant or similar equipment. Only a small percentage of common cause events are expected to be unidentified within the first day of the outage window which is less than the original Technical Specification window. If it is identified that an existing failure has a common impact on redundant or similar components, these components would be declared inoperable. This would require entry into a different Technical Specification.
NRC Request
: 6. The submittal discusses the development of a fire risk sensitivity study and presents its results. The staff requests additional clarifications regarding the fire risk of the proposed changes:
FPC Response
: 6. To respond to this RAI, a new fire assessment was needed. A fidl fire PRA is not availablefor evaluation of the AOT. However, sufficient information exists in the IPEEE to support this submittal. The IPEEE cable routing database was used in conjunction with the current PRA model. The initial screen numbers will be used where all unprotected cables are assumed to fail if fire suppression is unsuccessfid. Since detailed fire modeling is not credited,this will provide an alternate, conservativefire approach to assess the risk offire as it relates to the 7 day AOT for LPI and BS.
Assumptions:
* The cable spreading and control room are excluded from the analysis because both trains are impacted by fire, so the delta risk is expected to be zero with the simplified approachbeing used.
* Testing and Maintenance events were left at their nominal values.
* Fire wrap credit is given for the historicalIPEEEevaluated condition.
* Fire suppression credit is given for the historicalIPEEEevaluated condition.
* Fire ignition frequenciesfrom the historicalIPEEEare used.
0      All equipment that is not fire wrapped is assumed to fail for all fires when suppressionfails.

U.S. Nuclear Regulatory Commission                                                 Attachment A 3F0907-03                                                                           Page 8 of 16 The fire compartments were generally run at a truncation of approximately 1E-08. Some compartments would not quantify at 1E-08 because they contained a large amount of equipment that is assumed to fail. For these compartments, the truncation was raisedto yield results.
Results were generated by adding up the CDFfor all the fire compartments. Three CDFs were generated: a base CDF with nominal test and maintenancefraction, and two cases with Train A and B AOT removed from service. An ICCDP was calculatedusing a delta CDF and the 4 day increase. Train B yields a larger result with a change of ICCDP of 3.7E-07. The total 7 day ICCDP is 6.5E-07.
The following fire areasshowed they were sensitive to fire risk in conjunction with a DH, DC, BS, and RW-DC outage:
* The opposite train DH, DC, and RW-DC rooms.
* The opposite train MCC rooms (480 and 4160).
* Non-Safety 4160V and 480V rooms.
* The opposite train battery and chargerrooms.
* The opposite train Inverter rooms.
* Remote Shutdown Room.
Spurious actuations were not evaluated in the quantifiedfire assessment. However, they can be addressed qualitatively. Based upon a review of the PRA cutset results and the model, the opening of the pressurizer PORV would be the primary spurious event of concern when the DH system is unavailable. Although, the actual risk of this spurious event due to fire is not quantified, the impact can be mitigated by performing actions to protect cables associated with the PORV and the PORV block valve.
The List of Regulatory Commitments has been revised to reflect the above information.
The new list is provided in Attachment B.
: a. The submittal does not discuss the role of the LPI subsystem and its cooling water support systems for mitigation of fire events for achieving a safe shutdown. The licensee should discuss the anticipated role of these systems given a fire in the plant.
LPI is a function of the Decay Heat Removal System. The Appendix R Fire Study states the safe shutdown requirement for the Decay Heat Removal System is as follows:
The fimctions of the Decay Heat Removal System are:
: 1) Provide decay heat removal and temperature monitoring for cold shutdown; and
: 2) Assure RC outlet isolation until reaching hot shutdown conditions; and
: 3) Isolate the Borated Water Storage Tank from the ReactorBuilding sump.

U.S. Nuclear Regulatory Commission                                                 Attachment A 3F0907-03                                                                            Page 9 of 16 The PRA has an additional requirementfor Decay Heat Removal to provide a suction source and cool suction water from the emergency sump to makeup pumps for long tenn Feed and Bleed cooling.
