ML17212A056

From kanterella
Jump to navigation Jump to search
Final Safety Analysis Report, Rev. 35, Chapter 14, Safety Analysis
ML17212A056
Person / Time
Site: Millstone Dominion icon.png
Issue date: 06/29/2017
From:
Dominion Nuclear Connecticut
To:
Office of Nuclear Reactor Regulation
Shared Package
ML17212A038 List:
References
17-208
Download: ML17212A056 (497)


Text

{{#Wiki_filter:Millstone Power Station Unit 2 Safety Analysis Report Chapter 14

Table of Contents tion Title Page GENERAL............................................................................................................ 14.0-1

.1       Classification of Plant Conditions ............................................................ 14.0-1
.1.1     Acceptance Criteria................................................................................... 14.0-2
.1.2     Classification of Accident Events by Category ........................................ 14.0-3 0.2       Plant Characteristics and Initial Conditions.............................................. 14.0-3
.3       Power Distribution .................................................................................... 14.0-4
.4       Range of Plant Operating Parameters and States..................................... 14.0-4
.5       Reactivity Coefficients Used In The Safety Analysis .............................. 14.0-4
.6       Scram Insertion Characteristics ................................................................ 14.0-4
.7       Trip Setpoint Verification ......................................................................... 14.0-4
.7.1     Reactor Protection System........................................................................ 14.0-5
.7.2     Specified Acceptable Fuel Design Limits ................................................ 14.0-5
.7.3     Limiting Safety System Settings............................................................... 14.0-6
.7.3.1   Local Power Density................................................................................. 14.0-6
.7.3.2   Thermal Margin/Low Pressure ................................................................. 14.0-6
.7.3.3   Additional Trip Functions......................................................................... 14.0-6
.7.4     Limiting Conditions for Operation ........................................................... 14.0-6
.7.4.1   Departure From Nucleate Boiling............................................................. 14.0-6
.7.4.2   Linear Heat Rate ....................................................................................... 14.0-7
.7.5     Setpoint Analysis ...................................................................................... 14.0-7
.7.5.1   Limiting Safety System Settings............................................................... 14.0-7
.7.5.2   Limiting Conditions for Operation ........................................................... 14.0-8
.8       Component Capacities and Setpoints ....................................................... 14.0-8
.9       Plant Systems and Components Available For Mitigation of Accident Effects..

14.0-8

.10      Effects of Mixed Assembly Types and Fuel Rod Bowing ....................... 14.0-9
.11      Plant Licensing Basis and Single Failure Criteria .................................... 14.0-9
.12      Plot Variable Nomenclature.................................................................... 14.0-10
.13      References............................................................................................... 14.0-10 INCREASE IN HEAT REMOVAL BY THE SECONDARY SYSTEM............ 14.1-1
.1       Decrease in Feedwater Temperature......................................................... 14.1-1 14-i                                                               Rev. 35

tion Title Page

.1.1   Event Initiator ........................................................................................... 14.1-1
.1.2   Event Description ..................................................................................... 14.1-1
.1.3   Reactor Protection..................................................................................... 14.1-1
.1.4   Disposition and Justification..................................................................... 14.1-1
.2     Increase in Feedwater Flow ...................................................................... 14.1-2
.2.1   Event Initiator ........................................................................................... 14.1-2
.2.2   Event Description ..................................................................................... 14.1-2
.2.3   Reactor Protection..................................................................................... 14.1-2
.2.4   Disposition and Justification..................................................................... 14.1-2
.3     Increase in Steam Flow............................................................................. 14.1-3
.3.1   Event Initiator ........................................................................................... 14.1-3
.3.2   Event Description ..................................................................................... 14.1-3
.3.3   Reactor Protection..................................................................................... 14.1-3
.3.4   Disposition and Justification..................................................................... 14.1-4
.3.5   Definition of Events Analyzed ................................................................. 14.1-5
.3.6   Analysis Results........................................................................................ 14.1-5
.3.7   Conclusion ................................................................................................ 14.1-6
.4     Inadvertent Opening of a Steam Generator Relief or Safety Valve.......... 14.1-6
.4.1   Event Initiator ........................................................................................... 14.1-6
.4.2   Event Description ..................................................................................... 14.1-6
.4.3   Reactor Protection..................................................................................... 14.1-6
.4.4   Disposition and Justification..................................................................... 14.1-6
.5     Steam System Piping Failures Inside and Outside of Containment ......... 14.1-6
.5.1   Pre-Scram Analysis................................................................................... 14.1-7
.5.1.1 Event Initiator ........................................................................................... 14.1-7
.5.1.2 Event Description ..................................................................................... 14.1-7
.5.1.3 Reactor Protection..................................................................................... 14.1-7
.5.1.4 Disposition and Justification..................................................................... 14.1-7
.5.1.5 Definition of Events Analyzed ................................................................. 14.1-8
.5.1.6 Analysis Results...................................................................................... 14.1-13
.5.1.7 Conclusions............................................................................................. 14.1-15
.5.2   Post-Scram Analysis ............................................................................... 14.1-15
.5.2.1 Event Initiator ......................................................................................... 14.1-15
.5.2.2 Event Description ................................................................................... 14.1-15
.5.2.3 Reactor Protection................................................................................... 14.1-16
.5.2.4 Disposition and Justification................................................................... 14.1-16
.5.2.5 Definition of Events Analyzed ............................................................... 14.1-17
.5.2.6 Analysis Results...................................................................................... 14.1-21
.5.2.7 Conclusions............................................................................................. 14.1-25 14-ii                                                               Rev. 35

tion Title Page

.5.3     Radiological Consequences of a Main Steam Line Break...................... 14.1-26
.6       References............................................................................................... 14.1-27 DECREASE IN HEAT REMOVAL BY THE SECONDARY SYSTEM........... 14.2-1
.1       Loss of External Load............................................................................... 14.2-1
.1.1     Event Initiator ........................................................................................... 14.2-1
.1.2     Event Description ..................................................................................... 14.2-1
.1.3     Reactor Protection..................................................................................... 14.2-1
.1.4     Disposition and Justification..................................................................... 14.2-1
.1.5     Definition of Events Analyzed ................................................................. 14.2-2
.1.6     Analysis Results........................................................................................ 14.2-2
.1.7     Conclusion ................................................................................................ 14.2-3
.2       Turbine Trip .............................................................................................. 14.2-3
.2.1     Event Initiator ........................................................................................... 14.2-3
.2.2     Event Description ..................................................................................... 14.2-3
.2.3     Reactor Protection..................................................................................... 14.2-4
.2.4     Disposition and Justification..................................................................... 14.2-4
.3       Loss of Condenser Vacuum ...................................................................... 14.2-4
.4       Closure of the Main Steam Isolation Valves ............................................ 14.2-4
.4.1     Event Initiator ........................................................................................... 14.2-4
.4.2     Event Description ..................................................................................... 14.2-4
.4.3     Reactor Protection..................................................................................... 14.2-5
.4.4     Disposition and Justification..................................................................... 14.2-5
.4.5     Definition of Events Analyzed ................................................................. 14.2-6
.4.6     Analysis Results........................................................................................ 14.2-7
.4.7     Conclusion ................................................................................................ 14.2-8
.5       Steam Pressure Regulator Failure............................................................. 14.2-8
.6       Loss of Nonemergency AC Power to the Station Auxiliaries .................. 14.2-8
.7       Loss of Normal Feedwater Flow .............................................................. 14.2-8
.7.1     Event Initiator ........................................................................................... 14.2-8
.7.2     Event Description ..................................................................................... 14.2-8
.7.3     Reactor Protection..................................................................................... 14.2-9
.7.4     Disposition and Justification..................................................................... 14.2-9
.7.5     Definition of Events Analyzed ............................................................... 14.2-10
.7.5.1   Analysis Results...................................................................................... 14.2-10
.7.6     Conclusions............................................................................................. 14.2-11 14-iii                                                              Rev. 35

tion Title Page

.8       Feedwater System Pipe Breaks Inside and Outside Containment .......... 14.2-11
.9       References............................................................................................... 14.2-11 DECREASE IN REACTOR COOLANT SYSTEM FLOW................................ 14.3-1
.1       Loss of Forced Reactor Coolant Flow ...................................................... 14.3-1
.1.1     Event Initiator ........................................................................................... 14.3-1
.1.2     Event Description ..................................................................................... 14.3-1
.1.3     Reactor Protection..................................................................................... 14.3-1
.1.4     Disposition and Justification..................................................................... 14.3-1
.1.5     Definition of Events Analyzed ................................................................. 14.3-2
.1.6     Analysis Results........................................................................................ 14.3-2
.1.7     Conclusion ................................................................................................ 14.3-3
.2       Flow Controller Malfunction .................................................................... 14.3-3
.3       Reactor Coolant Pump Rotor Seizure ....................................................... 14.3-3
.3.1     Event Initiator ........................................................................................... 14.3-3
.3.2     Event Description ..................................................................................... 14.3-3
.3.3     Reactor Protection..................................................................................... 14.3-3
.3.4     Disposition and Justification..................................................................... 14.3-3
.3.5     Definition of Events Analyzed ................................................................. 14.3-4
.3.6     Analysis Results........................................................................................ 14.3-4
.3.7     Conclusion ................................................................................................ 14.3-4
.4       Reactor Coolant Pump Shaft Break .......................................................... 14.3-4
.5       References................................................................................................. 14.3-4 REACTIVITY AND POWER DISTRIBUTION ANOMALIES ........................ 14.4-1
.1       Uncontrolled Control Rod/Bank Withdrawal From A Subcritical or Low-Power Startup Condition ...................................................................................... 14.4-1
.1.1     Event Initiator ........................................................................................... 14.4-1
.1.2     Event Description ..................................................................................... 14.4-1
.1.3     Reactor Protection..................................................................................... 14.4-1
.1.4     Disposition and Justification..................................................................... 14.4-2
.1.5     Definition of Events Analyzed ................................................................. 14.4-2
.1.6     Analysis Results........................................................................................ 14.4-2
.1.7     Conclusion ................................................................................................ 14.4-3
.2       Uncontrolled Control Rod/Bank Withdrawal At Power........................... 14.4-3
.2.1     Event Initiator ........................................................................................... 14.4-3 14-iv                                                               Rev. 35

tion Title Page

.2.2   Event Description ..................................................................................... 14.4-3
.2.3   Reactor Protection..................................................................................... 14.4-3
.2.4   Disposition and Justification..................................................................... 14.4-4
.2.5   Definition of Events Analyzed ................................................................. 14.4-4
.2.6   Analysis Results........................................................................................ 14.4-4
.2.7   Conclusion ................................................................................................ 14.4-4
.3     Control Rod Misoperation ........................................................................ 14.4-5
.3.1   Dropped Control Rod/Bank ...................................................................... 14.4-5
.3.1.1 Event Initiator ........................................................................................... 14.4-5
.3.1.2 Event Description ..................................................................................... 14.4-5
.3.1.3 Reactor Protection..................................................................................... 14.4-5
.3.1.4 Disposition and Justification..................................................................... 14.4-6
.3.1.5 Definition of Events Analyzed ................................................................. 14.4-6
.3.1.6 Analysis Results........................................................................................ 14.4-6
.3.1.7 Conclusion ................................................................................................ 14.4-7
.3.2   Dropped Part-Length Control Rod ........................................................... 14.4-7
.3.3   Malpositioning of the Part-Length Control Rod Group............................ 14.4-7
.3.4   Statically Misaligned Control Rod/Bank .................................................. 14.4-7
.3.5   Single Control Rod Withdrawal ............................................................... 14.4-7
.3.5.1 Event Initiator ........................................................................................... 14.4-7
.3.5.2 Event Description ..................................................................................... 14.4-8
.3.5.3 Reactor Protection..................................................................................... 14.4-8
.3.5.4 Disposition and Justification..................................................................... 14.4-8
.3.5.5 Definition of Events Analyzed ................................................................. 14.4-9
.3.5.6 Analysis Results........................................................................................ 14.4-9
.3.5.7 Conclusion ................................................................................................ 14.4-9
.3.6   Reactivity Control Device Removal Error During Refueling .................. 14.4-9
.3.7   Variations in Reactivity Load to be Compensated by Burnup or On-Line Refu-eling .......................................................................................................... 14.4-9
.4     Startup of an Inactive Loop .................................................................... 14.4-10
.4.1   Event Initiator ......................................................................................... 14.4-10
.4.2   Event Description ................................................................................... 14.4-10
.4.3   Reactor Protection................................................................................... 14.4-10
.4.4   Disposition and Justification................................................................... 14.4-10
.5     Flow Controller Malfunction .................................................................. 14.4-10
.6     Chemical and Volume Control System Malfunction That Results in a Decrease In The Boron Concentration in the Reactor Coolant .............................. 14.4-11
.6.1   Event Initiator ......................................................................................... 14.4-11
.6.2   Event Description ................................................................................... 14.4-11 14-v                                                               Rev. 35

tion Title Page

.6.3     Reactor Protection................................................................................... 14.4-11
.6.4     Disposition and Justification................................................................... 14.4-11
.6.5     Definition of Events Analyzed ............................................................... 14.4-12
.6.6     Analysis Results...................................................................................... 14.4-12
.6.7     Conclusions............................................................................................. 14.4-13
.7       Inadvertent Loading and Operation of a Fuel Assembly in an Improper Position 14.4-13
.8       Spectrum of Control Rod Ejection Accidents......................................... 14.4-13
.8.1     Event Initiator ......................................................................................... 14.4-13
.8.2     Event Description ................................................................................... 14.4-13
.8.3     Reactor Protection................................................................................... 14.4-14
.8.4     Disposition and Justification................................................................... 14.4-14
.8.5     Definition of Events Analyzed ............................................................... 14.4-14
.8.6     Analysis Results...................................................................................... 14.4-15
.8.7     Conclusion .............................................................................................. 14.4-15
.8.8     Radiological Consequences .................................................................... 14.4-16
.9       Spectrum of Rod Drop Accidents (Boiling Water Reactor) ................... 14.4-17
.10      References............................................................................................... 14.4-17 INCREASES IN REACTOR COOLANT SYSTEM INVENTORY................... 14.5-1
.1       Inadvertent Operation of the Emergency Core Cooling System That Increases Reactor Coolant Inventory........................................................................ 14.5-1
.2       Chemical Volume and Control System Malfunction That Increases Reactor Coolant Inventory ..................................................................................... 14.5-1 DECREASES IN REACTOR COOLANT INVENTORY .................................. 14.6-1
.1       Inadvertent Opening of a Pressurized Water Reactor Pressurizer Pressure Relief Valve ......................................................................................................... 14.6-1
.1.1     Event Initiator ........................................................................................... 14.6-1
.1.2     Event Description ..................................................................................... 14.6-1
.1.3     Reactor Protection..................................................................................... 14.6-1
.1.4     Disposition and Justification..................................................................... 14.6-1
.1.5     Definition of Events Analyzed ................................................................. 14.6-2
.1.6     Analysis Results........................................................................................ 14.6-2
.1.7     Conclusions............................................................................................... 14.6-2
.2       Radiological Consequences of the Failure of Small Lines Carrying Primary 14-vi                                                               Rev. 35

tion Title Page Coolant Outside of Containment .............................................................. 14.6-3

