ML060620548

From kanterella
Jump to navigation Jump to search
Brown Ferry Nuclear EPU Containment Overpressure (COP) Credit Risk Assessment.
ML060620548
Person / Time
Site: Browns Ferry  Tennessee Valley Authority icon.png
Issue date: 02/27/2006
From: Andersen V
ERIN Engineering & Research
To:
Office of Nuclear Reactor Regulation, Tennessee Valley Authority
References
Download: ML060620548 (150)


Text

{{#Wiki_filter:ENCLOSURE 4 TENNESSEE VALLEY AUTHORITY (TVA) BROWNS FERRY NUCLEAR PLANT UNITS 2 AND 3 EXTENDED POWER UPRATE CONTAINMENT OVERPRESSURE CREDIT RISK ASSESSMENT

BFIN EPU Containment Overpressure (COP) Credit Risk Assessment Performed for: Tennessee Valley Authority Performed by: ERIN Engineering and Research, Inc. February 27, 2006

BFNEPUCOPProbabilisticRiskAssessment

      *--i--E-: --E-- -nes   a^mn...      .. .. .- - E--.........         - .....-.- .. .                             .. . . . . .                        ......
.'.:E.i. : .
                                    -.....                                             ws er.N c AO                          at3          N                   ...      i' -i'l        .. ..                   d...
                                                  .. ... ..                                                                        ..     . ..                    EE

..........- fif:f o , .z  :: :: lX

                                                                            .m.i .n,
                                                                          ........            .t.u......e,....-r.u s
                                                                                                                                                         .r...........
: : : : : 0 :...: : .. .:: ..........

i0i- f E

........... - :::::: :i
                                                                                                               ....................  :~~~iE.. i,:..::~..i: :: :' i:~ ::iEE.. ..........
~i :

00 ~. ~eaeiyE, ,,,..

                                               .0E...................

E . . ..... ....:: E -0D t

                                                                                                                                              ..................F¢.-..          ....... ....... -.      .i0--i
                                        .. . . . . . . . ... . .                            :fC fi~~S        .~ if----....   .....                 i . f.f..          :......
                                                                                  ........                              E..                ..         E E...                 ..-:....::....:

C1320503-6924 - 2/27/2006

BFNEPUCOPProbabilisticRisk Assessment Table Of Contents Section Paae EXECUTIVE

SUMMARY

................                                                                                     ii

1.0 INTRODUCTION

.1-1 1.1   Background .1-1 1.2   Scope .1-3 1.3   Definitions .1-4 1.4   Acronyms...............................................................................................1-6 2.0    APPROACH .2-1 2.1   General Approach .2-1 2.2   Steps to Analysis .2-3 3.0    ANAYSIS .3-1 3.1   Assessment of DBA Calculations .3-1 3.2   Probability of Plant State 1 and Plant State 2 .3-3 3.3   Pre-Existing Containment Failure Probability.3-5 3.4   Modifications to BFN Unit I PRA Models.                                                                 3-6 3.5   Assessment of Large-Late Releases .3-8 4.0    RESULTS .4-1 4.1   Quantitative Results .4-1 4.2   Uncertainty Analysis .4-1 4.3   Applicability to BFN Unit 2 and Unit 3 .4-13

5.0 CONCLUSION

S .5-1 REFERENCES Appendix A PRA Quality Appendix B Probability of Pre-Existing Containment Leakage Appendix C Assessment of River Water and SP Water Temperature Variation Appendix D Large-Late Release Impact Appendix E Revised Event Trees Appendix F Fault Trees i C1320503-6924 - 2/27/2006

BFNEPUCOP ProbabilisticRisk Assessment EXECUTIVE

SUMMARY

The report documents the risk impact of utilizing containment accident pressure (containment overpressure) to satisfy the net positive suction head (NPSH) requirements for RHR and Core Spray pumps during DBA LOCAs. The risk assessment evaluation uses the current BFN Unit 1 Probabilistic Risk Assessment (PRA) internal events model (including internal flooding). The BFN PRA provides the necessary and sufficient scope and level of detail to allow the calculation of Core Damage Frequency (CDF) and Large Early Release Frequency (LERF) changes due to the crediting of containment overpressure in determining sufficient NPSH requirements for the RHR system and Core Spray system emergency core cooling pumps. The steps taken to perform this risk assessment evaluation are as follows:

1) Evaluate sensitivities to the DBA LOCA accident calculations to determine under what conditions credit for COP is required to satisfy low pressure ECCS pump NPSH.
2) Revise all large LOCA accident sequence event trees to make low pressure ECCS pumps dependent upon containment isolation when other plant pre-conditions exist (i.e., SW high temperature, SP initial high temperature).
3) Modify the existing BFN PRA Containment Isolation System fault tree to include the probability of pre-existing containment leakage.
4) Quantify the modified PRA models and determine the following risk metrics:
  • Change in Core Damage Frequency (CDF)
  • Change in Large Early Release Frequency (LERF)
5) Perform modeling sensitivity studies and a parametric uncertainty analysis to assess the variability of the results.

ii C1320503-6924 - 2/27/2006

BFNEPUCOPProbabilisticRistkAssessment The conclusion of the plant internal events risk associated with this assessment is as follows.

1) Regulatory Guide 1.174 provides guidance for determining the risk impact of plant-specific changes to the licensing basis. Regulatory Guide 1.174 defines very small changes in risk as resulting in increases of core damage frequency (CDF) below 10/yr. Based on this criteria, the proposed change (i.e., use of COP to satisfy the net positive suction head (NPSH) requirements for RHR and Core Spray pumps) represents a very small change in CDF (1.53E-09/yr).
2) Regulatory Guide 1.174 provides guidance for determining the risk impact of plant-specific changes to the licensing basis. Regulatory Guide 1.174 defines very small changes in risk as resulting in increases of Large Early Release Frequency (LERF) below 10 7/yr. Based on this criteria, the proposed change (i.e., use of COP to satisfy the net positive suction head (NPSH) requirements for RHR and Core Spray pumps) represents a very small change in LERF (1.53E-09/yr).
                                           ...                       Ci1320503-6924 - 2/27/2006

BFNEPU COPProbabilisticRisk Assessment Section 1 INTRODUCTION The report documents the nsk impact of utilizing containment accident pressure (containment overpressure) to satisfy the net positive suction head (NPSH) requirements for RHR and Core Spray pumps during DBA LOCAs.

1.1 BACKGROUND

Tennessee Valley Authority (TVA) submitted the BFN extended power uprate (EPU) license amendment request (LAR) to the NRC in June 2004. In a October 3, 2005 letter to TVA, the NRC requested the following additional information on the EPU LAR:

      "SPSB-A. 11 As part of its EPU submittal, the licensee has proposed taking credit (Unit
1) or extending the existing credit (Units 2 and 3) for containment accident pressure to provide adequate net positive suction head (NPSH) to the ECCS pumps. Section 3.1 in Attachment 2 to Matrix 13 of Section 21.1 of RS-001, Revision 0 states that the licensee needs to address the risk impacts of the extended power uprate on functional and system-level success criteria. The staff observes that crediting containment accident pressure affects the PRA success criteria; therefore, the PRA should contain accident sequences involving ECCS pump cavitation due to inadequate containment pressure. Section 1.1 of Regulatory Guide (RG) 1.174 states that licensee-initiated licensing basis change requests that go beyond current staff positions may be evaluated by the staff using traditional engineering analyses as well as a risk-informed approach, and that a licensee may be requested to submit supplemental risk information if such information is not submitted by the licensee. It is necessary to consider risk insights, in addition to the results of traditional engineering analyses, while determining the regulatory acceptability of crediting containment accident pressure.

Considering the above discussion, please provide an assessment of the credit for containment accident pressure against the five key principles of risk-informed decisionmaking stated in RG 1.174 and SRP Chapter 19. Specifically, demonstrate that the proposed containment accident pressure credit meets current regulations, is consistent with the defense-11 C1320503-6924 - 2/27/2006

BFNEPUCOP ProbabilisticRiskAssessment in-depth philosophy, maintains sufficient safety margins, results in an increase in core-damage frequency and nsk that is small and consistent with the intent of the Commission's Safety Goal Policy Statement, and will be monitored using performance measurement strategies. With respect to the fourth key principle (small increase in risk), provide a quantitative risk assessment that demonstrates t'hat the proposed containment accident pressure credit meets the numencal risk acceptance guidelines in Section 2.2.4 of RG 1.174. This quantitative risk assessment must include specific containment failure mechanisms (e.g., liner failures, penetration failures, primary containment isolation system failures) that cause a loss of containment pressure and subsequent loss of NPSH to the ECCS pumps." Typical of other industry EPU LAR subrnittals, the BFN EPU LAR includes a request to credit containment accident pressure, also known as containment overpressure (COP), in the determination of net positive suction head (NPSH) for low pressure ECCS systems following design basis events. Also consistent with other industry EPU LAR submittals, the NRC is requesting risk information from licensees regarding the COP credit request. BFN Units 2 and 3 already have existing approvals for containment overpressure credit. The BFN EPU LAR requests containment overpressure credit for BFN Unit 1 for DBA LLOCA accidents. The need for COP credit requests is driven by the conservative nature of design basis accident calculations. Use of more realistic inputs in such calculations shows that no credit for COP is required. In any event, the request for containment accident pressure credit is a physical aspect that will exist during the postulated design basis accidents. The EPU LAR simply requests to include that existing containment accident pressure in the ECCS pump NPSH calculations. The NRC request is to investigate the impact on risk if the containment accident pressure is not present (e.g., postulated pre-existing primary containment failure) during the postulated scenarios. 1-2 C1320503-5924 - 2/27/2006

BEN EPUCOPProbabilisticRisk Assessment The Nuclear Regulatory Commission (NRC) has allowed credit for COP to satisfy NPSH requirements in accordance with Regulatory Guide 1.82 (RG 1.82). Specifically, RG 1.82 Position 2.1.1.2 addresses containment overpressure as follows:

        "For certain operating BWRs for which the design cannot be practicably altered conformance with Regulatory Position 2.1.1.1 may not be possible.

In these cases, no additional containment pressure should be included in the determination of available NPSH than is necessary to preclude pump cavitation. Calculation of available containment pressure should underestimate the expected containment pressure when determining available NPSH for this situation. Calculation of suppression pool water temperature should overestimate the expected temperature when determining available NPSH." The proposed change in the BFN license basis regarding credit for COP meets the approved positions of RG 1.82. However, developments between the NRC staff and members of the Advisory Committee on Reactor Safeguards (ACRS) in 2005 regarding proposed language to Revision 4 of RG 1.82 prompted the NRC to request performance of a 'risk-informed' assessment in accordance with NRC Regulatory Guide 1.174, "An Approach for Using Probabilistic Risk Assessment In Risk-informed Decisions on Plant-Specific Changes to the Licensing Basis". 1.2 SCOPE This risk assessment addresses principle #4 of the RG 1.174 risk informed structure. Principle #4 of RG 1.174 involves the performance of a risk assessment to show that the impact on the plant core damage frequency (CDF) and large early release frequency (LERF) due to the proposed change is within acceptable ranges, as defined by RG 1.174. The other principles (#1 43, and #5) are not addressed in this report. This analysis assesses the CDF and LERF risk impact on the BFN Unit I at-power internal events PRA resulting from the COP credit requirement for low pressure ECCS pumps during large LOCA scenarios. 1-3 C1320503-6924 - 2/27/2006

BFNEPU COP ProbabilisticRisk Assessment External event and shutdown accident risk is assessed on a qualitative basis. In addition, a review of the BFN Unit 2 and Unit 3 models is performed to show that the results from the Unit 1 BFN PRA apply to Units 2 and 3, as well. 1.3 DEFINITIONS Accident sequence - a representation in terms of an initiating event followed by a combination of system, function and operator failures or successes, of an accident that can lead to undesired consequences, with a specified end state (e.g., core damage or large early release). An accident sequence may contain many unique variations of events that are similar. Core damage - uncovery and heatup of the reactor core to the point at which prolonged oxidation and severe fuel damage is anticipated and involving enough of the core to cause a significant release. Core damage frequency- expected number of core damage events per unit of time. End State - is the set of conditions at the end of an event sequence that characterizes the impact of the sequence on the plant or the environment. End states typically include: success states, core damage sequences, plant damage states for Level 1 sequences, and release categories for Level 2 sequences. Event tree - a quantifiable, logical network that begins with an initiating event or condition and progresses through a series of branches that represent expected system or operator performance that either succeeds or fails and arrives at either a successful or failed end state. Initiating Event - An initiating event is any event that perturbs the steady state operation of the plant, if operating, or the steady state operation of the decay heat removal systems during shutdown operations such that a transient is initiated in the plant. Initiating events trigger sequences of events that challenge the plant control and safety systems. ISLOCA - a LOCA when a breach occurs in a system that interfaces with the RCS, where isolation between the breached system and the RCS fails. An ISLOCA is usually characterized by the over-pressurization of a low-pressure system when subjected to RCS pressure and can result in containment bypass. 1-4 C1320503-6924 - 2/27/2006

BFNEPUCOPProbabilisticRiskAssessment Large early release - the rapid, unmitigated release of airborne fission products from the containment to the environment occurring before the effective implementation of off-site emergency response and protective actions. Large early release frequency - expected number of large early releases per unit of time. Level I - identification and quantification of the sequences of events leading to the onset of core damage. Level 2 - evaluation of containment response to severe accident challenges and quantification of the mechanisms, amounts, and probabilities of subsequent radioactive material releases from the containment. Plant damage state - Plant damage states are collections of accident sequence end states according to plant conditions at the onset of severe core damage. The plant conditions considered are those that determine the capability of the containment to cope with a severe core damage accident. The plant damage states represent the interface between the Level 1 and Level 2 analyses. Probability - is a numerical measure of a state of knowledge, a degree of belief, or a state of confidence about the outcome of an event. Probabilistic risk assessment - a qualitative and quantitative assessment of the risk associated with plant operation and maintenance that is measured in terms of frequency of occurrence of risk metrics, such as core damage or a radioactive material release and its effects on the health of the public (also referred to as a probabilistic risk assessment, PRA). Release category - radiological source term for a given accident sequence that consists of the release fractions for vanrous radionuclide groups (presented as fractions of initial core inventory), and the timing, elevation, and energy of release. The factors addressed in the definition of the release categories include the response of the containment structure, timing, and mode of containment failure; timing, magnitude, and mix of any releases of radioactive material; thermal energy of release; and key factors affecting deposition and filtration of radionuclides. Release categories can be considered the end states of the Level 2 portion of a PRA. Risk - likelihood (probability) of occurrence of undesirable event, and its level of damage (consequences). Risk metrics - the quantitative value, obtained from a risk assessment, used to evaluate the results of an application (e.g., CDF or LERF). 1-5 C1320603-6924 - 2/27/2006

BFNEPUCOPProbabilisticRiskAssessment Severe accident - an accident that involves extensive core damage and fission product release into the reactor vessel and containment, with potential release to the environment. Split Fraction - a unitless parameter (i.e., probability) used in quantifying an event tree. It represents the fraction of the time that each possible outcome, or branch, of a particular top event may be expected to occur. Split fractions are, in general, conditional on precursor events. At any branch point, the sum of all the split fractions representing possible outcomes should be unity. (Popular usage equates "split fraction" with the failure probability at any branch [a node] in the event tree.) 1.4 ACRONYMS ACRS Advisory Committee on Reactor Safeguards ATWS Anticipated Transient without Scram BFN Browns Ferry Nuclear plant CCF Common Cause Failure CDF Core Damage Frequency CET Containment Event Tree COP Containment Overpressure CPPU Constant Pressure Power Uprate DBA Design Basis Accident DW Drywell ECCS Emergency Core Cooling Systems EPU Extended Power Uprate GE General Electric HEP Human Error Probability HPCI High Pressure Core Injection system HRA Human Reliability Analysis 146 C1320503-5924 - 2/27/2006

BFNEPUCOPProbabilisticRiskAssessment IPE Individual Plant Examination IPEEE Individual Plant Examination for External Events ISLOCA Interface System Loss of Coolant Accident La Maximum Allowable Primary Containment Leakage Rate LERF Large Early Release Frequency LOCA Loss of Coolant Accident LLOCA Large LOCA LOOP Loss of Offsite Power event LPCI Low Pressure Coolant Injection MAAP Modular Accident Analysis Program NPSH Net Positive Suction Head NRC United States Nuclear Regulatory Commission PRA Probabilistic Risk Assessment PSA Probabilistic Safety Assessment RCIC Reactor Core Isolation Cooling System RG Regulatory Guide RHR Residual Heat Removal System RPV Reactor Pressure Vessel SMA Seismic Margins Assessment SP Suppression Pool SPC Suppression Pool Cooling SW Service Water 1-7 C1320503-6924 - 2/27/2006

BFN EPU COPProbabilisticRisk Assessment TS Technical Specifications TVA Tennessee Valley Authority WW Wetwell 1-8 C1320603.6924 - 2/27/2006

BFNEPUCOP ProbabilisticRisk Assessment Section 2 APPROACH This section includes a brief discussion of the analysis approach and the types of inputs used in this risk assessment. 2.1 GENERAL APPROACH This risk assessment is performed by modification and quantification of the BFN PRA models. 2.1.1 Use of BFN Unit 1 PRA The current BFN Unit I PRA models (BFN model U1050517) are used as input to perform this risk assessment. The Browns Ferry PRA uses widely-accepted PRA techniques for event tree and fault tree analysis. Event trees are constructed to identify core damage and radionuclide release sequences. The event tree "lop events" represent systems (and operator actions) that can prevent or mitigate core damage. Fault trees are constructed for each system in order to identify the failure modes. Analysis of component failure rates (including common cause failures) and human error rates is performed to develop the data needed to quantify the fault tree models. For the purpose of analysis, the Browns Ferry PRA divides the plant systems into two categories:

1. Front-Line Systems, which directly satisfy critical safety functions (e.g.,

Core Spray and Torus Cooling), and

2. Support Systems, which are needed to support operation of front-line systems (e.g., AC power and service water).

2-1 C1320503.6924 - 2/27/2006

BFNEPUCOPProbabilisticRisk Assessment Front-line event trees are linked to the end of the Support System event trees for sequence quantification. This allows definition of the status of all support systems for each sequence before the front-line systems are evaluated. Quantification of the event tree and fault tree models is performed using personal computer version of the RISKMAN code. The Support System and Front-Line System event trees are "linked" together and solved for the core damage sequences and their frequencies. Each sequence represents an initiating event and combination of Top Event failures that results in core damage. The frequency of each sequence is determined by the event tree structure, the initiating event frequency and the Top Event split fraction probabilities specified by the RISKMAN master frequency file. RISKMAN allows the user to enter the split fraction names and the logic defining the split fractions (i.e., rules) to be selected for a given sequence based on the status of events occurring earlier in the sequence or on the type of initiating event. 2.1.2 PRA Quality The BFN PRA used as input to this analysis (BFN model U1050517) is of sufficient quality and scope for this application. The BFN Unit 1 PRA is highly detailed, including a wide variety of initiating events (e.g., transients, internal floods, LOCAs inside and outside containment, support system failure initiators), modeled systems., extensive level of detail, operator actions, and common cause events. The BFN Units 2 and 3 at-power internal events PRAs received a formal industry PRA Peer Review in 1997. All of the "A"and "B" priority comments have been addressed. Refer to Appendix A for further details concerning the quality of the BFN PRA. 2-2 C1 320503 8924 - 2/27/2006

BFN EPUCOPProbabilisticRisk Assessment 2.2 STEPS TO ANALYSIS The performance of this risk assessment is best described by the following major analytical steps:

  • Assessment of DBA calculations
  • Estimation of pre-existing containment failure probability
  • Analysis of relevant plant experience data
  • Manipulation and quantification of BFN Unit 1 RISKMAN PRA models
  • Comparison to ACDF and ALERF RG 1.174 acceptance guidelines
  • Performance of uncertainty and sensitivity analyses
  • Assessment of "Large Late" Release Impact
  • Review of BFN Unit 2 and Unit 3 PRAs Each of these steps is discussed briefly below.