: b. The fire zones of interest are identified as those which contain circuits of the systems for which the AOT extension is requested. However, fires which disable other systems resulting in a demand for the LPI and supporting systems would also be impacted by the AOT extension. For example, fire-induced failure of emergency feedwater components would increase the likelihood of requiring core cooling by bleed-and-feed once through cooling, which may ultimately require the use of LPI systems for long term decay heat removal. The licensee should justify the selection of fire scenarios.
The IPEEE fire analysis has been re-evaluated to address all originally scoped compartments. This ensures possible interactionswith support and redundant systems have been captured.
: c. The submittal states that fires which impact "manual actions that are credited in the Fire Study" are expected to have a minimal risk impact. It is not understood to what the term "Fire Study" refers, nor are the manual actions identified in any way, therefore the basis for this statement is not clear. The licensee should clarify the intent of this assumption and assure it has a proper basis.
The "Fire Study" referred to is the CR-3 Appendix R Safe Shutdown Analysis. The manual actions were those actions in the CR-3 Appendix R Safe Shutdown Analysis.
The fire analysis has been re-evaluated using IPEEE information instead of the "Fire Study" presented in the previous submittal.
: d. The development of the "compensated frequency" adjusts for "equipment which will not be operated without special precautions." It is not understood to what "special precautions" refers, to what equipment these precautions apply, and whether these are associated with any licensee commitments specific for this application. The licensee should clarify the scope and intent of this assumption.
Compensatedfrequency was an adjustment to the initiating event frequency for each fire compartment to reflect what equipment would be operating. If a motor was not going to be energized during the AOT, its contributionwas removedfrom the analysis.
This assumption is not used in the re-evaluatedfire analysis.
            "Special precautions" was used to refer to fire watches. This is no longer credited in the fire analysis.
: e. The licensee applied a 0.1 conditional core damage probability given a fire and identified this as a conservative assumption. However, no specific basis has been provided to justify this value as conservative. Further, the final results of the fire sensitivity study using this assumption do not demonstrate low sensitivity to fire risk, which would indicate that the use of a conservative screening value may be inappropriate. The licensee should justify the use of 0.1 as the conditional core

U.S. Nuclear Regulatory Commission                                                Attachment A 3F0907-03                                                                         Page 10 of 16 damage probability, or consider applying a more realistic value to better establish a less conservative estimate of fire risk during the extended AOT.
The IPEEEfire analysis has been re-evaluated with model quantification to generate estimates based on the current PRA model. Although detailedfire modeling has not been performed, this analysis (discussed in the response to 6, above) still provides conservative results. Initial screening values were used with all unprotected equipment in rooms assumed to fail because they containedafire.
: f. The results of the sensitivity study show that fire risk dominates the risk impact of the extended AOT, and the acceptance guidance of RG 1.177 of 5E-7 incremental conditional core damage probability is exceeded by almost a factor of
: 10. The licensee has not drawn any conclusions from these results, nor have specific compensatory measures been identified to offset this risk. (The licensee has proposed an undefined "periodic" fire watch in the decay heat pump room and seawater rooms, which contain the equipment subject to the extended AOT.
However, in their fire analysis the decay heat pump rooms have a compensated frequency of "0" since the pumps are not running, so this compensatory fire watch does not directly offset any specific source of the fire risk.) The licensee should discuss the implications of the fire risk sensitivity study to the overall conclusions of the risk impacts associated with the extended AOTs, and consider implementing specific compensatory measures to address fire risk as a significant contributor to the risk profile.
As indicatedin the response to 6.d. above, compensatedfrequency is no longer used in the fire analysis. As such, these fire watches are now compensatory measures that will directly offset some of the risk and not an assumption of the analysis. The compensatory measures, identified in the List of Regulatory Commitmnents, have been revised based on the re-evaluatedfire analysis discussed in the response to Question 6, above. The revised List of Regulatory Commitments is included as Attachment B to this RAI response.
