3F0312-01, Response to Request for Additional Information to Support NRC PRA Licensing Branch Technical Review of the CR-3 Extended Power Uprate LAR

From kanterella
Jump to navigation Jump to search

Response to Request for Additional Information to Support NRC PRA Licensing Branch Technical Review of the CR-3 Extended Power Uprate LAR
ML12086A107
Person / Time
Site: Crystal River Duke Energy icon.png
Issue date: 03/22/2012
From: Franke J
Progress Energy Florida
To:
Document Control Desk, Office of Nuclear Reactor Regulation
References
3F0312-01
Download: ML12086A107 (26)
v • d • e


Text

NProgress Energy Crystal River Nuclear Plant Docket No. 50-302 Operating License No. DPR-72 March 22, 2012 3F0312-01 U.S. Nuclear Regulatory Commission Attn: Document Control Desk Washington, DC 20555-0001

Subject:

Crystal River Unit 3 - Response to Request for Additional Information to Support NRC PRA Licensing Branch Technical Review of the CR-3 Extended Power Uprate LAR (TAC No. ME6527)

References:

1. CR-3 to NRC letter dated June 15, 2011, "Crystal River Unit 3 - License Amendment Request #309, Revision 0, Extended Power Uprate" (Accession No. MLI 12070659)
2. Email from S. Lingam (NRC) to D. Westcott (CR-3) dated January 3, 2012, "CR-3 EPU LAR - Draft RAIs from APLA Branch (PRA)"
3. NRC to CR-3 letter dated February 8, 2012, "Crystal River Unit 3 Nuclear Generating Plant - Request for Additional Information for Extended Power Uprate License Amendment Request (TAC No. ME6527)" (Accession No. ML12003A217)

Dear Sir:

By letter dated June 15, 2011, Florida Power Corporation, doing business as Progress Energy Florida, Inc., requested a license amendment to increase the rated thermal power level of Crystal River Unit 3 (CR-3) from 2609 megawatts (MWt) to 3014 MWt (Reference 1). On January 3, 2012, via electronic mail, the NRC provided a draft request for additional information (RAI) needed to support the Probabilistic Risk Assessment (PRA) Licensing Branch technical review of the CR-3 Extended Power Uprate (EPU) License Amendment Request (LAR) (Reference 2). By teleconference on January 11, 2012, CR 3 discussed the draft RAI with the NRC to confirm an understanding of the information being requested. On February 8, 2012, the NRC provided a formal RAI required to complete its evaluation of the CR-3 EPU LAR (Reference 3).

The attachment, "Response to Request for Additional Information to Support NRC PRA Licensing Branch Technical Review of the CR-3 EPU LAR," provides the CR-3 formal response to the RAI needed to support the PRA Licensing Branch technical review of the CR-3 EPU LAR.

This correspondence contains no new regulatory commitments.

If you have any questions regarding this submittal, please contact Mr. Dan Westcott, Superintendent, Licensing and Regulatory Programs at (352) 563-4796.

Jon . Franfke V, e President rystal River Nuclear Plant JAF/gwe A O Progress Energy Florida, Inc.

Crystal River Nuclear Plant 15760 W. Powerline Street Crystal River, FL 34428

U.S. Nuclear Regulatory Commission Page 2 of 3 3F0312-01

Attachment:

Response to Request for Additional Information to Support NRC PRA Licensing Branch Technical Review of the CR-3 EPU LAR xc: NRR Project Manager Regional Administrator, Region II Senior Resident Inspector State Contact

U.S. Nuclear Regulatory Commission Page 3 of 3 3F0312-01 STATE OF FLORIDA COUNTY OF CITRUS Jon A. Franke states that he is the Vice President, 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 e le Jo A. Franke ice President Crystal River Nuclear Plant The foregoing document was acknowledged before me this Z___t day of 14,arcA- , 2012, by Jon A. Franke.

/7 (Print, type, or stamp Commissioned Name of Notary Public)

Personally Produced Known __ -OR- Identification

FLORIDA POWER CORPORATION CRYSTAL RIVER UNIT 3 DOCKET NUMBER 50-302 /LICENSE NUMBER DPR-72 ATTACHMENT RESPONSE TO REQUEST FOR ADDITIONAL INFORMATION TO SUPPORT NRC PRA LICENSING BRANCH TECHNICAL REVIEW OF THE CR-3 EPU LAR

U.S. Nuclear Regulatory Commission Attachment 3F0312-01 Page 1 of 22 RESPONSE TO REQUEST FOR ADDITIONAL INFORMATION TO SUPPORT NRC PRA LICENSING BRANCH TECHNICAL REVIEW OF THE CR-3 EPU LAR By letter dated June 15, 2011, Florida Power Corporation (FPC), doing business as Progress Energy Florida, Inc., requested a license amendment to increase the rated thermal power level of Crystal River Unit 3 (CR-3) from 2609 megawatts (MWt) to 3014 MWt (Reference 1). On January 3, 2012, via electronic mail, the NRC provided a draft request for additional information (RAI) needed to support the Probabilistic Risk Assessment (PRA) Licensing Branch (APLA) technical review of the CR-3 Extended Power Uprate (EPU) License Amendment Request (LAR). By teleconference on January 11, 2012, CR 3 discussed the draft RAI with the NRC to confirm an understanding of the information being requested. On February 8, 2012, the NRC provided a formal RAI required to complete its evaluation of the CR-3 EPU LAR. The following provides the CR-3 formal response to the RAI needed to support the APLA Branch technical review of the CR-3 EPU LAR. For tracking purposes, each item related to this RAI is uniquely identified as APLA X-Y, with X indicating the RAI set and Y indicating the sequential item number.

APLA RAIs

1. (APLA 1-1)

In your August 11, 2011, response (ADAMS Accession No. ML 1123A05 1), Appendix 1, ASME 2007 self assessment, observation FnO-SC-B1-2 indicates that the peer reviewer identified an incongruity in the amount of time operators have to initiate feed-and-bleed after a small break loss-of-coolant accident (SBLOCA). The EPU resolution includes an analysis for transient-initiated accident scenarios when reactor coolant pumps (RCP) are not tripped as well as a station blackout (SBO) analysis. The resolutiondoes not indicate the maximum amount of time available for operators to initiate feed and bleed cooling for SBLOCA post-EPU. The resolution implies that feed and bleed cooling must be performed within 30 minutes; however, the SBO analysis suggests 60 minutes. Please verify the amount of time available for operators to perform feed and bleed cooling following SBLOCA for pre-EPU and post-EPU; and provide an explanation for differences or similarities in pre-EPU and post-EPU results.

Response

For pre-EPU conditions, the CR-3 PRA analysis indicates 60 minutes to initiate feed and bleed (F&B) cooling following a SBLOCA concurrentwith a loss of secondary cooling to prevent core damage. The 60-minute F&B cooling initiation timing is based on the F&B cooling initiation timing during event sequences with a loss of forced Reactor Coolant System (RCS) flow. To prevent core damage considering EPU conditions, the EPU PRA analysis indicates F&B cooling must be initiated within 55 minutes following an event with a loss of forced RCS flow. As a result, the SBLOCA was also updated to reflect this timing change. As described in Section 2.13, "Risk Evaluation," of the EPU Technical Report (TR) (Reference 1, Attachments 5 and 7),

the pre-EPU 60-minute timing was determined using MAAP 3.0B and the post-EPU 55-minute timing was determined using MAAP 4.0.6. For events concurrent with a loss of secondary cooling and forced RCS flow maintained, the pre-EPU and post-EPU analyses indicate F&B cooling must be initiated within 30 minutes to prevent core damage.

U.S. Nuclear Regulatory Commission Attachment 3F0312-01 Page 2 of 22 As discussed with the NRC staff during a teleconference on January 11, 2012 regarding the PRA RAI, Table 1-1, "F&B Cooling Initiation Timing Pre- and Post-EPU," is a list of pre- and post-EPU timing information related to F&B cooling initiation for SBO, loss of feedwater (LOFW),

and SBLOCA events; as modeled in the CR-3 PRA, including the RCP status associated with each analysis.