.3     Radiological Consequences of Steam Generator Tube Failure ................ 14.6-3
.3.1   Event Initiator ........................................................................................... 14.6-3
.3.2   Event Description ..................................................................................... 14.6-3
.3.3   Reactor Protection..................................................................................... 14.6-4
.3.4   Disposition and Justification..................................................................... 14.6-4
.3.5   Definition of Events Analyzed ................................................................. 14.6-5
.3.6   Analysis Results........................................................................................ 14.6-7
.3.6.1 Thermal-Hydraulic Calculation ................................................................ 14.6-7
.3.6.2 Radiological Calculation........................................................................... 14.6-8
.3.7   Conclusion ................................................................................................ 14.6-9
.4     Radiological Consequences of a Main Steam Line Failure Outside Containment 14.6-9
.5     Loss of Coolant Accidents Resulting From a Spectrum of Postulated Piping Breaks Within the Reactor Coolant Pressure Boundary ........................... 14.6-9
.5.1   Large Break Loss of Coolant Accidents ................................................... 14.6-9
.5.1.1 Event Initiator ........................................................................................... 14.6-9
.5.1.2 Event Description ................................................................................... 14.6-10
.5.1.3 Reactor Protection................................................................................... 14.6-10
.5.1.4 Disposition and Justification................................................................... 14.6-10
.5.1.5 Definition of Events Analyzed ............................................................... 14.6-11
.5.1.6 Summary of Results................................................................................ 14.6-14
.5.1.7 Post Analysis of Record Evaluations...................................................... 14.6-15
.5.1.8 Conclusions............................................................................................. 14.6-15
.5.2   Small Break Loss of Coolant Accident................................................... 14.6-15
.5.2.1 Event Initiator ......................................................................................... 14.6-15
.5.2.2 Event Description ................................................................................... 14.6-16
.5.2.3 Reactor Protection................................................................................... 14.6-16
.5.2.4 Disposition and Justification................................................................... 14.6-16
.5.2.5 Definition of Events Analyzed ............................................................... 14.6-17
.5.2.6 Analysis Results...................................................................................... 14.6-22
.5.2.7 Post Analysis of Record Evaluations...................................................... 14.6-23
.5.2.8 Conclusions............................................................................................. 14.6-23
.5.3   Post-LOCA Long Term Cooling ............................................................ 14.6-23
.5.3.1 The Post-LOCA Long Term Cooling Plan ............................................. 14.6-24
.5.3.2 Post-LOCA Long Term Cooling Equipment and Operator Actions ...... 14.6-24
.5.3.3 Assumptions Used in the Long Term Cooling Analysis ........................ 14.6-25
.5.3.4 Method of Analysis................................................................................. 14.6-26
.5.3.5 Parameters Used in the Long Term Cooling Analysis ........................... 14.6-26
.5.3.6 Results of the Long Term Cooling Analysis .......................................... 14.6-27 14-vii                                                              Rev. 35

tion Title Page

.5.3.7   Conclusions of the Long Term Cooling Analysis .................................. 14.6-28
.6       References............................................................................................... 14.6-28 RADIOACTIVE RELEASES FROM A SUBSYSTEM OR COMPONENT ..... 14.7-1
.1       Waste Gas System Failure ........................................................................ 14.7-1
.2       Radioactive Liquid Waste System Leak or Failure (Release to Atmosphere) ....

14.7-1

.3       Postulated Radioactive Releases Due To Liquid Containing Tank Failures .......

14.7-1

.4       Radiological Consequences Of Fuel Handling Accident ......................... 14.7-1
.4.1     General...................................................................................................... 14.7-1
.4.2     Method of Analysis................................................................................... 14.7-2
.4.2.1   Fuel Handling Accident in the Spent Fuel Pool ....................................... 14.7-3
.4.2.2   Fuel Handling Accident in Containment .................................................. 14.7-3
.4.3     Results of Analysis ................................................................................... 14.7-3
.4.3.1   Fuel Handling Accident in the Spent Fuel Pool ....................................... 14.7-3
.4.3.2   Fuel Handling Accident in Containment .................................................. 14.7-4
.4.4     Conclusions............................................................................................... 14.7-4
.5       Spent fuel cask drop accidents.................................................................. 14.7-4
.5.1     Spent Fuel Cask Tip Accident .................................................................. 14.7-4
.5.2     Method of Analysis................................................................................... 14.7-4
.5.3     Results of Analysis ................................................................................... 14.7-5
.5.4     Conclusions............................................................................................... 14.7-6 MILLSTONE UNIT 2 FSAR EVENTS NOT CONTAINED IN THE STANDARD REVIEW PLAN ................................................................................................... 14.8-1
.1       Failures of Equipment Which Provides Joint Control/Safety Functions .. 14.8-1
.2       Containment Analysis............................................................................... 14.8-1
.2.1     Main Steam Line Break Analysis ............................................................. 14.8-1
.2.1.1   Event Initiator ........................................................................................... 14.8-1
.2.1.2   Protective Systems .................................................................................... 14.8-1
.2.1.3   Method of Analysis................................................................................... 14.8-2
.2.1.4   Major Assumptions................................................................................... 14.8-2
.2.1.5   Initial Conditions and Input Data.............................................................. 14.8-4
.2.1.6   Results....................................................................................................... 14.8-4
.2.1.7   Conclusions............................................................................................... 14.8-6 14-viii                                                              Rev. 35

tion Title Page

.2.2   Loss of Coolant Accident Analysis .......................................................... 14.8-7
.2.2.1 Events Analyzed ....................................................................................... 14.8-7
.2.2.2 Method of Analysis................................................................................... 14.8-7
.2.2.3 Input and Assumptions ............................................................................. 14.8-7
.2.2.4 Results....................................................................................................... 14.8-8
.2.2.5 Conclusion ................................................................................................ 14.8-8
.3     Deleted ...................................................................................................... 14.8-8
.4     Radiological Consequences of the Design Basis Accident ...................... 14.8-8
.4.1   General...................................................................................................... 14.8-8 8.4.2   Release Pathways...................................................................................... 14.8-9 8.4.3   Control Room Habitability ..................................................................... 14.8-10
.4.4   Offsite Dose Computation ...................................................................... 14.8-10
.4.5   Conclusion .............................................................................................. 14.8-11
.5     References............................................................................................... 14.8-11 14-ix                                                               Rev. 35

List of Tables mber Title

-1    Reactor Operating Modes For Millstone Unit 2
-2    Disposition of Events Summary
.1-1  Accident Category Used For Each Analyzed Event
.2-1  Plant Operating Conditions
.2-2  Nominal Fuel Design Parameters
.3-1  Core Power Distribution (TABLE DELETED)
.4-1  Range of Key Initial Condition Operating Parameters
.5-1  Reactivity Parameters (TABLE DELETED)
.7-1  Analytical Trip Setpoints
.7-2  Uncertainties Applied at HFP Condition in Local Power Density Limiting Safety System Settings Calculations
.7-3  Uncertainties Applied at HFP Condition in the Thermal Margin/Low Pressure Limiting Safety System Settings Calculations 0.7-4  Uncertainties Applied at HFP Condition in the Local Power Density Limiting Condition for Operation Calculations
.7-5  Uncertainties Applied in Departure from Nucleate Boiling Limiting Condition for Operation Calculations
.8-1  Component Capacities and Setpoints
.9-1  Overview of Plant Systems and Equipment Available for Transient and Accident Conditions
.12-1 Nomenclature Used in Plotted Results
.1-1  Available Reactor Protection for the Decrease in Feedwater Temperature Event
.1-2  Disposition of Events for the Decrease in Feedwater Temperature Event 1.2-1  Available Reactor Protection for the Increase in Feedwater Flow Event
.2-2  Disposition of Events for the Increase in Feedwater Flow Event 1.3-1  Available Reactor Protection for the Increase in Steam Flow Event
.3-2  Disposition of Events for the Increase in Steam Flow Event
.3-3  Initial Conditions for the Increase in Steam Flow Event 14-x                                   Rev. 35

mber Title

.3-4   Event Summary for the Increase in Steam Flow Event
.3-5   Peak Reactor Power Levels for Increase in Steam Flow Event 1.4-1   Available Reactor Protection for the Inadvertent Opening of a Steam Generator Relief or Safety Valves
.4-2   Disposition of Events for the Inadvertent Opening of a Steam Generator Relief or Safety Valve Event 1.5.1-1 Available Reactor Protection for Steam System Piping Failures Inside and Outside of Containment
.5.1-2 Disposition of Events for Steam System Piping Failures Inside and Outside Containment
.5.1-3 S-RELAP5 Thermal-Hydraulic Input (Pre-Scram Steam Line Break)
.5.1-4 Actuation Signals and Delays (Pre-Scram Steam Line Break)
.5.1-5 S-RELAP5 Neutronics Input and Assumptions (Pre-Scram Steam Line Break) 1.5.1-6 MDNBR and Peak Reactor Power Level Summary (Pre-Scram Steam Line Break)
.5.1-7 LHGR-Limiting Pre-Scram Steam Line Break Sequence of Events: HFP 0.20ft2 Asymmetric Break Inside Containment with Offsite Power Available
.5.1-8 MDNBR-Limiting Pre-Scram Steam Line Break Sequence of Events:HFP 3.51ft2 Asymmetric Break Inside Containment with Loss of Offsite Power
.5.2-1 Available Reactor Protection for Steam System Piping Failures Inside and Outside of Containment, Post-Scram Analysis
.5.2-2 Disposition of Events for Steam System Piping Failures Inside and Outside of Containment, Post-Scram Analysis
.5.2-3 SRELAP5 -Thermal-Hydraulic Input (Post-Scram Steam Line Break)
.5.2-4 Actuation Signals and Delays (Post-Scram Steam Line Break)
.5.2-5 S-RELAP5 Neutronics Input and Assumptions (Post-Scram Steam Line Break)
.5.2-6 Post-Scram Steam Line Break Analysis Summary
.5.2-7 LHGR-Limiting Sequence of Events - HZP Offsite Power Available
.5.2-8 MDNBR-Limiting Post-Scram Steam Line Break Analysis Summary
.5.3-1 Assumptions Used in Main Steam Line Break Analysis
.5.3-2 Summary of Millstone 2 MSLB Accident Doses 14-xi                                    Rev. 35

mber Title

.5.3-3 Deleted by FSARCR PKG FSC 07-MP2-006
.1-1   Available Reactor for the Loss of External Load Event
.1-2   Disposition of Events for the Loss of External Load Event 2.1-3   Event Summary for the Loss of External Load Event (Primary Overpressurization Case)
.1-4   Event Summary for the Loss of External Load Event (Secondary Overpressurization Case)
.1-5   Event Summary for the Loss of External Load Event (Minimum Departure from Nucleate Boiling Ratio Case) 2.2-1   Available Reactor Protection for the Turbine Trip Event
.2-2   Disposition of Events for the Turbine Trip Event
.4-1   Available Reactor Protection for the Closure of the Main Steam Isolation Valves Events
.4-2   Disposition of Events for the Closure of the Main Steam Isolation Valves Events
.4-3   Event Summary for the Main Steam Isolation Valve Closure Event (Lower Steam Flow Case)
.7-1   Available Reactor Protection for the Loss of Normal Feedwater Flow Event
.7-2   Disposition of Events for the Loss of Normal Feedwater Flow Event
.7-3   Sequence of Events for Minimum Steam Generator Inventory Case: One Motor-Driven AFW Pump Fails to Start with Offsite Power and Steam Dumps 2.7-4   Sequence of Events for Maximum Pressurizer Level Case: Loss of Offsite Power, One Motor-Driven AFW Pump Fails to Start
.1-1   Available Reactor Protection for the Loss of Forced Reactor Coolant Flow Event
.1-2   Disposition of Events for the Loss of Forced Reactor Coolant Flow Event
.1-3   Event Summary for the Loss of Forced Reactor Coolant Flow
.3-1   Available Reactor Protection for the Reactor Coolant Pump Rotor Seizure Event
.3-2   Disposition of Events for the Reactor Coolant Pump Rotor Seizure Event
.3-3   Event Summary for the Reactor Coolant Pump Rotor Seizure 4.1-1   Available Reactor Protection for the Uncontrolled Control Rod/Bank Withdrawal from a Subcritical or Low-Power Startup Condition Event 14-xii                                     Rev. 35

mber Title

.1-2   Disposition of Events for the Uncontrolled Control Rod/Bank Withdrawal from a Subcritical or Low-Power Startup Condition Event
.1-3   Event Summary for the Uncontrolled Bank Withdrawal from Low-Power Event 4.2-1   Available Reactor Protection for the Uncontrolled Control Rod/Bank Withdrawal at Power Event 4.2-2   Disposition of Events for the Uncontrolled Control Rod/Bank Withdrawal at Power Event
.2-3   Event Summary for the Uncontrolled Rod/Bank Withdrawal Event for the Limiting 100% Power Case 4.3.1-1 Available Reactor Protection for the Dropped Control Rod/Bank Event
.3.1-2 Disposition of Events for the Dropped Control Rod/Bank Event
.3.1-3 Event Summary for the Limiting Dropped Control Rod/Bank Case
.3.5-1 Available Reactor Protection for the Single Control Rod Withdrawal Event
.3.5-2 Disposition of Events for the Single Control Rod Withdrawal Event
.4-1   Available Reactor Protection
.4-2   Disposition of Events for the Startup of an Inactive Loop Event
.6-1   Available Reactor Protection for Chemical and Volume Control System Malfunction that Results in a Decrease in the Boron Concentration in the Reactor Coolant Event
.6-2   Disposition of Events for the Chemical and Volume Control System Malfunction that Results in a Decrease in the Boron Concentration in the Reactor Coolant Event
.6-3   Summary of Results for the Boron Dilution Event Asymmetric Dilution Front Model
.6-4   Summary of Results for the Boron Dilution Event Instantaneous Mixing Mode 4.8-1   Available Reactor Protection for the Spectrum of Control Rod Ejection Accidents
.8-2   Disposition of Events for the Spectrum of Control Rod Ejection Accidents
.8-3   Event Summary for a Control Rod Ejection (Maximum Pressurization Case)
.8-4   Event Summary for a Control Rod Ejection Minimum Departure from Nucleate Boiling Ratio Case
.8-5   dnBounding Beginning of Cycle/End of Cycle Ejected Rod Analysis
.8-6   CREA Radiological Analysis Assumptions 14-xiii                                   Rev. 35

mber Title

.8-7   Radiological Consequences of a CREA 6.1-1   Available Reactor Protection for the Inadvertent Opening of a Pressurized Water Reactor Pressurizer Pressure Relief Valve Event
.1-2   Disposition of Events for the Inadvertent Opening of a Pressurized Water Reactor Pressurizer Relief Valve Event
.1-3   Event Summary for an Inadvertent Opening of a Pressurized Water Reactor Pressurizer Pressure Relief Valve 6.3-1   Available Reactor Protection for the Radiological Consequences of Steam Generator Tube Rupture Event
.3-2   Disposition of Events for the Radiological Consequences of Steam Generator Tube Rupture Event
.3-3   Sequence of Events for the Steam Generator Tube Rupture Event
.3-4   Mass Releases for the Steam Generator Tube Rupture Accident
.3-5   Assumptions for the Radiological Evaluation for the Steam Generator Tube Rupture Event
.3-6   Summary - Radiological Consequences of the Steam Generator Tube Rupture Event
.5.1-1 Available Reactor Protection for the Large Break Loss of Coolant Accident
.5.1-2 Disposition of Events for the Large Break Loss of Coolant Accident
.5.1-3 Millstone Unit 2 System Analysis Parameters (Large Break Loss of Coolant Accident Analysis)
.5.1-4 Millstone Unit 2 Large Break Loss of Coolant Accident Analysis
.5.1-5 Millstone Unit 2 Large Break LOCA Analysis
.5.1-6 Millstone Unit 2 Large Break LOCA Analysis
.5.1-7 Peak Clad Temperature Including All Penalties and Benefits - Large Break LOCA 6.5.2-1 Available Reactor Protection for the Small Break Loss of Coolant Accident
.5.2-2 Disposition of Events for the Small Break Loss of Coolant Accident
.5.2-3 Millstone Unit 2 Small Break Loss of Coolant Accident System Analysis Parameters
.5.2-4 Deleted by FSCR MPS-UCR-2016-016
.5.2-5 Calculated Event Times for Small Break Loss-of-Coolant Accident 14-xiv                                     Rev. 35

mber Title

.5.2-6 Analysis Results for Small Break Loss-of-Coolant Accident
.5.2-7 Peak Clad Temperature Including All Penalties and Benefits - Small Break LOCA
.5.3-1 Core and System Parameters Used in the LTC Analysis
.1-1   Deleted by FSARCR PKG FSC 07-MP2-006
.4-1   Assumption for Fuel Handling Accident in the Spent Fuel Pool
.4-2   Assumption for Fuel Handling Accident in Containment
.4-3   Deleted by FSARCR 02-MP2-015
.5-1   Assumptions for Spent Fuel Cask Tip Accident
.2-1   Containment Design Parameters
.2-2   Initial Conditions for Pressure Analyses
.2-3   Minimum Containment Heat Sink Data
.2-4   Sequence of Events, MP2-MSLB: Loss of Offsite Power and the Failure of Vital Bus VA-10 or VA-20 from 102% Power
.2-5   Engineered Safety Features Performance for MSLB Containment Analysis
.4-1   Loss of Coolant Accident (Off site Assumptions)
.4-2   Summary of Doses for Loss of Coolant Accident
.4-3   Loss of Coolant Accident (Control Room Assumptions) 8.4-4   Atmospheric Dispersion Data for Millstone Unit 2 Control Room
.4-5   Dose to Millstone Unit 2 Control Room Operators 14-xv                                   Rev. 35