2.2.1 Assessment of DBA Calcullations The purpose of this task is to develop an understanding of the BFN EPU design basis LLOCA calculations that result in the need to credit 3 psig containment overpressure credit. The need for COP credit requests is driven by the conservative nature of design basis accident calculations. The DBA LOCA calculations are reviewed and sensitivity calculations performed to determine under what conditions of more realistic inputs is there no need for COP credit in the determination of low pressure ECCS pump NPSH. 2-3 C1320503-E924 - 2/27/2006

BFN EPUCOPProbabilisticRisk Assessment 2.2.2 Estimation of Pre-Existina Containment Failure Probability This task involves defining the size of a pre-existing containment failure pathway to be used in the analysis to defeat the COP credit, and then quantifying the probability of occurrence of the un-isolable pre-existing containment failure. The approach to this input parameter calculation will follow EPRI guidelines regarding calculal:ion of pre-existing containment leakage probabilities in support of integrated leak rate test (ILRT) frequency extension LARs (i.e., EPRI Report 1009325, Risk Impact of Extended Integrated Leak Rate Testing Intervals, 12/03).[2] This is the same approach used in the recent Vermont Yankee EPU COP analyses presented to the ACRS in December 2005. The pre-existing unisolable containment leak probability is combined with the BFN PRA containment isolation failure on demand fault tree (CIL) to develop the likelihood of an unisolated primary containment at t=0 that can defeat the COP credit necessary for the determination of adequate low pressure ECCS pump NPSH. 2.2.3 Analysis of Relevant Plant Excerience Data An unisolated primary containment is not the only determining factor in defeating low pressure ECCS pump NPSH. The DBA calculations show that other extreme low likelihood plant conditions are required at t=0 to result in the need to credit COP in the determination of pump NPSH, such as high initial reactor power level and the following two key water temperature conditions:

  • High river water temperature
  • High initial torus water temperature 24 C1320503 2/27/2006

BFNEPUCOPProbabilisticRiskAssessment This step involves obtaining plant experience data for river water and torus water temperature and performing statistical analysis to determine the probabilities of exceedance as a function of water temperature. 2.2.4 Manipulation And Quantification of BFN Unit 1 RISKMAN PRA Models This task is to make the necessary modifications to the BFN Unit 1 RISKMAN-based PRA models to simulate the loss of low pressure ECCS pumps during PRA Large LOCA scenarios due to inadequate NFPSH caused by an unisolated containment and other extreme plant conditions (e.g., high service water temperature). All large LOCA initiated sequences in the BFN PRA are modified as appropriate (except ISLOCAs and LOCAs outside containment, because these LOCAs result in deposition of decay heat directly outside the containment and not into the suppression pool). This approach to manipulating only LLOCA scenarios is to mirror the DBA accident calculations requiring COP credit. This is consistent with the ACRS observations during the December 2005 Vermont Yankee EPU COP hearings, in which the ACRS commented that they did not prefer the approach of assigning COP credit to all accident sequence types in the PRA simply for the sake of conservatism. The modeling and quantification is performed consistent with common RISKMAN modeling techniques. 2.2.5 Comparison to ACDF and ALERF RG 1.174 Acceptance Guidelines The revised BFN Unit 1 PRA models are quantified to determine CDF and LERF. The difference in CDF and LERF between the revised model of this assessment and the BFN Unit 1 PRA base results are then compared to the RG 1.174 risk acceptance guidelines. The RG 1.174 ACDF and ALERF risk acceptance guidelines are summarized in Figures 2-1 and 2-2, respectively. The boundaries between regions are 2-5 CI 320503.6924 - 2/27/2006

BFN EPUCOPProbabilisticRisk Assessment not necessarily interpreted by the NRC as definitive lines that determine the acceptance or non-acceptance of proposed license amendment requests; however, increasing delta risk is associated with increasing regulatory scrutiny and expectations of compensatory actions and other related risk mitigation strategies. 2.2.6 Performance of Uncertainty and Sensitivity Analyses To provide context to the variability of the calculated deltaCDF and deltaLERF results, a parametric uncertainty analysis was performed using the RISKMAN software. 2.2.7 Assessment of "Large Late" Release Impact This task is to perform an assessment of the EPU COP credit impact on BFN Unit 1 PRA "Large Late" radionuclide releases. This task is performed because the ACRS questioned Entergy on this issue during the recent Vermont Yankee EEPU ACRS hearings in December 2005. This aspect of the analysis is for additional information, and does not directly correspond to the RG 1.174 risk acceptance guidelines shown in Figures 2-1 and 2-2. 2.2.8 Review of BFN Unit 2 and Unit 3 PRAs The base analysis uses the BFN Unit 1 PRA models. This task involves reviewing the BFN Unit 2 and BFN Unit 3 RISKMAN PRA models and associated documentation to determine whether the analysis performed for BFN Unit 1 is also applicable to Unit 2 and Unit 3. 246 C1320503.6924 - 2/27/2006

BFNEPUCOPProbabilisticRiskAssessment Figure 2-1 RG 1.174 CDF RISK ACCEPTANCE GUIDELINES

..-.-.f......l! -.  !

i., i...-....-.. ..000-i a..I.L....

ftq11l1 2-7 C1320503.6924 - 2/27/2006

BFNEPUCOPProbabilisticRiskAssessment Figure 2-2 RG 1.174 LERF RISK ACCEPTANCE GUIDELINES

                ...           ... i.                                                                        .....         ...         .
      -- ~ ~j;r  ~           ~ ll12 ~                           ~                ~   ~ .          .T

_- _ _ _ _ _ _ I _I _I __ ___ _ _ _ __ _ \ .~

                                             .-. .o.;. . .00
                                   . .0. .0.00               .         -    0-                         LER     F; 2-8                                                   C1320505-6924 - 2/27/2006

BFNEPUCOPProbabilisticRiskAssessment Section 3 ANALYSIS This section highlights the major qualitative and quantitative analytic sl:eps to the analysis. 3.1 ASSESSMENT OF DBA CALCULATIONS The purpose of this risk assessment is due to the fact that the conservative nature of design basis accident calculations result in the need to credit COP in determining adequate low pressure ECCS pump NPSH. Use of more realistic inputs in such calculations shows that no credit for COP is required. The GE DBA LOCA calculation makes the following conservative assumptions, among others, regarding initial plant configuration and operation characteristics:

  • Initial reactor power level at 1020/c EPU
  • Decay heat defined by 2 sigma uncertainty
  • 2 RHR pumps and 2 RHR heat exchangers in SPC
  • All pumps operating at full flow
  • River water temperature at 950 F
  • Initial suppression pool temperature at 950 F
  • No credit for containment heat sinks The GE DBA LOCA calculations were reviewed and the following input parameters were identified as those with a potential to significantly impact the D13A analytic conclusions regarding the need for COP credit in NPSH determination:
  • Initial reactor power level
  • Decay heat 3-1 C132050$3 924 - 2/27/2006

BFNEPUCOPProbabilisticRisk Assessment

  • Number of RHR pumps and heat exchangers in SPC
  • River water temperature
  • Initial suppression pool temperature
  • RHR heat exchanger effectiveness
  • Initial suppression pool water volume
  • Credit for containment heat sinks Based on knowledge of the calculations, other inputs such as initial containment air temperature and humidity, have non-significant impacts on the results.

It is recognized that there are numerous different combinations of more realistic calculation inputs that show that COP credit is not necessary for maintenance of low pressure ECCS pump NPSH. To simplify the risk assessment, the different combinations of realistic input sensitivities were maintained at a manageable number. Eleven sensitivity calculations were performed to identify key input parameters for use in this risk assessment. The results of these calculations are shown in Table 3-1 (the shaded cells show those parameters that changed from the base [BA LOCA calculation). [3] From the results of the sensitivity cases summarized in Table 3-1, the following general conclusions can be made:

  • Initial reactor power, decay heat level, and initial water temperatures are the key determining factors in the analytic conclusions
  • COP credit is not required for NPSH, even with the conservative DBA calculation inputs, if 3 or 4 RHR Ipumps and associated heat exchangers are in operation (refer to Cases 1 and l a in Table 3-1).
  • If the plant is operating at an unexpected 102% EPU initial power level with an assumed 2 sigma decay heat, only 2 RHR pumps and heat exchangers are placed in SPC operation, and initial torus water temperature is at the high temperature of 95 0F, then river water 3-2 C132050S36924 - 2/27/2006

BFNEPUCOPProbabilisticRisk Assessment temperature must be above 700 F to result in the need for COP credit (refer to Case 2b in Table 3-1). If the plant is operating at the expected nominal 100% EPU initial power level (2 sigma decay heat not assumed), only 2 RHR pumps and heat exchangers are placed in SPC operation, and initial torus water temperature is taken as 920 F, then river water temperature must be above 860 F to result in the need for COP credit (refer to Case 4c in Table 3-1). The analytic conclusions are used in this risk assessment to define two plant states that will result in failure of low pressure ECCS pumps on inadequate NPSH during large LOCAs if the containment is unisolated:

  • Plant State 1: 102% EPU initial power level, 2 sigma decay heat, 2 RHR pumps and heat exchangers in SPC, initial torus water temperature of 950F, and river water temperature above 700F
  • Plant State 2: 100% EPU initial power level, nominal decay heat, 2 RHR pumps and heat exchangers in SPC, initial torus water temperature of 920F, and river water temperatures above 860 F These two plant states are used in this iisk assessment to model the LLOCA scenarios that can result in loss of low pressure ECCS pumps due to inadequate NPSH when the containment is unisolated. The probability of being in Plant State 1 or Plant State 2 is discussed below in Section 3.2.

3.2 PROBABILITY OF PLANT STATE 1 AND PLANT STATE 2 This section discusses the estimation of the probability of being in Plant State 1 or Plant State 2. This assessment is based on the statistical analysis of BFN experience data. Refer to Appendix C for the statistical analysis of variations in BFN river water and torus water temperatures. 3-3 C1 320503.6924 - 2/27/2006

BFNEPUCOPProbabilisticRisk Assessment 3.2.1 Probability of Plant State 1 The probability of being in Plant State 1 is determined as follows:

  • The probability of being at 102% EPU power at the time of the postulated DBA LOCA is modeled as a miscalibration error of an instrument
  • If such a miscalibration error occurs, it is assumed that the plant will be operating at 102% and that the operator does not notice other differing plant indications that would cause the operator to re-evaluate the plant condition
  • If the plant is operating at 102% power, the decay heat level defined by 2 sigma uncertainty is assumed 1:o occur with a probability of 1.0 (this conservative assumption is to simplify the analysis).
  • The probability of river water temperature greater than 700 F is determined from the BFN experience data statistical analysis summarized in Appendix C.
   . If the above conditions are satisfied, it assumed that the torus water temperature is 950F, with a probability of 1.0 (this conservative assumption is to simplify the analysis).

Based on review of the pre-initiator human error probability calculations in the BFN Unit 1 PRA Human Reliability Analysis, this risk assessment assumes a nominal human error probability of 5E-3 for miscalibration of an instrument. As such, the probability of being at 102% power at t=0 is taken in this analysis to be 5E-3. As can be seen from Table C-1, the probability of river water temperature exceeding 700 F is 4.OE-1. Therefore, the probability of being in Plant State 1 is 5E-3 x 0.40 = 2E-3. 34 C1320503.6924 - 2/27/2006

BFN EPUCOPProbabilisticRisk Assessment 3.2.2 Probability of Plant State 2 The probability of being in Plant State 2 is determined as follows:

  • The probability of being at 100% EPU power at the time of the postulated DBA LOCA is reasonably assumed to be 1.0
  • The probability of river water temperature greater than 860 F is determined from the BFN experience data statistical analysis summarized in Appendix C.
   . If the above conditions are satisfied, it assumed that the torus water temperature is 920F, with a probability of 1.0 (this conservative assumption is to simplify the analysis).

As can be seen from Table C-1, the probability of river water temperature exceeding 860 F is 1.4E-1. Therefore, the probability of being in Plant State 2 is 1.4E-1 x 1.0 = 1.4E-1. 3.3 PRE-EXISTING CONTAINMENT FAILURE PROBABILITY As discussed in Section 2, the approach to this input parameter calculation follows the EPRI guidelines regarding calculation of pre-existing containment leakage probabilities in support of integrated leak rate test (ILRT) frequency extension LARs (i.e., EPRI Report 1009325, Risk Impact of Extended Integrated Leak Rate Testing Intervals, 12/03). [2] This assessment is provided in Appendix B of this report. As discussed in Appendix B, a pre-existing unisolable containment leakage path of 35La is assumed in the base case quantification of this risk assessment to result in defeating the necessary COP credit. As can be seen from Table B-1, the probability of the 35La pre-existing containment leakage used in this base case analysis is 9.86E-04. 3-5 C1320503.6924 - 2/27/2006

BFN EPUCOPProbabilisticRisk Assessment This low likelihood of a significant pre-existing containment leakage path is consistent with BFN primary containment performance experience. Neither BFN nor the BWR industry has experienced a 35La pre-existing containment leakage event. The BFN primary containment performance experience shows BFN containment leakages much less than 35La. Per Reference [1], the E3FN Unit 2 and Unit 3 primary containment ILRT results from the most recent tests are as follows: Unit Test Date j Containment Leakage (Fraction of La) 2 11/06/94 0.1750 2 03/17/91 0.1254 3 1 110/1/98 0.1482 3 1 1/i06/95 0.4614 Although the above results are for Units 2 and Units 3, given the similarity in plant design and operation and maintenance practices, the results are reasonably judged to be reflective of BFN Unit 1, as well. Sensitivity studies to the base case quantification (refer to Section 4) assess the sensitivity of the results to the pre-existing leakage size assumption. 3.4 MODIFICATIONS TO BFN UNIT 1 PRA MODELS As discussed in Section 2, all large LOCA initiated sequences in the BFN PRA are modified as appropriate (except ISLOCAs and LOCAs outside containment, because these LOCAs result in deposition of decay heat directly outside the containment and not into the suppression pool). The following Large LOCA initiated sequences in the BFN Unit 1 PRA were modified:

  • Large LOCA - Loop I Core Spray Line Break (LLCA)
  • Large LOCA - Loop II Core Spray Line Break (LLCB) 3-6 C132050a6924 - 2/27/2006

BFN EPUCOPProbabilisticRisk Assessment

  • Large LOCA - Loop A Recirc. Discharge Line Break (LLDA)
  • Large LOCA - Loop B Recirc. Discharge Line Break (LLDB)
  • Large LOCA - Loop A Recirc. Suction Line Break (LLSA)
  • Large LOCA - Loop B Recirc. Suction Line Break (LLSB)
  • Other Large LOCA (LLO)

The accident sequence modeling for the above LLOCA initiators was modified as follows:

     . A top event for loss of containment integrity (CIL) was added to the beginning of the Level 1 event tree structures
     . A top event modeling the additional Plant State pre-conditions (NP3SH) was added to the beginning of the Level 1 event tree structures, right after the CIL top event.
    . If top events CIL and NPSH are satisfied (i.e., occur), then the RHR pumps and CS pumps are directly failed Refer to Appendix E for print-outs of the revised large LOCA event trees.

The CIL top event is quantified using a fault tree. The fault tree is a modified version of the existing BFN Unit 1 Level 2 PRA containment isolation fault tree. The BFN Unit 1 Level 2 PRA containment isolation fault tree models failure of the containment isolation system on demand given an accident signal. Hardware, power and signal failures for all primary containment penetrations greater than 3" diameter are modeled in the fault tree. To this fault tree structure was added the probability of a pre-existing containment leak size of 35La. Refer to Appendix F for a print-out of the containment isolation fault tree used in this analysis for the CIL node in the large LOCA event trees. The NPSH top event is also quantified using a fault tree. The NPSH incorporates the fault tree logic to model the probability of being in Plant State 1 or Plant Stalte 2. Refer to Appendix F for a print-out of the fault tree used in this analysis for the NPSH node in the Large LOCA event trees. 3-7 C1320503-6924 - 2/27/2006

BFN EPU COPProbabilisticRisk Assessment The quantification of the revised model was performed to produce the new CDF. All the new CDF scenarios are those in which the containment is unisolated at t:=O, all RPV injection is lost early, and core damage occurs at approximately one hour. As such, the additional CDF contributions created by this model manipulation are also all LERF release sequences (i.e., deltaCDF equals deltaLERF). This is a conservative assumption as it assumes that the pre-existing containment leakage of 35La used in the base quantification is representative of a LERF release. Reference [2] determines that a containment leak representative of LERF is >600La. The quantification results and uncertainty and sensitivity analyses are discussed in Section 4. The revised BFN Unit 1 PRA RISKMAN model for this base case analysis is archived in file UICOP2-9 and saved on the BFN computers along with the other BFN PRA RISKMAN models. 3.5 ASSESSMENT OF LARGE-LATE RELEASES As discussed above in Section 3.3, all the deltaCDF resulting from this risk assessment also results directly in LERF. As such, there is no increase in Large-Late releases due to scenarios modeling in this risk assessment. Refer to Appendix D) for more discussion. 3-8 C1320503-6924 - 2/27/2006

BFN EPU COP ProbabilisticRisk Assessment Table 3-1

SUMMARY

OF COP DETERMINISTIC CALCULATIONS

                                                                                     .2                                         E n~      ~                      LL.-
                                                     ~3                    CO)       i        RE        ~        ~

Cl

  • ClCOP coCL E L gCl a. Peak SP Credit Case Case Descrition . o co zO O (0 . ~ epF eue Base Case EPU Ucensing Calculation - 102% ANSI 5.1 95 95 2 Full 2 2 4000 223 2 1Minimum Yes No 187.3 Yes DBA LOCA EPU w/l2c design Case 1 No Single Failure l[ 102% lANSI 5.11 95 1 95 Ful 4000 223 IYes NoT 166.4 1 No Case la 1

3 Pumps inSPC I EPU 102%

                                                  `22 IANSI 5.1 I           design Full                        4000    223 EPU     wI26                    design t   Yes I No     A175.0         No Case 2      DBA Calculation but SW         102%   ANSI 5.1                   Full                                223                      Yes   No      182.0        Yes Temperature = 85F              EPU     w/2a                    design Case 2a     DBA Calculation but SW         102% ANSI 5.1           95    2   Full        2    . 2        4000    223   2      Minimum     Yes   No      177.6        Yes Temperature = 75F              EPL     w/2a                    design                              I          I                                           N Case 2b     DBA Calculation but SW         102% ANSI 5.1 Temperature = 70F              EPU     wl2cr           95    2   FuNl        2        2      4000    223   2      Minimum     Yes   No      175.9         No Case 2c     DBA Calculation but SW         102% ANSI 5.1                 2   Full        2        2 14000 1223         2    lMinimum   I Yes    No      174.3         No Temperature = 65F              EPU     w/2cr                   design Case 3      DBA Calculation but SP                                       2   Full        2        2      4000    223   2      Minimum     Yes   No      183.8        Yes Temperature = 85F                                              design Case 4      100% Initial Power, Minimum                                  2  _Fu0          2 1     2 14000              2 IMnimum          Yes   No      177.0        Yes I oS Lava, tuU Nu ne= Sink Credit                                                         dvmign I                    1 I I Case 4a     100%Initial Power, Nominal l SP Level, and Heat Sink Credit 2   Full design 2

I 2 4000 174.7 No 3-9 C1320503-6924 - 2/27/2006

BFNEPUCOP ProbabilisticRisk Assessment Table 3-1

SUMMARY

OF COP DETERMINISTIC CALCULATIONS F Y Y P .2 Y P Y Y , Y U)

                                                           .c                                      0 0

E ~2 f.2 0 .' .2 .2 =5 0

a. =U) E a.8 0 0 E 0: .G E2 -LLffi t

0 a. C 3i = g 15 CL Cu 1B 0 Im o= a- a. 0ti a-0 v, I = E

                                        -a-       02        E .1                                             W.
                                                                                                              =a          O coc  Co S                            COP
                                         =e                                      = a.

tE 3 0 a. Co 0.E .0 Peak SP Credit Case Case Description is Co = CL =0 Z .6 00 ao Temp (F) Required

                                                     -_                  _-   I    _-  S   _        _                   _         _       _*

Case 4b 100% Initial Power, Minimum 92 2 Full 2 2 4000 2 Minimum Yes 178.9 Yes SP Level, and Heat Sink design Credit Case 4c 100% Initial Power, Minimum 92 2 Full 2 2 4000 2 Minimum Yes 175.8 No SP Level, Heat Sink Credit desin and SW Temp. that results in Peak SP Temp. equal to/less than 176F 3-10 C1320503-M924 - 2/27/2006

BEN EPU COPProbabilisticRisk Assessment Section 4 RESULTS 4.1 QUANTITATIVE RESULTS' The results of the base quantification of this risk assessment for the 35 L case are as follows:

    . deltaCDF: 1.42E-9/yr
    . deltaLERF: 1.42E-9/yr As discussed in Section 3, the additional CDF contributions created by this model manipulation are also all LERF release sequences (i.e., deltaCDF equals deltaLERF).