NRC Request
: 7. The submittal did not identify whether the risk analyses provided point estimates of the mean or actual means, nor was there any discussion of uncertainty analyses to support the calculations. The licensee is requested to address PRA model and parametric uncertainty using the guidance of RG 1.174 Section 2.2.5 and RG 1.177 Section 2.3.5.
FPCResponse
: 7. The results provided in thle submittal and in this RAI response are point estimates.
Additional uncertainty analysis has been performed.
UNCERT ParameterAnalysis Parameter uncertainty was evaluated using the Risk and Reliability Workstation's UNCERT software. The results are provided in Table 7-1.

U.S. Nuclear Regulatory Commission                                                                           Attachment A 3F0907-03                                                                                                    Page 1 of 16 Table 7 Parameter Uncertainty Results Total 7 Day ICCDP         4 Day Increase ICCDP Risk Measure           Base Model         Train B AOT          (Single AOT Risk) 3         (Single AOT Risk)
CDF (/yr)            4.70E-06            3.61 E-05                 6.OE-07                   3.4E-07 Mean               4.67E-06            3.61 E-05                 6.OE-07                   3.4E-07 5th Percentile          1.62E-06            1.22E-05                  2.OE-07                   1.2E-07 50th Percentile          3.65E-06            2.75E-05                  4.6E-07                   2.6E-07 9 5th Percentile          1.05E-05           8.1OE-05                  1.4E-06                   7.7E-07 Table Notes:
: 1. Based on Train B OOS which gives results that are more conservative. This result uses a 4 day period which is equal to the change from 3 to 7 days. All maintenance events were at nominal values.
: 2. Model truncation of 1 E-1 1 has been used to quantify level 1 results.
: 3. Based on Train B OOS which gives results that are more conservative. This result uses a 7 day period. All maintenance events were at nominal values and base model alignments.
Small LOCA ParameterUncertainty When reviewing the CCDP cutsets, small Loss of Coolant Accident (LOCA) was identified as a dominant initiatorfor cutsets including AOT failures. To evaluate the sensitivity of the initiating event frequency, the change in ICCDP has been recalculated at the 9 5 ih percentile.
Table 7 Comparison of Small LOCA Uncertainty Current Model                 Current Model             Current Model Results with                Results with 5 th         Results with 9 5 th Nominal Small                Percentile Small          Percentile Small LOCA                           LOCA                    LOCA Small LOCA Initiator Value                    5.OE-04                          1 E-04                  1 E-03 Base CDF (/yr)                       4.70E-06                       3.48E-06                6.22E-06 LPI Train B CDF (/yr)                     3.61 E-05                      2.58E-05                4.89E-05 Total 7 Day ICCDP                       6.OE-07 3                      4.3E-07 3               8.2E-07 3 (Single AOT Risk) 4 Day Increase ICCDP                       3.4E-07 1                      2.4E-07'                4.7E-07 1 (Single AOT Risk)
Table Notes:
: 1. Based on Train B OOS which gives results that are more conservative. This result uses a 4 day period which is equal to the change from 3 to 7 days. All maintenance events were at nominal values.
: 2. Model truncation of 1 E-1 1 has been used to quantify level 1 results.
: 3. Based on Train B OOS which gives results that are more conservative. This result uses a 7 day period. All maintenance events were at nominal values and base model alignments.
Component ParameterUncertainty Componentfailure rates are not always at their nomninalfailure rate. When one train of BS, DH, DC and RW-DC is removed from service, the opposite train becomes important for mitigatingplant events that require DH. The impact of varying failure ratesfor these individualpumps has been evaluated in Table 7-3. With the individualpumps at the 95/th percentile value, there is no significant increase in risk.                            Train B AOT was used to evaluate this case because it yields largernumbers.