List of Figures mber Title

.4-1   RCS Cold Leg Temperature as a Function of Power
.4-2   Not Used
.4-3   Not Used
.4-4   Not Used
.4-5   Not Used
.4-6   Not Used
.4-7   Linear Heat Rate Limiting Condition for Operation used in Local Power Density Limiting Condition for Operation Verification
.7-1   Verification of Local Power Density Limiting Safety System Setting
.7-2   Thermal Margin/Low Pressure Trip Function A1
.7-3   Thermal Margin/Low Pressure Trip Function QR1
.7-4   Not Used
.7-5   Verification of the Departure from Nucleate Boiling Limiting Condition for Operation
.7-6   Verification of Local Power Density Limiting Condition for Operation
.7-7   Linear Heat Rate Limiting Condition of Operation Used in Local Power Density Limiting Condition of Operation Verification
.3-1   Normalized Power and Heat Flux for the Increase in Steam Flow Event
.3-2   Reactivity Feedback for the Increase in Steam Flow Event
.3-3   Reactor Coolant Temperatures for Increase in Steam Flow Event
.3-4   Core Inlet Mass Flow Rate for the Increase in Steam Flow Event
.3-5   Pressurizer Pressure for the Increase in Steam Flow Event
.3-6   Steam Generator Pressures for the Increase in Steam Flow Event
.3-7   Steam Mass Flow Rates for the Increase in Steam Flow Event
.3-8   Main Feedwater Flow for the Increase in Steam Flow Event
.3-9   Main Feedwater Temperature for the Increase in Steam Flow Event
.5.1-1 Normalized Core Power (0.20 ft2 Asymmetric Break Inside Containment)
.5.1-2 Core Inlet Temperatures (0.20 ft2 Asymmetric Break Inside Containment) 14-xvi                                   Rev. 35

NOTE: REFER TO THE CONTROLLED PLANT DRAWING FOR THE LATEST REVISION. ure Title

.5.1-3   Reactivity Feedback (0.20 ft2 Asymmetric Break Inside Containment)
.5.1-4   Pressurizer Pressure (0.20 ft2 Asymmetric Break Inside Containment)
.5.1-5   Steam Generator Pressures (0.20 ft2 Asymmetric Break Inside Containment)
.5.1-6   Steam Mass Flow Rates (0.20 ft2 Asymmetric Break Inside Containment)
.5.1-7   Normalized Power and Heat Flux (Asymmetric 3.51 ft2 Break Inside Containment with Loss of Offsite Power)
.5.1-8   Reactor Coolant Temperatures (Asymmetric 3.51 ft2 Break Inside Containment with Loss of Offsite Power)
.5.1-9   Normalized Reactor Coolant System Flow Rate (Asymmetric 3.51 ft2 Break Inside Containment with Loss of Offsite Power)
.5.1-10  Pressurizer Pressure (Asymmetric 3.51 ft2 Break Inside Containment with Loss of Offsite Power)
.5.1-11  Steam Generator Pressures (Asymmetric 3.51 ft2 Break Inside Containment with Loss of Offsite Power)
.5.2-1    One Pump High Pressure Safety Injection System Delivery vs. Primary Pressure (Post-Scram Steam Line Break)
.5.2-2   Steam Generator Break Flow (HZP Post-Scram Steam Line Outside Containment Break with Offsite Power Available)
.5.2-3   Steam Generators' Secondary Pressures (HZP Post-Scram Steam Line Outside Containment Break with Offsite Power Available)
.5.2-4   Core Inlet Temperatures (HZP Post-Scram Steam Line Outside Containment Break with Offsite Power Available)
.5.2-5   Pressurizer Pressure (HZP Post-Scram Steam Line Outside Containment Break with Offsite Power Available)
.5.2-6   Pressurizer Level (HZP Post-Scram Steam Line Outside Containment Break with Offsite Power Available)
.5.2-7   Steam Generators' Secondary Mass (HZP Post-Scram Steam Line Outside Containment Break with Offsite Power Available)
.5.2-8   Reactivity Components (HZP Post-Scram Steam Line Outside Containment Break with Offsite Power Available)
.5.2-9   Reactor Power (HZP Post-Scram Steam Line Outside Containment Break with Offsite Power Available) 14-xvii                                  Rev. 35

NOTE: REFER TO THE CONTROLLED PLANT DRAWING FOR THE LATEST REVISION. ure Title

.5.2-10  Steam Generator Break Flow (HZP Post-Scram Steam Line Outside Containment Break with Loss of Offsite Power)
.5.2-11  Steam Generators' Secondary Pressures (HZP Post-Scram Steam Line Outside Containment Break with Loss of Offsite Power)
.5.2-12  Core Inlet Temperatures (HZP Post-Scram Steam Line Outside Containment Break with Loss of Offsite Power)
.5.2-13  Pressurizer Pressure (HZP Post-Scram Steam Line Outside Containment Break with Loss of Offsite Power)
.5.2-14  Pressurizer Level (HZP Post-Scram Steam Line Outside Containment Break with Loss of Offsite Power)
.5.2-15  Reactivity Components (HZP Post-Scram Steam Line Outside Containment Break with Loss of Offsite Power)
.5.2-16  Reactor Power (HZP Post-Scram Steam Line Outside Containment Break with Loss of Offsite Power)
.1-1      Reactor Power Level for Loss of External Load (Primary Overpressurization Case)
.1-2     Core Average Heat Flux for Loss of External Load (Primary Overpressurization Case)
.1-3     Reactor Coolant System Temperatures for Loss of External Load (Primary Overpressurization Case)
.1-4     Primary System Pressures for Loss of External Load (Primary Overpressurization Case)
.1-5     Total Reactivity for Loss of External Load (Primary Overpressurization Case)
.1-6     Reactor Power Level for Loss of External Load (Secondary Overpressurization Case)
.1-7     Core Average Hot Flux for Loss of External Load (Secondary Overpressurization Case)
.1-8     Reactor Coolant System Temperatures for Loss of External Load (Secondary Overpressurization Case)
.1-9     Pressurizer Pressure for Loss of External Load (Secondary Overpressurization Case)
.1-10    Total Reactivity for Loss of External Load (Secondary Overpressurization Case) 14-xviii                                   Rev. 35

NOTE: REFER TO THE CONTROLLED PLANT DRAWING FOR THE LATEST REVISION. ure Title

.1-11    Maximum Secondary System Pressures for Loss of External Load (Secondary Overpressurization Case)
.1-12    Reactor Power Level for Loss of External Load (MDNBR Case)
.1-13    Normalized Heat Flux for Loss of External Load (MDNBR Case)
.1-14    Reactor Coolant System Temperature for Loss of External Load (MDNBR Case)
.1-15    Pressurizer Pressure for Loss of External Load (MDNBR Case)
.1-16    Total Reactivity for Loss of External Load (MDNBR Case)
.1-17    Maximum Secondary System Pressure for Loss of External Load (MDNBR Case)
.4-1      Reactor Power Level for MSIV Closure (Lower Steam Flow Case)
.4-2     Reactor Coolant System Temperatures for MSIV Closure (Lower Steam Flow Case)
.4-3     Pressurizer Pressure for MSIV Closure (Lower Steam Flow Case)
.4-4     Isolated Steam Generator Pressure at Bottom of Boiler Region for MSIV Closure (Lower Steam Flow Case)
.4-5     Open MSIV Steam Generator Steam Dome Pressure for MSIV Closure (Lower Steam Flow Case)
.7-1     Reactor Coolant System Loop Temperatures for Minimum Steam Generator Inventory Case: Offsite Power Available, B Motor-Driven AFW Pump Fails to Start
.7-2     Steam Generator Dome Pressure for Minimum Steam Generator Inventory Case:

Offsite Power Available, B Motor-Driven AFW Pump Fails to Start

.7-3     Pressurizer Level for Minimum Steam Generator Inventory Case: Offsite Power Available, B Motor-Driven AFW Pump Fails to Start
.7-4     Steam Generator for Liquid Mass Inventory for Minimum Steam Generator Inventory Case: Offsite to Power Available, B Motor-Driven AFW Pump Fails to Start
.7-5     Steam Generator Collapsed Liquid Level for Minimum Steam Generator Inventory Case: Offsite to Power Available, B Motor-Driven AFW Pump Fails to Start
.7-6     Reactor Coolant System Loop Temperatures for Maximum Pressurizer Level Case:

Loss of Offsite Power, One Motor-Driven AFW Pump Fails to Start

.7-7     Steam Generator Dome Pressure for Maximum Pressurizer Level Case: Loss of Offsite Power, One Motor-Driven AFW Pump Fails to Start 14-xix                                  Rev. 35

NOTE: REFER TO THE CONTROLLED PLANT DRAWING FOR THE LATEST REVISION. ure Title

.7-8     Pressurizer Level for Maximum Pressurizer Level Case: Loss of Offsite Power, One Motor-Driven AFW Pump Fails to Start
.7-9     Steam Generator Liquid Mass Inventory for Maximum Pressurizer Level Case:

Loss of Offsite Power, One Motor-Driven AFW Pump Fails to Start

.7-10    Steam Generator Collapsed Liquid Level for Maximum Pressurizer Level Case:

Loss of Offsite Power, One Motor-Driven AFW Pump Fails to Start

.1-1      Reactor Power Level for Loss of Forced Reactor Coolant Flow
.1-2     Core Average Heat Flux for Loss of Forced Reactor Coolant Flow
.1-3     Reactor Coolant System Temperature for Loss of Forced Reactor Coolant Flow
.1-4     Pressurizer Pressure for Loss of Forced Reactor Coolant Flow
.1-5     Reactivities for Loss of Forced Reactor Coolant Flow
.1-6     Primary Coolant Flow Rate for Loss of Forced Reactor Coolant Flow
.1-7     Secondary Pressure for Loss of Forced Reactor Coolant Flow
.3-1      Reactor Power Level for Reactor Coolant Pump Rotor Seizure
.3-2     Core Average Heat Flux for Reactor Coolant Pump Rotor Seizure
.3-3     Reactor Coolant System Temperatures for Reactor Coolant Pump Rotor Seizure
.3-4     Pressurizer Pressure for Reactor Coolant Pump Rotor Seizure
.3-5     Reactivities for Reactor Coolant Pump Rotor Seizure
.3-6     Primary Coolant Flow Rate for Reactor Coolant Pump Rotor Seizure
.3-7     Secondary Pressure for Reactor Coolant Pump Rotor Seizure
.1-1      Reactor Power Level for Low Power Bank Withdrawal
.1-2     Core Average Heat Flux for Low Power Bank Withdrawal 4.1-3     Reactor Coolant Temperatures for Low Power Bank Withdrawal
.1-4     Pressurizer Pressure for Low Power Bank Withdrawal
.1-5     Reactivities for Low Power Bank Withdrawal 4.2-1      Reactor Core Power for an Uncontrolled Bank Withdrawal at Power
.2-2     Core Average Heat Flux for an Uncontrolled Bank Withdrawal at Power
.2-3     Reactor Coolant System Temperatures for an Uncontrolled Bank Withdrawal at Power 14-xx                                    Rev. 35

NOTE: REFER TO THE CONTROLLED PLANT DRAWING FOR THE LATEST REVISION. ure Title

.2-4     Pressurizer Pressure for an Uncontrolled Bank Withdrawal at Power
.2-5     Reactivities for an Uncontrolled Bank Withdrawal at Power
.2-6     Secondary Pressure for an Uncontrolled Bank Withdrawal at Power
.3.1-1    Reactor Power Level for the Limiting Dropped Control Rod/Bank Case 4.3.1-2   Reactor Coolant System Temperatures for the Limiting Dropped Control Rod/Bank Case
.3.1-3   Pressurizer Pressure for the Limiting Dropped Control Rod/Bank Case
.3.1-4   Secondary Pressure for the Limiting Dropped Control Rod/Bank Case
.8-1      Core Power for a CEA Ejection (Minimum Departure for Nucleate Boiling Ratio Case)
.8-2     Core Average Heat Flux for a CEA Ejection (Minimum Departure for Nucleate Boiling Ratio Case)
.8-3     Reactor Coolant System Temperatures for a CEA Ejection (Minimum Departure for Nucleate Boiling Ratio Case)
.8-4     Pressurizer Pressure for a CEA Ejection (Minimum Departure for Nucleate Boiling Ratio Case)
.8-5     Reactivities for a CEA Ejection (Minimum Departure for Nucleate Boiling Ratio Case)
.8-6     Secondary Pressure for a CEA Ejection (Minimum Departure for Nucleate Boiling Ratio Case)
.8-7     Core Power for a CEA Ejection (Overpressure)
.8-8     Core Average Heat Flux for a CEA Ejection (Overpressure)
.8-9     Primary System Temperatures for a CEA Ejection (Overpressure)
.8-10    Pressurizer Pressure for a CEA Ejection (Overpressure)
.8-11    Reactivities for a CEA Ejection (Overpressure)
.8-12    Secondary Pressure for a CEA Ejection (Overpressure)
.1-1      Reactor Power Level for an Inadvertent Opening of a Pressurized Water Reactor Pressurizer Pressure Relief Valve (Rated Power)
.1-2     Core Average Heat Flux for an Inadvertent Opening of a Pressurized Water Reactor Pressurizer Pressure Relief Valve (Rated Power) 14-xxi                                    Rev. 35

NOTE: REFER TO THE CONTROLLED PLANT DRAWING FOR THE LATEST REVISION. ure Title

.1-3     Reactor Coolant System Temperatures for an Inadvertent Opening of a Pressurized Water Reactor Pressurizer Pressure Relief Valve (Rated Power)
.1-4     Pressurizer Pressure for an Inadvertent Opening of a Pressurized Water Reactor Pressurizer Pressure Relief Valve (Rated Power)
.1-5     Reactivities for an Inadvertent Opening of a Pressurized Water Reactor Pressurizer Pressure Relief Valve (Rated Power)
.1-6     Secondary Pressure for an Inadvertent Opening of a Pressurized Water Reactor Pressurizer Pressure Relief Valve (Rated Power)
.3-1      Steam Generator Tube Rupture with the Loss of Offsite Power RCS Temperature Versus Time
.3-2     Steam Generator Tube Rupture with the Loss of Offsite Power Pressurizer Level Versus Time
.3-3     Steam Generator Tube Rupture with the Loss of Offsite Power Pressurizer Pressure Versus Time
.3-4     Steam Generator Tube Rupture with the Loss of Offsite Power Steam Generator Pressure Versus Time
.3-5     Steam Generator Tube Rupture with the Loss of Offsite Power Total Break Flow Rate Versus Time
.3-6     Steam Generator Tube Rupture with the Loss of Offsite Power Flashed Break Flow Rate Versus Time
.3-7     Steam Generator Tube Rupture with the Loss of Offsite Power Atmospheric Dump Valve Flow Rate per Steam Generator Versus Time
.3-8     Steam Generator Tube Rupture with the Loss of Offsite Power Main Steam Safety Valve Flow Rates per Steam Generator Versus Time
.3-9     Steam Generator Tube Rupture with the Loss of Offsite Power Auxiliary Feedwater Flow Versus Time
.5.1-1   Normalized Power (1.0 DECLG EOC Loss-of-Diesel)
.5.1-2   Safety Injection Tank (1.0 DECLG EOC Loss-of-Diesel)
.5.1-3   High Pressure Safety Injection Flow Rates (1.0 DECLG EOC Loss-of-Diesel)
.5.1-4   Low Pressure Safety Injection Flow Rates (1.0 DECLG EOC Loss-of-Diesel)
.5.1-5   Upper Plenum Pressure During Blowdown (1.0 DECLG EOC Loss-of-Diesel)
.5.1-6   Total Break Flow Rate During Blowdown (1.0 DECLG EOC Loss-of-Diesel) 14-xxii                                    Rev. 35