These very low results are expected and are well within the RG 1.174 guidelines (refer to Figures 2-1 and 2-2) for "very small" risk impact. If greater detail was included to address some of the conservative assumptive assumptions in this risk assessment (e.g., 2 sigma decay heat assumed wil:h a probability of 1.0 given 102% EPU power exists; refer to Section 3.2), the deltaCDF and deltaLERF would be even lower. 4.2 UNCERTAINTY ANALYSIS To provide additional information for the decision making process, the risk assessment provided here is supplemented by parametric uncertainty analysis and quantitative and qualitative sensitivity studies to assess the sensitivity of the calculated risk results. Uncertainty is categorized here into the following three types, consistent with PRA industry literature:

  • Parametric
  • Modeling 4-1 4-1 C1320503-.6924 - 2/27/2006

BFNEPUCOPProbabilisticRisk Assessment

  • Completeness Parametric uncertainties are those related to the values of the fundamental parameters of the PRA model, such as equipment failure rates, initiating event frequencies, and human error probabilities. Typical of standard industry practices, the parametric uncertainty aspect is assessed here by performing a Monte Carlo parametric uncertainty propagation analysis. Probability distributions are assigned to each parameter value, and a Monte Carlo sampling code is used to sample each parameter and propagate the parametric distributions through to the final results. The parametric uncertainty analysis and associated results are discussed further below.

Modeling uncertainty is focused on the structure and assumptions inherent in the risk model. The structure of mathematical models used to represent scenarios and phenomena of interest is a source of uncertainty, due to the fact that models are a simplified representation of a real-world system. Model uncertainty is addressed here by the identification and quantification of focused sensitivity studies. The model uncertainty analysis and associated results are discussed further below. Completeness uncertainty is primarily concerned with scope limitations. Scope limitations are addressed here by the qualitative assessment of the impact on the conclusions if external events and shutdown risk contributors are also considered. The completeness uncertainty analysis is discussed further below. 4.2.1 Parametric Uncertainty Analysis The parametric uncertainty analysis for this risk assessment was performed using the RISKMAN computer program to calculate probability distributions and determine the uncertainty in the accident frequency estimate. 4-2 C1320503 6924 - 2/27/2006

BFNEPUCOPProbabilisticRisk Assessment RISKMAN has three analysis modules: Data Analysis Module, System Analysis Module, and Event Tree Analysis Module. Appropriate probability distributions for each uncertain parameter in the analysis is determined and included in the Data Module. The System Module combines the individual failure rates, maintenance, and common cause parameters into the split fraction frequencies that will be used by the Event Tree Module. A Monte Carlo routine is used with the complete distributions to calculate the split fraction frequencies. Event trees are quantified and linked together in the Event Module. The important sequences from the results of the Event Tree Module are used in another Monte Carlo sampling step to propagate the split fraction uncertainties and obtain the uncertainties in the overall results. The descriptive statistics calculated by RISKMAN for the total core damage frequency of the plant caused by internal events include:

  • Mean of the sample
  • Variance of the sample
  • 5th, 50th, and 95th percentiles of the sample The parametric uncertainty associated with delta core damage frequency calculated in this assessment is presented as a comparison of the RISKMAN calculated CDF uncertainty statistics for the two cases (i.e., the Unit 1 base EPU PRA and the EPU COP Credit base case quantification). The results are shown in Table 4-1. Table 4-1 summarizes the CDF uncertainty distribution statistics for the Unit 1 PRA and for the COP credit base quantification.

As can be seen from the parametric uncertainty results summarized in Table 4-1, even when considering the parametric uncertainty the risk impact is small. The statistics show that CDF has not changed while the distribution of CDF for the COP study has narrowed slightly: the 5%ile increased slightly while the 95%ile decreased slightly. 4-3 C1320503-6924 - 2/27/2006

BFN EPUCOPProbabilisticRisk Assessment It should be cautioned that this distribution is developed via Monte Carlo (random) sampling, and as such it is dependent upon the number of samples and the initial numerical seed values of the sampling routine. Neither the initial seeds nor the number of samples used for the model of record are known. Consequently, some variation from the base model statistics is expected. Taking these cautions into consideration, a comparison of the distributions by percentiles shows little if any change. 4.2.2 Modeling Uncertainty Analysis As stated previously, modeling uncertainty is concerned with the sensitivity of the results due to uncertainties in the structure and assumptions in the logic model. Modeling uncertainty has not been explicitly treated in many PRAs, and is still an evolving area of analysis. The PRA industry is currently investigating methods for performing modeling uncertainty analysis. EPRI has developed a guideline for modeling uncertainty that is still in draft form and undergoing pilot testing. The EPRI approach that is currently being tested takes the rational approach of identifying key sources of modeling uncertainty and then performing appropriate sensitivity calculations. This approach is taken here. The modeling issues selected here far assessment are those related to the risk assessment of the containment overpressure credit. This assessment does not involve investigating modeling uncertainty with regard to the overall BFN PRA. The modeling issues identified for sensitivity analysis are:

  • Pre-existing containment leakage size and associated probability
  • Calculation of containment isolation system failure
  • Assessment of power and water temperature pre-conditions
  • Number of RHR pumps and heat exchangers in SPC 4-4 C1320503'6924 - 2/27/2006

BFNEPUCOPProbabilisticRisk Assessment Pre-Existina Containment Leakage Size/Probability The base case analysis assumes a pre-existing containment leakage pathway leakage size of 35La that would result in defeat of the necessary containment overpressure credit during a DBA LOCA. The following two modeling sensitivity cases are identified to assess the variability of the risk results to the assumed pre-existing containment leakage size:

  • A smaller, even more conservative, pre-existing leak size of 201a is assumed in this sensitivity to result in defeat of the necessary COP credit.

From EPRI 1009325, the probability of a pre-existing 20La containment leakage pathway is 1.88E-03.

  • A larger pre-existing leak size of 'IOLa, consistent with the EPRI 100,9325 recommended assumption for a "large" leak, is used in this sensitivity to defeat the necessary COP credit. From EPRI 1009325, the probability of a pre-existing 10OLa containment leakage pathway is 2.47E-04.

Calculation of Containment Isolation Sys tem Failure The base case quantification uses the containment isolation system failure fault tree logic to represent failure of the containment isolation system. The fault tree specifically analyzes primary containment penetrations greater than 3" diameter. This modeling sensitivity case expands the scope of the containment isolation fault tree! to include smaller lines as potential defeats of COP credit. This sensitivity is performed by increasing by a factor of 10 the failure probability associated with all the split fraction solutions for the containment isolation system fault tree. 4-5 CI320503~ 924 - 2/27/2006

BFNEPUCOP ProbabilisticRisk Assessment Assessment of Power and Water Temperature Pre-conditions This is a conservative sensitivity that assumes that all that is necessary for failure of the low pressure ECCS pumps due to inadequate NPSH during a large LOCA is an unisolated containment. This sensitivity is performed by assuming the other pre-conditions represented by the top event NSPH (e.g., river water temperature greater than 860F) exist with a probability of 1.0. Number of RHR pumps and heat exchangers in SPC The base case COP credit quantification addresses the situation in which 2 or less RHR pumps and heat exchangers are operating in SPC mode. The likelihood oiF failing any two RHR pumps is approximately 8.2E-3. The likelihood of an unisolated containment is approximately 1.4E-3 and the likelihood of other necessary extreme plant conditions (e.g., high river temperature, high reactor power) existing at the time of the LLOCA is approximately 0.14. As such, the base quantification results in an approximate 1.6E-6 conditional probability, given a LLOCA, of loss of low pressure ECCS pumps due to insufficient NPSH due to inadequate COP. This sensitivity discusses the risk impact of also explicitly quantifying scenarios with only 1 or no RHR pumps failed. Such scenarios are not explicitly included in the base quantification because their risk contribution is negligible, as shown by the sensitivities discussed here. As shown in Table 3-1, even with design basis conservative assumptions, if 3 or more RHR pumps and heat exchangers are operating in SPC, there is no need for containment overpressure. To result in a need for COP credit in such cases would require even more conservative input assumptions than the 2 RHR pump scenario. As such, the additional risk from such scenarios is negligible compared to the 2 RHR pump case explicitly modeled in this analysis. 4-6 C1320503-8924 - 2/27/2006

BFNEPUCOPProbabilisticRisk Assessment An estimate of the deltaCDF rsk contribution for the scenario with 3 RHR pumps in SPC operation can be approximated as follows:

  • Sum of BFN PRA Large LOCA initiator frequencies: 3.1OE-5/yr
  • Likelihood of failure of 1 RHR pump or 1 RHR heat exchanger: 1.OOE-2 (nominal estimate)
  • Probability of 102% EPU initial power level: 5E-3 (same as base analysis)
   . Probability of containment isolation failure:    7E-3 (nominal from base analysis)
  • Probability of river water temperature >-960 F: 9E-3 (nominal value based on Table C-1. Although the river temperature has not exceeded 900 F based on the collected plant data, statistically there is a non--zero likelihood of such a temperature). 960 F is assumed here as the temperature at which COP credit is required (refer to Case Ia of Table 3-1).
  • deltaCDF contribution for 3 RHR 1pump case: 3.1 E-5 x 1E-2 x 5E-3 x 9E-3
      = -1 E-1 3/yr This additional contribution to the calculated deltaCDF from a 3 RHR pump case is negligible in comparison to the 2 RHR pump case.

An estimate of the deltaCDF risk contribution for the scenario with 4 RHR pumps in operation can be approximated as follows:

  • Sum of BFN PRA Large LOCA initiator frequencies: 3.1OE-5/yr
  • Likelihood of 4 RHR pumps and 4 heat exchangers in SPC during L.arge LOCA: 1.0 (nominal estimate)
  • Probability of 102% EPU initial power level: 5E-3 (same as base analysis)
  . Probability of containment isolation failure:    7E-3 (nominal from base analysis)
  • Probability of river water temperature >-100F: 1E-3 (estimate based on Table C-1. Although the river temperature has not exceeded 900 F based 4-7 C1320503.6924 - 2/27/2006

BFNEPUCOPProbabilisticRisk Assessment on the collected plant data, statistically there is a non-zero likelihood of such a temperature). 100F is assumed here as the temperature at which COP credit is required (refer to Case 1 of Table 3-1). deltaCDF contribution for 3 RHR pump case: 3.1 E-5 x 1.0 x 5E-3 x 7E-3 x 1E-3 = -1 E-12/yr Similar to the 3 pump case discussed previously, this additional contribution to the calculated deltaCDF from a 4 RHR pump case is negligible in comparison to the 2 RHR pump case. Summary of Modeling Uncertainty Results The modeling uncertainty sensitivity cases are summarized in Table 4-2. 4.2.3 Completeness Uncertainty Analysis As stated previously, completeness uncertainty is addressed here by the qualitative assessment of the impact on the conclusions if external events and shutdown risk contributors are also considered. 4-8 C1320503.0924 - 2/27/2006

BFNEPUCOPProbabilisticRiskAssessment Table 4-1 PARAMETRIC UNCERTAINTY ANALYSIS RESULTS Statistic BFN Unit 1 Base CDF COP Risk Assessment 5% 4.71 E-7 4.73E-7 50% 1.23E-6 1.21 E-6 MEAN 1.77E-6 1.77E-6 95% 4.72E-6 4.69E-6 4-9 C1320503- 924 - 2/27/2006

BFNEPUCOPProbabilisticRiskAssessment Table 4-2

SUMMARY

OF SENSITIVITY QUANTIFICATIONS Case Description CDF LERF ACDF l ALERF Base(:) Base Case Quantification 1.77E-06 4.41 E-07 1.42E-09 1.42E-09 1(a) Pre-Existing Containment Leakage Sufficient to Fail COP Credit 1.77E-06 4.41 E-07 1.33E-09 1.33E-09 Defined by 1OOLa (probability = 2.47E-4) 2(1) (2) Pre-Existing Containment Leakage Sufficient to Fail COP Credit 1.77E-06 4.41 E-07 1.53E-09 1.53E-09 Defined by 2OLa (probability = 1.88E-3) 3(1) Expansion of Containment Isolation fault tree to Encompass Smaller 1.77E-06 4.42E-07 2.05E-09 2.05E-09 Lines (approximate by multiplying Cont. Isol. failure probability by 1Ox) 4(1) Assume Initial Power Level and Water Temperature Pre-Conditions 1.77E-06 4.42E-07 2.66E-09 2.66E-09 Exist 100% of the Time 5(1) Combination of Cases #2, #3 and #4 1.77E-06 4.48E-07 8.33E-09 8.33E-09 6 Incorporation of '3-RHR pumps in SPC" and '4-RHR pumps in SPC" 1.77E-06 4.41 E-07 1.42E-09 1.42E-09 loss of NPSH scenarios Notes:

      '.toueIcoUS wit Fatnure oi2 Ot mlte RnR pumipsi anu dsoIateu Iheat exi haUiib iif oSPC ade exp;icitly anaiyzed iII tiese .;ases. As sh-wt ii-Case 6, explicit incorporation of scenarios with 0 or I RHR pumps in SPC failed has a negligible impact on the results.

(2) Case 2, 20La containment leakage size, is the case used as the basis for the Conclusions of this study (refer to Section 5). 4-10 C1320503-6924 - 2/27/2006

BFNEPUCOPProbabilisticRisk Assessment Seismic The BFN seismic risk analysis was performed as part of the Individual Plant Examination of Extemal Events (IPEEE). BFN performed a seismic margins assessment (SMA) following the guidance of NUREG-1407 and EPRI NP-6041. The SMA is a deterministic evaluation process that does not calculate risk on a probabilistic basis. No core damage frequency sequences were quantified as part of the seismic risk evaluation. The conclusions of the SMA are judged to be unaffected by the EPU or the containment overpressure credit issue. The EPU has little or no impact on the seismic qualifications of the systems, structures and components (SSCs). Specifically, the power uprate results in additional thermal energy stored in the RPV, but the additional blowdown loads on the RPV and containment given a coincident seismic event, are judged not to alter the results of the SMA. The decrease in time available for operator actions, and the associated increases in calculated HEPs, is judged to have a non-significant impact on seismic-iriduced risk. Industry BWR seismic PSAs have typically shown (e.g., Peach Bottom NUREG-1150 study; Limerick Generating Station Severe Accident Risk Assessment; NUREG/CR-4448) that seismic risk is overwhelmingly dominated by seismic induced equipment and structural failures. Seismic induced failures of containment are low likelihood scenarios, and such postulated scenarios are moot for the COP question because they would be analyzed in a seismic PRA as core damage scenarios directly. Based on the above discussion, it is judged that seismic issues do not significantly impact the decision making for the BFN EPU and containment overpressure credit. 4-11 C1320503.6924 - 2/27/2006

BEN EPUCOPProbabilisticRisk Assessment Internal Fires The BFN fire risk analysis was performed as part of the Individual Plant Examination of External Events (IPEEE). BFN performed a screening methodology using the EPRI FIVE (Fire Induced Vulnerability Evaluatlion) methodology. Like most plants, BFN currently does not maintain a fire PRA. However, given the very low risk impact of the COP credit, even if fire risk was explicitly quantified the conclusions of this risk assessment are not expected to change, i.e., the risk impact is very small. Other External Hazards In addition to seismic events and internal fires, the BFN IPEEE Submittal analyzed a variety of other external hazards:

  • High WindslTomadoes
  • External Floods
  • Transportation and Nearby Facility Accidents
  • Other External Hazards The BFN IPEEE analysis of high winds, tornadoes, external floods, transportation accidents, nearby facility accidents, and other external hazards was accomplished by reviewing the plant environs against regulatory requirements regarding these hazards.

Based upon this review, it was concluded that BFN meets the applicable NRC Standard Review Plan requirements and therefore has an acceptably low risk with respect to these hazards. As such, these other external hazards are judged not to significantly impact the decision making for the BFN EPU and containment overpressure credit. 4-12 4-12 C1320503-6924 - 2/27/2006

BFNEPUCOP ProbabilisticRiskAssessment Shutdown Risk As discussed in the BFN EPU submittal, shutdown risk is a non-significant contributor to the risk profile of the proposed EPU. The credit for containment overpressure is not required for accident sequences occurring during shutdown. As such, shutdown risk does not influence the decision making for the BFN EPU containment overpressure credit. 4.3 APPLICABILITY TO BFN UNIT 2 AND UNIT 3 This risk assessment was performed using the BFN Unit 1 PRA. To assess the applicability of the Unit 1 results to BIFN Units 2 and 3, the BFN Unit i3 PRA was reviewed. The Unit 3 PRA was explicitly reviewed because it has a higher base CDF than the Unit 2 PRA due to fewer inter-unit crosstie capabilities than Unit 2. Review of the Unit 3 PRA models did not identify any differences that would make the Unit 1 PRA results and conclusions not applicable to Units 2 and 3. As further evidence, the Unit 3 PRA was modified in a similar manner as the Unit I sensitivity Case #2 and quantified to determine the ACDF impact. The result for Unit 3 was a deltaCDF of 1.9E-9/yr. The revised BFN Unit 3 PRA RISKMAN model supporting this review is archived in file U3COP2-9 and saved on the BFN computers along with the other BFN PRA RISKMAN models. Given the above, the results for the Unit 1 PRA risk assessment are comparable to the Units 2 and 3 PRAs. 4-13 C1 320503-6924 - 2127/2006 4-13

BFNEPUCOPProbabilisticRiskAssessment Section 5 CONCLUSIONS The report documents the risk impact of utilizing containment accident pressure (containment overpressure) to satisfy the net positive suction head (NPSH) requirements for RHR and Core Spray pumps during DBA LOCAs. The need for COP credit requests is driven by the conservative nature of design basis accident calculations. Use of more realistic inputs in such calculations shows that no credit for COP is required. The conclusions of this risk assessment are based on the conservative 201.a assumed containment leakage size (refer to Case 2 of Table 4-2). The conclusions of the plant internal events risk associated with this assessment are as follows.

1) Regulatory Guide 1.174 provides guidance for determining the risk impact of plant-specific changes to the licensing basis. Regulatory Guide 1.174 defines very small changes in risk as resulting in increases of core damage frequency (CDF) below 104/yr. Based on this criteria, the proposed change (i.e., use of COP to satisfy the net positive suction head (NPSH) requirements for RHR and Core Spray pumps) represents a very small change in CDF (1.53E-09/yir).
2) Regulatory Guide 1.174 provides guidance for determining the risk impact of plant-specific changes to the licensing basis. Regulatory Guide 1.174 defines very small changes in risk as resulting in increases of Large Early Release Frequency (LERF) below 10-7/yr. Based on this criteria, the proposed change (i.e., use of COP to satisfy the net positive suction head (NPSH) requirements for RHR and Core Spray pumps) represents a very small change in LERF (1.53E-O9ftr).

These results are well within the guideline of RG 1.174 for a "very small" risk increase. Even when modeling uncertainty and parametric uncertainty, and external event scenarios are considered, the risk increase is small. As such, the credit for COP in 5-1 C1320503-0924 - 2/27/2006

BFN EPUCOPProbabilisticRisk Assessment determining adequate NPSH for low pressure ECCS pumps during DBA LOCAs is acceptable from a risk perspective. The general conclusions that the risk impact from the COP credit for DBA LOCAs is very small, applies to BFN Unit 1 as well as BFN Units 2 and 3. 5-2 C1320503 6924 - 2/27/2006

BFNEPUCOPProbabilisticRisk Assessment REFERENCES [1] "Browns Ferry Nuclear Plant (BFN) - Units 2 and 3 - Technical Specifications (TS) Change 448 - One-Time Frequency Extension For Containment Integrated Leakage Rate Test (ILRT) Interval", TVA-BFN-TS-448, July 8, 2004. [2] Risk Impact Assessment of Extended Intearated Leak Rate Testing Intervals, EPRI Report 1009325, Final Report, December2003. [3] "Project Task Report - Browns Ferry Units 1, 2 & 3 EPU, RAI Response - NPSH Sensitivity Studies", GE Nuclear Energy, GE-NE-0000-0050-0044.3-RO-Draft, February 2006. [4] Letter from G.B. Wallis (Chairman, ACRS) to N.J. Diaz (Chaimian, NRC),

   "Vermont Yankee Extended Power Uprate", ACRSR-2174, January 4, 2006.