U.S. Nuclear Regulatory Commission                                                                        Attachment A 3F0907-03                                                                                                 Page 12 of 16 Table 7 Comparison of Individual Co ponent Uncertainty Results with         Results with          Results with Component                 Risk Measure                   Nominal           5 Percentile 95 Percentile Value                Value                Value Basic Event Value                 1.98E-03             6.43E-04             4.34E-03 Base CDF (/yr)                  4.70E-06             4.65E-06             4.78E-06 LPI Train B CDF (/yr)               3.61 E-05            3.51 E-05            3.78E-05 DHP-1A            4 Day Increase ICCDP                                     1                     3.6E-07 (Single AOT Risk)
Total 7 Day ICCDP                  6.0E_073             5.8E_07'             6.3E-073 (Single AOT Risk)
Basic Event Value                 2.05E-03             9.76E-04             3.66E-03 Base CDF (/yr)                  4.70E-06             4.66E-06             4.75E-06 LPI Train B CDF (/yr)               3.61 E-05            3.53E-05             3.73E-05 DCP-1 A             Incremental change                 3.4E-071             3.4E-07 1            3.6E-071 CCDP (Single AOT Risk)
Total 7 Day ICCDP                  6.0E-07 3            5.9E-07 3            6.2E-07 3 (Single AOT Risk)
Basic Event Value                 9.26E-04             3.93E-04             1.77E-03 Base CDF (/yr)                  4.70E-06             4.68E-06             4.72E-06 LPI Train B CDF (/yr)               3.61 E-05            3.57E-05             3.67E-05 RWP-3A              Incremental change                 3.4E-071              3.4E-071            3.5E-071 CCDP (Single AOT Risk)
Total 7 Day ICCDP                  6.OE-07 3            5.9E-07 3            6.1 E-07 3 (Single AOT Risk)
Table Notes:
: 1. Based on Train B OOS which gives results that are more conservative. This result uses a 4 day period which is equal to the change from 3 to 7 days. All maintenance events were at nominal values and base model alignments.
: 2. Model truncation of 1E-1 1 has been used to quantify level 1 results.
: 3. Based on Train B OOS which gives results that are more conservative. This result uses a 7 day period. All maintenance events were at nominal values and base model alignments.
Loss of Offsite Power ParameterUncertainty A Loss of Offsite Power sensitivity was performed which showed results to be insensitive to changes in the initiatingevent value.
Assumption of Expected Unavailability The expected increase in unavailability due to the increased AOT was 13 hours. This assumption of expected unavailability has a direct impact on delta CDF and delta LERF results. The level 1 model is more sensitive than the level 2 model. In the level 1 model, the delta CDF has been calculatedusing an average yearly increase of 26 hours. The 26 hour increase is based on historical data for maintenance and surveillances, including rebuildingpumps online.
Table 7-4 provides modified test and maintenance terms calculatedfor the 13 hour and 26 hour expected average yearly assumption. Table 7-5 provides the results which show values that are still lower than those identified in the topical report. The delta CDF increased linearly to the increased hours.

U.S. Nuclear Regulatory Commission                                                                         Attachment A 3F0907-03                                                                                                  Page 13 of 16 Table 7 Unavailability Values for Expected Increase Unavailability Sensitivity "NominalModel Mean. Duration..               Mean Duration Description                            VaUe., -             Increase of 13    increase of 26 Hours              Hours 1        1.33E-02  2 DHP-1A Train In Maintenance                      1.03E-02                 1.18E-02 RB Spray Train A in Maintenance                    9.4E-03                  1.09E-021          1.24E-02 2 RWP-3A in Maintenance                        8.6E-03                  1.01 E-021         1.16E-02 2 1        6.97E-03 2 DC Train A in Maintenance                        4E-03                  5.48E-03 1        1.33E-02 2 DHP-1 B Train in Maintenance                     1.03E-02                 1.18E-02 RB Spray Train B in Maintenance                     9.4E-03                  1.09E-02 1         1.24E-02 2 RWP-3B in Maintenance                        8.6E-03                  1.01 E-02 1        1.16E-02 2 1        6.97E-03 2 DC Train B in Maintenance                        4E-03                  5.48E-03 Iable Notes:
: 1. The mean duration values are calculated by adding 13 hours of unavailability per year to the current PRA model unavailability value.
: 2. The mean duration values are calculated by adding 26 hours of unavailability per year to the current PRA model unavailability value.