Table 1-1 F&B Cooling Initiation Timing Pre- and Post-EPU Event Pre EPU (min.) Post EPU (min.) RCP Status SBO 60 55 RCP tripped LOFW 30 30 RCP running SBLOCA 60 55 RCP tripped

2. (APLA 1-2)

Please verify that all outstanding issues from the individual plant examinations (IPE) and individual plant external events examinations (IPEEE) highlighted in the NRC safety evaluation have been satisfied in subsequent PRA updates. If there are outstanding items from the IPE and IPEEE, explain the impact of the proposed EPU for those particular issues.

Response

FPC has performed a re-review of the CR-3 IPE and IPEEE and confirmed that the outstanding issues highlighted in the applicable NRC safety evaluations have been resolved or reflected in the current PRA model updates and there are no further outstanding items.

Outstanding issues for the CR-3 IPE identified in the NRC safety evaluation dated April 28, 1997 (Reference 2) were completed as documented in the NRC Safety Evaluation Report (SER)

Supplement dated June 30, 1998 (Reference 3).

As noted in the NRC SER dated January 11, 2001 (Reference 4), the NRC and associated contractors conducted a screening review of the CR-3 IPEEE. The NRC staff concluded that (1) the CR-3 IPEEE process is capable of identifying the most likely severe accidents and severe accident vulnerabilities from external events and (2) the CR-3 IPEEE has met the intent of Generic Letter 88-20, Supplement 4, "Individual Plant Examination of External Events (IPEEE) for Severe Accident Vulnerabilities - 10CFR 50.54(f)," and the resolution of specific generic safety issues discussed in the SER.

3. (APLA 1-3)

Per NRC RS-001, "Review Standard for Extended Power Uprates" (ADAMS Accession No. ML033640024), the licensee is required to provide findings of any industry or independent peer review. Please explain the focus of the 2009 peer review.

Response

The focus of the 2009 peer review was Initiating Events, Quantification, and Large Early Release Frequency (LERF). The Quantification element is a typical element included in peer reviews.

The Initiating Events and LERF elements were reviewed based on three issues: (1) the scope of PRA changes since the previous peer review; (2) changes in methodology used to close gaps

U.S. Nuclear Regulatory Commission Attachment 3F0312-01 Page 3 of 22 between the CR-3 PRA and the guidance of Regulatory Guide 1.200, "An Approach for Determining the Technical Adequacy of Probabilistic Risk Assessment Results for Risk-Informed Activities," Revision 1; and (3) the scope of the previous peer review did not include the LERF element.

4. (APLA 1-4)

Section 2.13.1 of the original license amendment request (LAR) dated June 15, 2011 (ADAMS Accession No. ML112070659), under section titled, "CR-3 Current Licensing Basis," paragraph four, states, "The current PSA [probabilistic safety analysis] model of record is based on internal events and internal flooding only." However, the introduction in Section 2.13.2 lists an at-power PSA model that includes "Fire." Please clarify this discrepancy.

Response

The sentence in paragraph four under heading titled, CR-3 Current Licensing Basis, in Section 2.13 of the EPU TR (Reference 1, Attachments 5 and 7), is referring to the CR-3 PRA model-of-record (MOR) and, as stated, consists of internal events and internal flooding. The PRA MOR is the current PRA model utilized for plant programs such as Maintenance Rule and on-line risk programs. The statement that CR-3 has an at-power PSA model that includes fire; under the heading titled, Introduction, in subsection 2.13.2 of Section 2.13 of the EPU TR (Reference 1, Attachments 5 and 7); is referring to a revised PRA model which evaluated the impact of EPU conditions and includes the recently developed Fire PRA and is not yet released as the CR-3 MOR.

5. (APLA 1-5)

The EPU may impact the frequency of a stuck open pressurizer power-operated relief valve (PORV) or safety relief valve, thus impacting the consequential loss-of-coolant accident (LOCA) frequency. In addition, pressurizer level can have larger variations due to the EPU. Please perform a sensitivity study that increases the PORV challenged frequency and provide the resulting delta risk metrics.

Response

It can be postulated that operation at EPU conditions could increase the potential for a transient that challenges a pressurizer PORV or pressurizer safety valve (PSV) thereby increasing the risk contribution of accident sequences involving a stuck open relief valve. However, operation at EPU conditions does not impact the underlying failure data associated with valves of this type failing to re-close since no changes are proposed to the valves themselves. The effect of EPU operation on sequences that involve a stuck open PORV or PSV can be shown by increasing the initiating event (IE) frequency in accident sequences that challenge these valves. Accident sequences that model a stuck open PORV or PSV were quantified with a 25% increase in the associated base IE values to show the effect of these sequences on core damage frequency (CDF). Accident sequences that include a stuck open PORV or PSV are listed in Table 5-1, "Accident Sequences with Stuck Open PORV or PSVs."

U.S. Nuclear Regulatory Commission Attachment 3F0312-01 Page 4 of 22 Table 5-1 Accident Sequences with Stuck Open PORV or PSVs Sequence Description TQU A transient with loss of RCS integrity and failure of high pressure injection (HPI); resulting in core damage.

TQX A transient with loss of RCS integrity. HPI is successful, but long-term cooling fails due to failure of high pressure recirculation (HPR); resulting in core damage.

TBQU A transient with an interruption of primary to secondary heat removal and loss of RCS integrity. HPI fails; resulting in core damage.

TBQX A transient with an interruption of primary to secondary heat removal and loss of RCS integrity. HPI succeeds, but HPR fails; resulting in core damage.

TBLWX A transient with an extended loss of heat removal via the steam generators. F&B cooling succeeds and feedwater is eventually restored.

At least one of the PSV sticks open and HPR fails. This leads to core damage.

Table 5-2, "Sensitivity Results," provides the calculated CDF contribution by each accident sequence involving a stuck open PORV or PSV and the overall calculated CDF at EPU conditions (OPCD). The accident CDF contributions were calculated using the post-EPU model with the base IE values and with the base IE values increased by 25%.

Table 5-2 Sensitivity Results EPU 25% IE Sequence Results Increase OP CD 3.38E-06 NA TQU 8.40E-08 1.05E-07 TQX 1.69E-08 2.12E-08 TBQU 1.61E-09 2.01E-09 TBQX 0* 0*

TBLWX 0* 0*

% of CDF 3 3.8

  • Results are below the truncation threshold.

As indicated in Table 5-2, the sensitivity study confirms that accident sequences that challenge the PORV and PSVs have an insignificant impact on the overall CR-3 CDF at EPU conditions.

6. (APLA 1-6)

Installation of the low pressure injection (LPI) cross-tie and the boron precipitation line result in an additional pathway with the potential for an interfacing systems LOCA (ISLOCA). Therefore the ISLOCA frequency was increased by seven percent. Please provide the basis for determining a 7 percent increase and perform a sensitivity study that shows the impact on core damage frequency and large early release frequency for a larger frequency.

U.S. Nuclear Regulatory Commission Attachment 3F0312-01 Page 5 of 22

Response

The ISLOCA analysis for the CR-3 PRA model explicitly evaluates potential pathways between the high pressure RCS piping and low pressure piping outside containment. The PRA analyses consider the following three potential ISLOCA pathways associated with EPU modifications to the LPI System:

1. Decay Heat Removal (DHR) System drop line,
2. LPI Train A injection, and
3. LPI Train B injection.

The DHR System drop line design currently contains two normally-closed, motor-operated valves (MOVs) in series between the high and low pressure boundary. As shown in Figure 3 of , "LPI Cross-Tie Modification," in Appendix E of the EPU TR (Reference 1, Attachments 5 and 7), the proposed LPI System hot leg injection (HLI) line connects to the drop line between the RCS hot leg and the first pressure isolation boundary MOV. As a result, the ISLOCA frequency for the DHR System drop line pathway is not affected by the LPI cross-tie modification. The DHR System drop line pathway continues to be the predominate contributor to the overall ISLOCA frequency with a value of 4.82E-09 per year.

For the pre-EPU configuration, each of the two LPI lines is independent of the other when considering an ISLOCA. For each LPI line, two series check valves along with the associated normally-closed LPI pump discharge MOV forms the high/low pressure boundary. Failure of both check valves and the LPI pump discharge MOV is needed for an ISLOCA. The total ISLOCA frequency for these two independent pathways, one per LPI train, is 3.35E-10 per year in the pre-EPU configuration.