R-1 C1320503-6924 - 2/27/2006

BFNEPUCOP ProbabilisticRisk Assessment Appendix A PRA QUALITY The BFN Unit 1 EPU PRA was used in this analysis for the base case quantification as it was recently updated consistent with the ASME PRA Standard and it is representative of each of the three BFN unit PRAs. The following discusses the quality of the BFN Unit 1 PRA models used in performing the risk assessment crediting containment overpressure for RHR and Core Spray pump NPSH requirements:

  • Level of detail in PRA
  • Maintenance of the PRA
  • Comprehensive Critical Reviews A. 1 LEVEL OF DETAIL The BFN Unit 1 PRA modeling is highly detailed, including a wide variety of initiating events, modeled systems, operator actions, and common cause events.

The PRA model (Level 1 and Level 2) used for the containment overpressure risk assessment was the most recent internal events risk model for the BFN Unit 1 plant at EPU conditions (BFN model U1050517). The BFN PRA models adopts the large event tree / small fault tree approach and use the support state methodology, contained in the RISKMAN code, for quantifying core damage frequency. The PRA model contains the following rmodeling attributes. A. 1.1 Initiating Events The BFN at-power PRA explicitly models a large number of internal initiating events: A-1 A-i C1320503-6924 - 2/27/2006

BFN EPU COPProbabilisticRisk Assessment

  • General transients
  • LOCAs
  • Support system failures
  • Internal Flooding events The initiating events explicitly modeled in the BFN at-power PRA are summarized in Table A-1. The number of internal initiating events modeled in the BFN at-power PRA is similar to or greater than the majority of U.S. BWR PRAs currently in use.

A.1.2 System Models The BFN at-power PRA explicitly models a large number of frontline and support systems that are credited in the accident sequence analyses. The BFN systems explicitly modeled in the BFN at-power PRA are summarized in Table A-2. The number and level of detail of plant systems modeled in the BFN at-power PRA is equal to or greater than the majority of U.S. BWR PRAs currently in use. A. 1.3 Operator Actions The BFN at-power PRA explicitly models a large number of operator actions:

  • Pre-Initiator actions
  • Post-Initiator actions
  • Recovery Actions
  • Dependent Human Actions Approximately fifty operator actions are explicitly modeled in the BFN PRA. A summary table of the individual actions modeled is not provided here.

A-2 C1320503-6924 - 2/27/2006

BFNEPU COPProbabilisticRisk Assessment The human error probabilities for the actions are modeled with accepted industry HRA techniques. The BFN PRA includes an explicit assessment of the dependence of post-initiator operator actions. The approach used to assess the level of dependence between operator actions is based on the method presented in the NUREG/CR-1278 and EPRI TR-1 00259. The number of operator actions modeled in the BFN at-power PRA, and the level of detail of the HRA, is consistent with that of other U.S. BWR PRAs currently in use. A. 1.4 Common Cause Events The BFN at-power PRA explicitly models a large number of common cause component failures. Approximately two thousand common cause terms are included in the BFN Unit 1 PRA. Given the large number of CCF terms modeled in the BFN at-power internal events PRA, a summary table of them is not provided here. The number and level of detail of common cause component failures modeled in the BFN at-power PRA is equal to or greater than the majority of U.S. BWR PRAs currently in use. A.1.5 Level 2 PRA The BFN Unit 1 Level 2 PRA is designed to calculate the LERF frequency consistent with NRC Regulatory Guidance (e.g. Reg. Guides 1.174 and 1.177) and the PRA Application Guide. The Level 2 PRA model is a containment event tree (CET) that takes as input the core damage accident sequences and then questions the following issues applicable to LERF: A-3 A-3 C1320503-6924 - 2/27/2006

BFN EPUCOPProbabilisticRisk Assessment

  • Primary containment isolation
  • RPV depressurization post-core damage
  • Recovery of damaged core in-vessel
  • Energetic containment failure phenomena at or about time of RPV breach
  • Injection established to drywell for ex-vessel core debris cooling/scrubbing
  • Containment flooding
  • Drywell failure location
  • Wetwell failure location
  • Effectiveness of secondary containment in release scrubbing The following aspects of the Level 2 model reflect the more than adequate level of detail and scope:
1. Dependencies from Level 1 accidents are carried forward directly into the Level 2 by transfer of sequences to ensure that their effects on Level 2 response are accurately treated.
2. Key phenomena identified by the NRC and industry for inclusion in 13WR Level 2 LERF analyses are treated explicitly within the model.
3. The model quantification truncation is sufficiently low to ensure adequate convergence of the LERF frequency.

A.2 MAINTENANCE OF PRA The BFN PRA models and documentation are maintained living and are routinely updated to reflect the current plant configuration following refueling outages and to reflect the accumulation of additional plant operating history and component failure data. The PRA Update Report is evaluated for updating every other refueling outage. The administrative guidance for this activity is contained in a TVA Procedure. Ax4 C1320503-6924 - 2/27/2006

BEN EPU COPProbabilisticRisk Assessment In addition, the PRA models are routinely implemented and studied by plant PRA personnel in the performance of their duties. Potential model modifications or enhancements are itemized and maintained for further investigation and subsequent implementation, if warranted. Potential modifications identified as significant to the results or applications may be implemented in the model at the time the change occurs if their impact is significant enough to warrant. A.2.1 History of BFN PRA Models The current BFN Unit 1 PRA is the model used for this analysis. The BFN Unit 1 PRA was initially developed in June 2004 using the guidance in the ASME PRA Standard, and to incorporate the latest plant configuration (including EPU) and operating experience data. The Unit 1 PRA was then subsequently updated in August 2005. The Unit 1 PRA was developed using the BFN Unit 2 and Unit 3 PRAs as a starting point. The BFN Unit 2 and Unit 3 PRAs have been updated numerous times since the original IPE Submittal. The BFN Unit 2 PRA revisions are summarized below: Original BFN IPE Submittal 9/92 Revision to address plant changes and 8/94 incorporate BFN IE and EDG experience data Revision to ensure consistency with the 4/95 BFN Multi-Unit PRA Revision to address PER BFPIER 970754 10/97 2002 PRA Update 3/02 2004 PRA Update (includes conditions to 6/04 reflect EPU) 2005 Update 8/05 A-5 C1320503.6924 - 2127/2006

BFN EPUCOPProbabilisticRisk Assessment A.3 COMPREHENSIVE CRITICAL REVIEWS As described above, the BFN Unit 1 PRA used in this analysis was built on more than 10 years of analysis effort and experience associated with the Unit 2 and 3 PRAs. During November 1997, TVA participated in a PRA Peer Review Certification of the Browns Ferry Unit 2 and 3 PRAs administered under the auspices of the BVVROG Peer Certification Committee. The purpose of the peer review process is to establish a method of assessing the technical quality of the PRA for its potential applications. The elements of the PRA reviewed are summarized in Tables A-3 through A-4. The Peer Review evaluation process utilized a tiered approach using standardized checklists allowing a detailed review of the elements and the sub-elements of the Browns Ferry PSAs to identify strengths and areas that need improvement. The review system used allowed the Peer Review team to focus on technical issues and to issue their assessment results in the form of a "grade" of 1 through 4 on a PRA sub-element level. To reasonably span the spectrum of potential PRA applications, the four grades of certification as defined by the BWROG document "Report to the Industry on PRA Peer Review Certification Process - Pilot Plant Results" were employed. During the Unit 2 and 3 PSAs updates in 2003, the significant findings (i.e., designated as Level A or B) from the Peer Certification were resolved, resulting in the PRA elements now having a minimum certification grade of 3. The Unit 1 PRA used in this analysis has incorporated the findings of the Units 2 and 3 PSA Peer Review. The previously conducted Peer Review was effectively an administrative and technical Peer Review of the Unit 1 PRA. Similar models, processes, policies, approaches, reviews, and management oversight were utilized to develop the Unit 1 PRA. Ax6 C1320503-6924 - 2/27/2006

BFNEPUCOPProbabilisticRisk Assessment A.4 PRA QUALITY

SUMMARY

The quality of modeling and documentation of the BFN PRA models has been demonstrated by the foregoing discussions on the following aspects:

  • Level of detail in PRA
  • Maintenance of the PRA
  • Comprehensive Critical Reviews The BFN Unit 1 Level 1 and Level 2 PRAs provide the necessary and sufficient scope and level of detail to allow the calculation of CDF and LERF changes due! to the risk assessment requiring containment overpressure for sufficient NPSH for the low pressure ECCS pumps.

A-7 C1320503-6924 - 2/27/2006

BFNEPUCOPProbabilisticRisk Assessment Table A-1 INITIATING EVENTS FOR BFN PRA Initiator l Mean Frequency Category l (events per yea) Transient Initiator Categories Inadvertent Opening of One SRV 1.36E-2 Spurious Scram at Power 8.76E-2 Loss of 500kV Switchyard to Plant .. 02E-2 Loss of 500kV Switchyard to Unit 2.37E-2 Loss of Instrumentation and Control Bus 1A 4.27E-3 Loss of Instrumentation and Control Bus 1B 4.27E-3 Total Loss of Condensate Flow 9.45E-3 Partial Loss of Condensate Flow 1.93E-2 MSIV Closure 5.52E-2 Turbine Bypass Unavailable 1.95E-3 Loss of Condenser Vacuum 9.70E-2 Total Loss of Feedwater 2.58E-2 Partial Loss of Feedwater 2.47E-1 Loss of Plant Control Air 1.20E-2 Loss of Offsite Power 7.87E-3 Loss of Raw Cooling Water 7.95E-3 Momentary Loss of Offsite Power 7.57E-3 Turbine Trip 5.50E-1 High Pressure Trip 4.29E-2 Excessive Feedwater Flow 2.78E-2 Other Transients 8.60E-2 ATWS Categories Turbine Trip ATWS 5.50E-1 LOSP ATWS 7.87E-3 Loss of Condenser Heat Sink ATWS 1.52E-1 Inadvertent Opening of SRV ATWS 1.36E-2 Loss of Feedwater ATWS 3.02E-1 LOCA Initiator Categories Breaks Outside Containment 6.67E-4 Excessive LOCA (reactor vessel failure) 9.39E-9 Interfacing Systems LOCA 3.15E-5 A-8 C1 320503-5924 - 2/27/2006

BFNEPUCOPProbabilisticRiskAssessment Table A-1 INITIATING EVENTS FOR BFN PRA Initiator Mean Frequency Category (events per year) Large LOCA - Core Spray Line Break Loop I 1.68E-6 Loop II 1.68E-6 Large LOCA - Recirculation Discharge Line Break Loop A 1.1 8E-5 Loop B 1.18E-5 Large LOCA - Recirculation Suction Line Break Loop A 8.39E-7 Loop B 8.39E-7 Other Large LOCA 8.39E-7 Medium LOCA Inside Containment 3.80E-5 Small LOCA Inside Containment 4.75E-4 Very Small LOCA Inside Containment 5.76E-3 Internal Flooding Initiator Categories EECW Flood in Reactor Building - shutdown units 1.20E-3 EECW Flood in Reactor Building - operating unit 1.85E-6 Flood from the Condensate Storage Tank 1.22E-4 Flood from the Torus 1.22E-4 Large Turbine Building Flood 3.65E-3 Small Turbine Building Flood 1.65E-2 A-9 C1320503-6924 - 2/27/2006

BFN EPU COPProbabilisticRisk Assessment Table A-2 BFN PRA MODELED SYSTEMS 120V and 250V DC Electric Power AC Electric Power ARI and RPT Condensate Storage Tank Condensate System Containment Atmospheric Dilution Control Rod Drive Hydraulic Core Spray System Drywell Control Air Emergency Diesel Generators Emergency Equipment Cooling Water Feedwater System Fire Protection System (for alternative RPV injection) Hardened Wetwell Vent High Pressure Coolant Injection Main Steam System Plant Air Systems Primary Containment Isolation Raw Cooling Water Reactor Building Closed Cooling Water Reactor Core Isolation Cooling Reactor Protection System Recirculation System Residual Heat Removal System RHR Service Water Secondary Containment Isolation Shared Actuation Instrumentation System SRVs/ADS Standby Gas Treatment System Standby Liquid Control System A-10 A-b C1320503-6924 - 2/27/2006

BFNEPUCOPProbabilisticRiskAssessment Table A-2 BFN PRA MODELED SYSTEMS Suppression Pool / Vapor Suppression Turbine Bypass and Main Condenser A-1 1 C1320503-6924 - 2/27/2006

BFNEPUCOPProbabilisticRiskAssessment Table A-3 PRA PEER REVIEW TECHNICAL ELEMENTS FOR LEVEL 1 PRA ELEMENT CERTIFICATION SUB-ELEMENTS Initiating Events

  • Guidance Documents for Initiating Event Analysis
  • Groupings
                                -   Transient
                                -   LOCA
                                -   Support System/Special
                                -   ISLC)CA
                                -   Brea8k Outside Containment
                               -    Internal Floods
  • Subsumed Events
  • Data
  • Documentation Accident Sequence Evaluation
  • Guidance on Development of Event Trees (Event Trees)
  • Event Trees (Accident Scenario Evaluation)
                               -    Transients
                               -    SBO
                               -    LOCA
                               -   ATWS
                               -    Special
                               -    ISLOCAIBOC
                               -    Internal Floods
  • Success Criteria and Bases
  • Interface with EOPs/AOPs
  • Accident Sequence Plant Damage States
  • Documentation A-12 C1320503-3924 - 2/27/2006

BFNEPUCOPProbabilisticRiskAssessment Table A-3 PRA PEER REVIEWTECHNICAL ELEMENTS FOR LEVEL 1 PRA ELEMENT l CERTIFICATION SUB-ELEMENTS Thermal Hydraulic Analysis

  • Guidance Document
  • Best Eslimate Calculations (e.g., MAAP)
  • Generic Assessments
  • FSAR - Chapter 15
  • Room Heat Up Calculations
  • Documentation System Analysis
  • System Analysis Guidance Document(s)

(Fault Trees)

  • System Models
                              -  Structure of models
                              -  Level of Detail
                              -  Success Criteria
                              -  Nomenclature
                              -  Data (see Data Input)
                              -  Dependencies (see Dependency Element)
                              -  Assumptions
  • Documentation of System Notebooks A-13 C1320503-6924 - 2/27/2006

BFNEPUCOPProbabilisticRiskAssessment Table A-3 PRA PEER REVIEW TECHNICAL ELEMENTS FOR LEVEL 1 PRA ELEMENT CERTIFICATION SUR-ELEMENTS Data Analysis

  • Guidance
  • Component Failure Probabilities
  • System/Train Maintenance Unavailabilities
  • Common Cause Failure Probabilities
  • Unique Uinavailabilities or Modeling Items
                             -   AC Recovery
                             -   Scram System
                             -   EDG Mission Time
                             -   Repair and Recovery Model
                            -    SORV
                            -    LOOP Given Transient
                            -    BOP Unavailability
                            -    Pipe Rupture Failure Probability
  • Documentaflon Human Reliability Analysis
  • Guidance
  • Pre-Initiator Human Actions
                            -    Idenlification
                            -   Analysis
                            -   Quantification
  • Post-Initiator Human Actions and Recovery
                            -   Idenlification
                            -   Analysis
                            -   Quantification
  • Dependence among Actions
  • Documentation A-14 C1320603-6924 - 2/27/2006

BFNEPUCOPProbabilisticRiskAssessment Table A-3 PRA PEER REVIEW TECHNICAL ELEMENTS FOR LEVEL 1

                                                                                   =

PRA ELEMENT CERTIFICATION SUB-ELEMENTS Dependencies

  • Guidance Document on Dependency Treatment
  • Intersysltem Dependencies
  • Treatment of Human Interactions (see also HRA)
  • Treatment of Common Cause
  • Treatment of Spatial Dependencies
  • Walkdowvn Results
  • Documentation Structural Capability
  • Guidance
  • RPV Capability (pressure and temperature)
                                  -    ATWS
                                  -    Transient
  • Containment (pressure and temperature)
  • Reactor Building
  • Pipe Overpressurization for ISLOCA
  • Documentation Quantification/Results
  • Guidance Interpretation
  • Computisr Code
  • Simplified Model (e.g., cutset model usage)
  • Dominant Sequences/Cutsets
  • Non-Dominant Sequences/Cutsets
  • Recovery Analysis
  • Truncation
  • Uncertainty
  • Results:Summary A-15 C1320503.6924 - 2/27/2006

BFNEPUCOPProbabilisticRiskAssessment Table A-4 PRA CERTIFICATION TECHNICAL ELEMENTS FOR LEVEL 2 PRA ELEMENT I CERTIFICATION SUB-ELEMENTS Containment Performance Analysis

  • Guidance Document
  • Success Criteria
  • L1'L2 Interface
  • Phenomena Considered
  • Important HEPs
  • Containment Capability Assessment
  • End state Definition
  • LERF Definition
  • CETs
  • Documentation A-16 C1320503-6924 - 2/27/2006

BFNEPUCOPProbabilisticRisk Assessment Table A-5 PRA CERTIFICATION TECHNICAL ELEMENTS FOR MAINTENANCE AND UPDATE PROCESS PRA ELEMENT CERTIFICATION SUB-ELEMENTS Maintenance and Update Process

  • Guidance Document
  • Input - Monitoring and Collecting New Information
  • Model Control
  • PRA Maintenance and Update Process
  • Evaluation of Results
  • Re-evaluation of Past PRA Applications
  • Documentation A-17 C1320503 6924 - 2/27/2oo6

BFNEPUCOP ProbabilisticRisk Assessment Appendix B PROBABILITY OF PRE-EXISTING CONTAINMENT LEAKAGE Containment failures that may be postulated to defeat the containment overpressure credit include containment isolation system failures (refer to Appendix D) and pre-existing unisolable containment leakage pathways. The pre-existing containment leakage probability used in this analysis is obtained from EPRI 1009325, Risk Impact of Assessment of Extended Integrated Leak Rate Testing Intervals.[2] This is the same approach as used in the recent 200'5 Vermont Yankee EPU COP analyses, and accepted by the NRC and ACRS. [4] EPRI 1009325 provides a framework for assessing the risk impact for extending integrated leak rate test (ILRT) surveillance intervals. EPRI 1009325 includes a compilation of industry containment leakage events, from which an assessment was performed of the likelihood of a pre-existing unisolable containment leakage pathway. A total of seventy-one (71) containment leakage or degraded liner events were compiled. Approximately half (32 of the 71 events) had identified leakage rates of less than or equal to 1La (i.e., the Technical Specification containment allowed leakage rate). None of the 71 events had identified leakage rates greater than 21 La. EPRI 1009325 employed industry experts to review and categorize the industry events, and then various statistical methods were used to assess the data. The resulting probabilities as a function of pre-existing leakage size are summarized here in Table B-1. The EPRI 1009325 study used 100-La as a conservative estimate of the leakage size that would represent a large early release pathway consistent with the LERF risk measure, but estimated that leakages greater than 600La are a more realistic representation of a large early release. B-1 C1320503-6924 - 2/27/2006

BEN EPUCOPProbabilisticRisk Assessment This analysis is not concerned per se about the size of a leakage pathway that would represent a LERF release, but rather a leakage size that would defeat the containment overpressure credit. Given the low likelihood of such a leakage, the exact size is not key to this risk assessment, and no detailed calculation of the exact hole size is performed here. The recent COP risk assessment for the Vermont Yankee Mark I BWR plant, presented to the ACRS in November and December 2005, determined a leakage size of 27La using the conservative 10CFR50, Appendix K containment analysis approach. Earlier ILRT industry guidance (NEI Interim Guidance - see Ref. 10 of EPRI 1009325) conservatively recommended use of 10-La to represent "small" containment leakages and 35La to represent "large" containment leakages. Given the above, the base analysis here assumes 35La as the size of a pre-existing containment leakage pathway sufficient to defeat the containment overpressure credit. Such a hole size does not realistically represent a LERF release (based on EPRI 1009325) and is also believed (based on the VY hole size estimate) to be on the low end of a hole size that would preclude containment overpressure credit. As can be seen from Table B-1, the probability of the 35La pre-existing containment leakage used in this base case analysis is 9.86E-04. Sensitivity studies to the base case quantification (refer to Section 4) assess the sensitivity of the results to the preexisting leakage size assumption. B-2 C132W503,6924 - 2/27/2006