NOTE: REFER TO THE CONTROLLED PLANT DRAWING FOR THE LATEST REVISION. ure Title

.5.1-7   Average Core Inlet Flow Rate During Blowdown (1.0 DECLG EOC Loss-of-Diesel)
.5.1-8   Hot Channel Inlet Flow Rate During Blowdown (1.0 DECLG EOC Loss-of-Diesel)
.5.1-9   Peak Cladding Temperature Node Fluid Quality During Blowdown (1.0 DECLG EOC Loss-of-Diesel)
.5.1-10  Peak Cladding Temperature Node Fuel (Average), Cladding and Fluid Temperatures During Blowdown (1.0 DECLG EOC Loss-of-Diesel)
.5.1-11  Peak Cladding Temperature Node Heat Transfer Coefficient During Blowdown (1.0 DECLG EOC Loss-of-Diesel)
.5.1-12  Peak Cladding Temperature Node Heat Flux During Blowdown (1.0 DECLG EOC Loss-of-Diesel)
.5.1-13  Containment Pressure (1.0 DECLG EOC Loss-of-Diesel)
.5.1-14  Upper Plenum Pressure (1.0 DECLG EOC Loss-of-Diesel)
.5.1-15  Downcomer Mixture Level (1.0 DECLG EOC Loss-of-Diesel)
.5.1-16  Core Effective Flooding Rate (1.0 DECLG EOC Loss-of-Diesel)
.5.1-17  Core Mixture Level (1.0 DECLG EOC Loss-of-Diesel)
.5.1-18  Core Quench Level (1.0 DECLG EOC Loss-of-Diesel)
.5.1-19  Peak Cladding Temperature Node and Ruptured Node Cladding Temperatures (1.0 DECLG EOC Loss-of-Diesel)
.5.2-1   Peak Cladding Temperature Versus Break Size (SBLOCA Break Spectrum)
.5.2-2   Reactor Power - 3.78-Inch Break
.5.2-3   Primary and Secondary System Pressures - 3.78-Inch Break
.5.2-4   Break Mass Flow Rate - 3.78-Inch Break
.5.2-5   Break Vapor Void Fraction - 3.78-Inch Break
.5.2-6   Loop Seal Void Fraction - 3.78-Inch Break
.5.2-7   Total Core Inlet Mass Flow Rate - 3.78-Inch Break
.5.2-8   Downcomer Collapsed Liquid Level - 3.78-Inch Break
.5.2-9   Inner and Outer Core Collapsed Liquid Level - 3.78-Inch Break
.5.2-10  Reactor Vessel Mass - 3.78-Inch Break 14-xxiii                               Rev. 35

NOTE: REFER TO THE CONTROLLED PLANT DRAWING FOR THE LATEST REVISION. ure Title

.5.2-11  RCS Loop Mass Flow Rates - 3.78-Inch Break
.5.2-12  Steam Generator Main Feedwater Mass Flow Rates - 3.78-Inch Break
.5.2-13  Steam Generator Auxiliary Feedwater Mass Flow Rates - 3.78-Inch Break
.5.2-14  Steam Generator Total Mass - 3.78-Inch Break
.5.2-15  (Steam Generator Narrow Range Level % - 3.78-Inch Break
.5.2-16  (High Pressure Safety Injection Mass Flow Rates - 3.78-Inch Break
.5.2-17  (Low Pressure Safety Injection Mass Flow Rates - 3.78-Inch Break
.5.2-18  (Safety Injection Tank Mass Flow Rates - 3.78-Inch Break
.5.2-19  (Integrated Break Flow And Eccs Flow - 3.78-Inch Break
.5.2-20  (Hot Assembly Collapsed Liquid Level - 3.78-Inch Break
.5.2-21  (Hot Assembly Mixture Level - 3.78-Inch Break
.5.2-22  (Peak Cladding Temperature At Pct Location (11.02 Ft) - 3.78-Inch Break
.5.2-23  (DELETED by FSARCR 00-MP2-023)
.5.2-24  (DELETED by FSARCR 00-MP2-023)
.5.3-1    Long Term Cooling Plan
.5.3-2   Reactor Coolant System Refill Time vs. Break Area
.5.3-3   Core Flush by Hot Side Injection for a Double-Ended Guillotine Cold Leg Break
.5.3-4   Inner Vessel Boric Acid Concentration vs. Time for a Double-Ended Guillotine Cold Leg Break
.2-1     Main Steam Line Break Analysis - 102% Power with Loss of Offsite Power and Failure of Vital Bus Cabinet VA-10 or VA Containment Pressure vs. Time
.2-2     Main Steam Line Break Analysis - 102% Power with Loss of Offsite Power and Failure of Vital Bus Cabinet VA-10 or VA Containment Temperature vs. Time
.2-3     Main Steam Line Break Analysis - 102% Power with Loss of Offsite Power and Failure of Vital Bus Cabinet VA-10 or VA Mass Flow Rate vs. Time
.2-4     Main Steam Line Break Analysis - 102% Power with Loss of Offsite Power and Failure of Vital Bus Cabinet VA-10 or VA Energy Release Rate vs. Time
.2-5     Main Steam Line Break Analysis - 102% Power with Loss of Offsite Power and Failure of Vital Bus Cabinet VA-10 or VA Integrated Mass Flow vs. Time 14-xxiv                                   Rev. 35

NOTE: REFER TO THE CONTROLLED PLANT DRAWING FOR THE LATEST REVISION. ure Title

.2-6     Main Steam Line Break Analysis - 102% Power with Loss of Offsite Power and Failure of Vital Bus Cabinet VA-10 or VA Integrated Energy Release vs. Time
.2-7     Main Steam Line Break Analysis - 102% Power with Loss of Offsite Power and Failure of Vital Bus Cabinet VA-10 or VA Affected Steam Generator Pressure vs. Time
.2-8     Main Steam Line Break Analysis - 102% Power with Loss of Offsite Power and Failure of Vital Bus Cabinet VA-10 or VA Unaffected Steam Generator Pressure vs. Time
.2-9     Main Steam Line Break Analysis - 102% Power with Loss of Offsite Power and Failure of Vital Bus Cabinet VA-10 or VA Affected Steam Generator Liquid Mass vs. Time
.3-1     Deleted by FSARCR 04-MP2-018.
.3-2     Deleted by FSARCR 04-MP2-018.
.3-3     Deleted by FSARCR 04-MP2-018.
.3-4     Deleted by FSARCR 04-MP2-018.
.3-5     Deleted by FSARCR 04-MP2-018.
.3-6     Deleted by FSARCR 04-MP2-018.

14-xxv Rev. 35

GENERAL ceding sections of this report describe and evaluate the reliability of major systems and ponents of the plant. The purpose of this section is to assume that certain accidents occur withstanding the precautions taken to prevent their occurrence and to evaluate the capability of installed safety equipment to mitigate the potential consequences of such accidents. The lyses show that the health and safety of the public are assured in the event of even the most ere of the hypothetical accidents analyzed. events described in the Standard Review Plan (SRP) (Reference 14.0-1) have been reviewed placed (dispositioned) into one of the following four categories:

1. The event needs to be analyzed.
2. The event is bounded by another event which is analyzed.
3. The event is not in the licensing basis for the plant.
4. The event is not applicable to Millstone Unit 2.

he event disposition, all of the reactor operating conditions allowed by the plant Technical cifications (Reference 14.0-2) are examined to ensure that the bounding subevents are tified for each SRP event category. This ensures that the safety analysis will support the plete range of allowable operating conditions. Events which are not bounded by other events y existing accepted analyses, and are in the plant licensing basis, are dispositioned to be lyzed. In the event disposition process, the event initiator is identified for each event. The nitude of the initiator for each event is calculated and compared to the magnitude of the ator for other events. The comparison basis includes all the plant operating conditions. This ws, in several cases, a ranking of the event initiators as to severity, allowing the lesser events e dispositioned as bounded by the greater event. Similar logic is applied in determination of applicability and bounding nature for existing accepted analyses. reactor operating modes allowed for Millstone Unit 2 by the plant Technical Specifications listed in Table 14.0-1. Table 14.0-2 presents a summary of results of the event disposition. s chapter presents the basis and justification for the disposition of events, and analysis of those nts dispositioned as requiring analysis. 0.1 CLASSIFICATION OF PLANT CONDITIONS nt operations are placed in one of four categories. The categories are:

1. Normal Operations and Operational Transients - Events which are expected to occur frequently in the course of power operation, refueling, maintenance, or plant maneuvering.

14.0-1 Rev. 35

3. Infrequent Faults - Events which may occur once during the lifetime of the plant.
4. Limiting Faults - Events which are not expected to occur but which are evaluated to demonstrate the adequacy of the design.
.1.1 Acceptance Criteria acceptance criteria for the four categories of events are as given below:
1. Operational Events This condition describes the normal operational modes of the reactor. As such, occurrences in this category must maintain margin between operating conditions and the plant setpoints. The setpoints are established to assure maintenance of margin to design limits. The set of operating conditions, together with conservative allowances for uncertainties, establish the set of initial conditions for the other event categories.
2. Moderate Frequency Events
a. The pressures in reactor coolant and main steam systems should be less than 110% of design values.
b. The fuel cladding integrity should be maintained by ensuring that fuel design limits are not exceeded. That is, the minimum calculated departure from nucleate boiling ratio (DNBR) is not less than the applicable limits of the DNBR correlation being used.
c. The radiological consequences should be less than 10 CFR 20, Sections 105 and 106 and Appendix B (version prior to January 1, 1994).
d. The event should not generate a more serious plant condition without other faults occurring independently.
3. Infrequent Faults
a. The pressures in reactor coolant and main steam systems should be less than 110% of design values.
b. A small fraction of fuel failures may occur, but these failures should not hinder the capability of the core to be cooled.

14.0-2 Rev. 35

d. The event should not generate a limiting fault or result in the consequential loss of the reactor coolant or containment barriers.
4. Limiting Fault Events
a. Radiological consequences should be within the guidelines of 10 CFR 50.67 and Regulatory Guide 1.183.
b. The event should not cause a consequential loss of the required functions of systems needed to cope with the reactor coolant and containment systems transients.
c. Additional criteria to be satisfied by specific events are:
1. Loss-of-Coolant Accident (LOCA) - 10 CFR 50.46 and Appendix K.
2. Rod Ejection - Radially averaged fuel enthalpy deposition < 280 cal/gm.
.1.2 Classification of Accident Events by Category le 14.0.1-1 lists the accident category used for each event analyzed in this report. This sification is used in evaluating the acceptability of the results obtained from the analysis.
.2 PLANT CHARACTERISTICS AND INITIAL CONDITIONS operational modes have been considered in the analysis and are shown in Table 14.0-1. These rational modes have been considered in establishing the subevents associated with each event ator. A set of initial conditions is established for the events analyzed that is consistent with the ditions for each mode of operation.

nominal plant rated operating conditions are presented in Table 14.0.2-1 and principal fuel gn characteristics in Table 14.0.2-2. The uncertainties used in the accident analysis applicable he operating conditions are: (1) Core Power, HFP Calorimetric +/- 2% (2) Primary Coolant Cold Leg Temperature +/- 2.25°F (3) Primary Coolant Pressure +14/-25 psi (4) Primary Coolant Flow +/- 4% 14.0-3 Rev. 35

Technical Specification (Reference 14.0-2) power peaking factors are used in the accident lysis. The Technical Specification Limiting Conditions of Operation (LCO) assure that the er distribution is maintained within these limits during normal operation.

 .4 RANGE OF PLANT OPERATING PARAMETERS AND STATES le 14.0.4-1 presents the range of key plant operating parameters considered in the analysis. A ader range of power, core inlet temperature, and primary pressure is considered in establishing trip setpoints verified by the analysis results presented in this document.

range of operating states of the reactor is also considered in the analysis. The effect of osure on fuel thermal performance and neutronics parameters is considered. State values are cted for the event analyzed to provide the greatest challenge to the acceptance criteria for that nt. Several calculations may be required to bound the range of the state variable. For example, nge of neutronic parameters is used in the analysis of rod withdrawal events in order to verify range of protection of the challenged trip setpoints. range of initiating events is also considered in formulating the analysis conditions for an nt. The initiating conditions are examined to identify a set which conservatively challenges the eptance criteria. Where not obvious, sensitivity studies are performed. For example, analyses performed for uncontrolled rod withdrawal events throughout the range of reactivity insertion possible from boron dilution to maximum withdrawal rate of the highest worth control banks.

 .5 REACTIVITY COEFFICIENTS USED IN THE SAFETY ANALYSIS reactivity coefficients used in the analysis are consistent with the AREVA approved hodology and the Technical Specification limits. The set of parameters used in each analysis is d in the appropriate section for that event.

0.6 SCRAM INSERTION CHARACTERISTICS am reactivity insertion as a function of axial shape index (ASI) was used in the analysis for tor trip. The insertion worth includes the most reactive rod stuck out. The shutdown margin of % delta rho and a control rod drop time of 2.75 seconds (to 90% insertion) have been ported by the transient analysis. 0.7 TRIP SETPOINT VERIFICATION rating limits for the Millstone Unit 2 nuclear plant are summarized below. Methods of lysis for determining or verifying the operating limits are detailed in Section 14.0.7.5 and erence 14.0-4. Axial power distributions and other core neutronics related parameters used in setpoint verification analyses were generated with AREVA approved core simulator code SM (Reference 14.0-5). This data was generated on a three-dimensional core basis, as cribed in Reference 14.0-4. With this methodology, the values of FQ used in the setpoint 14.0-4 Rev. 35

hnical Specification on FrT limits FQ, the need for an FxyT limit is eliminated. ults of the analyses indicate that operating limits established for Millstone Unit 2 are eptable. 0.7.1 Reactor Protection System reactor protection system (RPS) is designed to assure that the reactor is operated in a safe and servative manner. The input parameters for the RPS are denoted as limiting safety system ings (LSSS). The values or functional representation of the LSSSs are calculated to ensure erence to the specified acceptable fuel design limits (SAFDL) during steady state and cipated operational occurrences (AOO). The safe operation of the reactor is also maintained estricting reactor operation to conform with the LCOs, which are administratively applied at reactor plant. The LSSS and LCO parametric values are presented in the following sections. 0.7.2 Specified Acceptable Fuel Design Limits SAFDLs are limits on the fuel and cladding established in order to preclude fuel failure. se limits may not be exceeded during steady-state operation or during AOOs. The SAFDLs used to establish the reactor setpoints to ensure safe operation of the reactor. The specific FDLs used to establish the setpoints are:

1. The local power density (LPD) which coincides with fuel centerline melt.
2. The minimum departure from nucleate boiling ratio (MDNBR) corresponding to the accepted criterion which protects against the occurrence of departure from nucleate boiling (DNB).

minimum power level required to produce centerline melt in Zircaloy clad uranium fuel rods efined as the Fuel Centerline Melt Linear Heat Rate (FCMLHR) limit and is expressed in KW/ his FCMLHR is determined using the methodology of Reference 14.0-4. The LPD limit for lstone Unit 2 is the FCMLHR limit. It is noted that reload fuel may contain gadolinia-bearing rods which, for a given LPD, will operate with a higher fuel temperature and will sequently have a lower LPD limit. The methodology used to determine the limit considers h uranium fuel rods and gadolinia-bearing fuel rods in establishing the FCMLHR limit. High Thermal Performance (HTP) critical heat flux correlation (Reference 14.0-8) is used in thermal margin analysis with statistical parameters to support the upper 95/95 limit. ervance of the LCO will protect against DNB with 95% probability at a 95% confidence level ng an AOO. A penalty on DNBR was included in the calculations to account for the sibility of a mixed core that includes both fuel assemblies with an all HTP spacer grid design fuel assemblies with both HTP and High Mechanical Performance (HMP) spacer grids. 14.0-5 Rev. 35

 .7.3.1 Local Power Density LPD trip limit is the locus of the limiting values of core power level versus ASI that will duce a reactor trip to prevent exceeding the FCMLHR limit. The correlation between allowed power level and peripheral ASI was determined using methods which take into account the l calculated nuclear peaking and the measurement and calculational uncertainties associated h power peaking. The LPD barn for operation at 2700 MWt is shown in Figure 14.0.7-1 as a s of power and ASI pairs which conservatively bounds the calculated power and ASI pairs.

is defined as the difference between the core power in the bottom half of the core and the top divided by the sum of the top and bottom halves.