For the post-EPU configuration, the LPI cross-tie line ensures that pressure is equalized between the two LPI lines. As shown in Figure 2 of Enclosure 1 in Appendix E of the EPU TR (Reference 1, Attachments 5 and 7), the proposed cross-tie line connects upstream of the two in-series check valves on each LPI line. The addition of this cross-tie line results in multiple ISLOCA pathways; failure of the two check valves in one LPI line exposes both LPI pump discharge MOVs to high pressure. Thus, failure of either LPI pump discharge MOV could result in an ISLOCA. Also, the addition of the HLI creates a new ISLOCA pathway from the RCS via two series check valves, two parallel normally-closed MOVs, and the LPI cross-tie line. Failure of both HLI check valves and either parallel MOV exposes both LPI pump discharge MOVs to high pressure. Thus, failure of either LPI pump discharge MOV could result in an ISLOCA. For the post-EPU configuration, the total frequency for these additional ISLOCA pathways is calculated to be 6.70E-10 per year.

The total frequency reported for the ISLOCA is the sum of the frequency values for the pathways as discussed above. For the pre-EPU configuration, the overall ISLOCA frequency is 5.16E-09 per year and for the post-EPU configuration, the overall ISLOCA frequency is 5.49E-09 per year. These values are the basis for the 7% increase shown for the ISLOCA IE frequency. This 7% IE increase was included in the PRA analysis performed for EPU and is accounted for in the CDF and LERF risk results provide in Table 2.13-2, "Risk Results Without Risk Reduction Modifications," of the EPU TR (Reference 1, Attachments 5 and 7). As shown in Table 2.13-3, "Contribution to CDF by Initiating Events," of the EPU TR (Reference 1, Attachments 5 and 7),

the contribution to overall CR-3 CDF from an ISLOCA remains 0.2% even with an IE increase of 7% for the ISLOCA event sequence.

U.S. Nuclear Regulatory Commission Attachment 3F0312-01 Page 6 of 22 Additionally, since an ISLOCA directly contributes to CDF and LERF, a 20% sensitivity case for ISLOCA IE would result in a 20% increase in the ISLOCA contribution to CDF and LERF.

Yet, the resulting ISLOCA contribution to CDF and LERF would continue to be approximately 0.2% and 5.0%, respectively. As such, a sensitivity study is unnecessary to show the impact an ISLOCA IE increase of 7% has on CDF and LERF.

7. (APLA 1-7)

Installation of the fast cooldown system (FCS) introduces an additional avenue for inadvertently opening the atmospheric dump valves (ADVs). Provide an explanation for why the value of the secondary line break and spurious actuation frequency remain the same between pre-EPU and post-EPU with the addition of FCS and inadequate core cooling mitigation (ICCM) systems.

Response

The frequency of secondary line breaks for the CR-3 PRA is based on a Bayesian update of generic failure data. Although spurious operation of the FCS could result in an event similar to an increase in steam flow, such an event is included in the existing data implicitly through spurious opening of an ADV or main steam safety valve (MSSV). Also, the Inadequate Core Cooling Mitigation System (ICCMS) design of a two-out-of-three instrument logic scheme minimizes the probability of spurious FCS/ADV actuation. Therefore, it is concluded that no changes to the frequency of the secondary line break or ADV spurious actuation is expected as a result of the addition of the FCS and ICCMS.

Nevertheless, a sensitivity study was performed to evaluate the combined uncertainty associated with various transient events including spurious opening of an ADV or MSSV. In this sensitivity study, the following IE values were simultaneously increased by 20 percent:

Reactor Trip, Loss of Main Feedwater, Loss of Offsite Power - Grid Centered, Loss of Offsite Power - Plant Centered, Loss of Offsite Power - Switchyard Centered, Loss of Offsite Power - Weather Centered, Excessive Feedwater, and Main Steam/Feedwater Line Break.

These changes resulted in a combined 3% increase in CDF. Since the main steam/feedwater line break event represents a small part of the total 3% increase, it is concluded that any uncertainty related to inadvertent ADV actuation, as a result of adding the FCS and the ICCMS, does not change the overall risk conclusions of the CR-3 EPU PRA analysis as cited in Section 2.13, of the EPU TR (Reference 1, Attachments 5 and 7).

8. (APLA 1-8)

The following questions address excessive feedwater transient:

A. In Section 2.13.2 of the original LAR under subject, "Excessive Feedwater," the licensee states, "The spurious actuation of the new ICCM system could impact emergency feedwater but has no impact on ICS [integrated control system] or the

U.S. Nuclear Regulatory Commission Attachment 3F0312-01 Page 7 of 22 MFW [main feedwater] System." Also in Section 1.2.7 of Appendix E of the original LAR, the licensee states that emergency feedwater requirements will be raised from 275 gallons per minute (gpm) to 330 gpm to satisfy increase in decay heat. The licensee is requested to explain why the addition of ICCM increases the actual excessive feedwater initiating event frequency and why the licensee chooses not to model this frequency change. Please provide a sensitivity study showing the effect on risk metrics.

B. If there were a transient that involved an increase in feedwater flow to the steam generator (SG), the SG water level may exceed the aspirator port level, thus preventing the preheating of the feedwater. If the increased thermal stresses on the tubes or shell wall are excessive, an SG tube rupture or steam generator shell failure accident could occur creating a LOCA. The initiating transient of concern involves an increase in feedwater flow. According to Appendix E of the original LAR, replacement generators were originally specified and designed to 3010 MWt and were reevaluated to EPU conditions at 3030 MWt, thereby raising additional concerns on the adequacy of the generators to address this transient. The NRC staff requests sensitivity studies for the conditional probability of failure given a feedwater overfeed event based on updated excessive feedwater frequencies for post-EPU, and current excessive feedwater frequencies for pre-EPU. In addition, characterize the impact on operator actions for excessive feedwater for post- and pre-EPU conditions.

Response

A. The excessive feedwater event noted in Section 2.13 of the EPU TR (Reference 1, Attachments 5 and 7) is and remains a function of the Main Feedwater (MFW) System and is not a function of the Emergency Feedwater (EFW) System. The flow rates for MFW are much higher than EFW and the flow-paths are different.

In addition, the new ICCMS does not interface with the MFW System. As a result, the addition of the ICCMS has no impact on the excessive feedwater IE frequency.

Nevertheless, a sensitivity study was performed to evaluate the combined uncertainty associated with various transient events including an excessive feedwater event. In this sensitivity study, the IE values were simultaneously increased by 20% and resulted in a combined 3% increase in CDF. Therefore, it is concluded that any uncertainty related to the frequency of an excessive feedwater event does not change the overall risk conclusions of the CR-3 EPU PRA analysis as cited in Section 2.13, of the EPU TR (Reference 1, Attachments 5 and 7).

B. The original specification for the CR-3 replacement once-through steam generators (OTSGs) contemplated an EPU, and specified a power of 3010 MWt. Subsequently, a power of 3030 MWt (core power of 3014 MWt plus RCP heat) was selected and the replacement OTSGs have been fully qualified for operation at 3030 MWt pursuant to a subsequent contract with the replacement OTSG supplier. In addition, the CR-3 replacement OTSGs are qualified for a normal tube-to-shell differential temperature of 907F with tubes hotter than the shell and the secondary side normal cooldown rate is 100°F per hour, for 240 design cycles of heatup and cooldown.

A transient that causes an increase in feedwater flow and causes reduction or loss of aspirating steam is not a CR-3 specific design transient. This type of transient is neither a

U.S. Nuclear Regulatory Commission Attachment 3F0312-01 Page 8 of 22 limiting event nor one that introduces significant stress on the OTSGs. During normal operation, feedwater is heated to approximately 460'F prior to entering the OTSGs and aspirating steam mixes with the feedwater in the OTSG shroud to heat the water to saturation; approximately 536°F at EPU conditions. If a loss of aspirating steam occurred due to a feedwater overfeed event, the water in the shroud area, downcomer, and the lower shell could be cooled to the equilibrium feedwater temperature; 4600 F.

The upper shell is at or near steam temperature; approximately 600'F. In this case, the average shell temperature would be approximately 530'F.

With an RCS TAVG of 582°F at EPU conditions, a 530'F average shell temperature resulting from a loss of aspirating steam is within the tube-to-shell temperature limit.

Also, since the average shell cooldown rate would be less than the OTSG secondary cooldown rate limit, the shell stress resulting from a feedwater overfeed transient would not exceed the qualified thermal stress limits of the replacement OTSGs.