BFNEPUCOPProbabilisticRiskAssessment Table B-1 PROBABILITY OF PRE-EXISTING UNISOLABLE CONTAINMENT LEAK [2] (as a Function of Leakage Size)(') Leakage Size Mean Probability of (La) Occurrence I 2.65E-02 2 1.59E-02 5 7.42E-03 10 3.88E-03 20 1.88E-03 35 9.86E-04 50 6.33E-04 100 2.47E-04 200 8.57E-05 500 1.75E-05 600 1.24E-05 Notes: (1) Reference [2] recommends these values for use for both BWRs and PWRs. Reference 12] makes no specific allowance for the fact that inertedl BWRs, such as BFN, could be argued to have lower probabilities of significant pre-existing containment leakages. B-3 C1 3205036924 - 2/27/2006

BFNEPU COPProbabilisticRisk Assessment Appendix C ASSESSMENT OF RIVER WATER A1ND SP WATER TEMPERATURE VARIATION The BFN river and torus water temperatures were analyzed to statistically model variability in temperature. The purpose of this data assessment is to estimate for use in the risk assessment the realistic probability that these temperatures will exceed a given value, i.e. the probability of exceedance. C.1 BFN EXPERIENCE DATA The following sets of river water inlet and torus water daily temperature data were obtained and reviewed: Unit l Data Period Years 2 001/01 J00 - 01/31/06 6.1 3 T 02/01/03 - 01/31/06 j 3.0 Data for suppression pool water level for the above time periods were also obtained. However, statistical assessment of the variation in pool level was not pursued as the small variation in pool level has a non-significant impact on the COP / NPSH calculations. The river water temperature data from the above units is not pooled because river temperature is dependent upon the seasonal cycle in weather and is not independent between the units. Use of data for SW\ inlet temperatures from multiple units would incorrectly assume the sets of data are independent when in fact they are directly dependent upon weather and the common river source. As such, the statistical assessment of the river water temperature variation uses the largest set of data (i.e., the 6.1 years of data from the Unit 2 river water inlet). c-1 C1320503-6924 - 2/27/2006

BFNEPU COPProbabilisticRisk Assessment As the torus water temperature has a high dependence on river water temperature for most of the year, the assessment of the torus temperature variability also is based on the 6.1 year data set from Unit 2. C.2 STATISTICAL ANALYSIS OF TEMPERATURE DATA The chronological variation in river water temperature and torus water temperature is plotted together on the graph shown in Figure C-1. As can be seen from Figure C-1, the torus water temperature is always equal to or higher than the river water temperature. Also, the river water temperatures and torus temperatures are closely correlated in the warmer months when river water temperature is above approximately 700 F. The 6.1 years of temperature data was categorized into 5-degree temperature bins ranging from 500 F to 990 F degrees. The resulting histograms are shown in Figures C-2 and C-3. Figure C-2 presents histogram for the river water temperature and Figure C-3 presents the histogram for the torus water temperature. The histogram information was then used in a statistical analysis software package (Crystal Ball, a MS Excel add-in, developed by Decisioneering, Inc. of Denver, CO) to approximate a distribution of the expected range in temperature. The Crystal Ball software automatically tests a number of curve fits. The best fit for the temperature data is a normal distribution that is truncated at user-defined upper and lower bounds. If upper and lower bounds are not defined, the tails of the curve fit distribution extend to unrealistic values (e.g., river water and torus water temperatures below 0F degrees). To constrain the distributions, the following user-defined upper and lower bounds were used: C-2 C132050a.6924 - 2/2712006

BFNEPUCOPProbabilisticRisk Assessment

  • River water temperature lower bound of 320 F (no data points in the 6.1 years of data reached 32 0F, only a single data point reached 35 0F)
  • River water temperature upper bound of 950 F (no data points in the 6.1 years of data exceeded 900F)
  • Torus water temperature lower bound of 550 F (no data points in the 6.1 years of data reached lower than 57 0F)
  • Torus water temperature upper bound of 950 F (only a single data point in the 6.1 years of data reached 930F)

The Crystal Ball software statistical results for the river water temperature and torus water temperature variations are provided in Figures C-4 and C-5, respectively. The statistical results are also summarized in the form of exceedance probability as a function of temperature in Figures C-6 and C-7. The information is also presented in tabular form, Tables C-1 and C-2. As discussed previously, the river water and the torus water temperature variations are not independent; as such, the exceedance frequencies are not independent (i.e., they should not be multiplied together directly to determine the probability of exceeding a particular temperature in the river AND at the same time exceeding particular temperature in the torus). C-3 C1 320503 6924 - 2/27/2006

BFN EPU COP ProbabilisticRisk Assessment Figure C-1 CHRONOLOGICAL VARIABILITY IN RIVER WATER AND TORUS WATER TEMPERATURES

         - PoolTemp     -   River Ter 95 85 75 0
0. Jf f I iff f 9 1 lff I A 65 I!

55 45 35 01/01/99 01/01/00 12/31/00 12/31/01 12/31/02 12/31/03 12/30/04 12/30/05 12/30/06 Date Cal C1320503-6924 - 2/27/2006

BFNEPUCOP ProbabilisticRiskAssessment Figure C-2 RIVER WATER TEMPERATURE HISTOGRAM 400-350 300 250 520 - a 150 -  : 32.5 37.5 4A2.5 47.5 52.5 A75 6 25 AT75 779 R 775. P95 P7 9 Q9,5 Temperature C-5 C1320503-6924 - 2/27/2006

BFN EPUCOP ProbabilisticRisk Assessment Figure C-3 TORUS TEMPERATURE HISTOGRAM 700 600 500 400 a 300 200 100 0 lympm"M . F.M.N. S2.S5 S7.S 0-.5 a, 5 One v0 .5 V1.5 07.5 Temperature C-6 C1320503-6924 - 2/27/2006

BFN EPUCOPProbabilisticRisk Assessment Fiigure C4 STATISTICAL RESULTS FOR RIVER WATER TEMPERATURE VARIATION Crystal Ball Report Simulation started on 2/6/06 at 7:09:56 Simulation slopped on 2/6/06 at 7:11:44 Forecast: Pool Temperature Cell: C15 Summary: Display Range is from 55.00 to 95.00 F Entire Range is from 55.00 to 95.00 F After 50,000 Trials, the Std. Error of the Mean is 0.05 Statistics: value Trials 50000 Mean 75.75 Median 76.06 Mode Standard Deviation 11.30 Variance 127.65 Skewness -0.08 Kurtosis 1.85 Coeff. of Variability 0.15 Range Minimum 55.00 Range Maximum 95.00 Range Width 40.00 Mean Std. Error 0.05 Forecast Pool Tenoperatwre

            ,000 Tdals                   FrequenqcChart                    0 Outliers
                             .011                                          573
         .:    .ca3
  • 571 AM0 - 143.2 M50OD 75.00 Wm 75.07 F

C-7 C1320503-6924 - 2/27/2006

BFNEPUCOPProbabilisticRisk Assessment Fligure C-5 STATISTICAL RESULTS FOR TORUS WATER TEMPERATURE VARIATION Crystal Ball Report Simulation started on 2/6/06 at 7:09:56 Simulation slopped on 2/6/06 at 7:11:44 Forecast: River Temperature Cell: G18 Summary: Display Range is from 30.00 to 100.00 F Entire Range is from 32.00 to 95.00 F After 50,000 Trials, the Std. Error of the Mean is 0.08 Statistics: Value Trials 50000 Mean 63.50 Median 63.41 Mode Standard Deviation 18.07 Variance 326.51 Skewness 0.00 Kurtosis 1.81 Coeff. of Variability 0.28 Range Minimum 32.00 Range Maximum 95.00 Range Width 63.00 Mean Std. Error 0.08 Forecast River Temperature 50,000 MHals FrequencyChart 0 Outliers D12 013

         .in   me6                                                        306.6
         .01      .                                                       ZG     C6Il111111101lIL M0 36.0      47.5          8606          8260     106.00 F

C-8 C1320503aW924 - 2/27/2006

BFNEPUCOP ProbabilisticRiskAssessment Figure C-6 RIVER WATER TEMPERATURE EXCEEDANCE PROBABILITY 1.0E40

                          ...........- I...........      ............ .......... .......
                                                    ...---l-                                                                                                      .................................
                                                                                                                                                                                                     .............. I.. . .............................
                                                                                                                                                                                                                                                                 ........... . . ......-....     .............I
             ............................................................. .............................. ...............                                                                                         .......... I...................................................................................
                                                 ..........I......11......I.............. ...11.....-.....                                                                                                                                           ..........................................................
               ........... . . . I............... ............. 11.1-             1..................... 1-11,        ......................................                      ....... .                                                                         ...... ...-. . . .                            - I..I 1.OE-1    ........................................................................................................................................................................................................ .................\.......................................

-I m .......... .......................... .. ................................................................................... ......... ...................... ...... ...... w ....... ........I............... . ....................-. .............................................-. ............ U z 0 LU xli 1.OE-2 ............................

              ..........               ...... ........            ...... .... ......             ... .......... . .... ..........                      ....... .. ........................... ..I........... ...... ...
                                      ........... . ....... I... .... ........ .. ...... - ...........- ............... ....... .............                        .............. ...1- 11   ........        ...11...1-11I-- ...... .......                                    ...... ............ .... . ....... .
                   ...................I... ............................ I...... .......I.... .....' ' ......I......... ............................................I..............1 ........- .11......I.... ....... - ...I........ I......... ....................... ...........
                                                       ............... - .... 11- 1......................- .......... ........ .- - - ..............                            .................................... I...I.... ...-.1-1.1.1....., ....- .............                         ...................-

1.OE-3 25 34 37 41 44 48 51 55 58 62 65 69 72 76 79 83 86 90 93 97 100 RIVER WATER TEMPERATURE (F) C-9 C-9 C1320503-M94 - 212712006

BFNEPUCOPProbabilisticRiskAssessment Figure C-7 TORUS WATER TEMPERATURE EXCEEDANCE PROBABILITY 1.OE+O 1777 7777777 ................ .......... i I0. IL ........................................................................................................................ ............................................. ................................................................. .... ........... z .................................................................................................... ...................................... ......................... ...................................................... ................... ...... ......... ,a4 w w x uJ 1.OE ................................ ................................... ............................................................................................................................ ........................................ 1.OE 50 57 59 61 63 65 67 69 71 73 75 77 79 81 83 85 87 89 91 93 95 TORUS WATER TEMPERATURE (F) C-1 0 C1 320503-6924 - 2127/2006

BFNEPU COPProbabilisticRisk Assessment Table C-1 RIVER WATER TEMPERATURE EXCEEDANCE PROBABILITIES Temperature (F) l Exceedance Probability 30 1.OOE+00 35 9.55E-01 40 8.80E-01 45 8.02E-01 50 7.24E-01 55 6.45E-01 60 5.64E-01 65 4.74E-01 70 3.97E-01 75 3.17E-01 80 2.41 E-01 85 1.64E-01 86 1.40E-01 90 8.46E-02 95 9.15E-03 100 o.OOE+00 C-1 1 C1320503-6924 - 2/27/2006

BFN EPUCOP ProbabilisticRisk Assessment Table C-2 TORUS WATER TEMPERATURE EXCEEDANCE PROBABILITIES Temperature (OF) l Exceedance Probability 30 1.OOE+00 35 1.OOE+00 40 1.OOE+00 45 1.OOE+00 50 1.OOE+O0 55 1.OOE+00 60 8.90E-01 65 7.79E-01 70 6.63E-01 75 5.28E-01 80 4.01 E-01 85 2.62E-01 90 1.35E-01 92 8.25E-02 95 1.01 E-02 100 O.OOE+oo C-12 C13205036924 - 2/27/2006

BFNEPUCOPProbabilisticRisk Assessment Appendix D LARGE-LATE RELEASE IMPACT In the November-December 2005 ACRS meetings concerning the Vermont Yankee EPU and COP credit risk assessments, the! ACRS questioned the impact on Large-Late releases from EPU and COP credit. The following discussion is provided to address this question for the BFN COP credit risk assessment. D.1 OVERVIEW OF BFN PRA RELEASE CATEGORIZATION The spectrum of possible radionuclide release scenarios in the BFN Level 2 PRA is represented by a discrete set of release categories or bins. Typical of industry PRAs, the BFN release categories are defined by the following two key attributes:

  • Timing of the release
  • Magnitude of the release D.1.1 Timing Categorization Three timing categories are used, as follows:
1) Early (E) Less than 6 hours from accident initiation
2) Intermediate (I) Greater than or equal to 6 hours, but less than 24 hours
3) Late (L) Greater than or equal to 24 hours.

The definition of the timing categories is relative to the timing of the declaration of a General Emergency and based upon past experience concerning offsite accident response: D-1 D-I C1320503-6924 - 2/27/2006

BFNEPUCOPProbabilisticRisk Assessment

  • 0-6 hours is conservatively assumed to include cases in which minimal offsite protective measures have been observed to be performed in non-nuclear accidents.
  • 6-24 hours is a time frame in which much of the offsite nuclear plant protective measures can be assured to be accomplished.
   *    >24 hours are times at which the offsite measures can be assumed 1:o be fully effective.

Magnitude Categorization The BFN Level 2 PRA defines the following radionuclide release magnitude classifications:

1) High (H) - A radionuclide release of sufficient magnitude to have the potential to cause prompt fatalities.
2) Medium or Moderate (M) - A radionuclide release of sufficient magnitude to cause near-term health effects.
3) Low (L) - A radionuclide release with the potential for latent health effects.
4) Low-Low (LL) - A radionuclide release with undetectable or minor health effects.
5) Negligible (OK) - A radionuclide release that is less than or equal to the containment design base leakage.

The definition of the source terms levells distinguishing each of these release severity categories is based on the review of existing consequence analyses performed in previous industry studies, PRAs and NRC studies containing detailed consequence modeling. The BFN Level 2 PRA uses cesium as the measure of the source term magnitude because it delivers a substantial fraction of the total whole body population dose. This approach is typical of most industry PRAs. In terms of fraction of core inventory Csl released, the BFN release magnitude classification is as follows: D-2 C1320503e9624 - 2127/2006

BFNEPUCOPProbabilisticRisk Assessment Release Magnitude Fraction of Release Csl Fission Products High greater than 10% Medium/Moderate 1 to 10% Low 0.1 to 1.0% Low-Low less than 0.1% Negligible much less than 0.1% D.2 LLOCA COP CREDIT IMPACT ON LARGE-LATE Based on the preceding discussions, it can be seen that "Large-Late" scenarios are termed High-Late releases in BFN Level 2 PRA terminology and are defined as releases occurring after 24hrs and with a magnitude of >10% CsI. For this risk assessment it is not necessary to perform any explicit quantification of the Level 2 PRA to determine the effect on large-late releases, i.e., the scenarios of interest in this analysis are never late releases, in fact they are all always Early releases. The scenarios of interest in this risk assessment are very low frequency postulated scenarios that were not explicitly incorporated into the BFN base PRA. These scenarios are defined by containment isolation failure at t=0, leading to assumed loss of NPSH to the ECCS pumps in the short term and leading to core damage in approximately one hour. In summary, there is no change in the frequency of Large-Late releases due to the credit of COP in DBA LOCA scenarios. D-3 C1320503-6924 - 2/27/2006

BEN EPUCOPProbabilisticRisk Assessment Appendix E REVISED EVENT TREES This appendix provides print-outs of the! BFN Unit 1 PRA modified event trees used in this analysis. In' addition, the RISKMAN software event tree "rules" and "macros" for these revised event trees are also provided in this appendix. E.1 MODEL CHANGES The following are details of the changes made to the BFN Unit 1 PRA RISKMAN models for this risk assessment. The BFN Unit 1 PRA model of record was modified for this risk assessment to question the status of containment integrity first in the Level 1 large LOCA event trees. In addition, a second node was added to the large LOCA event trees to question the probability of extreme plant conditions (e.g., high river water temperature). These nodes are then used to fail the RHR and CS pumps for scenarios with 2 or less RHR pumps in SPC. The scope of the analysis is limited to large LOCA accidents. In order to ensure that only the large LOCA initiators are affected by the event tree changes, several of the existing event trees were renamed. In addition, because the containment isolation top event CIL is located in the containment event tree CET1, it too was renamed. The event tree names were revised as follows: Original Event New Event Tree Tree Description CET1 CETNI Containment event tree 1 LLCS LLCSN Core spray LLOCA event tree LLRD LLDSN Recirc discharge LLOCA event tree LLO LLON Other large LOCA event tree LLRS LLSN Recirc suction LLOCA event tree E-1 C1 320503 e924 - 2/27/2006

BFN EPUCOPProbabilisticRiskAssessment In the containment event tree, top event CIL was replaced with a dummy top event, CILDUM, which is a switch whose branches depends on CIL, now moved into the large LOCA event trees. Two split fractions were developed for CILDUM, one for success (CILDS) and one for failure (CILDF). The branches of CILDUM depend on C:IL, which is traced via macro CILFAIL. Macro CIL.FAIL is a logical TRUE if top event CIL=F, otherwise it is FALSE. If CILFAIL is TRUE, that is if CIL fails, then the failed branch of CILDUM is assigned via split fraction CILDF (1.OOE+00). Otherwise, the success branch is assigned via split fraction CILDS (0.OCIE+00). The purpose of installing dummy top event CILDUM is to preserve the containment event tree structure (i.e., the RISKMAN software allows use of a specific top event name only once in an accident sequence structure). All top events that are asked in the base model if CIL fails are still asked; those that are not normally asked are not asked in this sensitivity case. In each of the large LOCA event trees, top event CIL was added as the left most top event. Top event NPSH was added as the next top event to the right. In this way, the original event tree structure is preserved because CIL transfers to NPSH which transfers to the original first top of each event tree. CIL models containment isolation penetrations greater than 3 inches, and top event NPSH models the probability of reactor power at 102% as well as river water temperature greater than 86F. Top event NPSH has two split fractions NPSH1 and NPSHS (success, equal to 0.OOE+00). The latter is applied for all initiators other than those modeling large LOCAs. The existing CIL fault tree was modified to add the probability of a pre-existing containment leak; a basic event was inserted just under the top 'OR' gate of the CIL fault tree. The lbasic event is set to different values depending on the size of the leak rate assumed. See Table 4-2 for the sensitivity cases and associated pre-existing leak size. The values used and the resultant CIL split fraction values are listed below: E-2 c1 320503-6924 - 2/27/2006

BFNEPUCOP ProbabilisticRisk Assessment Sensitivity Leak CIL Split Case Leak Size Probabiit Fractions(') Base 35 La 9.86E-04 1.36E-03 1 100 La 2.47E-04 6.22E-04 2 20 La 1.88E-03 2.25E-03 3 Base CIL split fractions X 10, 9.86E-04 6.37E-03 plus pre-existing leak 35 La 4 35 La 9.86E-04 1.36E-03 5 Base CIL split fractions X 10, 1.88E-03 7.37E-03 plus pre-existing leak 20 La Note: 7rpAiI support split fraction. Degraded state split fraction is also affected but not shown. Top event NPSH models the probability that the plant is at 102% reactor power with 86F river water, 'OR' the reactor is at the nominal 100% reactor power level with river water greater than 70F. The probability that the plant is at 102% power is modeled using a miscalibration human error probability taken from a similar action documented in the existing BFN Unit 1 PRA Human Reliability Analysis (see event ZHECCL, instrument calibration error, Control Room). The probability that the river water is either greater than 70F or greater than 86F is developed in the data analysis (refer to Appendix C). Top event NPSH has two split fractions, NPSH1 and NPSHS. The latter is Used to filter out sequences where greater than 3 RHR pumps are running. This latter pass-through split fraction is used to exclude the cases where sufficient RHR pumps are cooling the torus such that containment overpressure is not necessary (per DBA calculations) for the success of the RHR and CS pumps. The status of the RHR pumps and heat exchangers is tracked via an existing macro in the event tree RHRET. Split fraction NPSH1 is the default split fraction. Relfer to Section 4.2.2 where scenarios with more than 2 RHR pumps in SPC are analyzed as a sensitivity case. When both top events CIL and NPSH fail, conditions are present such that the model assumes there is insufficient NPSH for the low pressure pumps to operate during a large LOCA. RISKMAN rules were added to assign guaranteed failure split fractions for E-3 C1320503-6924 - 2/27/2006

BFNEPUCOPProbabilisticRisk Assessment top events: CS, LPCI, LPCII, SPI and SPII. A macro was created (NPSHLOST, defined as CIL=F*NPSH=F) and defined in each large LOCA event tree. The macro was then added to the split fraction rule for each guaranteed failed split fraction for the desired top event. Note that drywell spray failure is captured by the event tree structure i.e., if LPCI loops I and 11are failed, then drywell spray is never asked in the event trees). E-4 C1320503.6924 - 2/27/2006

MODELDWnn:UIEtN Page No. Iof 2 EvesMTrc LCSN.ETI t.yS Fft".14,2W aE CIL NPSN RPSM RPME TOR TTP NOe Ulm LPCII Cs 9 m I1I 1I I1I1 K .-Lz .to 0 C) I_- - Lw i cry 3., At

                                                                            ._.........._.........~~~.                                         _ ..............................
                                            ........... .. .....      . ., .....    .  .,,....I,.....                 ..........      ,........<..I...,...           ,.-   ., ..........              -  , .. ,,..      .   ..............                          . .. .   ......                    .............
                            ............                    .................                         .........                               ......                               ....      .   .......         I.........                ...        .............                 I.. ..  ..........            ,...............