 .7.3.2 Thermal Margin/Low Pressure thermal margin/low pressure (TM/LP) trip protects against the occurrence of DNB during dy state operations and for many, but not all, AOOs. This reactor trip system monitors primary em pressure, core inlet temperature, core power and ASI. A reactor trip occurs when primary em pressure falls below the computed limiting core pressure, Pvar. A statistical setpoint hodology (Reference 14.0-4) is used to verify the adequacy of the existing TM/LP trip. The hodology for the TM/LP trip accounts for uncertainties in core operating conditions, HTP B correlation uncertainties, and uncertainties in power peaking. The existing TM/LP trip ction is given by:

Pvar = 2215 x A1 (ASI) x QR1 (Q) + 14.28 x Tin - 8240 [psia], re Q is the higher of the thermal power and the nuclear flux power, Tin is the inlet temperature F and A1 and QR1 are shown in Figures 14.0.7-2 and 14.0.7-3, respectively.

 .7.3.3 Additional Trip Functions ddition to the LPD and TM/LP trip functions, other reactor system trips have been determined rovide adherence to reactor system design criteria. The analytical setpoints for these trips are wn in Table 14.0.7-1.
 .7.4 Limiting Conditions for Operation
 .7.4.1 Departure From Nucleate Boiling validity of the existing LCO for allowable core power as a function of ASI was verified to ure adherence to the SAFDL on DNB during a postulated loss-of-flow operational occurrence.

statistical analysis accounted for the effects of uncertainties associated with core operating meters, the HTP critical heat flux correlation, and power peaking. The allowed core power as nction of ASI for the existing LCO is shown to conservatively bound the present analysis in ure 14.0.7-5. 14.0-6 Rev. 35

he event that the in-core detector system is not in operation, the linear heat rate (LHR) will be ted through the use of an LPD LCO. The verification of this LCO was performed in a fashion ilar to that used in verifying the LPD LSSS (Section 14.0.7.3.1). The verification plot is wn in Figure 14.0.7-6. The LPD LCO limits core power so that the LHR LCO based on CA considerations is not exceeded. The LHR LCO protected by the LPD LCO is depicted in ure 14.0.7-7.

 .7.5 Setpoint Analysis
 .7.5.1 Limiting Safety System Settings
 .7.5.1.1    Local Power Density LPD trip monitors core power and ASI in order to initiate a reactor scram which precludes eeding fuel centerline melt conditions. In the analysis for this trip function a large number of l power distribution cases typical of the cycle were examined to establish bounding values of l power peaking, FQ, versus ASI. These cases were generated in a manner consistent with that ussed in Reference 14.0-4. Statistical methods were then employed to account for the ertainties in the parameters that are given in Table 14.0.7-2.

peak LHR in the core occurs at the position of the maximum total peaking factor, FQ, which e ratio of the maximum linear heat generation rate (LHGR) in the core to the average LHGR he core. allowed power for each ASI was calculated statistically by incorporating the uncertainties d in Table 14.0.7-2 as described in Reference 14.0-4. The results in Figure 14.0.7-1 are nded by the existing Millstone Unit 2 LPD trip and thus verify the adequacy of the existing function.

 .7.5.1.2    Thermal Margin/Low Pressure Limiting Safety System Settings TM/LP trip is designed to shut the reactor down should the reactor conditions (ASI, inlet perature, core power and pressure) approach the point where DNB might occur during either mal operation or an AOO. This analysis uses the HTP critical heat flux correlation and the core mal-hydraulic methodology described in References 14.0-8 and 14.0-11. The analysis hodology is consistent with the NRC's SRP in requiring DNB to be avoided with 95%

bability at a 95% confidence level. uncertainties shown in Tables 14.0.7-2 and 14.0.7-3 are included in the verification of the /LP trip as described in Reference 14.0-4. An excess margin of protection is provided by the 14.0-7 Rev. 35

 .7.5.2.1   Departure from Nucleate Boiling TM/LP trip system does not directly monitor reactor coolant flow. Thus, the TM/LP trip erally does not provide DNB protection for the four pump coastdown AOO. The analysis of transient is given in Section 14.3.1. The LCO presented here administratively protects the B SAFDL for this transient.

method used to establish the DNB LCO involved simulations of the loss-of-flow transient g the core thermal hydraulic code XCOBRA-IIIC (Reference 14.0-11) to determine the initial er, as a function of ASI, which provides protection from DNB with 95% probability. The ertainties listed in Tables 14.0.7-4 and 14.0.7-5 are applied using the methodology described Reference 14.0-4. The results of the statistical analysis for the loss-of-flow transient are marized by the points in Figure 14.0.7-5. The points are bounded by the existing DNB LCO , thus, verify the adequacy of the existing DNB LCO for Millstone Unit 2, which is shown in same figure by the straight line segments.

 .7.5.2.2   Local Power Density plant Technical Specifications allow plant operations for limited periods of time with the ore detectors out of service. In this situation, the LPD barn provides protection in steady-state ration against penetration of the LPD limit established by LOCA considerations. The istical methodology for the LPD LCO is essentially the same as that for LPD LSSS except:
1. The peak LPD limit is reduced, and
2. The uncertainties listed in Table 14.0.7-4 are used, as opposed to the values in Table 14.0.7-2.

allowed power versus ASI was statistically analyzed to account for the appropriate ertainties. The points in Figure 14.0.7-6 represent the statistical calculation of the 15.1 kw/ft R curve depicted in Figure 14.0.7-7. The LCO curve is shown by the straight line segments in ure 14.0.7-6, and conservatively bounds the calculated verification points. 0.8 COMPONENT CAPACITIES AND SETPOINTS le 14.0.8-1 presents the component setpoints and capacities used in the analysis. 0.9 PLANT SYSTEMS AND COMPONENTS AVAILABLE FOR MITIGATION OF ACCIDENT EFFECTS le 14.0.9-1 is a summary of trip functions, engineered safety features (ESF), and other ipment available for mitigation of accident effects. These are listed for all SRP Chapter 15 nts. A more detailed listing of available reactor protection for each event in each operating de is given in the individual event descriptions. 14.0-8 Rev. 35

MPS2 UFSAR 14.0.10 EFFECTS OF MIXED ASSEMBLY TYPES AND FUEL ROD BOWING To account for the possibility of a mixed core that includes both fuel assemblies with all HTP spacer grids and fuel assemblies with both HTP and HMP spacer grids in the core, a penalty was included in the AREVA MDNBR Calculations. In accordance with AREVA rod bow methodology (Reference 14.0-12), the magnitude of rod bow for the AREVA assemblies has been estimated. The calculations indicate that 50% closure of the rod-to-rod gap occurs at an assembly exposure of about 76,450 MWd/MTU for the AREVA 14 x odp14 design. Significant impact to MDNBR due to rod bow does not occur until the gap closures exceed 50%. Since the maximum design exposure for AREVA reload fuel in Millstone Unit 2 is significantly less than that at which 50% closure occurs, rod bow does not significantly impact the MDNBR for AREVA fuel. Also, total peaking is not significantly impacted. 14.0.11 PLANT LICENSING BASIS AND SINGLE FAILURE CRITERIA All event scenarios considered in the safety analysis depend on the following single failure criteria in the RPS: The RPS is designed with redundancy and independence to assure that no single failure or removal from service of any component or channel of a system will result in the loss of the protection function. For each event, the reactor trips occur at the specified setpoint within the specified delay time assuming a worst single active failure. Except for the steam generator tube rupture, design basis accident (limiting fault event) scenarios considered in the Millstone 2 safety analysis depend on one of the following additional single failure criteria:

1. Each ESF is designed to perform its intended safety function assuming a failure of a single active component. For these events, the ESFs required to function in an event are assumed to suffer a worst single failure of an active component.
2. The onsite power system and the offsite power system are designed such that each shall independently be capable of providing power for the ESF assuming a failure of a single active component in either power system.

The assumptions for concurrent loss of offsite power are as follows:

1. The following postulated accidents are considered assuming a concurrent loss of offsite power: main steam line break, control rod ejection, steam generator tube rupture, and LOCA.
2. The loss of normal feedwater, an anticipated operational occurrence, is analyzed assuming a concurrent loss of offsite power.

14.0-9 Rev. 35

MPS2 UFSAR The requirements of 10 CFR 50, Appendix A, Criteria 10, 20, 25 and 29 require that the design and operation of the plant and the RPS assure that the SAFDLs not be exceeded during AOOs. As per the definition of AOO in 10 CFR 50, Appendix A, Anticipated Operational Occurrences mean those conditions of normal operation which are expected to occur one or more times during the life of the nuclear power unit and include but are not limited to loss of power to all recirculation pumps, tripping of the turbine generator set, isolation of the main condenser, and loss of all offsite power. The SAFDLs are that: 1) the fuel shall not experience centerline melt; and 2) the DNBR shall have a minimum allowable limit such that there is a 95% probability with a 95% confidence interval that DNB has not occurred. 14.0.12 PLOT VARIABLE NOMENCLATURE Some of the plotted results presented in Sections 14.1 through 14.6, use PTSPWR2 (Reference 14.0-13) output variable nomenclature. Specific variables plotted are listed and defined in Table 14.0.12-1. 14.0.13 REFERENCES 14.0-1 Standard Review Plan for the Review of Safety Analysis Reports for Nuclear Power Plants, NUREG-0800, U.S. Nuclear Regulatory Commission, July 1981. 14.0-2 Technical Specifications for Millstone Unit 2, Docket Number 50-336. 14.0-3 Deleted. 14.0-4 Statistical Setpoint/Transient Methodology for Combustion Engineering Type Reactors, EMF-1961(P)(A), Revision 0, Siemens Power Corporation, July 2000. 14.0-5 Reactor Analysis Systems for PWRs, Volume 1 - Methodology Description, Volume 2 - Benchmarking Results, EMF-96-029-(P)(A), Siemens Power Corporation, January 1997. 14.0-6 Deleted. 14.0-7 Deleted. 14.0-8 HTP: Departure from Nucleate Boiling Correlation for High Thermal Performance Fuel, EMF-92-153(P)(A), Revision 1, Siemens Power Corporation, January 2005. 14.0-9 Deleted. 14.0-10 Deleted. 14.0-11 XCOBRA-IIIC: A Computer Code to Determine the Distribution of Coolant During Steady-State and Transient Core Operation, XN-NF-75-21(A), Revision 2, Exxon Nuclear Company, January 1986. 14.0-10 Rev. 35

MPS2 UFSAR 14.0-12 Computational Procedure for Evaluating Fuel Rod Bowing, XN-NF-75-32(A), Supps. 1, 2, 3 & 4, Exxon Nuclear Company, Richland, WA 99352, October 1983. 14.0-13 Description of the Exxon Nuclear Plant Transient Simulation Model for Pressurized Water Reactors (PTS-PWR), XN-NF-74-5(A), Rev. 2 and Supplements 3-6, Exxon Nuclear Company, Richland, WA 99352, October 1986. 14.0-11 Rev. 35

TABLE 14.0-1 REACTOR OPERATING MODES FOR MILLSTONE UNIT 2 Reactivity  % Rated Average Coolant Mode Condition, Keff Thermal Power

  • Temperature Power Operation 0.99 > 5% 300°F Startup 0.99 5% 300°F Hot Standby < 0.99 0 300°F Hot Shutdown < 0.99 0 300°F > Tavg > 200°F Cold Shutdown < 0.98 0 200°F Refueling ** 0.95 0 140°F
  • Excluding decay heat.
   **     Fuel in the reactor vessel with the vessel head closure bolts less than fully tensioned or with the head removed.

14.0-12 Rev. 35

TABLE 14.0-2 DISPOSITION OF EVENTS

SUMMARY

RP Event Bounding esignation Name Disposition Event 1 INCREASE IN HEAT REMOVAL BY THE SECONDARY SYSTEM 15.1.1 Decrease in Feedwater Temperature Bounded 15.1.3 15.1.2 Increase in Feedwater Flow

1) Power Bounded 15.1.3
2) Startup Bounded 15.1.3 15.1.3 Increase in Steam Flow Analyze 15.1.4 Inadvertent Opening of a Steam Generator Relief Bounded 15.1.3 or Safety Valve 15.1.5 Steam System Piping Failures Inside and Outside Analyze of Containment 2 DECREASE IN HEAT REMOVAL BY THE SECONDARY SYSTEM 15.2.1 Loss of External Load Analyze 15.2.2 Turbine Trip Bounded 15.2.1 15.2.3 Loss of Condenser Vacuum Not in Licensing Basis 15.2.4 Closure of the Main Steam Isolation Valves Analyze 15.2.5 Steam Pressure Regulator Failure Not applicable; BWR Event 15.2.6 Loss of Nonemergency AC Power to the Station Not in Licensing Basis Auxiliaries 15.2.7 Loss of Normal Feedwater Flow Analyze 15.2.8 Feedwater System Pipe Breaks Inside and Not in Licensing Basis Outside Containment 3 DECREASE IN REACTOR COOLANT SYSTEM FLOW 15.3.1 Loss of Forced Reactor Coolant Flow Analyze 15.3.2 Flow Controller Malfunction Not Applicable 15.3.3 Reactor Coolant Pump Rotor Seizure Analyze 15.3.4 Reactor Coolant Pump Shaft Break Not in Licensing Basis 4 REACTIVITY AND POWER DISTRIBUTION ANOMALIES 14.0-13 Rev. 35

RP Event Bounding esignation Name Disposition Event 15.4.1 Uncontrolled Control Rod/Bank Withdrawal Analyze from a Subcritical or Low-Power Condition 15.4.2 Uncontrolled Control Rod/Bank Withdrawal at Analyze Power 15.4.3 Control Rod Misoperation