Since a thermal transient as a result of feedwater overfeed event at EPU conditions is within the normal thermal design limits of the replacement OTSGs, the thermal stresses associated with the OTSG tubes and shell are not impacted by a feedwater control failure of this nature. Additionally, the event IE frequency and operator actions associated with excessive feedwater events are not altered as a result of EPU operation. Therefore, the probability of an OTSG failure is not increased as a result of a feedwater overfeed event at EPU conditions. Nevertheless, a sensitivity study was performed to evaluate the combined uncertainty associated with various transient events including an excessive feedwater event. In this sensitivity study, the IE values were simultaneously increased by 20% and resulted in a combined 3% increase in CDF. Therefore, it is concluded that any uncertainty related to the frequency of an excessive feedwater event does not change the probability of an OTSG failure.

9. (APLA 1-9)

In Section 1.3.1 of Appendix E of the original LAR, the licensee notes that the decay heat removal system performs LPI functions and supports active boron precipitation controls.

However, at EPU conditions, these two scenarios are not adequately supported by the current system design and performance. Therefore, the licensee plans to modify the system by cross-connecting the two trains inside the reactor building. Please explain how the licensee plans to address the decrease in defense-in-depth associated with this modification and provide additional detail on how success criteria was changed in the PRA to reflect two out of two requirements instead of one out of two requirements.

Response

Operation at EPU conditions requires modifications to the Decay Heat System to meet the LOCA analyses. Specifically, the LPI cross-tie ensures at least one LPI train is available for certain LOCA scenarios. The CR-3 safety analysis and PRA analysis success criteria continue to require one of two LPI subsystems at EPU conditions. The CR-3 PRA model has been updated to reflect plant modifications to the LPI System and consists of new basic events to reflect the potential failure of the new LPI cross-tie and the potential ISLOCA flow path associated with the long term HLI boron precipitation mitigation flow path.

U.S. Nuclear Regulatory Commission Attachment 3F0312-01 Page 9 of 22 The LPI cross-tie facilitates a more effective and less operator reliant boron precipitation mitigation strategy by replacing two existing partial-scope, active mitigation means with the HLI System. The general subject of boron precipitation mitigation was addressed by the NRC staff prior to acceptance of the CR-3 EPU LAR and is expected to be actively discussed during review by other technical branches. FPC proposes deferring additional information related to defense-in-depth associated with the LPI cross-tie modification until deterministic considerations are more fully resolved concerning boron precipitation mitigation methods.

10. (APLA 1-10)

In Section 2.13.2 of the original LAR under label, "Emergency Feedwater and Auxiliary Feedwater (PSA System Model EF)," the licensee notes that the EPU will require changes to ensure that the EFW pumps can deliver the required rated flow to the steam generators.

However, PRA best-estimate analyses performed for the EPU show that additional flow is not required to prevent core damage. The licensee is requested to provide additional detail that summarizes how the conclusion for not changing success criteria was reached. Additionally, under label, "Fast Cooldown System," the licensee also establishes that plant-specific PRA analysis does not require the FCS system for medium break LOCAs. Please provide the NRC staff a summary of the delta between the design basis analysis and the plant-specific PRA analysis for the success criteria determination of the FCS system.

Response

Emergency Feedwater and Auxiliary Feedwater In the EPU design basis analyses, the higher EFW System flow is assumed following a LOFW for the event duration. This additional flow is needed to meet applicable safety analysis acceptance criteria at EPU conditions. The design basis analysis acceptance criteria for a LOFW event are: less than 110% of reactor vessel design pressure, minimum departure from nucleate boiling ratio (DNBR) remains above the DNBR limit, and prevention of a liquid-solid pressurizer and subsequent relief of liquid through the PSVs. The EFW System flow increase needed for the design basis analyses is achieved by isolating EFW pump recirculation flow when recirculation is not needed to maintain minimum pump flow. The design basis analyses use the degraded EFW pump performance curve.

The success criteria for the PRA analyses were developed to show that core damage is prevented using best-estimate conditions. The PRA analyses use the nominal EFW pump performance curve. The EPU PRA success criteria were determined using the MAAP code and a core damage criterion of the hottest core node temperature exceeding 1800'F for a sustained period of time. Another difference between the EPU PRA and design basis analyses that affect the success criteria is the instrument uncertainty assumption. Using these criteria, the MAAP code analysis showed that nominal EFW System flow is adequate to prevent core damage. Therefore, the existing success criteria of one EFW pump to one steam generator for a LOFW event is not altered in the CR-3 PRA analysis and modeling EFW pump recirculation line isolation is not required as a result of EPU operation.

Fast Cooldown System The design basis LOCA analyses credit the FCS during specific spectrums of SBLOCAs to demonstrate compliance with 10 CFR 50.46 criteria by using the ADVs on the secondary side to

U.S. Nuclear Regulatory Commission Attachment 3F0312-01 Page 10 of 22 rapidly reduce RCS temperature and pressure assuring acceptable Emergency Core Cooling System (ECCS) performance. The LOCA analyses assume a core decay heat of 1.2 times the American Nuclear Society (ANS) Standard decay heat values as required by Paragraph 1.A.4 of 10 CFR 50, Appendix K and the best-estimate PRA analysis assumes a core decay heat of 1.0 times the ANS Standard decay heat values resulting in differing success criteria. Other differences between the EPU PRA and design basis analyses that affect LOCA success criteria include instrument uncertainty and ECCS pump degradation assumptions. As a result, the MAAP code analyses show that the FCS is not required to prevent core damage following a LOCA at EPU conditions. However, as noted in Section 2.13 of the EPU TR (Reference 1, Attachments 5 and 7), the FCS is conservatively included in the EPU PRA model for the medium break LOCA, which is considered a SBLOCA in the CR-3 safety analysis. Therefore, the FCS is included in the PRA model as a success criterion consistent with the EPU safety analysis.

11. (APLA 1-11)

EPU associated conditions may reduce the time to reach boron solubility limits in the core for medium and large LOCAs. This condition can result in boron precipitation on the fuel assemblies, which reduces heat transfer rates, and may lead to core damage. The potential impact on risk is from an increase in probability of the operators failing to initiate safety injection within this time period. Please characterize how this issue is addressed for CR-3 and any associated risk implications.

Response

Initiation of safety injection is automatic. Manual initiation of HLI for boron precipitation mitigation occurs later in the LOCA accident sequence so as to not interfere with safety injection timing. The proposed boron precipitation mitigation strategy is less complex and operator dependent than the current active boron precipitation control methods and thus has a positive impact on EPU risk considerations. Refer to Appendix D, "Core Boric Acid Dilution Control For CR-3 At EPU Conditions," (Reference 1, Attachment 5) and Enclosure 1, "LPI Cross-Tie Modification," in Appendix E (Reference 1, Attachments 5 and 7) of the EPU TR for additional details associated with HLI boron precipitation mitigation.

Operator timing and other aspects will be vetted as part of other NRC staff technical branch reviews of the proposed boron precipitation mitigation strategy and will be the subject of additional dialogue with these branches. Thus, FPC proposes deferring quantitative probabilistic efforts associated with boron precipitation mitigation strategy until deterministic considerations are more fully resolved.

12, (APLA 1-12)

Of 70 unique post-initiator operator actions developed for the CR-3 PRA, the licensee states that only 7 were impacted by the EPU. Based on previous EPU precedent, the NRC staff finds the number of human error probabilities (HEPs) impacted by the EPU to be extremely low. The licensee is requested to provide a change in HEP assessment from pre- to post-EPU for all operator actions impacted by the EPU that are either under 30 minutes; have Fussel Vessly (FV)

> .005; or Risk Achievement Worth >2.

U.S. Nuclear Regulatory Commission Attachment 3F0312-01 Page 11 of 22

Response

The following tables provide the post-initiator human failure events (HFEs) that exceed the importance measures criteria; Fussell-Vesely (F-V) > 0.005 and Risk Achievement Worth (RAW) > 2.0; and the HFEs that must be completed within 30 minutes of the initial reactor trip.

Table 12-1, "HFEs - 30 Minutes or Less," provides a list and description of HFEs that met the timing threshold for high risk and includes the associated required completion time of each HFE.