C.,

z Ob

MOM. HneM UWERIN Pmge No 2of2 Evmt Tef LLCNITI 12t" F 14,KM OSPC SPI SRI SPo WM D8 SE 1 1 2 2 3 3 4 4 Xi 5 5-8 9-12 xi 7 13-18 _~~~~~~~~~~~. _.............., ,,.<W,, 8 17

                                               ..                                                                                                                ....................                         9      18 10     19 II     20 A    - -            -------          ...... . .,...                          . .. .. .. . ..... . ..                                    X2 12   2148 m                                                                                                                                                                                                              13     39 X2 14   4-57 15     58 1B     69 X2 17   eD-77 tz t8     78
                                                                                                                                                                                                           )a 19                              C~

20 97 21 8

                                                                                                                                     ..................                                                    X2 22   99-118 23     117 24     118
                                           ........... I. . .........                                               ..._._.... .......... ....... I...............                                         X3 25  119-3 28     237 2?     238 28     239 29     240
                                                                                                                                                                                                              .AJ                             tm 31     242                      Cl Xs 32  243488                      :s
   .......................... ..... I..........................,,.......,,,...I. ............. ,..                                          _-            ..                                               X6 33  4885N8

BFNEPUCOPProbabilisticRisk Assessment Mo~del Rmi~ UICOPI-9 Top Z3v~ztn fo2- Event Wxae.- L=SX~ S;O(; &X af$/goes 1uge I~ CJL RIt4AI =AU~WNT ICAU-0C MX~jPZ LRM~I (-~3 ZJOMS) COND1ITXC0iS'P1O'tIPNO WnN - FRUM R PimirL mZCS Of mP au=UFUL RPSE KLac?&1am Mmam or It?8 REURBQ-SS% BASIS) TOR PRESS'3KE SUP2951(w POOL TTrP TURBINS TM? !VC CE.-guV4 or. =~Vs x OOP T

    ~6n~z   LV    tOOP it Cs           coae SPSAY $Vs~sPe 6T           LMIC WZI'CX FOR BtJMWROIBT 1INOWTCOC OSPC         OPSE"MCR ALIMIS SUPRUUN?MLI            coW.I" OPT ~ ~   Sul PMI104 MODLCOUOLN       FADK       - LOOp I OFP"                   V=

OO coOOrum~ YPA~m= im LO x UEIZ OGM MTCH FOR VUPFRB82Ct POOL, WOLIT4?1 1t~r- uI. OLTU ONT.R??'OA ALIGRZ 13KLZ. GSSRAY

    -DK3     myW911.. SPMA VAPOWt E-7

BFNEPUCOPProbabilisticRiskAssessment Model bema tCOOPZ-9 Split Fraction Assignaeint Rule for Evemnt Tree: LLC 5:1S IN 2/9/2006 SW pl-t fzaotioa Aaap"wt Rule CILIDs<= + tw-s' GI ~ RPG-r(1D~-s t- tV*-34 CXL?2 # we-S 4 LVP' fl))4*3 * + fH*3HJtp.2C*f4 4 uuft*Al*Wuf + 8aR1*U.ft3*(3 4 + Coments IF 3 OR WURZPUMPS AM AflIABLWS DW? MD COP FOR 9C35 VPS3?SWU lZZT.C + INITsLLV + InTaLLr 4 tKI-eLLD

  • 4 Wfl-X.S -lll'L 1VTMLLSB KPSMS I 3RPMs IPC1 I P722 fl5*sstuXs?

..flP 133-u-LPCr -MtPISP T 32ps9 T LPCI2 LPCISUP Coments YfMAE LPKC STA MOCRDUUM LWC; 000 $PUT fCTICO LPCOtr -LCZBUU 4 E3iIL ZC172 LMT-S L011*4 -LCISUP LPCIIE L~~PCI~A 1*018 Car ICT-fLOR (cf4AOsW+0A0.Wfl>74NPXtc C&9S=SSG

                                                                                +tarlhVwr+E&F4 -*EE).
                 + XtvLB           (ne?*nubAE4Ss?+O+wXl?+arva3fl.P+ -ncw) +
52 IRIzf - (11L341A440RX-?+3i444 21t3*LVSF4fCMP4 -32031j 5r221 -(ftVO-F+ DII-?4RD-Y400-f4'SPU'4 ChSZG*tV*F4Ru!?4 -2C)

CSP 1 E-8

BFNEPU COPProbabilisticRisk Assessment Model Namse: UXCOW2-9 Split Fraction .JAsigrant tAe for Evet Tree LCSWN 5: c6 2/9/nat Saga 2 St tYtioa Aausgnaewt Jl as 4

  • 81 LPcIa3'RvA-s3.n8oS 4 X.PSlS'fS-S4'ZPO-S 4' LwCX-B LPCU.E"( (u&-S+Br4 t jftP8-tiWD4 t 084S Ccahnts f l WIR PUMPS OR Cs FRl$M UBDED xOP Sir 1 MSCI
  • RP~St~w
Pl 0310.! 4 Rat-t + Wt5iST 81S2 Sz-C4 a- gRlfrs'ch + 3-tS*'fCl0)

Sfl! 1 StIlt 082.? + R?** f fPS2fgT 4 SPI14 fL8'SIM-5 + 4'SPD-8 tC0-S) *$PS-S 82116 (RPaSAUIX8..S 4 Pbc*ia8*SflnrS ODWSI 1' nis-2XI-FX2-r + ta-ravC-r +SRNS t+SS)

  • rn-tV+ K050
 .DWS         PX1F-S*MFl*   IRPA-*SWC-S) **       3* (fS-8+RPD-8) '-WOD DWS2         ttPAflRPC't 4RHRF4SOGPXlS)
  • t{raPn*rP+RXSI 4 P+

ft) W~SV 1 E-9

BFNEPUCOP ProbabilisticRisk Assessment Vadal Mm: UlCO12-9 vac=0 for ]vmnt Tree: LLCObT 35-.6 Mt WV20ZO6 A_ THIS M.-RPO IS NEE IN MCET? aCt4-L RnzSM-0t x ~zi n ~c czlN-3 RMI~o, CLASS2-A CLASl2T S'PsID&VOSPa CLW8AJ~1 RPR§B CLASS82 E-1 0

BFNEPUCOPProbabilisticRisk Assessment WoI Naze.: CUCOP2-9 URC-o fox z~ent True LL=cW 5'.0f: VI 2/9/2004 Psa a Kaazo Rule cou~ontsx _

           -,sies)i, - (ThN*B+XI ')
          - ITCAvS) 1anxAY VLAS83D RPJM0 THIS HACRO Xg bfr-DMb IN UZ             21.

SIH L04 ThXC14A.ORO 1S XM =0ZIN4 THICCZB

         *RPSM-B
          -&~*        i3(8                 4 LV'5S I-fI!

LOlf TK tL-XsiT.LLCB LPCISUP '

 ~ros    Loop I LPOZ SW1POT THRIs NPcrO IS           D3D Ut TE2S TKSS KI0R       is Xnam IN THE CK.S XORVt rATNA0 159RIPSMO3IN 'fl0 CVS S N03xY T-13       4MACRO is NUMD X    #TSIN xPSeOS.T  ctL-8,brd 1       ;

E-1 I

BFNEPUCOP ProbabilisticRisk Assessment Model. 1Name; IOC02-9 Kacro for Spaxt Tree: LLSI

  • S*4¢c sot X Tais VA0 *18 NfUM W TIM C=S RHEER Gus; YryOOL LMARX~sE,3 M AtRf3 Y Dl£REDr-t E-12

MODL RMmU1Euta pagent I o2 gvetTrue LoEm1 I3t,;2 F Myto 2 IE .L Wm Rpsm RPSM TM TTP rte LUV LPQ es Si We am Spit _ P ows . I (A) 0 C)

                                                                                              , I, Cz.

Q1

t

MOMfiLN~aW U15M Pap t&2Of2 EwntTMt I+/-ONAETI 1t3?:12 Fdran 1IS.

            $     elf         83f I          1 2           2 3          3 4           4 Xi       5          54 Xi       6         942 Xi       7         MI-1S
                        *IT 3           17 10          18 II          20 fl       12       21-3 13          39
          )l       14        40-5 15          55
          )e v

to 17 W77 is 7 gp 19

  -&20                         St 4'.             ~21          9
           )Z      22       09416 23         IlTZ 24         118 m       25       119-236 26         237 28         23o                        e 29         so 30         241 58      32       243-44
                                                         'S G.I RI

BFNEPUCOP ProbabilisticRisk Assessment m~del Eawe: U1COP2-9 Top Events for Zvent Tmees LLOW 5:07 IM 2/3/2006 Pa~g I. CIL PRIM4ARY OOMTAtt4M?9W' GOIMrCK FAILURE LAWMCA..-> INC0XWS1 NP99 ~ COt MSTT.qPRELWIMS 3KPS FM3 LE"MP RPM' WCAMICAL M'VTION OF RN SUOCCESSFUL RPSE Sumfu.a POKOZG? OF RPS INFJRG-5500 ~amsk VCR PRESSURE SU22RZ~OSI POOL T-. P.FTURBINE Tf MULPC L.OOP ~ es COME SP'RAY SYSMn4

     $1          LOGIC 8$ITC MFR SUFFICIEN7 INklBcIlou Ospc              OPRAO      ALIGNS suPPES~oS:M     MgO CoOMING SPIsupSSZON                POOl.

WO 0001.K gm~ax LOG? I LF SUMMR81Z00 POWl. c003,ING KRAMIM'IM0 LOOP 11. LOGIC 8'fl7O FOP $U? ~ST0. MYL COOLIK WITH UI WERU ODWS, OVERATOR A~LIGNS DRY)SLL SPRAY DWSDRYh'ELL 37SMY Humus~ E-1 5

BFNEPUCOPProbabilisticRiskAssessment blodeol Name: ti10co2-9 Split Fraction Assigrmetet Rule for Zvnnt Tree: LLQ Se7 1K 2V9S/MOs page 1I St Split traction lnignnent. Rfle C1L2 pcr - + tvp-sj CILF wt?aF'LVFWr aR1-RMRR RFSS + R llQR4+ R tt t4+ .m M* t Canmats 27 3 OR KM PUMPS Al AVAILABLE WE nOw ? MAg CDP FO ZCs tWSH,. fl?-LLOA 4. Z$IflR 4 LLDA + I?-L=8O + Izle-t.O + XVIT'ILS& + ZUfP'ILS RPISS 1 TORI RIS . I . TT1P3 . 3SBS-POX-S lyi I Xmzr 1 gqw-LPC!F -LP-lStJP + NPSHLOSI' LPC12 LPCZuP. Coamnwts mA1u3I MI STA? MOT CREDITED LLOCS; ODD SLI ThfOK BICOMO APPLI LPCZZF LX5P~fS' ES hCII6 -LPCISOP (aE-rAAt+DA-rAa-Fuc:r*Etzt4D-rLY-r5c-r -Hitc + EISKLS 0525 - (KF-F+AC-40-PADFDO.W+WOIZ-rF4lt>=~br -ASGOfV~tRa4-H Cer -EMU) C3F + srs+ w^+~S :CI*RbzP"5iCw +tts E-16

BFNEPUCOPProbabilisticRiskAssessment Model b;mea: U1012-9 Splt Fraotion Aszsiagment Rule for Zvent 2?ree: LLO' 5:01y aM 2/912005 pa" 2 SF Sp3+/-t naottoa fAint Ru2 Comment AtY tO it VwMS OR CB Tvwxn atiaawtn LOO DSfPC1 Sir. RP3M4S*2S>Ss oswcr 1 SSF un + OsPcn

  • mwprn SPl2 3.

sPnr ossC-T + PI-F + )IFSHLOE? SPI}4 t&P+SaSXBwS 4 RPD"SIHKEw5)*SP4IS SPITS (ftPSflfXZ-S tP0S*KXC)S)*SPI-I*R-S SPUt6 .(.PS-09*ncB-S ., RPC-SfCTh-5) *flPI-VflZ-SPIT? 1. (s SBPI- 3 - (SBPI"E SPCF 1 OWS1 1 Owva w PXlwV*PXvaF + tCR2-ItRP:t-+Mwi?+NOGB)

  • Isr c + 3 )

DWSa pXlinS*flteS* (RPA*3+PJCM'3) *'.4 E

  • R?-SB 0P5 -bu
  • 5 wsa :RPA.F'RPC-F +aa-+N053.frp-r)
  • CSASs*P>fl?-F + $OSQ+?X2F) 50SF 1 E-1 7

BFNEPUCOPProbabilisticRiskAssessment Moel Nam.Q tTlCOP2-9 Macro for li'ment Tree: LL 5:07 af s/s/2009. l PKC8/..

                                        , 1..../......O....

Hae:=o hue / Cadet urssJlo Vnis MA010 is troMdd INtTH ETO ALTINt3= AW~JM2X THIS MACPO- ED Nm dET S CILIA-L R8>3 CLASSIA CLAS3;B CLAM31% C'2SS1BL "M,. CLASS2$ CLAS82A O p-+SPC-F R#sg44 CLAMS2V CLXAUfA E-1 8

BFN EPUCOPProbabilisticRisk Assessment bXdal xsma: M1COI2-9 aoro fox lIvent Trees LLMlE

                                                'SzO'w WMa/nI2006 Vag* t Karc                      Mao~o RUle   / coMMts ClmsA;3c                  -ISI-9 ) s - t TPB+lYC-!'

S. 3:}-iTCR-) CLASS4 CLASSS

                         *HU; MACRO     P xs(&=8D LN tug CWV Kt~lfERPgiWo3 LOW                       XflXLLO LPOU8UPS(                     tWP-2*DW-S      +  LV-S LOOP I K~dr S!3PE'OR W.,

ACd33 RPSB THIS wvmo8 is 1N831D IN Th3s Cm7 N000 M 3

  • tTTP-SIvc-8 *s-s~gpc-s THIS M4ACRO IS WE=l~ IN~TR8 0T8
                         ,ma1 mmo IS NSS!1O IN ?&          CST NOSAV                     RPX-s
                         ?1f8 MARO IS NTEMD IN THEMS TS NPSHLOSTr E-1 9

BFNEPUCOPProbabilisticRisk Assessment mode1 ha:-; U1COP2-9 Mcro fEr 9vouit TrLee:l cawc~~ag musae/ OIPDESP R:S3P Vae MACRO is NFZDmo IE CT LGE LiWAS ARE ALWAYS iMMURIZZEWE E-20

MODEL ftme UIERIN Page No.I of 2 Event Tree: LLRDN.Er1 13:37:46 Febwary 16,2006 IE AL NPSH RPSM RPSE TR TTP IVC DI DV2 LPCI LPCII Cs Si _..........Z....Z....

MODEL Name: U1ERIN Page No. 2 of2 Event Tree: LLRDN.ETI 13:37:46 February 16, 2006 OSPC SPI Spil SPC ODWS MS X# f# S

  • 1 I 2 2 3 3 4 4 Xi 5 5.8 Xl S 9-12 Xi 7 13-16 8 17 9 18 X2 10 19-3B-X2 11 37V-54 X2 12 55.72 X2 13 73.W q

X2 14 91-108 X2 18 109126 t X2 la 127-144 C.

                                                                                                                                                                                                                                            *.1T X4    17  145-2B8 X4   le   289432                               Zil X3    19  433.468
         .. _ - _-e. ...............................................................                                ......................                   ....................           . Xs   20   469-936
                                                                                                                                                  ,,          ,.,     ................             21     937 22      938 23     939 24     940                              4..

X6 P25 94141600 X7 20 18813?70 -2 cle m z

BFNEPUCOPProbabilisticRiskAssessment Model 1am: U1CO2-9 Top Events fr bvent Tree: LLEN 5:05 P 2/3/ZaOO

                                                   *ag* 1 VW Evet siai      P CILI              PRIMRY CONTAINMENT ISQLATIOt         rAILU      - LARGA ('3 1NO3E
                     *-COITIONSpflSfMINk    NtPi- F'O LLOCA zmcm;zN        PCRxon or ApI9 8UOCS mSiL APSE           . ESCTRICAL POTION OF RPIR (NREa-53Q0 BUIS)

TOR patFnsst SUPPRISSION noC

  • wE9 URSINE TRIP IvO CLOSURE OF KlSIY VI LOOP IX4ECTRCATmIWr axSaQS, vALV C
                ?LOOP 1    RE!ORZUAION
                            !               DISC bURR5 VALYSN      CLOSU LNcI              SCI LOOP I LPCII             LP    LOFPI Cs                CORl   SAY SYM t               LOGIC SCTCZ        R SUFFIuCIT        x145CIC)

OPERATOR ALIWS SUPPISU POOL COLNG SI UAPRZSSC*w POOL =COQN4 uNAKs - LOt I SPMI SUPfUSIO OOL COOL:= HARM= - LOOP S;p  ? lOC swrr'c;t FM SUPPRESSiON POOL COOIJO W11THUt3l WI COmS OERAIOI AL--WS DZIWELL S3RAY 3*s D&XI* smAm E-23

BFNEPUCOPProbabilisticRiskAssessment Model NaM: (flCOP2-9 Split Fraction Aasigwinvt Rule for Event Tree: LLRlNq Stag PH 2/9/2005 Panea t SZ Split Fractio Aaatqsmrt.AuI& cati * ?CAS*VMW-$

  • LWR-S) 0CL2 iAar* (ONPRs + kV?-)

CILF MP-~r*LVI-t 4 MPSHdS RJIR~ 3Ii2tP'WS 9+ =f! kpJml*mta4 + P%1 81413*RXa4 + RR2S2R3*L94 , RfMI*PX&2*tR2P3*9RP1 CormstB !f 3 OR IfOR! PUPS ARE AVAXLtALS E DON"t KS5 COP rOv 8005 NPSHI IKVtlZ-LL + zun-xzc .

                                          +  ltWLLD + INITLLOD + IXT*LtO t KNITLLUWA +

nTWI-LLSZ NpSmS I nnaVsS I

"RS0.