Table 7 Comparison of BAW-2295-A to Current Model Results BAW-2295-A Values                  Mean Duration           Mean Duration Risk Measure                for CR3 with Mean                  Increase of 13          Increase of 26 Duration    Increase of                    H ro
              ,...          .....               l3ouHours                            HoursH Base CDF (/yr)                     8.18E-06                        4.70E-06                4.70E-06 Modified Unavailability CDF                  Not provided                      4.89E-06               5.09E-06 1
(/yr)
Incremental Yearly CDF (Mean duration)                    4.4E-07                          1.9E-07                 3.9E-07
(/yr)
Table Notes:
: 1. Quantified using values from Table 7-4.
: 2. Model truncation of 1 E-1 1 has been used to quantify level 1 results.
Assumption of EFIC Failure Without Room Coolingz The EFW system provides long term decay heat removal and therefore, plays an important role as a redundantsystem to the DH system. On a loss of EFIC room cooling the failure mode of the EFW valves is uncertain. Although these valves fail open on a loss of power, there is a possibility when the control cabinets over heat with power still available the EFW valves may fail to the closed position. The PRA model assumes a 50%
probability of failure for EFW valves on a loss of cooling to the EFIC control panels.
The impact of varying the assumed state of EFW valves on a loss of EFIC room cooling has been evaluated in Table 7-6. With the individual valves assumed to fail on a loss of EFIC room cooling, there is not a significant increase in risk. Train B AOT was used to evaluate this case because it yields largernumbers.

U.S. Nuclear Regulatory Commission                                                                          Attachment A 3F0907-03                                                                                                   Page 14 of 16 Table 7 EFIC Failures on Loss of Room Cooling EFW Valves Failing         EFW Valves Failing Risk Measure                 Nominal Values                 Open on Loss of           Closed on Loss of EFIC Room Cooling EFIC Room Cooling EFIC failure on a Loss Of Room                         50%                             0%                       100%
Cooling Base CDF (/yr)                    4.70E-06                        3.57E-06                  5.81 E-06 LPI Train B CDF (/yr)                 3.61 E-05                       2.33E-05                  4.75E-05 Total 7 Day ICCDP                    6.OE-07   3                    3.78E-07 3                 8.OE-07 3 (Single AOT Risk) 4 Day Increase ICCDP (Single AOT                    3.4E-07'                       2.16E-07'                  4.6E-07 1 Risk)
Table Notes:
: 1. Based on Train B OOS which gives results that are more conservative. This result uses a 4 day period which is equal to the change from 3 to 7 days. All maintenance events were at nominal values and base model alignments.
: 2. Model truncation of 1E-1 1 has been used to quantify level 1 results.
: 3. Based on Train B OOS which gives results that are more conservative. This result uses a 7 day period. All maintenance events were at nominal values and base model alignments.
Transition Risk Completeness Uncertainty A transition model that can be used to evaluate the impact of removing DHfrom service has not been developed.
Although maintenance is not normally planned during transitionmodes, it is possible for unplanned unavailabilityto exceed the AOT limit prompting a shutdown with one train of DH unavailable. Increasing the AOT from. 3 to 7 days would decrease the frequency of this occurrence and presents a decrease in risk to the public.
Shutdown Risk Completeness Uncertainty A shutdown model that can be used to evaluate the impact of removing DHfrom service has not been developed. This risk can only be considered on a qualitative level. This would decrease the risk impact if it were perfonned.
Fire Risk Completeness Uncertainty The fire contributionfor external events is based on routing informationfrom the IPEEE.
Using the IPEEE information to perform fire analysis provides a level of insight but does not provide a thoroughfire assessment.
NRC Request
: 8. The submittal did not address the truncation level used for this analysis. The licensee is requested to provide this information consistent with RG 1.177 Section 2.3.3.4.