For listed HFE sequences with multiple iterations, a suffix of (30) indicates the specific HFE sequence iteration that assumed an operator action of 30 minutes in the PRA model.

Table 12-2, "HFEs with F-V > 0.005," and Table 12-3, "HFEs with RAW > 2.0," provide a list and description of HFEs that exceeded the importance measures threshold for high risk.

Selection of these HFEs was based on the results of the EPU PRA cutset report.

With the exception of HFE sequence QHUEFW9Y, the CR-3 EPU human reliability analysis (HRA) concluded that EPU operation does not alter HEPs for the existing pre-EPU HFE sequences that met the timing and importance measures threshold for high risk. Table 12-4, "High Risk HFEs - EPU Impact," provides a discussion for each HFE identified with F-V > 0.005, RAW > 2.0, or action time of < 30 minutes and includes the logic for concluding the HFE was not affected by the EPU, as applicable. Also, Table 2.13-1 of the EPU TR (Reference 1, Attachments 5 and 7) shows an HEP increase of 5% for HFE sequence QHUEFW9Y. Although the existing operator action to trip the RCPs on a loss of subcooling margin (SCM) (RHURCP2Y) was not included in the pre-EPU PRA model, HFE sequence RHURCP2Y is added to the post-EPU PRA model for completeness and is included in applicable tables below.

Table 12-1 HFEs - 30 Minutes or Less Required HFE Sequence Completion Description Time (min)

AHU4KVXY (30) 30 Operators Fail To Power 4kv ES Bus From Alternate Offsite Source AHUE3ABY (30) 30 Operators Fail To Switch Power Source To ES MCC 3AB AHUEG1CY (30) 30 Operators Fail To Start And Align EDGD-1C To Rx Aux Bus 3 AHUEGDGY (30) 30 Operators Fail To Start EGDG Manually AHUMT2HY (30) 30 Operators Fail To Align MTSW-2G To MTSW-2H HHUINJAY 13 Operators Fail To Switch MUV-23/24 To Backup Power HHUINJBY 13 Operators Fail To Switch MUV-25/26 To Backup Power HHUMANUY 13 Operators Fail To Manually Initiate HPI HHUMBACY (30) 30 Operators Fail To Switch MUP-IB Power Source HHUMPSBY 13 Operators Fail To Start Non-ES Selected Makeup Pump Operators Fail To Start Non-ES Selected Makeup Pump HHUMPSBY (T)(30) 30 (Transients)

HHUTHR1Y 12 Operators Fail To Control HPI Following Spurious HPI (T7)

HHUTHR2Y 25 Operators Fail To Control HPI Before Liquid Relief

U.S. Nuclear Regulatory Commission Attachment 3F0312-01 Page 12 of 22 Required HFE Sequence Completion Description Time (min)

HHUTHRTY 15 Operators Fail To Control HPI Flow After ES Actuation Due To Overcooling LHULPRCY 30 Operators Fail To Switch From Low Pressure Injection To Recirculation Following Large LOCA PHUMTBVY (30) 30 Operators Fail To Manually Control TBVs QHUEFP1Y (30) 30 Operators Fail To Start EFP-1 QHUEFWMY (30) 30 Operators Fail To Manually Operate EFW Control Valves QHUFW7EY 15 Operators Fail To Start Fwp-7 Before PORV Lifts QHUMSIVY (30) 30 Operators Fail To Manually Close MSIV RHUMRXTY 0.67 Operators Fail To Manually Trip Reactor RHURCP2Y I Failure To Trip RCP Upon Loss Of SCM RHURCPTY 10 Operators Fail To Trip RCPs Given No Seal Cooling Or Injection VHUAHFSY 30 Operators Fail To Manually Start Standby RB Cooler Table 12-2 HFEs with F-V > 0.005 HFE Sequence Description AHUEGICY (30) Operators Fail To Start And Align EGDG-1C AHUMT2HY (30) Operators Fail To Align MTSW-2G To MTSW-2H FHUF6A1Y Operator Fails To Isolate Flood F6a (Case 1)

FHUF6A2Y Operator Fails To Isolate Flood F6a (Case 2)

FHUF6A3Y Operator Fails To Isolate Flood F6a (Case 3)

HHUHPRCY Operators Fail To Switch From High Pressure Injection To Recirculation HHUINJBY Operators Fail To Switch MUV-25/26 To Backup Power HHUMPSBY Operator Fails To Start Standby Makeup Pump HHUXTYSR Operators Fail Crosstie MU Suction HHUTHR2Y Operators Fail To Control HPI Before Liquid Relief LHULPRCY Operators Fail To Go To Low Pressure Recirculation QHUEFT2Y Operators Fail To Crosstie EFW Sources QHUFW7EY Operators Fail To Start FWP-7 Before PORV Lifts QHUFWP7Y Operators Fail To Start FWP-7 RHUCOOLY Operators Fail Cooldown Via OTSG RHUDHRCY Operators Fail To Initiate DHR Following SGTR RHUPORVY Operators Fails To Open PORV For Pressure Relief RHURCPTY Operators Fail To Trip RCPs Given No Seal Injection Or Cooling SHURWP1Y Operators Fail To Start RWP Manually WHUBWSTY Operators Fail To Attempt BWST Refill

U.S. Nuclear Regulatory Commission Attachment 3F0312-01 Page 13 of 22 Table 12-3 HFEs with RAW > 2.0 HFE Sequence Description AHUEGICY (30) Operators Fail To Start And Align EGDG-1C FHUF101Y Operator Fails To Isolate Flood F10 (Case 1)

FHUF6A1Y Operator Fails To Isolate Flood F6a (Case 1)

FHUF6A2Y Operator Fails To Isolate Flood F6a (Case 2)

HHUHPRCY Operators Fail To Switch From High Pressure Injection To Recirculation JHUCHPARRRM Operators Fail To Start Ventilation Systems LHULPRCY Operators Fail To Go To Low Pressure Recirculation QHUAFSUY Operators Fail To Switch AFW (FWP-7) Suction QHUEFT2Y Operators Fail To Crosstie EFW Sources QHUEFW9Y Operators Fail To Raise OTSGs Level RHUCOOLY Operators Fail Cooldown Via OTSG RHURCP2Y Failure To Trip RCP Upon Loss Of SCM WHUBWSTY Operators Fail To Attempt BWST Refill Table 12-4 High Risk HFEs - EPU Impact HFE Sequence EPU Impact These HFEs represent failure to restore power Engineered Safeguards AHU4KVXY (30) (ES) buses from alternate or emergency sources. Timing for these AHUE3ABY (3 0) actions is based on the time needed to restore core cooling after an AHUEGICY (30) initial loss. Automatic actuation of HPI on high reactor building AHUEGDGY (30) pressure ensures adequate core cooling through F&B cooling; therefore, AHUMT2HY (3 0) restoration of power to a de-energized ES bus will prevent core damage.

AHUMT2HY (30) MAAP analyses performed for EPU conditions show that core damage HHUMBACY (30) will not occur when F&B cooling is initiated within 30 minutes H(UMP3BY consistent with the pre-EPU MAAP analyses. Therefore, the timing for (T)(30) these events and the associated HEPs are not altered as a result of EPU operation.

These HFEs represent failure to isolate various flooding scenarios before level exceeds the specified level value. The volume of water, FHUF1O1Y which is the basis for the time available for operator action, is FHUF6A1! Y determined by the physical size of the plant buildings. Since the FHUF6A2Y dimensions of the buildings do not change as a result of EPU operation, the timing for these events and the associated HEPs are not altered as a result of EPU operation.

These HFEs represent failure to transfer the ECCS from injection mode to recirculation mode following a LOCA. The timing for these events is derived from the time for the associated ECCS pumps to drain the HHUHPRCY borated water storage tank (BWST) from 20 ft. to 7 ft. The timing to LHULPRCY drain the BWST is based on the design flow of the ECCS pumps and is conservative with respect to the ECCS flow at EPU conditions.

Therefore, the timing for these events and the associated HEPs are not altered as a result of EPU operation.

U.S. Nuclear Regulatory Commission Attachment 3F0312-01 Page 14 of 22 HFE Sequence EPU Impact These HFEs represent failure to supply alternate power to HPI System injection valves in order to maximize flow delivered to the reactor after a LOCA. For pre-EPU conditions, the HEPs were analyzed using a 13-HHUINJAY minute window. For post-EPU conditions, the MAAP analyses were HHUINJBY performed using a newer version of the code and the results showed a time window slightly longer. Since the time window available to perform these actions is relatively short and the updated MAAP analyses showed a slight increase in time available, no change is made to these HEPs as a result of EPU operation.