1) Dropped Control Rod/Bank Analyze
2) Dropped Part-Length Control Rod Not Applicable
3) Malpositioning of the Part-Length Control Not Applicable Rod Group
4) Statically Misaligned Control Rod/Bank Not in Licensing Basis
5) Single Control Rod Withdrawal Analyze
6) Reactivity Control Device Removal Error Not Applicable During Refueling
7) Variations in Reactivity Load to be Not Applicable Compensated by Burnup or On Line Refueling 15.4.4 Startup of an Inactive Loop Not Applicable (Tech Specs Preclude Significant Consequences) 15.4.5 Flow Controller Malfunction Not applicable; No Flow Controller 15.4.6 Chemical and Volume Control System (CVCS) Analyze, Modes 1-6 Malfunction that Results in a Decrease in the Boron Concentration in the Reactor Coolant 15.4.7 Inadvertent Loading and Operation of a Fuel Not in Licensing Basis Assembly in an Improper Position 15.4.8 Spectrum of Control Rod Ejection Accidents Analyze 15.4.9 Spectrum of Rod Drop Accidents Not applicable; BWR (BWR) Event 5 INCREASES IN REACTOR COOLANT INVENTORY 15.5.1 Inadvertent Operation of the Emergency Core Not in Licensing Basis Cooling System that Increases Reactor Coolant Inventory 14.0-14 Rev. 35

RP Event Bounding esignation Name Disposition Event 15.5.2 CVCS Malfunction that Increases Reactor Not in Licensing Basis Coolant Inventory 6 DECREASES IN REACTOR COOLANT INVENTORY 15.6.1 Inadvertent Opening of a PWR Pressurizer Analyze Pressure Relief Valve 15.6.2 Radiological Consequences of the Failure of Not Applicable Small Lines Carrying Primary Coolant Outside of Containment 15.6.3 Radiological Consequences of Steam Generator Analyze Tube Failure 15.6.4 Radiological Consequences of a Main Steamline Not applicable; BWR Failure Outside Containment Event 15.6.5 Loss-of-Coolant Accidents Resulting from a Analyze Spectrum of Postulated Piping Breaks Within the Reactor Coolant Pressure Boundary 7 RADIOACTIVE RELEASE FROM A SUBSYSTEM OR COMPONENT 15.7.1 Waste Gas System Failure Analyze 15.7.2 Radioactive Liquid Waste System Leak or Not in Licensing Basis Failure (Release to Atmosphere) 15.7.3 Postulated Radioactive Releases due to Liquid Not in Licensing Basis Containing Tank Failures 15.7.4 Radiological Consequences of Fuel Handling Analyze Accidents 15.7.5 Spent Fuel Cask Drop Accidents Analyze AR EVENTS NOT CONTAINED IN THE STANDARD REVIEW PLAN (1) Failures of Equipment Which Provide Not Applicable Joint Control/Safety Functions (2) Containment Pressure Analysis Analyze (3) Deleted (4) Radiological Consequences of the Design Analyze Basis Accident 14.0-15 Rev. 35

ABLE 14.0.1-1 ACCIDENT CATEGORY USED FOR EACH ANALYZED EVENT Accident Event Category 1.3 Increase in Steam Flow Moderate 1.5 Steam System Piping Failures Inside and Outside of Containment Limiting Fault 2.1 Loss of External Load Moderate 2.4 Single Main Steam Isolation Valve Closure Moderate 2.7 Loss of Normal Feedwater Flow Moderate 3.1 Loss of Forced Reactor Coolant Flow Moderate 3.3 Reactor Coolant Pump Rotor Seizure Limiting Fault 4.1 Uncontrolled Bank Withdrawal at Subcritical or Low Power Moderate 4.2 Uncontrolled Bank Withdrawal at Power Moderate 4.3 Control Rod Misoperation

1) Dropped Control Rod/Bank Moderate
5) Single Control Rod Withdrawal Infrequent 4.6 Chemical and Volume Control System Malfunction Resulting in Moderate creased Boron Concentration 4.8 Control Rod Ejection Limiting Fault 6.1 Inadvertent Opening of a Pressurizer Pressure Relief Valve Moderate 6.3 Radiological Consequences of Steam Generator Tube Failure Limiting Fault 6.5 Loss-of-Coolant Accidents Limiting Fault 7.1 Waste Gas System Failure 7.4 Radiological Consequences of Fuel Handling Accidents 7.5 Spent Fuel Cask Drop Accidents 8.2 Containment Analysis 8.4 Radiological Consequences of the Design Basis Accident 14.0-16 Rev. 35

Core Thermal Power 2700 MWt Pump Thermal Power (Total) 17.1 MWt System Pressure 2225-2280 psia Reactor Coolant System Flow Rate (Minimum) 360,000 gpm

  • Core Inlet Coolant Temperature at full power 541.0-549.0°F **

Flow reductions to 349,200 gpm are compensated for by reductions in the FrT and linear heat rate limits. A full power coastdown to an indicated RCS cold leg temperature of 537°F at EOC is supported. 14.0-17 Rev. 35

al Number of Fuel Assemblies 217 l Assembly Design Type 14 x 14 embly Pitch 8.180 inches l Rods per Assembly 176 de Tubes per Assembly 4 rument Tubes per Assembly 1 Pitch 0.580 inches d Outside Diameter 0.440 inches de and Instrument Tube OD (Above Dashpot) 1.115 inches ive Fuel Length 136.70 inches l Rod Length 146.25 inches mber of Spacers 9 14.0-18 Rev. 35

14.0-19 Rev. 35 Core thermal power Subcritical to 2754 MWt (1) Core inlet temperature (power operation) Figure 14.0.4-1 Reactor coolant system pressure 2225-2280 psia +14/-25 psi Pressurizer Water Level Programmed +/- 7.5 inches Feedwater flow and temperature Range consistent with power level (1) 102% of 2700 MWt 14.0-20 Rev. 35

MPS-2 FSAR FIGURE 14.0.4-1 RCS COLD LEG TEMPERATURES AS A FUNCTION OF POWER 555 551.25 550 Maximum Cold Leg Temperature Including Instrument Uncertainty 549 545 541 540 Values @ HZP 534.25 Operating Band 538.75 535 Values @ HFP 532 530 RCS Cold Leg Temperature (degrees F) Minimum Cold Leg Temperature Including Instrument Uncertainty 524 525 521.75 520 0 10 20 30 40 50 60 70 80 90 100 Reactor Power (percent of rated power - 2700 MWt)

  • A full power coastdown to an indicated RCS cold leg temperature of 537 °F at End of Cycle is supported.

Rev. 26.1

FIGURE 14.0.4-2 NOT USED 14.0-22 Rev. 35

FIGURE 14.0.4-3 NOT USED 14.0-23 Rev. 35

FIGURE 14.0.4-4 NOT USED 14.0-24 Rev. 35

FIGURE 14.0.4-5 NOT USED 14.0-25 Rev. 35

FIGURE 14.0.4-6 NOT USED 14.0-26 Rev. 35

MPS-2 FSAR FIGURE 14.0.4-7 LINEAR HEAT RATE LIMITING CONDITION FOR OPERATION USED IN LOCAL POWER DENSITY LIMITING CONDITION FOR OPERATION VERIFICATION UNACCEPTABLE OPERATION ACCEPTABLE OPERATION Allowable LHR (kw/ft) Fraction of Active Fuel Height Rev. 21

14.0-28 Rev. 35 TABLE 14.0.7-1 ANALYTICAL TRIP SETPOINTS Parameter Setpoint w steam generator pressure 658 psia w steam generator water level 43% riable high power 111.6% ceiling on nuclear indicated power; 114% ceiling on thermal power 27.22% - Floor w reactor coolant flow 89.7% of Tech Spec minimum gh pressurizer pressure 2422 psia gh containment pressure 5.83 psig 14.0-29 Rev. 35

TABLE 14.0.7-2 UNCERTAINTIES APPLIED AT HFP CONDITION IN LOCAL POWER DENSITY LIMITING SAFETY SYSTEM SETTINGS CALCULATIONS Parameter Value gineering tolerance +/- 3% a clear flux power measurement uncertainty at full power +/- 2.34% a, b ermal Power measurement uncertainty at full power +/-4.19% a, c aking uncertainty 7% d cal power density trip transient offset e I uncertainty +/- 0.039 a, f Two-sided 95% tolerance. The nuclear power measurement uncertainty used in the setpoints analyses account for calorimetric and flux signal uncertainties. The thermal power measurement uncertainty used in the setpoints analyses account for calorimetric and thermal power signal uncertainties. Included in this power measurement uncertainty are 2 uncertainties on the RCS hot and cold leg temperature signals of 2.2°F and 1.725°F. One-sided 95% tolerance. Not treated statistically, treated as an event specific bias. Events where this trip is credited include the inadvertent opening of a pressurized water reactor pressurizer pressure relief valve, the uncontrolled rod/bank withdrawal at power, and the increase in steam flow transients. An additional ASI bias for the INPAX-II / shape annealing factor ASI measurement uncertainty was applied. 14.0-30 Rev. 35

TABLE 14.0.7-3 UNCERTAINTIES APPLIED AT HFP CONDITION IN THE HERMAL MARGIN/LOW PRESSURE LIMITING SAFETY SYSTEM SETTINGS CALCULATIONS Parameter Value

/LP trip uncertainty                                      +/- 90.30 psi a, b
/LP trip bias                                            c et coolant temperature                                    +/- 2.25°F a w measurement uncertainty                                 + 4% d uncertainty                                                +/- 6% e Two-sided 95% tolerance.

Includes both pressure measurement and trip processing uncertainties. A 7.5 psi pressure measurement bias is applied along with an event specific set of transient biases on power, pressurizer pressure and RCS hot and cold leg temperatures. The events where this trip is credited include the inadvertent opening of a pressurized water reactor pressurizer pressure relief valve, the uncontrolled rod/bank withdrawal at power, and the increase in steam flow transients. One-sided 95% tolerance. Stated tolerance is percentage of design volumetric flow of 324,800 gpm. One-sided 95% tolerance. 14.0-31 Rev. 35

BLE 14.0.7-4 UNCERTAINTIES APPLIED AT HFP CONDITION IN THE LOCAL OWER DENSITY LIMITING CONDITION FOR OPERATION CALCULATIONS Parameter Value gineering tolerance +/- 3% a wer measurement uncertainty at full power +/- 2.34% a, b aking uncertainty 7% c I uncertainty +/- 0.045 a, d Two-sided 95% tolerance. The nuclear power measurement uncertainty used in the setpoints analyses account for calorimetric and flux signal uncertainties. One-sided 95% tolerance. An additional ASI bias for the INPAX-II / shape annealing factor ASI measurement uncertainty was applied. 14.0-32 Rev. 35

ABLE 14.0.7-5 UNCERTAINTIES APPLIED IN DEPARTURE FROM NUCLEATE BOILING LIMITING CONDITION FOR OPERATION CALCULATIONS Parameter Value ssure measurement uncertainty +/- 17.9 psi a et coolant temperature +/- 2.25 °F a w measurement uncertainty +/- 4% b I uncertainty +/- 0.045 a, d wer measurement uncertainty (at full power) +/- 2.34% a, c Two-sided 95% tolerance. One-sided 95% tolerance. Stated value is percentage of design volumetric flow of 324,800 gpm. The nuclear power measurement uncertainty used in the setpoints analyses account for calorimetric and flux signal uncertainties. An additional ASI bias for the INPAX-II / shape annealing factor ASI measurement uncertainty was applied. 14.0-33 Rev. 35

MPS-2 FSAR FIGURE 14.0.7-1 VERIFICATION OF LOCAL POWER DENSITY LIMITING SAFETY SYSTEM SETTING (1) 200 o Minimum Allowed Power 180 --ea* LPD LSSS Analyzed Ba rn _ ..,..._ LPD LSSS TS Barn 160

       ~

0 140 0, 125 Q.

      .5 120 E                                                            0, 125 z0
      ~ 100
                                      -0.222, 100
       ~
       ~
       ~

fl 80 ,,' l

      ~

60 40

               -0.445, 65 er

20 **

                   -0.445, Ot
                              '         -0.4, 0                                            0.4,0   0.453, 0
             -0.6      -0.5        -0.4     -0.3    -0_2   -0.1       0      0_1   0.2 0.3     0.4 0.5      0.6 A><ial Sh ape Index {ASI)

(1) The analyzed barn, which includes an additional bias for INPAX-II/shape annealing factor ASI measurement, is represented by the dashed line. The analyzed barn is not meant for direct use in the plant. Rev. 33.2

MPS-2 FSAR FIGURE 14.0.7-2 THERMAL MARGIN/LOW PRESSURE TRIP FUNCTION A1 1.5

                                                                                         -e- Installed Al Ai : -0.5143 Yi + 1.1029                          --8--Adjusted Ai c

0,

               ~
                § 1.3 u..

c0; u

                 ..f o

u 1.2 c.

               ~                                                                              Ai: -0.5 Yi + 0.9 c.'
               ~

f§ 1.1 1

                         -0.65        -0.45          -0.25          -0.05           0.15           0.35         0.55 6-Axial Shape Index (ASI) a An additional bias was added for INPAX-II/shape annealing factor uncertainties.

Rev. 33.2

MNPS-2 FSAR MAY, 1990 ~ rot P:: a La ,;' Z u ., -: 0 foot 1.0 U ,;' Z 0.1

J p..
                                                                       ./

0.1 ./~ Z 0.7- ., 0 E--. 0.8 ,;' U PJ 0.6 n: ,,--: Cl:: 0 0.4-u ,;' c, 0.3 ~ ,-. ./ ~ 0 .8 - f-4 n, OJ

                      ./

.....l ',- 0.0 ~ 0 .0 0.1 0 .2 0 .* 0.5 0 .8 0.7 0.8 1.0 1.1 1.a f-4 FRAC1101t ..= RA1ED 1~1IW. PlIlER FIGURE 14.0 .7 - 3 THERMAL MARGIN/LOW PRESSURE TRIP FUNCTION QRl

FIGURE 14.0.7-4 NOT USED 14.0-37 Rev. 35

MPS-2 FSAR FIGURE 14.0.7-5 VERIFICATION OF DEPARTURE FROM NUCLEATE BOILING LIMITING CONDITION FOR OPERATION (1) 120.0

                       ~ COLRBarn 0
                       *~*An alyzed       Barn                                                                               0 0

o Data 0 110.0 - 0 0 0 0 0.174, 100 100.0

                                            -0.107, lOOL. ,~-"""""".......,,,,,,_..,..~"""""""""""'"l'lfi. i!i
                                                                                                              \
    ~                     ()

0

                                                             -0.1, 100,
                                                                                                                ' \
                                                                                                                    \
     ~                                                                                                                \
                                                                                                                        \

e.

     ~

0

                                                                                                                          \
                                                                                                                            \

90.0 '\

                                                                                                                                 \
                                                                                                                                   \
                                                                                                                                     \
                                                                                                                                       't:i 0.3, 85.1 I

I I 80.0

                                -0.3, 80 70.0   ..;--------~-----~---~---~---~-----'1'----
             -0.400       -0.300        -0.200        -0.100             0.000             0.100             0.200                   0 .300 Axial Shape Index (ASI}

(1) The analyzed barn, which includes an additional bias for INPAX-II/shape annealing factor ASI measurement un-certainty, is represented by the dashed line. The analyzed barn is not meant for direct use in the plant. Rev. 33.2

MPS-2 FSAR FIGURE 14.0.7-6 VERIFICATION OF LOCAL POWER DENSITY LIMITING CONDITION FOR OPERATION (1) (2)

                                                                                   ~ COLRBarn 120.0 Note 1: Data points with ASI values beyond width of the o   Data barn have been omitted.

110.0 100.0

         ~

QJ

         ~   90.0 0

OOo 0 80.0 2.05, 71.6 70.0 -0.3, 69.9 60.0 + - - - - - *

                -0.400       -0.300     -0.200      -0.100      0.0-00       0.100 0.200       0.300 Axial Shape Index {ASI)
1) The center point on the LPD LCO barn is shown at (0.0, 101.0); however, that point is used only for setting plant instrumentation. Operation is not permitted above 100% of rated power. The analysis was actually performed with that point at (0.0, 100.0).
2) The analyzed barn, which includes additional bias for the INPAX-II/shape annealing factor ASI measurement uncertainty, is represented by the dashed line. The analyzed barn is not meant for direct use in the plant.