FPC Response
: 8. The original submittal results were quantified with a IE-12 truncation. DH, DC, and RW-DC events are in the top 50 cutsets. However at this truncation level, the number of cutsets is found to increase above the capability of the recovery program when LPI is removed from service. This means some cutsets .for these cases do not recover. The CCDP values generatedfor these cases are higher than they would be otherwise. This

U.S. Nuclear Regulatory Commission                                                  Attachment A 3F0907-03                                                                           Page 15 of 16 causes delta CDF and CCDP to be conservative. Since this recovery failure has a conservative bias, re-quantification of results at a lower truncation is unnecessary, provided results could be obtained. However, results for some cases can not be quantified at a truncation level of 1E-12. These cases are run at an increasedtruncation level, the lowest the code would support.
The truncation levels are presented in the footnotes for tables containing results.
NRC Request
: 9. The submittal (Attachment A, page 16) ends the first bulleted compensatory measure with a phrase "These protective measures .. ," which is not completed. The licensee is requested to provide the missing information.
FPCResponse
: 9. This is the last sentence of the first of five bullets in that paragraph. The complete sentence should be:
            "These protective measures will continue after the proposed AOTs have been implemented.
This is immediatelyfollowed by the second offive bullets in that paragraph.
NRC Request
: 10. The submittal (Attachment A, page 16) identifies a commitment to "avoid an AOT... that results in 'Higher Risk' (Orange Color Code)." The licensee is requested to clarify the intended meaning of "avoiding an AOT," and to specify the basis for the "Higher Risk (Orange Color Code)" in terms of risk impact.
FPCResponse
: 10. To avoid an AOT that results in "HigherRisk" (Orange Color Code) during an extended outage (greaterthan 72 hours) of the LPI, BS, DC or RW-DC System means CR-3 will not plan any maintenance that results in "Higher Risk" (Orange Color Code) when operating under those conditions. This is why it was identified as the first of the Regulatory Commitments identified in Attachment F of the submittal. The Orange Color Code is an administrative guideline that indicates risk significant SSCs have been removed from service, and that more stringent risk management actions need to be taken.
It is defined when the ICCDP is greater than or equal to 1.OE-05 for a 7 day period or greaterthan or equal to 1.OE-06for a 36 hour period. Under these conditions, the Work Week Manager and Superintendent Shift Operations determine what necessary actions should be taken to protect redundanttrains and to return the SSC to service.
NRC Request
: 11. RG 1.177 Section 2.3.7.2 identified the key components of a configuration risk management program. The licensee has not identified whether the program applied under plant procedure CP-253 satisfies all aspects of RG 1.177, specifically:

U.S. Nuclear Regulatory Commission                                                 Attachment A 3F0907-03                                                                          Page 16 of 16
             "  The scope of structures, systems, and components (SCCs) included.
* Applicability if additional SSCs become inoperable or non-functional while in a risk-informed AOT.
* Consideration of external events and level 2 issues.
The licensee is requested to clarify this issue.
FPCResponse
: 11. CP-253, "Power Operation Risk Assessment and Management," applies to all on-line work activities and manages the resultant risk from these activities. The scope of SSCs included in CP-253 is consistent with that identified in Regulatory Guide 1.177.
Specifically, this scope is all SSCs modeled in the CR-3 PRA in addition to all SSCs considered to be highly safety significant that are not modeled in the PRA. Work activities are screened for impact through quantitative and qualitative methods. The quantitative assessment is perfonned using the Equipment Out of Service (EOOS) program, a computer model that has the PSA as its basis. This assessment method provides risk impact by reviewing various combinations of EOOS in accordancewith the Work Week Schedule. The qualitative assessment is performed on the equipment that is not modeled in the PSA and also takes into account other aspects that are not modeled, such as containment integrity and plant environmental conditions. Containment integrity is listed as one of the key safety functions, which address level 2 impacts. Qualitative assessments for external events address predicted adverse weather conditions such as hurricanes, internalflooding and maintenance external to the plant (i.e., maintenance to fossil plants at the Crystal River Energy Complex). When emergent work or an unplanned degradation is identified, a Senior Reactor Operatorreview is performed and an EOOS risk assessment (quantitative assessment) is performed if the affected coniponent(s) are part of the EOOS model. If the component(s) are not part of the EOOS model, a qualitative risk assessment will be performed and documented in accordance with CP-253. Additional assessments against AOTs of Technical Specifications, Fire ProtectionPlan and the Offsite Dose CalculationManual are also made, as applicable.