These HFEs represent failure to either manually actuate HPI following a failure of automatic actuation or to manually start the non-ES selected HPI pump following initiation of a LOCA. For pre-EPU conditions, the HEPs were analyzed using a 13-minute window. For post-EPU HHUMANUY conditions, the MAAP analyses were performed using a newer version HHUMPSBY of the code and the results showed a slightly longer time window. Since the time window available to perform these actions is relatively short and the updated MAAP analyses showed a slight increase in time available, no change is made to these HEPs as a result of EPU operation.

This HFE represents failure to control HPI flow after a spurious HPI actuation. Timing of this action is based on the time needed to prevent challenging the PORV on steam overpressure. For pre-EPU conditions, the HEP was analyzed using a 12-minute window. For post-EPU HHUTHR1Y conditions, the MAAP analyses were performed using a newer version of the code and the results showed a slightly longer time window. Since the time window available to perform the action is relatively short and the updated MAAP analyses showed a slight increase in time available, no change is made to this HEP as a result of EPU operation.

This HFE represents failure to control HPI flow after a spurious HPI actuation. Timing of this action is based on the time needed to prevent liquid release through the PORV. For pre-EPU conditions, the HEP was analyzed using a 25-minute window. For post-EPU conditions, the HHUTHR2Y MAAP analyses were performed using a newer version of the code and the results showed a slightly longer time window. Since the time window available to perform the action is relatively short and the updated MAAP analyses showed a slight increase in time available, no change is made to this HEP as a result of EPU operation.

This HFE represents failure to control HPI flow after an overcooling event to prevent challenging the PORV. The timing used for this event during development of the HRA was based on discussions with HHUTHRTY experienced operators and is judged to be conservative; 15 minutes to control HPI flow following ES actuation during an RCS overcooling event. The timing for this event and the associated HEP are not altered as a result of EPU operation.

This HFE represents a recovery action for the standby HPI pump by HHUXTYSR swapping the suction supply from the makeup tank (MUT) to the BWST. Since the standby HPI pump does not automatically swap to the BWST, the pump will eventually drain the MUT and fail due to loss of

U.S. Nuclear Regulatory Commission Attachment 3F0312-01 Page 15 of 22 HFE Sequence EPU Impact suction, unless operators stop it upon receiving the MUT low-low level alarm. The timing to drain the MUT prior to failure of the standby HPI pump is classified as short and is not reclassified as a result of EPU operation. As a result, the associated HEP for this event is not altered as a result of EPU operation.

This HFE represents failure to close stuck open turbine bypass valves.

If the turbine bypass valves do not close, then loss of steam pressure to the turbine-driven EFW pump could occur reducing the pump flow required for mitigation. Timing for the action is based on the time PHUMTBVY (30) needed to restore core cooling after an initial loss. MAAP analyses performed for EPU conditions show that core damage will not occur when F&B cooling is initiated within 30 minutes consistent with the pre-EPU MAAP analysis. Therefore, the timing for this event and the associated HEP are not altered as a result of EPU operation.

This HFE represents failure to swap the auxiliary feedwater pump (FWP-7) suction to a long-term water supply when the condensate storage tank (CST) inventory is depleted. The timing for this event is the approximate time to drain the CST from 20 feet to the minimum QHUAFSUY required level and is estimated to be at least six hours. Although some change in this time could be expected because of the higher core decay heat at EPU conditions, the current action time is considered acceptable for the HRA. Therefore, the timing for this event and the associated HEP are not altered as a result of EPU operation.

These HFEs represent failure to manually start the motor-driven EFW pump and failure to manually operate EFW flow control valves. Timing for these actions are based on the time needed to restore core cooling QHUEFP1Y (30) after an initial loss. MAAP analyses performed for post-EPU conditions QHUEFWMY (30) show that show that core damage will not occur when F&B cooling is initiated within 30 minutes consistent with the pre-EPU MAAP analysis. Therefore, the timing for these events and the associated HEP are not altered as a result of EPU operation.

This HFE represents failure to swap the EFW pump suctions to ensure a long-term water supply to the EFW pumps. The timing for this event is based upon the approximate time to drain the emergency feedwater tank QHUEFT2Y and is estimated to be more than four hours. Although some change in this time could be expected due to the higher core decay heat at EPU conditions, the current action time is considered acceptable for the HRA. Therefore, the timing for this event and the associated HEP are not altered as a result of EPU operation.

This HFE represents a failure to raise the steam generator level to the target value following failure of the automatic feature upon a loss of QHUEFW9Y SCM. This HFE sequence was modified due to the automation of this feature and the associated HEP results are reported in Section 2.13 of the EPU TR (Reference 1, Attachments 5 and 7).

This HFE represents, following a transient event, failure to manually QHUFW7EY start FWP-7 before an RCS overpressure event results in a challenge to the PORV. For pre-EPU conditions, the HEP was analyzed using a 15-

U.S. Nuclear Regulatory Commission Attachment 3F0312-01 Page 16 of 22 HFE Sequence EPU Impact minute window. For post-EPU conditions, the MAAP analyses were performed using a newer version of the code and the results showed a slightly longer time window. Since the time window available to perform the action is relatively short and the updated MAAP analyses showed a slight increase in time available, no change is made to this HEP as a result of EPU operation.

This HFE represents, following a transient event, failure to manually start FWP-7 after secondary cooling has been maintained for one hour using the EFW pumps. After EFW flow is lost, FWP-7 must be started before RCS overpressure results in a challenge to the PORV. For pre-EPU conditions, the HEP was analyzed using a 32-minute window. For QHUFWP7Y post-EPU conditions, the MAAP analyses were performed using a newer version of the code and the results showed a slightly longer time window. Since the time window available to perform the action is relatively short and the updated MAAP analyses showed a slight increase in time available, no change is made to this HEP as a result of EPU operation.

This HFE represents a failure to close the main steam isolation valves if the main turbine fails to trip following a reactor trip. Timing for the action is based on the time needed to restore core cooling after an initial QHUMSIVY (30) loss. MAAP analyses performed for EPU conditions show that core damage will not occur when F&B cooling is initiated within 30 minutes consistent with the pre-EPU MAAP analysis. Therefore, the timing for this event and the associated HEP are not altered as a result of EPU operation.

These HFEs represent failure of actions following a steam generator tube rupture event; i.e., failure to: cooldown and depressurize the RCS, establish decay heat removal, open the PORV, or provide makeup to the RHUCOOLY BWST. The timing for these events is based on the time to deplete the RHUDHRCY BWST and is assumed to be several hours. This draindown time is RHUPORVY governed by HPI System flow and cooldown rates. Although some WHUBWSTY change in this draindown time could be expected at EPU conditions, the current action times are considered acceptable for the HRA. Therefore, the timing for these events and the associated HEPs are not altered as a result of EPU operation.

This HFE represents a failure to manually trip the reactor if an automatic trip fails. Timing for this HRA is based upon BAW-10099, RHUMRXTY Revision 1, "Analysis of B&W NSSS Response to ATWS." Therefore, the timing for this event and the associated HEP are not altered as a result of EPU operation.

This HFE represents a failure to trip the RCPs following a loss of seal RHURCPTY cooling. RCP seal cooling requirements are not affected by operation at EPU conditions. Therefore, the timing for this event and the associated HEP are not altered as a result of EPU operation.

This HFE represents a failure to start the non-ES raw water pump after SHURWP1Y failure of associated ES powered raw water pumps. For pre-EPU conditions, the timing for this action is based upon plant experience,

U.S. Nuclear Regulatory Commission Attachment 3F0312-01 Page 17 of 22 HFE Sequence EPU Impact assumed to be greater than three hours, and is considered conservative.

Therefore, the timing for this event and the associated HEP are not altered as a result of EPU operation.

This HFE is added to the CR-3 PRA model and represents a failure to RHURCP2Y trip the RCPs following failure of the new automated feature upon a loss of SCM. The HFE was not included in the pre-EPU model. The HEP calculated for this new HFE is 3.OE-4.