I-TORI TTP flP2 B15-S*'01 335.. *D*a mrS TTPT

            .1 zVCI          I DliF          n'r+nar*RCar+N3isrMj2ar+w*rtLvar UVII              aS*LWSfl#MIeS*NU2at*flSiItO.iS DV}2          0VaS*LVaS*VMIsS*lQH2S*       taaar+tC-r, DV13 OV14          Dv*VaS*Nat.4.2.4*RSaS*3tsnS DW'.SLVeS*     (NEF-t441N2*) *9j.5*U.5 1

DV2F RW-f4RE-F20-F+WI1-P KR2.!+DWnF*LV*F 0V21 flt*DVI.WnDWuS* LV-fl" RfluS*5K2w3*fls4*8CinS 0V21 DV1.4*DWaS*LY*S*REia$*NM2.B*RSSS*50s8 DV22 DVIsVtWaStLVsS*BKI.4*tfll2aS*flnS&P.>S DV24 I&at*Pflt*DV$ *Lt)..SuMgI..seNMS. (RS-EWRC-t) E-24

BFNEPUCOPProbabilisticRiskAssessment Mtobol llama: U1OD22-9 Split Traction Assignment Rule for Event Trae; LLWN StO3P fl 2/9/2006 7&Qe 2 SW t rract assgnut tul. 0v223 avle*Dw>s*Lviwrkaas*:a2"s* (UBsf+Cf) 4 0fl4 nlr~rav$F-3*=*ut-82us3.- (BB.WatCs DV28 l.8*DVbS.LVg*NltN8*fl2aSNlS*TC-S 3V29 . va-mw-s*LVfl.8Bl..S2w*iR~9 OV2A; nPfn** rLvS*E1leS 2W3*n ** 0V2t .nDV-rS ~Nsrw**'vtastn  :~2s*nAs-*cS8 2VZD R3..DVla-t*w-StLVS C1B1Y4flQ'4 *4S5*J*S t DV25 &VI'W"*"wS W.f.* (llab-tS#2-Fl *n.*Rcb8 0720 DVflwPDW.8'SW5 (Nuarfl2zcr) *)&3$c*jS DV2? 1 2 r-LtCZSUit rPwS* (r-n .wFlfslsh 1:X2 XLPCISO? LCZZ? -LCflOUvvP 4D-1372-W. + wewtos LPcII4 -LPCTsCP LPt~flS 4lt~tI A-Fr*na-rWs:Dr + n-r+l- F + SY3*SC~.Fx~flnF 'jRAr+io-+nart.as-r4013-rnPnl- CASIGN-Ffl-t4'.? ZMPw-F*&srF 4 flfT*OGPI0W + .flsrncrEDF + K8-rpt-PMn-) + 09 1 .- (PflWr+AWF+30M+flnf+O"f+Ny-Fwr D-F*LVert+Rta+g :nrKasr *rt t

                  *E-+*!c-gD-Fl          *-F*

CS2 DEsVfl£F*S+ Ef+

                   - {CAsszatow-r4v-+DED-r46a-rswa-~:rEW-r            + s&-rssrssn-*3St*X~  aror~4 t~g+8-F+ANFi D~^P~<F E~scPr}*

E-25

BFNEPUCOP ProbabilisticRisk Assessment bxodel zramn: UlCOP2-9 Split Fract:on Asgignent Rule for Event Tree: LLRD& 6:tI9 Ng 2/912006 page 3 Sput 'xactioa husvmuent 1421. CSZR (R3nF4-A&1+DA~t+Aan7.r">:rpZ..f~f~r~tvartKfarg*3trnflsc-r, KAn~ifs atvrs:t 4 BaF*lic~ri:D] + Z8 :kCaCfg'f *- tRZ.V- AC-+DBV4AD-t+D-1+fl-+ eA5*tGott;v-ra5r+Icr~s-rfl-r32.

  • za-rn-rsoe t E~wvF*3C+R~DwE t Msr Cosmenta Core pwray Lap II Pipe Break X4argw LO srs casS + IJCIFS*'CAflS BPB-S) + WC+n$*(tP$4 +flfr) air oSIE' 1 SPXF BP-?+ OC-F + VISSIN.O3 5112 1.

51111 oSpcqr + ar-r +'pataliT 9e114 (Res.Skax*ns + S:-$WS)214 SP1S5 (fP*Btlw I tP3>*&fl5)*jPI-f*l.f s SPIZO (B-3'HKUMS + ftVO-5'sp '-Fl-F SPIT 1 SPth' -t+s.{noa s~ts . is~ts (BFlaS*BXPAZS 4 ftP>sStflC4} 4 Sf514 t Ifl..*j.5+ftf..9*gSa9 $1cr 1 Onwal 1 D57SF uxp*tX2sF + (R;tA-t*aPC-e +PPnrXOGz,

  • fR-rt+a-r + BOGD) own vn rpxa* (-S 4 *-N0* (StP54Pflw5) *-WOf DWS2 CRa-F*RaC-r 4BS"P4tflX.rF * (WP*K r 4 lOGDtX2-F)
                                                                          +aiwE E-26

BFNEPUCOPProbabilisticRisk Assessment Macrzo for 3Iv.et Tree: LZU= ALTIOB ZM CI= .3 NCED N TI Cer THIS AKTNo P PSM / O&3 CLASSISP~' A. "M18 CUU318LRew'-!- C--WAS 3 CLAJ82L Q8LKO.' +

                              *I CLAW23MMA MA~s802               tM-E-27

BFNEPUCOPProbabilisticRisk Assessment

         -                      model    Wane:        t71C012-9 Macrfor        lytat       free I     D Page 2 1       HCWO XACRO     C9NED         T      CT OLASS4  &PSM-WoAcari
4n ftPSF~

RP-BSY low tRIT-LItA 4 lt.T-t+/-DS

                  -
  • j PIl- -) +

pju4- .W8 . . ToluPRDMCO-I VEM.It MM

              -LOOP5 I *-;E IS$      8             0S8 711S MAO     1S X$505     XN THE CTS

$000RE R?h144 NOCY THIS (AsO I3 SE5010 IN 7811E7 tOSRV Rfl4-8 TH2S MACa XS NE E4D0It Tlflit MgTS (*S$PLO-S' OILWNPSE-F E-28

BFNEPUCOPProbabilisticRisk Assessment Koade trame UiCOP2-S Macro fcar stuent Tzea: LLR3 M*aa oBy *s / mmmts x s 1Atts P.OCPSto AmhmYSN Tim CEIS Chaos CAS M LAXYS LEPRE&MUISgEE E-29

MODEL Nwaw: UIER Pap No. i 1 Ewent TDee: LRSEIn 13:38:20 FetW1fl M IE CLa NPSH RPSM RPSE TOR TIP WC DVI DV2 LPCI LPUII Ca S1 O$PC SPI m or Q~ to C) zi CI Ce? (t C-1 9

MoEL Nmnw-UERIN Page No. 2o4 veat TfWm LJRSN.E1 13:38:2D Felxy 18, 20E SR W OM" .W )X P S# 1 I 2 2 3 3 4 4

                                              -. . . . _. .                                5-8
       .,...                   .,_ .                                             X9   8    9-12 XQ   7   13-18 8     17 m

4) 9 18 10 19

            .... . ~--............                                              X10 11   3754
        .          .-........-. -.                             -X               Kill
                                                                               *'-16 12    S5-72 13   73M0.

14 914107

           .. ............. '..........                                         X1$

15 18124

                                                                   #....         X11
                   ..........                      .-                            X   I$  125-141
             ....... _                . ..-.----- ........                       XS  17  142-M                          (VI,
                                                         .  ----             *Xl     18  28423
                         ...........................            -    -           X2  19  424-450
                                                       .   ,^W      ,, e,-2          20  4691 21    917 a-I

_2

MODEl. eNamUIERIN Pape No.3 of 4 Ewvet Tmee: LLRSN.En 13:3&:20 Fetusy 16, 200E E CL NPl RPSM RPSE TOR. TlY Wu MI DV2 LPC1 LPCII Cs Si O0PO SF1 I L.. 9' 0 N C', t9 m CI it

mOLAm: QUIRN Pape No 4 of4 EwrtTncc LLRSN.ETI 1&3&2 Fewa 16, 200 1.1 l SPII SPO ODWS X W l# so

  ....... -23                                                                    919 24   920
     . ...... I............. .... ...... I.;....I...-..........         X12 25 921-1540
   ...........I........... . ..............           .         ..-. X13 25 184M0 C;

Cu Ca 0 tv Cd Ct

                                                                                                              !-S.

t9-Q CY E.:z C11 tt Zs

BFNEPUCOPProbabilisticRisk Assessment Modl Sames U;LcOPZ- 9 Tap Zventg f(o Event Tree: LLBN _pae I TOP Event bwa Dsx&bc CIL ER3Y COOA2AN4T XBSOLM104N flZ - LARUG (s>3 I GCHrnsCAL. OR7TI4 Of RPS 3SUcsSSrJL RPSE E-tlC- M PORTION or RPS (NURE-5500 aSws.. TOR PRSSSURT suttPSSm POL t TP .tras^xn TRf IVC CLOoSre OF MKZYS

  • V L*OP I REaCtLATIU4 VXwZonB vAva CoS0 .

DuV LOOP 1 RSOIRCUIATfON DXSCReRGS VALVE CLOSU LPVI LPCX LOOP I LPI:O LP? LOOP X CS GM1 SE Y SYStEM S: Ltooc SWIMO SuOR UlWgc3NflCINJCTEoN OS9P OPERATOR AL26S SUPPESSIC POOL CO0LZMG OPT SUPoRmSSZoW POOL CO LDJG HARISMU - LOOP I Spri StIPFRtiCNm POOL COOLZK kmAowAI* LOOP.1I PrS L4GIC UAiTCN FIN euMnznX POOL UCLNM WITH V _S 00$ . O:PEATR ALZ*4I ORYHtLL SPRAY DRS GRYNELL SORAY IWARS E-34

BFNEPUCOP ProbabilisticRiskAssessment model Nme: U2.0P2-9 Split Frctin Assignment Rile for Event Tree: LLZIIS 3:l04 PM 21//2006 sr split Fxaction Asuinm.t Rule CIM -CA-s5t wr-S + &w-s) 01L2 PCA'-{W?0S + tfl-S) §FPSEG R3IR*1tm2flE3 + ERnItR2'M4 4 anR1"A3a M2*R *R4 4 RnX*RM*RE3*M.4 Comments IF 3 OR 3MV rMHSO AM AV ABLa WE OM'I RE3 COP noR EDS tP8R2. lI2T-T' k + IN=-=f + flCT-.LA + INIT-LDS 4 INXTII= :El-LUA 4 NIEWMS 'A RPSSO. 1

       ?OtI,       2 TTP1                  B5-3*Dz-s 1

TP3 DB5SF*D -3 IN1I

DV1, v21 DS;-S*LViS*1'2s*S*RC1.5 OV13 5*l1SflC4iaS*Kfl2 (af+s 7)

DV14 DvarVWs*nHls*vS2se"SE*RcO3 avis mf-S*zbvs* (Nel-F+NUI-3 *BZ~aSt *S DY1F I ' ¢raY BFt-il"rSFtcl 1-rtn'.rXgEF*'tff 4 S-25t1*DflF*1ES ^ALVaS*f 1-V1C42-835-5 f0 E2V21 DV1_3* PgnS*LY-SN3El3 *zi21S*PSW3*aO>S Dv22 OV1.*PDWaS *LV"SMiw :H15*MI2aS *R*nwECS DVZ4 s1*fl"in*flw!.S*2fi*SXYwt l*Sa*NfhS*:CRE~FR~ofJ E-35

BFNEPUCOPProbabilisticRisk Assessment Mfdel Name: UrCOP2-9 Split Wraction AssignQrt Rule f or Evart Tre: LLRS

                                                *3:09 P 2/9/2006
                                                       *Pae 2 sW                    split   lraoNtion Asa+/-gmaent RuJlm Dv23                    w                                   (uwF+3CF)

DV2JiV1-?*OW3*LV.9h1R1.Si flfl8* (flF+R0T DVa PfVtFtW Dw~s*Lvsrw ngo m l~s.3*"s flo.393

  • Ca8 Dv2e DVn8Sfl.\A1tVM1,cS*tdi2*SeVRtu3 DVII Dvlt.*Dwts$*Lqe?*u1o4*uest32aStn'a(

uvna gt*>r*Dnflt*Dt;8?*LV~lz-S*#2$aWs5a3S*3CwS OV2B W1.rSVLV*S*Snt8aS* a f3B*f DVID ' DV1.PqNFl.*h'gnStfEhS (IH1-r+312-r *nra DV2t DVPnI*DWfY QQtWZ-W) *p.3.SRC.B8 Dv20 EvI-?*DserS*LV-s* ImIl-T~vH2-FN*TI wD DV2G D¶Jsrwm-NSBLv.* IKUWV4)IflZ'E *pS.)j,. rI 31-rF + DV1MF + ESL .C, L9C12 1 LPC1Z a + nV2-r + nSIMs7 LXC:I2 LI-S-LWCXI^4 FUPF LPCXII L901F*t~w$ CSF (,zor+Aun+DkXr+nFrocwpn^Fla+Wtt*LVar+Rs R+=Es}o d{R-t4A~f403'4t AD>+D D-F.Pflw~-r+ C&5SG+DttXLvr+tw-r -StOP) t IPUL 051 -;(RE.>lF*r+D?&sW+Ahst1wrfltri s~f+DesiPL~v.+Rme1F t

         '            -StElg  *- (1FbE+AC4+D~f1'+kDFtD.V+WP1Iat            ASS1+t*Ftx LV-bf.T     43C#)

C33Z _ (RS-F+MFfiarb3?ABrIhFF L.S+tPIF+Dwwr-~v*F+Rt2-*F+.

                 *    -StCVw)*,(Rt4+Ata+zFl4AD.W4OD-V+V9U-+              cS1G>nfr*LVtirtaSv.-Stew CSF                    1 Conmets         Cor Spray Loop U3 wpLe Sink Mr             LOa t

313 LPI.S&t9 ^.flP3- + LSCII.S*3wBn3*tPgnS

  • LPCZ.SLPCEts4 t.tP&a.#-Siz * {-SM~

E-36

BFNEPUCOPProbabilisticRiskAssessment Ymodal R~aen: UICOV2-9 Split fWactiorz Msiqmamt Rae fr Event Tes. LLRS 3:Ot PK,2/9/gOC6 Wac 3 sW Spt Wraction Asuipgua31t e1& OSPC 1 $Pu%, SP12 IRZ-V" + smrr + Nt3O0S.M . sprtA SPII4 tRP3-S~X3-S 05FC- i3.NS~l:*~SP1FS

                           + Iaxa- ++ sskHLMT         s;-a~

SPI~ SP115 1(RtB*¢S*K stB^g

                                   + R        ReeSODS)Kos           2rst~s*Ssa~kx S:          -jtSP.~(D4 S~tt.S         4                   4-      .                   .-

Lspd W2-PF atws iv- 43 V4MRPG2s 4 GD t~l ' X1.-S^PX2-S~ t'PA.S+32C-S) *-3C* {aPbS n43NS~ *.44O (isP~PR..RC-F l4.tE-SE~+FXl'., * (32-FlA-P43I.. + NQGD+?2-?s~ MVWST E-37

BFN EPUCOPProbabilisticRiskAssessment Model timet VcoP2-4 Umcro for I~veant Tes LIUSS 5sa; VW 2/gI2006 paim 1 X1aa. - Maco Rule Cmmts AYrnXJRR5$lr cO is mpm~4 t4SWmf - 'm4 CFTB RPSOMff THIS ],=RC 1s WtU1E' IN TM3 CETS CILFA;IL CLASSgISA CLASSLBE RP-5B RtI*4 CXASSIEL R-rSv; ' CLASSIE -CLAS61E BM-CLS32 amE-w CLASS2A OOEK-F ZRO~-T KPs348 C1,SSZT CL1W33+/- CLAS3A E-38

BFNEPUCOPProbabilisticRiskAssessment Model N .ame: uICOPZ-9 Macro for Xmut Tree., LT*SU

            -                             5:+/-oe       aM
                                                       /siao6                        -

page 2 Xao Kao Ruas / canm CIASS3C - 8Z -{(TTP-SelvC-eI 633W -jT--ns-8 CLASS4 CMAS$5 - TTFI*- IV 3 DWSPeAY M$5Y5 THPtS iJOAO _S Nt'990D IN THE CXTS 690%, EP-VizaB8 + &c48 + -P (C8*4EC-

                                                                   + D-SU + ICS*3D8S EHDSIPHDh'AR~b THIS.A0CRC is        N        IN 'RE CTs 1DkD HI:GH             S-;
                              +

Le.RS t tr= (+l~44 4 I.V.S I~ LP>C.--SI, R^*l lPSFtfS 4- Lvs I LCISU PRZ-S* t tPx-SMDNSi LY-S LOOP I LIP= SUPPORT 81-8 HCAR23RC RPSMX THIS HAX^0 is S2EDEfD IN rile C91 MOeD RPSK-B

  • TORPS*4TS,!:VC *s s*gpCs THIS xw to mmDag I TRZ CETS RORV THIS oMO IS EM4D : TIs CK-TS TH's NMaO r8 WU=D f Tl2 CGS E-39

BFNEPUCOPProbabilisticRisk Assessment Model lIfmae: UICOP2-9 Usro oar Event, Tax: ILtR Page 3 he~o

            ..   ~

Kacv fl. ... Rule / Gents~

                                  .t I'PAHLOS1   ~   ~    R~>F.

THIS MACQ IS N4UDOD IN THS COS Soav gau-S LRGCE LO0US &REi:LAYE tPRESSM2XIJ E-40

MOWEL Name: UlEPiN PogoNo. It 4 Ever4 Tre ETNI.ETI 1.3:6.50 Febnrty 16, M I IE L2 . AL :CALDUM of IR CZ TQ F Dw WR RME

                       .~~              -----__                                                                                                 I I                                                                                                                    2 I------     -----           -   -          -.                                               3 4

L-- 5 B 7 b _ .S _ 8

                                                                                                             -Li                                9 10 11 12 mp PI I                                                   =

14 15 I l6 17 18 Is .t 29 I 21 C~ L 22 I L- 23 24 1 t - . .---- - ------ ----- r------ 25 ____ I__-__,. _ 26 2.7

                                                                                                                ...          E                  28   Qt.

29

                                                                                                                                               .31    Z-
                                                                          §            .                 f            ----7    '   ' '

32

                                                                                                                       -L--

34 i..... 35 36 C12 C-m

MODEL Name: UIERIN Page No. 2*14 E:vat Tie CETNtETI 1336:50 FebRay 16, 200

             $1 3

4 5 6 7 9 11 11 13 m 14

4. 15 16 17 vt 18 19 20 C~

21 22 23 24 25 ti 26 Q (Zi 27 28 29 30 31 32 33 ft 34 3S cle 36 Is

MODEL amMe UIERIN Papa No. 3 of 4 Event Trwee CETUC.E17 13:36:.D Febuaiy 16f, 2g0 X2 AL CILDUM of -R Z FO DM WR RBMEIX B# I_ ._ 37 I L- 38 i 39 40 41 C', C., Z' "lo C',4

MOL Nww. UIERIN Page No,4of4 Event Tme CETN1.ETI 13:36:50 xuasy 16,20M So 37 38 39 40 41 I'll

                                         .4..

BFNEPUCOPProbabilisticRisk Assessment Model Name: UlCOP2-9 Top Eaots for Efrnt Tres: CeTnI 5:03 DC 2/9/S2006 P &i"UtKam1q 1 I UVEL 2 /nRt RtSlU$LT Al. cmtU LtC MNOS Pak CtUS 2 ASI CLASS1BL cmLam CIL Dw ..lQP ox OPrtPRKv uPnSSURIzz RgP (L2, ZR IZ-VESSEL R&CVSY Cz 0CMADSNT 1tOA! AND INTACT TD 2NJrmzo* CShAB3L!s81D EDtonwre nooDIN' V DW NO DIRECT DRAYWE4 REUASE PATtI 'W WsT AIR SaCE fAlWLR "3O *"AZNMsN? BJILDING' FFTBCTIV$ E-45

BFNEPUCOPProbabilisticRisk Assessment Model Ease: UMC012-9 Split Wrx*tioa Assignment Rule for Event Tree+/- CETW2.