==References:==
: 1. NRC to Crystal River Unit 3 Letter dated July 18, 2007, "Crystal River Unit No. 3, Request for Additional Infonnation Regarding License Amendment Request No. 295 Rev.
0, Extension of Allowed Outage Time to Seven Days and Elimination of Second Completion Times (TAC No. MD5241)
: 2. Crystal River Unit 3 to NRC Letter dated April 13, 2007, "Crystal River Unit License Amendment Request #295, Revision 0, Extension of Allowed Outage Time to Seven Days and Elimination of Second Completion Times Limiting Time"

PROGRESS ENERGY FLORIDA, INC.
CRYSTAL RIVER UNIT 3 DOCKET NUMBER 50-302 / LICENSE NUMBER DPR-72 LICENSE AMENDMENT REQUEST #295, REVISION 0 EXTENSION OF ALLOWED OUTAGE TIME TO SEVEN DAYS AND ELIMINATION OF SECOND COMPLETION TIMES ATTACHMENT B LIST OF REGULATORY COMMITMENTS

U.S. Nuclear Regulatory Commission                                                 Attachment B 3F0907-03                                                                            Page 1 of 1 List of Regulatory Commitments The following table identifies those actions committed to by Florida Power Corporation (FPC) in this document. Any other actions discussed in the submittal represent intended or planned actions by FPC. They are described to the NRC for the NRC's information and are not regulatory commitments. Please notify the Supervisor, Licensing and Regulatory Programs of any questions regarding this document or any associated regulatory commitments.
Commitment                                        Due Date CR-3 will perform procedure CP-253, "Power Operation Risk         During extended (greater than Assessment and Management," which requires both a                 72 hours) preplanned outage deterministic and probabilistic evaluation of risk for the        on the LPI, BS, DC or RW-performance of all maintenance activities. This procedure uses    DC System the Level 1 PSA model to evaluate the impact of maintenance activities on core damage frequency. CR-3 will not plan any maintenance that results in "Higher Risk" (Orange Color Code) during an extended outage (greater than 72 hours) of the LPI, BS, DC or RW-DC System.
The Remote Shutdown Panel, the Appendix R Cooler and the          During extended (greater than opposite train of LPI, BS, DC, RW-DC, EFW, Auxiliary              72 hours) preplanned outage Feedwater System, Emergency Feedwater Initiation and              on the LPI, BS, DC or RW-Control System, HPI, and their power supplies (AC and DC)         DC System will be administratively designated as "protected" (i.e., no planned maintenance or discretionary equipment manipulation).
CR-3 will not initiate an extended preventive maintenance         During extended (greater than outage (greater than 72 hours) on the LPI, BS, DC or RW-DC        72 hours) preplanned outage System if adverse weather, as designated by Emergency             on the LPI, BS, DC or RW-Preparedness procedures, is anticipated.                          DC System When extended maintenance (greater than 72 hours) is              During extended (greater than scheduled on a train of the LPI or BS System, CR-3 will limit     72 hours) preplanned outage transient combustibles in the decay heat pump vault of the        on the LPI or BS System opposite train and establish a periodic fire watch of the decay heat pump vault of the opposite train.
When extended maintenance (greater than 72 hours) is              During extended (greater than scheduled on a train of the DC or RW-DC System, CR-3 will         72 hours) preplanned outage limit transient combustibles in the seawater room and establish   on the DC or RW-DC System a periodic fire watch in the seawater room.
When extended maintenance (greater than 72 hours) is During extended (greater than scheduled on a train of the LPI, BS, DC or RW-DC System, 72 hours) preplanned outage CR-3 will limit transient combustibles and establish a periodic on the DC LPI, BS, DC or fire watch in the fire zones containing routed cables associated RW-DC System with the pressurizer PORV and PORV Block Valves.}}