This HFE represents a failure to start the standby reactor building cooler. This action is assumed to be required within 30 minutes of the initial signal for high reactor building pressure. However, the time to receive the initial signal is only weakly dependent on power level.

Therefore, the timing for this event and the associated HEP are not altered as a result of EPU operation.

13. (APLA 1-13)

Please describe any new credited operator actions as a result of the EPU. (i.e., new operator action to lock out the ADVs actuation in event of fire).

Response

The following new operator actions are added as a result of CR-3 operation at EPU conditions:

1. Terminate fluid additions to the Makeup and Purification System from the boric acid storage tanks, reactor coolant bleed tanks, and Demineralized Water System following a loss of SCM;
2. Bypass FCS during an SBO;
3. Initiate HLI boron mitigation flowpath during a LOCA; and
4. Override ADV upon confirmation of a fire in the control complex.

Refer to Section 2.11.1, "Human Factors," of the EPU TR (Reference 1, Attachments 5 and 7) for a general description of these operator actions. The boric acid storage tanks, reactor coolant bleed tanks, and Demineralized Water System are not modeled flowpaths in the CR-3 PRA model; therefore, operator action to isolate these water sources is not included in the EPU PRA model. The operator action of bypassing the FCS during an SBO is not modeled in the PRA because the bypass action is an action needed to reach cold shutdown conditions and is subsumed by the current PRA operator action (RHUCOOLY) to cool down to DHR conditions.

Boron precipitation mitigation is not considered a LOCA success criterion in the CR-3 PRA; therefore, initiation of the HLI boron mitigation flowpath is not included in the EPU PRA model.

As discussed in Section 2.13 of the EPU TR (Reference 1, Attachments 5 and7), the timing analysis of operator action to override ADV actuation has not been performed since the cable routing and component location has not been finalized. Therefore, a fire HRA has not been performed for this HFE. Since a formal HRA has not been performed for the ADV override feature, the ADV override feature will be preemptively actuated by control room operators: (1) upon confirmation of a control complex fire; or (2) prior to a control room evacuation.

Also, the current operator actions of tripping the RCPs and raising the steam generator levels to the inadequate SCM value on a loss of SCM are being automated to actuate via the new ICCMS.

U.S. Nuclear Regulatory Commission Attachment 3F0312-01 Page 18 of 22 As noted in Table 2.13.1 of the EPU TR, the operator action of raising steam generator level following a loss of SCM is considered a recovery HFE in the EPU PRA analysis. The existing operator action to trip the RCPs on a loss of SCM is added to the EPU PRA model as a recovery HFE. Although not included in the current pre-EPU PRA model, the action is added to the EPU PRA model for completeness with the new automatic trip provided by ICCMS.

14. (APLA 1-14)

The human reliability analysis (HRA) assumes a reduction in system time window from 60 to 55 minutes (approximately 9 percent). This is less than the approximately 15 percent increase in power sought by the EPU. Please describe the analysis for how the system time window is uniformly reduced for all impacted HEPs by approximately 9 percent.

Response

For each of the HFEs where the system time window was reduced from 60 to 55 minutes, the modeled actions are associated with restoring systems used to remove core decay heat. The latest time that F&B cooling could be initiated to prevent core damage following a loss of offsite power and a loss of EFW is the basis for this system time window. As described in Section 2.13 of the EPU TR (Reference 1, Attachments 5 and 7), a pre-EPU system time window of 60 minutes was determined using MAAP 3.OB and a post-EPU system time window of 55 minutes was determined using MAAP 4.0.6. Core power is one of several factors that influence the system time window value. Therefore, the percent increase in core power is not equivalent to the percent of decrease in the system time window value and a 55-minute time window (9%

reduction from pre-EPU timing) is considered acceptable for use in the HRA for the affected HFEs.

15. (APLA 1-15)

Section 2.13.2 of the original LAR under label, "HRA Dependency Evaluation", notes that none of the timing changes expected for post-EPU conditions affected the dependency level. The following sentence depicts that some of the dependent HEPs were updated and included in the quantification process. Please provide the pre- and post-EPU dependent HEPs that changed; similar to Table 2.13-1 of the original LAR.

Response

The statement concerning the dependency level refers to how the success of the second HFE in a series of multiple events is dependent on the success of the first HFE event. Some of the attributes that determine the level of dependency include common procedures or common cues for the operator action. When the second HFE failure is assured to occur given failure of the first HFE in time, the two HFEs are considered to be dependent. Likewise, if there are no common attributes, the two HFEs have a low or zero dependency. Different relationship factors are applied for each dependence level and these are the factors referred to as not changing as a result of EPU operation.

The following sentence in Section 2.13 of the EPU TR (Reference 1, Attachments 5 and 7) that the updated dependent HEPs were included intended to convey that the HEP calculations for dependent HFEs were also updated to account for the new HEP values for the HFEs listed in Table 2.13-1.

U.S. Nuclear Regulatory Commission Attachment 3F0312-01 Page 19 of 22 Table 15-1, "HRA Recovery HEPs Pre- and Post-EPU," below provides the change in HEP for recovery actions as a result of EPU operation. Each recovery listed in the table represents the combined failure probability of some combination of events in an individual cutset. These HEP changes associated with the recovery actions are due to the updated HEPs listed in Table 2.13-1.

The HFE that resulted in a change to the dependent recovery action HEP is indicated in BOLD text in the "HFE Combinations" column.

Table 15-1 HRA Recovery HEPs Pre- and Post-EPU Recovery Post EPU Pre EPU  % Change HFE Combinations AHU4KVNZ 1.40E-02 8.70E-03 37.9% AHU4KVXY AHUE3ANZ 1.60E-02 1.1 OE-02 31.3% AHUE3ABY AHUEGDNZ 1.40E-01 9.90E-02 29.3% AHUEGDGY AHUMT2NZ 2.20E-01 1.60E-01 27.3% AHUMT2HY HHUMBANZ 2.40E-01 1.l0E-01 54.2% HHUMBACY HHUMPSNZ 5.00E-03 4.20E-03 16.0% HHUMPSBY ZHUAA02Z 4.42E-03 4.40E-03 0.5% AHUEGICY, AHUE3ABY ZHUAH02Z 3.16E-03 2.17E-03 31.3% AHUE3ABY, HHUINJAY ZHUAH04Z 9.92E-03 6.11E-03 38.4% AHUE3ABY, HHUMBACY ZHUAH06Z 2.35E-03 1.61E-03 31.5% AHUE3ABY, HHUMPSBY ZHUAH1OZ 2.44E-03 1.68E-03 31.1% AHUE3ABY, HHUTHR1Y ZHUAH12Z 2.68E-03 1.84E-03 31.3% AHUE3ABY, HHUTHR2Y ZHUAH14Z 5.17E-03 3.55E-03 31.3% AHUE3ABY, HHUTHRTY ZHUAH16Z 4.35E-02 3.16E-02 27.4% AHUMT2HY, HHUINJBY ZHUAH18Z 3.24E-02 2.34E-02 27.8% AHUMT2HY, HHUMPSBY ZHUAH22Z 3.69E-02 2.68E-02 27.4% AHUMT2HY, HHUTHR2Y ZHUAH24Z 7.1OE-02 5.17E-02 27.2% AHUMT2HY, HHUTHRTY ZHUAH26Z 4.35E-05 3.65E-05 16.1% AHUEGICY, HHUMPSBY ZHUAH3IZ 1.32E-02 9.1 1E-03 31.0% AHUE3ABY, HIHUXTYSR ZHUAS02Z 1.80E-03 1.24E-03 31.1% AHUE3ABY, SHUMCNSY ZHUAS04Z 8.61E-04 5.92E-04 31.2% AHUE3ABY, SHURWP1Y ZHUAS06Z 1.18E-02 8.55E-03 27.5% AHUMT2HY, SHUMADCY ZHUAS08Z 1.18E-02 8.61E-03 27.0% AHUMT2HY, SHURWP1Y ZHUH102Z 4.75E-02 2.17E-02 54.3% HHUMBACY, HHUINJAY ZHUHH05Z 2.66E-03 2.23E-03 16.2% HHUMPSBY, HHUINJAY ZHUHH08Z 2.66E-03 2.23E-03 16.2% HHUMPSBY, HHUINJBY ZHUHS 11Z 5.6 1E-04 4.73E-04 15.7% HHUMPSBY, SHUMCNSY ZHUHS14Z 2.69E-04 2.26E-04 16.0% HHUMPSBY, SHURWPIY ZHUHS17Z 2.51E-04 2.11E-04 15.9% HHUMPSBY, SHUSWP1Y ZHUJA02Z 3.26E-04 2.55E-04 21.8% JHUCHPARRRM, AHUMT2HY ZHUJA04Z 3.26E-04 2.55E-04 21.8% JHUCHPARRR M, AHUMT2HY ZHUJZO5Z 1.36E-04 1.14E-04 16.2% JHUCHPARRR M, HHUMPSBY, HHUINJBY ZHUJZ06Z 1.36E-04 1.14E-04 16.2% JHUCHPARRR M, HHUMPSBY, HHUINJBY ZHUQA02Z 3.46E-04 3.20E-04 7.5% QHUFWP7Y, AHUE3ABY