Rev. 33.2

MPS-2 FSAR FIGURE 14.0.7-7 LINEAR HEAT RATE LIMITING CONDITION FOR OPERATION USED IN LOCAL POWER DENSITY UNACCEPTABLE OPERATION 15.1 Kw/ft ACCEPTABLE OPERATION Allowable LHR (kw/ft) Fraction of Active Fuel Height Rev. 21

TABLE 14.0.8-1 COMPONENT CAPACITIES AND SETPOINTS Component Setpoint Response Time Capacity Turbine main throttle valve NA NA 110% of flow at rated conditions Turbine stop valve NA 0.020 sec NA Main steam line isolation valves NA 6.0 sec NA Feedwater flow regulating valves NA 14 sec 120% of flow at rated conditions Pressurizer safety valves 2500 psia +/- 3% NA 294,000 lbm/hr/valve Steam line safety valves 2 at 1000 psia +/- 3% NA 794,060 lbm/hr/valve 2 at 1005 psia +/- 3% 2 at 1015 psia +/- 3% 2 at 1025 psia +/- 3% 2 at 1035 psia +/- 3% 2 at 1045 psia +/- 3% 4 at 1050 psia +/- 3% Auxiliary feedwater pumps NA 240 sec for Events 14.2.7, 300 gpm/MDAFP 14.6.3 and 14.6.5.2 180 sec for Events 14.8.2 and 600 gpm for TDAFP 14.1.5 Pressurizer relief valves 2397 psia (+)14 (-)25 psi 2.0 sec 153,000 lbm/hr/valve 14.0-41 Rev

Component Setpoint Response Time Capacity Pressurizer sprays Off - 2300 psia (+)14 (-)25 psi NA 375 gpm Full On - 2350 psia (+)14 (-)25 psi Pressurizer proportional heaters Off - 2275 psia (+)14 (-)25 psi NA 320 kW (may be less due to heater unavailability) Full On - 2225 psia (+)14 (-)25 psi Pressurizer backup heaters Off - 2225 psia (+)14 (-)25 psi NA 1280 kW (may be less due to heater unavailability) On - 2200 psia (+)14 (-)25 psi 14.0-42 Rev

TABLE 14.0.9-1 OVERVIEW OF PLANT SYSTEMS AND EQUIPMENT AVAILABLE FOR TRANSIENT AND ACCIDE CONDITIONS Event Reactor Trip Functions Other Signals and Equipment

  • 14.1, Increase in Heat Removal by the Secondary System Feedwater System Malfunctions High Power Trip Steam Generator Water Level Signals Thermal Margin / Low Pressure Trip Feedwater Isolation Valves Low Steam Generator Pressure Trip Main Steamline Isolation Valves Safety Injection Actuation Signal Turbine Trip on Reactor Trip Increase in Steam Flow Low Steam Generator Pressure Trip Steam Generator Water Level Signals Thermal Margin / Low Pressure Trip Main Steamline Isolation Valves High Power Trip Turbine Trip on Reactor Trip Safety Injection Actuation Signal Atmospheric Steam Dump Controller Steam Bypass to Condenser Controller Auxiliary Feedwater System Inadvertent Opening of a Steam Low Steam Generator Pressure Trip Steam Generator Water Level Signals Generator Relief or Safety Valve Thermal Margin / Low Pressure Trip Main Steamline Isolation Valves High Power Trip Turbine Trip on Reactor Trip Safety Injection Actuation Signal Atmospheric Steam Dump Controller Steam Bypass to Condenser Controller Auxiliary Feedwater System 14.0-43 Rev

Event Reactor Trip Functions Other Signals and Equipment

  • Steam System Piping Failure Low Steam Generator Pressure Trip Steam Generator Water Level Signals Thermal Margin / Low Pressure Trip Main Steamline Isolation Valves High Power Trip Turbine Trip on Reactor Trip Safety Injection Actuation Signal Atmospheric Steam Dump Controller High Containment Pressure Trip Steam Bypass to Condenser Controller Low Reactor Coolant Flow Trip Auxiliary Feedwater System Containment Spray Containment Isolation Containment Air Coolers Feedwater Isolation 14.2, Decrease in Heat Removal by the Secondary System Loss of External Load / Turbine Trip / High Pressurizer Pressure Trip Steam Generator Water Level Signals Loss of Condenser Vacuum High Power Trip Turbine Trip on Reactor Trip Thermal Margin / Low-Pressure Trip Atmospheric Steam Dump Controller Low Steam Generator Water Level Trip Steam Bypass to Condenser Controller Steam Generator Safety Valves Pressurizer Safety Valves Pressurizer Sprays 14.0-44 Rev

Event Reactor Trip Functions Other Signals and Equipment

  • Closure of the Main Steam Isolation Low Steam Generator Pressure Trip Atmospheric Steam Dump Controller Valves Low Steam Generator Water Level Trip Steam Bypass to Condenser Controller High Power Trip Steam Generator Safety Valves Thermal Margin / Low Pressure Trip Pressurizer Safety Valves High Pressurizer Pressure Trip Pressurizer Sprays Loss of Nonemergency AC Power to Low Reactor Coolant Flow Trip Steam Generator Water Level Signals the Station Auxiliaries High Pressurizer Pressure Trip Steam Generator Safety Valves Thermal Margin / Low Pressure Trip Pressurizer Safety Valves Low Steam Generator Water Level Trip Auxiliary Feedwater System Loss of Normal Feedwater Flow Low Steam Generator Water Level Trip Steam Generator Water Level Signals High Pressurizer Pressure Trip Steam Generator Safety Valves Thermal Margin/Low-Pressure Trip Pressurizer Safety Valves Auxiliary Feedwater System Pressurizer Sprays and Level Control Feedwater System Pipe Break High Pressurizer Pressure Trip Steam Generator Water Level Signals Thermal Margin / Low Pressure Trip Steam Generator Safety Valves Low Steam Generator Water Level Trip Pressurizer Safety Valves Low Steam Generator Pressure Trip Auxiliary Feedwater System Pressurizer Sprays and Level Control 14.3, Decrease in Reactor Coolant System Flow 14.0-45 Rev

Event Reactor Trip Functions Other Signals and Equipment

  • Loss of Forced Reactor Coolant Flow Low Reactor Coolant Flow Trip Atmospheric Steam Dump Controller Thermal Margin / Low Pressure Trip Steam Bypass to Condenser Controller High Pressurizer Pressure Trip Steam Generator Safety Valves Pressurizer Safety Valves Reactor Coolant Pump Rotor Seizure/ Low Reactor Coolant Flow Trip Atmospheric Steam Dump Controller Shaft Break High Pressurizer Pressure Trip Steam Bypass to Condenser Controller Thermal Margin / Low Pressure Trip Steam Generator Safety Valves Pressurizer Safety Valves 14.4, Reactivity and Power Distribution Anomalies Uncontrolled Control Rod Bank Thermal Margin / Low Pressure Trip Rod Withdrawal Prohibit Action on Pretrip Alarms Withdrawal from a Subcritical or High-Power Trip Shutdown Margin is Maintained as Required by Low-Power Startup Condition Technical Specifications for Modes 2-6 High Pressurizer Pressure Trip Uncontrolled Control Rod Bank High-Power Trip Pressurizer Safety Valves Withdrawal at Power Operation Thermal Margin / Low Pressure Trip Steam Generator Safety Valves Conditions High Pressurizer Pressure Trip Pressurizer Spray and Level Control Control Rod and Bank Deviation Alarms which Initi Rod Withdrawal Prohibit Rod Withdrawal Prohibit on Pretrip Alarms 14.0-46 Rev

Event Reactor Trip Functions Other Signals and Equipment

  • Control Rod Misoperation High Power Trip Pressurizer Safety Valves Thermal Margin / Low Pressure Trip Steam Generator Safety Valves Low Steam Generator Water Level Trip Pressurizer Spray and Level Control Safety Injection Actuation Signal Control Rod and Bank Deviation Alarms Start up of an Inactive Loop High Power Trip Administrative Procedures for Start up of an Idle Pu Thermal Margin / Low-Pressure Trip Plant Operation with Less than All Four Primary Coolant Pumps is Controlled by Technical Specifications Chemical and Volume Control System High Power Trip Administrative Procedures (CVCS) Malfunction that Results in a Thermal Margin / Low Pressure Trip Sufficient Operator Response Time Decrease in the Boron Concentration in the Reactor Coolant High Pressurizer Pressure Trip Inadvertent Loading and Operation of (Technical Specification Measurement a Fuel Assembly in an Improper Requirement and Administrative Position Procedures Preclude Occurrence)

Spectrum of Control Rod Ejection High Power Trip Accidents High Pressurizer Pressure Trip Thermal Margin / Low Pressure Trip Long Term, Safety Injection Actuation Signal 14.5, Increases in Reactor Coolant System Inventory 14.0-47 Rev

Event Reactor Trip Functions Other Signals and Equipment

  • Inadvertent Operation of the ECCS / High Power Trip Pressurizer Safety Valves CVCS Malfunction that Increases Thermal Margin / Low Pressure Trip Overpressurization Mitigation System Reactor Coolant Inventory High Pressurizer Pressure Trip 14.6, Decreases in Reactor Coolant Inventory Inadvertent Opening of a PWR High Power Trip High Pressure Safety Injection System Pressurizer Pressure Relief Valve Thermal Margin / Low Pressure Trip Pressurizer Heaters Steam Generator Tube Failure Thermal Margin/Low Pressure Trip Steam Generator Safety Valves Safety Injection Actuation Signal Main Steamline Isolation Valves Atmospheric Steam Dump Controller Steam Bypass to Condenser Controller Auxiliary Feedwater System Small-Break Loss-of-Coolant Thermal Margin/Low Pressure Trip Emergency Core Cooling System Accidents Resulting from a Spectrum Safety Injection Actuation Signal Auxiliary Feedwater System of Postulated Piping Breaks within the Reactor Coolant Pressure Boundary Low Reactor Coolant Flow Trip Containment Isolation Containment Spray and Air Cooler 14.0-48 Rev

Event Reactor Trip Functions Other Signals and Equipment

  • Large-Break Loss-of-Coolant No Credit taken for a Reactor Trip by Emergency Core Cooling System Accidents Resulting from a Spectrum the Reactor Protection System due to Auxiliary Feedwater System of Postulated Piping Breaks within the the Rapid Depletion of the Moderator Reactor Coolant Pressure Boundary which Shuts Down the Reactor Core Containment Isolation Almost Immediately, Followed by Containment Spray and Air Cooler ECCS Injection Which Contains Sufficient Boron to Maintain the Reactor Core in a Subcritical Configuration
  • All of the systems and equipment listed for an event would normally be available to mitigate the event consequences. However, t event analysis only takes credit for those systems and equipment which are safety-grade (such as safety valves). In addition, the ev analysis assumes that any non safety-grade systems or equipment which would worsen the event consequences (such as condense steam dump, for certain events) are available.

14.0-49 Rev

TABLE 14.0.12-1 NOMENCLATURE USED IN PLOTTED RESULTS Variable Number Definition LEVA1 Steam Generator Downcomer Level, Loop 1 Total Reactivity DOP Doppler Reactivity MOD Moderator Temperature Reactivity 01 Steam Generator Dome Pressure, Loop 1 Core Power Level R Pressurizer Pressure GSA1 Steam Generator Pressure, Loop 1 A Core Average Heat Flux VEC Core Average Coolant Temperature VG1 Average Coolant Temperature, Loop 1 IO Core Inlet Coolant Temperature L1 Cold Leg Temperature, Loop 1 L1 Hot Leg Temperature, Loop 1 WPR Pressurizer Liquid Volume DOSLT Total Steamline Steam Flow Rate FWT Total Feedwater Flow Rate LPCR Vessel Flow Rate 14.0-50 Rev. 35

1.1 DECREASE IN FEEDWATER TEMPERATURE 1.1.1 Event Initiator ecrease in feedwater temperature may be caused by loss of one or more feedwater heaters. The could be due to the interruption of steam extraction flow or to an accidental opening of a water heater bypass line. The worst loss of feedwater heaters incident would occur if all of the pressure heaters were bypassed. The effects of any decrease in the feedwater temperature due ow increases (Main or Auxiliary Feedwater (AF)) are discussed in Section 14.1.2. 1.1.2 Event Description to a malfunction in the feedwater heater system, the enthalpy of the feedwater being injected the steam generators is reduced. The increased subcooling of the feedwater reduces the ondary system average fluid enthalpy and increases the energy removal rate from the primary em. The increase in primary to secondary heat transfer causes the reactor coolant temperature e outlet of the steam generator to decrease. This causes a corresponding decrease in the core t coolant temperature. With a negative moderator temperature coefficient (MTC), the reactor power will begin to increase as the cooler moderator fluid reaches the core. 1.1.3 Reactor Protection ctor protection is provided by the variable overpower, thermal margin/low-pressure (TM/LP), l power density (LPD), and low steam generator pressure trips. Reactor protection for the rease in feedwater temperature event is summarized in Table 14.1.1-1.

 .1.4 Disposition and Justification operating Modes 1-3, the response of the nuclear steam supply system (NSSS) is governed by magnitude of the overcooling introduced by the initiating event. There is no extraction to the water heaters for operating Modes 4-6. As such, there is not a credible event for these reactor rating conditions.

most limiting case for Mode 1 is from rated power conditions because the feedwater flow rate heater duty decrease with load. Also, at rated power conditions, the initial Departure from leate Boiling Ratio (DNBR) margin is minimized. The consequences of the event in Modes 2 3 are bounded by those of Mode 1 because the magnitude of the initiating event in Modes 2 or much smaller than in Mode 1. s cooldown rate due to bypassing the feedwater heaters is bounded by that of the maximum ldown event postulated in Section 14.1.3. As such, the consequences of the Increase in Steam w (Event 14.1.3) bound the consequences for the Decrease in Feedwater Temperature event ussed in this section. The disposition of events for the Decrease in Feedwater Temperature nt is summarized in Table 14.1.1-2. 14.1-1 Rev. 35

1.2.1 Event Initiator s event is initiated by a failure in the feedwater system which causes an increase in the water flow to the steam generators. The initiators considered are complete opening of the water control valves, overspeed of the feedwater pumps, inadvertent start of a second water pump at low power, startup of the auxiliary feedwater system (AFWS), and inadvertent ning of the feedwater control valve bypass lines. 1.2.2 Event Description increased flow to the steam generators causes an increase in the energy removal capability of steam generators by reducing the average fluid enthalpy in the steam generators. The eased energy removal from the primary system causes the reactor coolant temperature at the et of the steam generator to decrease. The core inlet temperature will correspondingly be uced, which will cause the core power to increase if the moderator temperature coefficient TC) is negative. ause this event is characterized as a primary system overcooling event, the primary system sure initially decreases along with the core inlet temperature. There is also a possibility for a power increase in the presence of a negative moderator reactivity feedback coefficient. eased reactor power reduces the core Departure from Nucleate Boiling (DNB) margin. A ential exists that the net effect of these three factors will represent a challenge to the core DNB gin. 1.2.3 Reactor Protection ctor protection for the rated power and power operation conditions (Mode 1) is provided by variable overpower trip, LPD trip, TM/LP trip, low steam generator pressure trip and by the ty injection actuation signal (SIAS) on low pressurizer pressure. Additional protection is vided by the control grade reactor trip on turbine trip due to high steam generator water level. Modes 2 and 3, protection is provided by the low steam generator pressure trip, safety ction actuation signal (SIAS), and the variable overpower trip. Reactor protection for the ease in Feedwater Flow event is summarized in Table 14.1.2-1.