09 W. 2ft/2006 LZOg* tlit v Fati*= evsigfl: Sul*

Comments s20-O XPLII K LkVZL 1 l KPLWS LVSS;. MP Xrr TO CHAtWE Ai CLASSIA + CASSIBE + C:lSSiC + CLASSID + CLASSIE + CLASS3A + CLASS3B t CLASS3. ALO WOCO + CcassUL + aassn + CXASS2L 3 CLASS2T + CLXSS2V + (CLASS3D

  • CLASS4
                    + CLAWSS)     + SucJxs Co-wents        CLASS 3D AND CLASS 4 ARU EVaLUATED FOE DSL Crtug                CILAIL 015                  CLASS3A    4 CLASS3E + CLASSS3c + Lof 1                 CLASS2A + CLLSS2T 4 RORV*(CLASS1A       +   2.SlBS + CLUSaSs   + CLASSICt +

CL~aSla (NOACRO tOD 04 CLASSIS 013 -OPOPL1I (CLASslA + CLASSCW 4 CLSSD1) Commaents change! hiGH PRSSaSu tar 012 OPOE-PLI (CLASSIA + CLASSIC + CW9S1iD) Comments change htCii PPRSSORE Lg9W OnF*ICCLASSIA + CLmASIC) ZR3 CLASS1S' IR4 C1ASlIlx IRS ,O1-CLASSlD IR6 OZ-S*CLASSID Coecsats the izginal CiiU model IR7 Or-1c&aIE IRB 0I-S*CLUSSLS

15. or-s Cossets LM PROSUM xB19Er:ON iXPD=C1T IRF I CZ2 zRa-TAO-s c-CrRtF*=t q

CZl ZR-50D4 CZ.3 l~za-sOnr E-46

BFNEPUCOP ProbabilisticRisk Assessment Modal WNae: U10oP2-9 Split Fraction Assiqnzz.t Rule for Erent Tree:. CZ= 5:01 5'3/s/200V page 2 ST pir action asa fnL TMCL*ASS~lt TD2 0t~s~ffsPtAY 0 TD3 -(01-b) CLAS~S1M TS4 - (o0-9 'OL-ASSIBL TDF ml L0l3RS + D5i~ VDI T0-9'(CLASStA + CL3USB1B CL&ASS1L + CL&81 + cZASs3A + CLASS3B + C01SA3C3 It) tt3-1(CLM$1A + CLASSIC + CMASIlD 4 04833A v CLAsS3S + CEASS3c) F'D4 ?D.8* LS1BIBf+CLS1LL) D#IF I Conuer.ts T0XS*flW5MXAt*WtSPC00'i This s4Ac an easx;;i0n that resulted iis zo RMfi BME7Ss 3S 0ISf'TDS*Fl$S*DSF Irea) o~t~mrt 1.201 Coxtawnts 1.2"-0 ZKIVtS VZ. 1;W2-1 01LT.S1 L=2lg CUS WV TO cca-om AMr ASIIa+ 4CA*1S8 + CIASsIC + CdLAsiD + Clasi + CLA83PA .4 CASS3) + CLASS3C AL) IC= + CiASS3iL, + CASS 4 CLS2L 4 CtAa2? 4 CXASSZV 4 (CLAB + CfA S4

                     + CkSS5)       4 LucKS!

Cormnuts CasS 30 AND CASS 4 n EVmTW.U Dhrt naM CILD.? ' 1 E-47

BFNEPUCOPProbabilisticRisk Assessment Model Name: UlCOP2-9 Split Fraction Assignment Rutle .for Event Tree: C:.E= Sz:@ TX 2(9/2006 Pag 3 sra txactioan &Autgmeeot tuA 0i CLMSS3A + CLS3*3 + MAS3aC + tO 011 CLASS2a + CLASSZt + NtR* (CLSS + CUslSs + CCA1SL ASSC) + CLASBB~lrCNQ{fl + NIC 014 CLASSlB n3 -oP#OPLI*CCLA8SS + CLAISiC + CLUSiD) Comm.ets Chang? hbIWM PRISStI3r LEMt 012 OPD#LI*(CASla + CLUSUO + CLASSiD Commnts tngs I tICK m=ssuag tafl l1l OIzcf(CLASS} + CLASSLCJ In CLSSISEI IRI CLACSS1lD XRS rtas3s51a Commntts the isginal V1L2 to el 1t2-*CtASSIE IRS ODS*S5I cwrents LOW MN=SSW3 INJECtO 21KICIS IRF Cal IRES*O1S-CZ1 IZ-S*O-CZ^3 lR-5*01-v TDI CIB) 'OLASSI TD3 -tIa asa VDa -l(OIBi*CLASS13L TnUB Ot-rACLASSt To r E-48

BFNEPUCOPProbabilisticRiskAssessment M-del iane: ut2COP2-9 Split Fraction. Assigimtat Rule for lvent Tree: CEWNi 5:09 3t 2/5/200>6

                                          )aga 4 SI         SpIut rcaIticn                   Rulet 2       tDaS*(CLSSIA + CXASS            + CLtASSIL + CLASSD 1   ' CIASS3A + CLS38 +       1ASS3C1 FD3        TDeP {(CLS.StA + CLASSIC + CLASSID           CLSS3A + CLAWSf    +

4 CLS) FD4 .nreLt ssas + ASiEL) P.Etl CLSSlLB Coamments TbOS#SP~rYfRRSPCOO Thin was an .ssut iLon that resulted in 1N0 NSE RP=7 Olzttg+ B ESS4 01eS*?flaS*rD0S*:wSc RliE3 OzwS*rDTbrDar LE~O 1 Com=ents L20-Zo' DMVIES VEL 1;i20s ' lKPlLISS LEVEL: USE SF.F TO C.SIGE AL CTASSIA + CtASsiSE + CLASSIC

  • CUSSID + CLASSIE + CASS3A 4 CWASiS13 +

CLAS33C LSO MOD + CLAS815L

  • CLAS912A , ,CIAS32L + WMS32T + LaS2Y 4 (CLASSS3 + 0L3SS4
           + CLASSSJ   +JCSwr B

Cor=ants CLASS 3D AKO CLASS 4 AME EVAWA2tD FOR LEaF CILOF CILFAIL GILDS 1 O:S CLASS& 4' CDM583B 4 CASSS34 LOW o1n CLASE2A 4 CLASS2T + JO9S(CLASSIA, + CLASSIDE i- CLSSIBEL4 CLASSIC) 4 C1ASI3~(E:itSC + NODc) 0X4 -CLSIS t 013 -OPDLl (CLAZSSA + CJiSSIC + CLASSID) Coments ZhangeIt bxIw PmsESS LItY 012 ORDl1,'+*{CLSSIA + CLASsIC + CLASSiD) Cao ents change I hIO# PnESSUIW LEXR E-49

BFNEPUCOPProbabilisticRisk Assessment Model Name: UlCOP2-9 Split Fraction Assignentt Rule foe Zvent Tree: CETlT 5:0! PH 2i9/2006 Page S Sr Split Fraotioa Asaignment Rule IR1 OS-F*(CLASSlA + CLAS81C) IR3 CLASS11E IR4 CLASSlBL IR5 OI-F*CLASS1D IR6 OI-S*CLASSlD Conruents the irginal J1 h2 model IR7 OI-F*CLASS1E IRE OIS*CLASSIE IR2 01-S Conments LOW PRESSURE INJECTION IMPLICIT IRF 1 CS2 IR-F'OI-S CZ4 IR-F*oI-F CZ1 IR-S*OI-S CZ3 IR-S*OI-P CZF 1 TD1 CLASSJE TD2 OI-8*DWSPRAY TD3 -(OI-3) *CLASSlBE TD4 - (OI-B)*CLASS1BL TD0 O1-F*CLASSlA TDF I FOl ALTINJRHSW + DWSPRAY PD2 TD-S* (CLASSlA + CLAS81B.3 + CLASS1BL + CLASSID + CLASS3A + CLASS3B + CLASS3C) FD3 TD-r*CCLASs1A + CLASS1C + CLASS1D + CLASS3A + CLAS53B + CLASS3C) PD4 TD-F*(CLASSIBE + CLASS1:3L) DWIF1 WRK DW-S RM33 CLASS18L E-50

BFNEPUCOPProbabilisticRisk Assessment Model Name: UlCO.2-9 Split Fraction Assignment Rule for Event free: CETNI 5:0 PK 2/9/2006 Wage6 Sr Split Fraction Assignmen: Rule Comments TD-S*DWSPiA"*R*HRSPCOOL This was an assumption that resulted in 100 RBE RME7 OI-t RN36 OI-S*TD-S*FD-S*DWS-S RME5 OI-S*TD-3*FD-S*DWS-F RME4 O1-S*TD-S*FD=F PME3 0I-S*TD-F*FD-F RYEF 1 L20 1 Comments L20-0 NPLIECS LEVEL 1; L20-1 IMPLtES LEVEL2; USE IFr TO C4ANGE ALF CLASS1A + CLASS1BE + CIASSIC + CLASSlD + CLASSlE + CLASS3A + CLASS3B + CLASS3C ALO NOCD + CLASS1BL + CLASS2A + CLASS2L + CLASS2T + CLASS2V + (CLASS3D + CLASS4

            + CLASS5)     + BUCKET Conmments       CLASS 3D AND CLASS'4 ARE EVALUATED FOR LERF CILDF        CILFAIL CILDS        1 OIS          CLASS3A + CLASS3B + CLAES3C.+ LOW OIl   .      CLASS2A + CLASS2T + NORi'*tCLASSlA'+ CLASSiSE + CLASS18L+ CLASS1C)       +

CLASS1B* (OACREC + NODC)' 014 CLASSlB 013 -OPDEPLl*(CLAS5IA + CLASElC + CLASSID) Comments changel hIGi PRESSURE LERF 012 OPDEPLI*(CLASSIA + CLAS81C + CLASSID) Comments change I hIGA PRESSURE LERJ IR1 OI-F*(CLASS1A + 'CLASSC) IR3' CLASSIBE IR4 CLASSI13 IR I6-Fr*cLAss1D IR6 OI-S*CLASS1D Comments the irginal Vi L2 model IR7 OI-F*CLASS1E IRS OI-S'CLASSIE E-51

BFNEPUCOPProbabilisticRiskAssessment Model braze: UlCOP2-9 Split Fraction, Assignmeint Rule for Event Tree: CETNI I3:09 WP2/9/2006 Page 7 SPI *splitb Wraction asaignment Rule ,IR2 OI-S Comments LOW PRESSURE INJECTION IXPLICIT IRF CZ2 IR-F*OI-S CZ4 Cz1 IR-S*OI-S CZ3 IR-S*OI-F CZF 1 TDO CLASSIZ TD2 01-S DOW9PRAY T03 -(OI-B)*CLASSlBE' TD4 -(OI-B)*CLASS1BL TDB OI-F*CLASSlA TDF 1 ED1 ALTINURHSW + DWSPRAY FD2 TD=S*(CLASSLA + CLASS11BE + CLASSlBL + CLASSlD + CLASS3A + CLASS3B + CLASS3C) ?FD3 TD-F*(CLASSlA + CLASS1t: + CLASSlD + CLAS53A + CLASS3E + CLASS3C) FD4 TO-F*(CLASSIBE + CLASS:LIL) 3WIF 1 WRL DW-S RMEB CLASSIBL Comments TD-S*DWSPBAY*RHRSPCOOL This was an assumption that res'ilted in 100 RBE RME7 0I-F RME6 OI-S*TD-S*FDOS*DWS-S RME5 OI=S*TD-S*FD-S*DWS-F RME4 OI-S*TD-S*FD-F RME3 OI-S*TD-F*FD-F RMEF 1 L20 1 E-52

BFNEPUCOPProbabilisticRisk Assessment Model Name: UlCOP2-9 Split Fraction Assignumnt Rule for Event Tree: CETN1 5:09 NX 2/9/200. Page 8 SP Split ftaction As-igiment Rule Corments L20-0 IMPLIES LEVEL 1; L20-1 IMPLIES LEV3L2; USE MFF TO CHANGE ALF CLASSlA + CLAS91BE + CLASSIC + CLASSID + CLASSlE + CLASS3A + CLlSS3B + CLASS3C ALO NacD + CLASSIBL + CLASS2A + CLASS2L + CLASS2T + CLASS2V + (CLASS3D + CLASS4

           + CLASSS)    + BUCKET Conments      CLASS 3D AND CLASS 4 ARE EVALUATED FOR LERF CILDF       CILFAIL CILDS       1 OS          CLASS3A + CLASS3B + CLASS3C + LOW OIl         CLASS2A + CLASS2T + N6RV*(CLASS1A + CLASSIS8        + CLASS1BL+ CLASSIC) +

CLASS1B*(NOACREC + NODC) 014 CLASSlB C13 -OPDEPL1*(CLASS1A + CLASSIC + CLASS1D) Comnments change! hIGG PRESSURE LERN 012 OPDEPL1* (CLASSlA + CLASS1C + CLASS1D) Corments change I hIGH PRESSURE LERF IRI OI-F*(CLASSIA + CLASSiC) _R3 CLASSIBS IR4 CLASS1BL IR5 OI-F*CLASS1D IR6 0I0S*CLASS1D Comments the Irgi:al1 Ul I2 model IR7 OI-F*CLASSIE IRS O1-S*CLASS9E IR2 01-S Connents LOW PRESSURE INJOCTION IMPLICIT IRF 1 CZ2 IR-F*OI-S CZ4 IR-F*OI-F CZ1 IR-S*OI-S CZ3 IR-S*OI-F CZF 1 E-53

BFNEPUCOP ProbabilisticRisk Assessment Modal Name: UlCOP2-9 Split Fraotion Assignment Rule for Ivent Tree: CETNI. 5:05' PM 2/9/2006 Page 9 SF Split rwactiorn Assigmnant Rule TD1 CLP.SSIE TD2 OI-S*DWSpRAY TD3 - (OX-B) *CLASS1BE TD4 -(OI=B)*CLASS1BL TDS OIF*CLASSlA TDF 1 FD1 ALTIJR9SW + DWSPRAY FD2 TD-S*(CLASSIA + CLASSlBlE + CLASSlBD + CLASSlD + CLASS3A + CLASS3UB + CLASS3C) FD3 TD-F*(CLASSIA + CLASS1C + CLASS1D + CLASS3A + CLASS33 + CLASS3C) FD4 TD-F*(CLASS1BE + CLASuwnL) DWIF 1 WRi D=S . RME8 CLASS1BL Commnents TD-S*UWsPRAY*RHRPCOOL This was an assumption that resulted In 100 REM RME7 01-F RME6 OIS*TD-S*FD-S*DWS-S RMES OI S*TD-S*FD-S*DWS-F RE:OI-S*T-FrF tMEW 1-E-54

BFNEPUCOPProbabilisticRiskAssessment Model Nazme: UlCOP2-9 Macro for Event Tree: CETN1 5:09 :PM2/9/2006

                                   ;~age X.

Macro Macro RUe / Coments C1C3LERF CZ=F + RME-F*(CILF}L+DWI:F+IR-F*TD-S*FOS) CZ-F + RME-F* (CILFAIL+DWI"-+IR-F*TD-S*D8-S) CZ-F + RME-h* (CIL9AIL+DWI -F+IR-F*TD-S*FD-S) CZ=L + RME-F*(CILPAIL+DWI-F+IR-F*TD-S*r-SI E-55

BFNEPUCOPProbabilisticRisA Assessment Appendix F FAULT TREES This appendix provides print-outs of the BFN Unit 1 PRA modified containment isolation fault tree and the NPSH fault tree used in this analysis. F-1 C1320503-6924 - 2/27/2006

ThpEMICL CwoI*L-AWbmFiuIUri1 __W _nh%

P1ap lt9
                                                                     -d FARAME3 WaI)

ITHOREQUALWm3vml _ CAaMMPRE4MM I - i PR I-, '1 N~)

2 IPage~l TopEvent CILC rfnt bSoaonFle UinI OWNj0 Die

                                                                     -asi

[wOMD Anlys,.S! i Lt aWksi1:

                !              I ToPOw I     . ,.

l-RAm0I I toG PIMGE SUPPLY IT f

                   'A    OMWMinEAlEWtpT i         Fv 1IRE     I!

I Ge 11 I VALVEFDY4tFALTOLI CLMSEONDEM^ i AFCM--I a?

TcpF"CIL Cantdwnad bbdUm F&Wtd 1 ~~e A2X~ EXHAMAW5~ II

IP 3t4dS NtDw---

                                 . I  1madfe.

A.rtL- nCN - - I

              . 2   ,,1 I-n L"

Fl-*IVju .1I ALTO

                              !   dOSEORDPMAD I            .

_ IPapGSd9 TopEvat CILC uabhtsOM FUiN I.

                                                                       .j tps2,0bo he1tubl Up                                - i -      _ _

1 4 I SEGHA __ To.Pa q*'. .. ToP2 I; ,, '1

t TopEventCLOoantint 1IW FdmeU4t _ heft Uio DL9' - LOMAofntm: DM2VM-PE 4 TRA1O# X19biTh I 1R. lCIX.,DRAIN I P-EEWJlI4T R W5. l 1 i, I

                                                         .,J _

-n

TcPEACL Wen ut lxk3b2 Fifte hidI jPage7ofg AVptUno*qmbg3 ILama-ciat 0 1 Date m CO

Corn eitbdl i FneUnit1 ;Page8.of0 TopEventCIL

Dae; WJ AndyL kh OPM Las!tWifeati:D n

TopEnt&CL Cmwtiatblsda Fbe Urit. -.Papgol An~x

  • Ledn _ . M . _ ,, _
                                                                             - :O -lam.n.V.

SYfOLNAIAE Pt SYfEIM.YPE SYTBOLBAME PS SYIUWTYPE AOYFCfNIT 2 AS 2 W COALKIE09410110 2 eAUCVENr AOWlMFCV4h 2 GASWEWNT CVLbt...W NI S BASC.EWI OMYIFNCWW041i 2 RAW-PM D I H1JSE_EWIT AOWCIPC S 2 . 2 BAIJV IES I ANl (ATE AcmFClV w4aW2 2 PASII:YI AOVFFC1VO422 2 BAsc- pow Slime I COR-ATE AOVPIFCVIS40 2 EAE JrW - .

                                                                                                              *z  1         LGA   1E ADFC1WCWO4W           3      aMfCJEYer AOW FC1VCW    0033      2    BAUCJWWVER                    W~tiINN            2       ARDtGATE AOWCIRl               32I      MM    E                    (tmmlm              2       OR-GATE A OVFCM  if IVIN        2    IA SJYEPW                      amcm              2       OR-tGATE AOW IFCYPIH2g4        I       BMIJWNT.                    O VttA              2       TRAMSFBRBJT t    DCIFCVO7A       4       WJVG     T                  G"BOA               I       TRAuSFEVI ADVFCIt!2             4      BANC-EFOl                    UOAbE               2       OlIGATE AOWCIPVWMSA           4      BASIC-PEM                       HMEBt
  • 2 AR3_GAlE AC*CtFC TIBIU 4 BAMCVEWT 00tYED. 2 OGATE AXO iPCV 4C01 I BASICJVUOTf SO t 2 .* NA *! .

A 1X)lU40d16 2 SAMtPAW OSGa 2 OLGAlE AoMltVuhttI 2 .ASOJA` -W 2 TRAMSFERLJOUT ADV'ZFCNCM 2 gm~mw aGi I TRA N AiV_ 2tVMI 2 BASIIECPA I t 3 ANDWE AoWxoiPCWI4U23 BAVQWT S3 OILGAtE AMOKVM 3 PAICP OI2 A 3 ROI !ATE -ni _ 2001411103 r BANCfi W3 A14DATE 0 cl ,1290 32

  • _..SATE AOVXOIFC WTIfoO 2 BAKIQJABI Gn
  • 32 OLGATE AOVXOIFCVIW 4 BASICFJFNT Gram 2 URANMBWSG AWOIFCWQWU2B 3 BASICPEN0T GIJIA I TRAI A SX1CVIB 4 BASICEIlift GS1VDB 2 CRAGATE AOYXOFCWIUgM 4 psAWENT A _3VTClM 4 OAICr CO11EA 2 AND-GATE AOYMlt5MA 4 S"51cHEtt OS3 2 OROATE-O AOYAOitF CYPTI/St 4 SAW-I IBEIA 2 TRA NIROWJ CLiJcViU 4 ORGATE GoOKIA 2 *RAMJN.

C 7jcv 2I S1tIATE. OD ESI 3 C.R5 1!TE CI7tfS2A 4 ORAGATE SODA 2 T h3NSFEILOtIT CIeCt rs 4 WATE S4MEA 2 IRANIB.LIt asCj`v772 4 OR-ATE. 04Emas 3 ORLGAtE CtsFcVnlR 4 OPGAvTE LWP I H8U02WEENT 3W34J3 2 TIlFElCtT PCA 2 HOUSEOM

  • au tIJWPTRSJKt PCA 2 IOU V40lT CoISII-sOPR I3 AND-S1ATE CLXIs 4 tRANIAIT a.lon s TRtusmm ULJEiB 4 AND-IATE CL X19 4 tRANRI OM CL_,Xl9 4 ANO-SIAtE CJIU I TIANWEP IN COalFN t I BASIC-PEfT GCWTL.5- 401011 a BASCJVSITf

CONDITIONS PREVENTING ECCS NPSH FOR LLOCA CASES NPSH CASE1:RXPOWERATI RIVER WATER 1o2% A GREATER THAN 89F

                    =]iP.       RVER89.

Lj ' IASCAUBRATION I RIVER WATER ERROR RESULTING IN'l GREATER THAN 70F ACTUAL POWER 102% '1 ZHECdL RIVER70 I{ I I N

SYMBOL NAME P# SYMBOL TYPE 1 ANDGATE NPSH 1 OR.GATE RIVER70 1 BASICJEVENT RIVER89 1 BASIC-EVENT ZHECCL 1 BASIC.EVENT l1 p.)}}