U.S. Nuclear Regulatory Commission Attachment 3F0312-01 Page 20 of 22 Recovery Post EPU Pre EPU  % Change HFE Combinations ZHUQH03Z 1.96E-02 1.95E-02 0.5% QHUA204R, HHUMPSBY ZHUQH07Z 2.80E-04 2.78E-04 0.7% QHUEFP1 Y, HHUMPSBY ZHUQH12Z 1.85E-03 1.26E-03 31.9% QHUFWP7Y, HHUMBACY ZHIUQH15Z 2.90E-04 2.86E-04 1.4% QHUFWP7Y, HHUMPSBY ZHURA02Z 1.49E-02 1.45E-02 2.7% RHURCPTY, AHUE3ABY ZHURA04Z 8.07E-04 5.55E-04 31.2% RHURCP1Y, AHUE3ABY ZHUZH1OZ 4.35E-05 3.65E-05 16.1% HHUINJBY, AHUEGICY, HHUMPSBY

16. (APLA 1-16)

Table 2.13-1 of the original LAR provides the base and updated values for impacted human error probabilities. The NRC staff is requesting the licensee to review their analyses and explain why a 5-minute reduction in system time (approximately 9 percent) increases the operator failure probability in some cases by over 115 percent. Also, basic event QHUEFW9Y requires the operators to complete the action within a 10-minute time window. How does that correspond to the 60-minute base value and 55-minute updated value?

Response

The HRA associated with the HFEs identified in Table 2.13-1 of the EPU TR (Reference 1, Attachments 5 and 7) were updated and evaluated using standard HRA methods as implemented in the Electric Power Research Institute HRA calculator. The final value of any HEP value is a function of many variables, one of which is time. However, an HEP does not directly correlate to the total time available to perform an action. The HEP is strongly influenced by the time required to perform an action (execution time) and the time needed for operators to respond to the event (median response time). Additional time beyond the sum of the execution and median response times may be used for recovery of failures. The events listed in Table 2.13-1 have different execution and median response times. For events where minimal additional time is available for recovery beyond the execution and median response times, a small change in timing (e.g., a reduction from 60 minutes to 55 minutes) results in a larger increase in the associated HEP.

Specifically regarding the HFE with an increase of 118%, a failure to switch the power source of Makeup Pump 11B (HHUMBACY (60)), the median response time is 10 minutes and execution time is 40 minutes. Therefore, as a result of EPU operation, the additional time available for recovery is reduced; 10 minutes to 5 minutes. This HRA is more sensitive to the change in timing than others because the execution portion takes most of the overall time window not allowing much credit for a recovery on the cognition errors.

Regarding basic event QHUEFW9Y, the EPU safety analyses assume 10 minutes to raise steam generator level to the target value in response to a loss of SCM. However, failure of this action modeled in QHUEFW9Y does not result in core damage for 55 minutes from event initiation.

As a result, approximately 45 minutes is available for recovery. Note that the operator action represented by QHUEFW9Y is obviated by the new ICCMS, which automates the 10 minute action. Therefore, this operator action is not required in the EPU PRA model. However, the QHUEFW9Y operator action is retained in the model as a recovery action upon a failure of automatically raising steam generator level to the target value on a loss of SCM. The timing for this HFE is based on the updated F&B cooling initiation timing of 55 minutes.

U.S. Nuclear Regulatory Commission Attachment 3F0312-01 Page 21 of 22

17. (APLA 1-17)

Due to increased decay heat during EPU operations, additional PORVs may be required for successful feed-and-bleed, especially if charging is unavailable. Describe changes in success criteria for successful feed-and-bleed pre-EPU and post-EPU, and the resulting risk implications.

Response

At CR-3, the make-up, charging, and HPI functions are performed by the same pumps; makeup pumps. Thus, if the charging function is unavailable, F&B cooling is also unavailable since the same pumps and injection flowpath that are used for the charging function are also used for the F&B cooling function. An F&B cooling analysis, assuming the existing CR-3 systems and assuming the higher core decay heat as a result of EPU conditions, was performed and shows that core damage is prevented with the current pressurizer PORV configuration. Therefore, no changes to the success criteria of the pressurizer PORV are required for successful F&B cooling as a result of EPU operation.

18. (APLA 1-18)

The licensee has not performed a probabilistic seismic analysis for the plant, therefore, please verify that all structural plant modifications and anchoring of all replacement components for EPU will have the same or greater seismic capability than the current design basis.

Response

As required by the CR-3 engineering change process; structures, systems, and components modified or replaced as a result of EPU will meet or exceed the applicable seismic design and qualification requirements specified by the current CR-3 design and licensing basis as described in the CR-3 Final Safety Analysis Report and applicable plant design control documents. These engineering change process requirements ensure that all structural plant modifications, including the anchoring of components, will continue to meet the seismic capability required by the current design and licensing basis, as applicable.

19. (APLA 1-19)

Table 2.13-2 describes risk results without risk reduction modifications. Briefly describe all risk reduction modifications implemented as part of the EPU. If available, the NRC staff requests a table showing risk results with risk reduction modifications.

Response

There are no specific risk reduction modifications planned for the CR-3 EPU. Several EPU modifications have positive risk reduction impacts; however, each of these modifications are being installed for deterministic reasons and have not been quantitatively assessed for CR-3 risk reduction impacts. Automation of the actions to trip the RCPs and raise OTSG levels to the target level following a loss of SCM is estimated to reduce CDF as follows:

Automated Action CDF Reduction Trip RCP on loss of SCM 2E-08 Raise OTSG level on loss of SCM 4E-07

U.S. Nuclear Regulatory Commission Attachment 3F0312-01 Page 22 of 22

20. (APLA 1-20)

Describe how the EPU affects the ability of the operator to close containment during an outage in the event of loss of shutdown cooling. Please provide any changes to the existing guidance/procedures.

Response

The effect of EPU operation on a loss of DHR event was evaluated and reported in Section 2.8.7.1, "Loss of Decay Heat Removal at Mid-loop," of the EPU TR (Reference 1, Attachments 5 and 7). A reassessment of existing plant operational curves and operator actions is performed prior to each refueling outage to ensure compliance with the requirements of NRC Generic Letter 88-17, "Loss of Decay Heat Removal," (Reference 5). As stated in Section 2.8.7.1, this reassessment will be performed prior to draining the RCS to mid-loop following EPU operation.

References

1. FPC to NRC letter dated June 15, 2011, "Crystal River Unit 3 - License Amendment Request #309, Revision 0, Extended Power Uprate." (Accession No. ML112070659)
2. NRC to FPC letter dated April 28, 1997, "Crystal River Unit 3 - Review of Individual Plant Examination Submittal - Internal Events (TAC No. M74401)."
3. NRC to FPC letter dated June 30, 1998, "Crystal River Unit 3 - Supplemental Staff Evaluation Report Regarding Individual Plant Examination Report -Internal Events (TAC No. M74401)."
4. NRC to FPC letter dated January 11, 2001, "Review of Crystal River Unit 3 Individual Plant Examination of External Events (IPEEE) Submittal (TAC No. M83612)."
5. NRC Generic Letter 88-17, "Loss of Decay Heat Removal," October 17, 1988.
Retrieved from "https://kanterella.com/w/index.php?title=3F0312-01,_Response_to_Request_for_Additional_Information_to_Support_NRC_PRA_Licensing_Branch_Technical_Review_of_the_CR-3_Extended_Power_Uprate_LAR&oldid=2463538"