 .2.4 Disposition and Justification event consequences at rated power operating conditions will bound the consequences from ther power operating conditions. At rated power operating conditions, the initial thermal gin (DNBR) is minimized. Maximizing the increase in feedwater flow maximizes the load and. This results in the maximum rate of moderator cooldown which, in the presence of a ative MTC, results in the maximum challenge to the specified acceptable fuel design limits FDLs). Therefore, the limiting consequences of the increase in feedwater flow will occur at full load rated power conditions and will bound all other power operating conditions due to 14.1-2                                     Rev. 35

s cooldown is larger than that due to inadvertent startup of the three AFW pumps but less than calculated for Event 14.1.3, Increased Steam Flow. main feedwater system is off-line in Modes 4-6 but may be on-line in Mode 3. For Mode 3 rating conditions, the potential cooldown in conjunction with a negative MTC may result in a rn to power at reduced primary pressure, elevated all-rods-in peaking, and less than four tor coolant pump (RCP) conditions. This case may pose a greater challenge to the SAFDLs the full power case, and would bound zero power operation in Mode 2 if the cooldown vides sufficient reactivity insertion to overcome the shutdown margin. This is due to the ential for prompt criticality in Mode 3. The greatest increase in feed flow would result from the tup of an idle pump with both control valves full open. The cooldown rate is less than the rate puted for Event 14.1.3 in Mode 3, and consequently Event 14.1.2 in Modes 2 and 3 is nded by Event 14.1.3 initiated from Mode 3. odes 4-6, the only increased feed flow event initiator is inadvertent startup of one or more AF ps since the main feedwater system is off-line. The startup of all three AF pumps results in an eased energy removal rate, less than the maximum possible for the Increase in Steam Flow ent 14.1.3) for Modes 4-6. disposition of events for the Increase in Feedwater Flow event is summarized in Table 14.1.2-1.3 INCREASE IN STEAM FLOW 1.3.1 Event Initiator s event is initiated by a failure or misoperation in the main steam system which results in an ease in steam flow from the steam generators. This event could be caused by the rapid ning of the turbine control valves (TCVs), the atmospheric steam dump valves (ADVs), the ine bypass valves (TBVs), the steam dump to condenser valves (SDVs), a safety relief valve V), or the turbine feed pump control valves.

 .3.2 Event Description increased steam flow resulting from the failure creates a mismatch between the heat being erated in the core and that being extracted by the steam generators. As a result of this power match, the primary-to-secondary heat transfer increases and the primary system cools down. If MTC is negative, the cooldown of the primary system coolant would cause an insertion of itive reactivity and the potential erosion of thermal margin.
 .3.3 Reactor Protection main steam system is designed to accommodate a 10% increase in load (step increase).

ctor protection against a main steam flow increase greater than a 10% step is provided by the 14.1-3 Rev. 35

ety Injection Actuation Signal (SIAS) on low pressurizer pressure. Reactor protection for the ease in Steam Flow event is summarized in Table 14.1.3-1. 1.3.4 Disposition and Justification s event is predominantly a cooldown event characterized by a primary pressure and perature drop with a power rise. Therefore, the most limiting event for power operation is one ch results in the highest power, the highest core inlet temperature, and the lowest primary sure. Core flow remains essentially unchanged. The magnitude of the pressure drop for a n increase in steam flow is about the same regardless of the initial power level. The core inlet perature will be maximized at HFP. The margin to DNB is the smallest at full power since the duct of reactor power and peaking factor decreases as power drops. Additionally, the Variable rpower trip setpoint will decrease as power decreases, thus providing greater margin to the FDLs at lower powers. Therefore, for Mode 1 and for nonzero power operation in Mode 2, the nding event is one initiated from HFP. maximum possible steam release results from the simultaneous opening of the steam dump to denser valves, the atmospheric dump valves and the turbine bypass valves. Furthermore, ulating the turbine control valves as operating in the automatic mode, rather than the nual mode, is limiting. Therefore, a spectrum of HFP cases, with steam releases ranging up hat for the steam dump to condenser valves and turbine bypass valves fully open with the ine control valves operating in the automatic mode, were analyze. For Cycle 18, the itional opening of the atmospheric dump valves was evaluated and determined to be bounded he cases previously analyzed and discussed in Section 14.1.3.5. The effects of power alibration were also included in the analysis. ADVs are sized to accommodate 15% of steam flow at 2700 MWt. The SDVs and the TBV sized to accommodate 40.5% of steam flow at 2700 MWt. Each SRV will pass 6.75% of steam at 2700 MWt. The TCVs are sized to accommodate 111.3% of steam flow at 2700 MWt. The acities of the control valves for the main feedwater and AF pump turbines are significantly bound the allowable plant operation with the TCVs in automatic control mode, the TCVs were ned fully, simultaneous with the SDVs and the TBV opening. This energy removal rate nds those of the rated power operating conditions for Events 14.1.1 to 14.1.2, and 14.1.4. refore, this event is analyzed as part of the plant transient analysis for Millstone Unit 2. The sequences of this event for all other operating conditions are bounded by the rated power rating condition due to the increased margin to DNB at the other power operating conditions. disposition of events for the Increase in Steam Flow event is summarized in Table 14.1.3-2. 14.1-4 Rev. 35

pectrum of HFP cases were analyzed, with steam flows ranging from approximately 11% ess steam flow (turbine control valves fully open) to the excess steam flow associated with the m dump to condenser valves, turbine bypass valve, and turbine control valves fully opening. effects of power decalibration were also included in the analysis. end-of-cycle (EOC) Doppler feedback coefficient was selected to maximize the challenge to specified fuel design limits. The consequences of this event are bounded at EOC conditions n the MTC is at its maximum negative value. Therefore, the core operating limits report most ative MTC limit (-32 pcm/°F) was used. y full power cooldown and low power events which credit power-dependent reactor trips have potential to be adversely affected by power decalibration. Power decalibration is caused by sity induced changes in the reactor vessel downcomer shadowing the power-range ex-core ctors during heatup or cooldown transients. The nuclear power levels indicated by those ruments are lower than the actual reactor power levels when the coolant entering the reactor sel is cooler than the normal temperature for full power operation. The Variable Overpower , the TM/LP trip function, and the LPD trip all depend on the indicated nuclear power level. power decalibration effect was included in the modeling of any power-dependent reactor trips ited in this analysis. initial conditions for the Increase in Steam Flow event is summarized in Table 14.1.3-3. 1.3.6 Analysis Results transient for the limiting case (approximately 11% excess steam flow) is initiated by a failure ch causes the turbine control valves to open fully at initiation. The responses of key system ables are given in Figures 14.1.3-1 to 14.1.3-9. The sequence of events is given in le 14.1.3-4. The peak reactor power level calculated for each of the Increase in Steam Flow s analyzed are listed in Table 14.1.3-5.The increased steam flow (see Figure 14.1.3-7) creates ismatch between the core heat generation rate and the steam generator heat removal rate. This er mismatch causes the primary-to-secondary heat transfer rate to increase, which in turn ses the primary system to cool down (see Figure 14.1.3-3). With a negative MTC (see ure 14.1.3-2), the primary system cooldown causes the reactor power level to increase (see ure 14.1.3-1. However, due to power decalibration, the indicated nuclear power level does not ease along with the reactor power level. Eventually, the indicated thermal power level reaches Variable Overpower reactor trip ceiling, and the reactor is tripped. This terminates the power ursion. minimum DNBR for the limiting Increase in Steam Flow case (with approximately 11% ess steam flow) is bounded by the Section 14.3.1 four RCP loss of flow event, and is above the 95 DNBR safety limit. Moreover, the TM/LP trip is designed to protect DNB limits for this nt. The LPD LSSS trip ensures that the maximum linear heat rate is below the FCMLHR limit. se results demonstrate that fuel failures do not occur for the Increase in Steam Flow event and the event acceptance criteria are satisfied. 14.1-5 Rev. 35

results of the analysis demonstrate that the event acceptance criteria are met since the imum DNBR predicted for the full power case is greater than the safety limit. The correlation t assures that with 95% probability and 95% confidence, DNB is not expected to occur; efore, no fuel is expected to fail. The FCMLHR limit is not violated during this event. 1.4 INADVERTENT OPENING OF A STEAM GENERATOR RELIEF OR SAFETY VALVE

 .4.1 Event Initiator s event is initiated by an increase in steam flow caused by the inadvertent opening of a ondary side safety or relief valve.
 .4.2 Event Description resulting mismatch in energy generation and removal rates results in an overcooling of the ary system. If the MTC is negative, the reactor power will increase.
 .4.3 Reactor Protection ctor protection is provided by the variable overpower trip, LPD trip, TM/LP trip, low ondary pressure trip, and low steam generator water level trip. In Modes 1, 2, and 3, protection lso provided by the safety injection actuation signal (SIAS) on low pressurizer pressure.

ctor protection for the Inadvertent Opening of a Steam Generator Relief or Safety Valve event ummarized in Table 14.1.4-1.

 .4.4 Disposition and Justification inadvertent opening of a steam generator safety valve would result in an increased steam flow pproximately 6.75% of full rated steam flow. Each dump (relief) valve is sized for roximately 7.50% steam flow with the reactor at full rated power. As such, the consequences ny of these occurrences will be bounded by the events in Section 14.1.3. The disposition of nts for the Inadvertent Opening of a Steam Generator Relief or Safety Valve event is marized in Table 14.1.4-2.

1.5 STEAM SYSTEM PIPING FAILURES INSIDE AND OUTSIDE OF CONTAINMENT o separate analyses have been performed for the Steam Line Break (SLB) event. tion 14.1.5.1 describes the pre-scram analysis performed to determine Departure from leate Boiling Ratio (DNBR) and Linear Heat Generation Rate (LHGR) up to and including tor trip. This time period represents the highest reactor power condition and the assumptions e been selected to minimize DNBR and maximize LHGR during this time frame. tion 14.1.5.2 describes the post-scram analyses performed to determine MDNBR and LHGR ng the return to power caused by the overcooling. A different set of assumptions and single 14.1-6 Rev. 35

1.5.1 Pre-Scram Analysis

 .5.1.1 Event Initiator pre-scram SLB analysis is initiated by a rupture in the main steam piping which results in an ontrolled steam release from the secondary system.
 .5.1.2 Event Description increase in energy removal through the secondary system results in a severe overcooling of primary system. With a negative MTC, the primary system cooldown causes the reactor power l to increase. If the break is not large enough to trip the reactor on a Low Steam Generator ssure signal, the cooldown will continue until the reactor is tripped on a Variable Overpower or

/LP signal (for breaks outside containment) or a High Containment Pressure signal (for breaks de containment) or until the reactor reaches a new steady-state condition at an elevated power l. hough the SLB calculation is typically a cooldown event, for the pre-scram analysis the ldown event is not significant for the limiting pre-scram case. The case with a loss of off site er, also known as a pumps off case, credits the low reactor coolant flow trip for harsh ditions. In this case, the Reactor Coolant Pumps (RCPs) are tripped shortly after the initiation he transient. The sharp reduction in reactor coolant flow causes the pre-scram pumps off ulation to become a heat up transient very similar to a Loss of Coolant Flow (LOCF). refore, the conditions for this case are biased as if it were a LOCF (i.e. BOC neutronics). This becomes a combination of an MSLB and an LOCF event.

 .5.1.3 Reactor Protection ctor protection is provided by the low steam generator pressure and water level trips, variable rpower trip, LPD trip, TM/LP trip, high containment pressure trip, low reactor coolant flow, SIAS. Reactor protection for the Steam System Piping Failures Inside and Outside of tainment event is summarized in Table 14.1.5.1-1.
 .5.1.4 Disposition and Justification P initial conditions are limiting for the pre-scram SLB cases since this is the highest power dition.

outside containment breaks do not cause harsh conditions inside containment, and therefore, ot cause the Low Reactor Coolant Flow trip to be degraded. If a loss of off site power were current with an outside containment break, the primary coolant flow rate would coastdown ilar to an LOCF event, without the Low Reactor Coolant Flow trip being degraded. The 14.1-7 Rev. 35

inside containment breaks do cause harsh conditions inside containment, and therefore, an eased allowance for instrument uncertainty was applied for the Low Reactor Coolant Flow . Therefore, only the inside containment breaks will be analyzed with a loss of off site power. following pre-scram HFP Steam Line Break cases were analyzed for a range of MTCs (up to most negative core operating limits report limit) and for break sizes ranging up to a double-ed guillotine break in a main steam line. The effects of power decalibration and harsh tainment conditions (where applicable) were included in the analysis:

1. Breaks outside containment and downstream of the check valves (symmetric cases)
2. Breaks outside containment and upstream of a check valve (asymmetric cases)
3. Breaks inside containment with RCPs on (asymmetric cases)
4. Breaks inside containment with RCPs off (asymmetric cases) event is analyzed to support a range of MTCs up to the most negative core operating limits ort limit. This event must be analyzed both with and without a coincident loss of off site er.

single failure assumed in this analysis is the loss of one channel of Nuclear Instrumentation ) which provides power indication to the RPS. If one channel is out of service, the three aining NI safety channels will be in a 2-out-of-3 coincidence mode. With the assumption of a ure in one of these channels, both of the remaining channels are required for a trip, relying on lowest power indication for the safety function. disposition of events for the Steam System Piping Failures Inside and Outside of tainment event is summarized in Table 14.1.5.1-2.

 .5.1.5 Definition of Events Analyzed pre-scram SLB event is initiated by a rupture in the main steam piping. The break location is nstream of the steam generator integral flow restrictor and either:
1. outside containment and upstream of the main steam line check valves (asymmetric break), or
2. outside containment and downstream of the main steam line check valves (symmetric break), or 14.1-8 Rev. 35

m released through a break located downstream of the main steam line check valves flows to break from both steam generators and, therefore, results in a symmetric transient. However, m released through a break located upstream of one of the check valves flows to the break m the upstream steam generator only (because the check valve precludes backflow to the break m the other steam generator) and, therefore, results in an asymmetric transient. er decalibration is caused by density induced changes in the reactor vessel downcomer dowing of the power range ex-core detectors during heatup or cooldown transients. The lear power levels indicated by those instruments are lower than the actual reactor power levels n the coolant entering the reactor vessel is cooler than the normal temperature for full power ration (and higher when the vessel inlet coolant is warmer than the normal full power perature). This effect is included in the modeling of any power-dependent reactor trips ited in the analysis of full power cooldown events and low power events. The Variable rpower trip, the Thermal Margin/Low Pressure (TM/LP) trip function, and the Local Power sity (LPD) trip all depend on the indicated nuclear power level. sh containment conditions can be caused by the release of steam within the reactor tainment. Under such conditions, only those trips which have been qualified for harsh ironments are credited, and increased uncertainties are included in the setpoints of all ironmentally qualified trips which are credited. outlined in Reference 14.1-1, three computerized calculations are required prior to the final ulation of the Minimum Departure From Nucleate Boiling Ratio (MDNBR) values and the imum Linear Heat Generation Rate (LHGR) values utilized in the determination of fuel ure. The NSSS response of the pre-scram SLB with AC power available is computed using the EVA S-RELAP5 code (Reference 14.1-1), the detailed core and hot assembly power ributions are calculated using the PRISM code (Reference 14.1-3), and the detailed core and assembly flow and enthalpy distributions are calculated using the XCOBRA-IIIC code ference 14.1-4). The HTP correlation (Reference 14.1-5) was utilized to calculate MDNBR. previous pre-scram SLB case with the concurrent loss of off site power analyzed using the F RELAP (Reference 14.1-2) code was determined to remain bounding. The results of this are described in Section 14.1.5.1.6.2.

 .5.1.5.1    Analysis of Results S-RELAP5 analysis provides the NSSS boundary conditions for the PRISM and the OBRA-IIIC calculations. This section presents a description of the treatment of factors which have a significant impact on NSSS response and resultant MDNBR and LHGR values. The t specific parameters used in this analysis are listed in Tables 14.1.5.1-3 to 14.1.5.1-5.

servatisms are included in parameters or factors known to have significant effects on the SS performance and resulting MDNBR and LHGR values. 14.1-9 Rev. 35

pre-scram SLB event analyzes breaks outside containment both downstream (symmetric s) and upstream (asymmetric cases) of the main steam line check valves and breaks inside tainment (asymmetric cases). A full range of break sizes, up to the double-ended guillotine k of a main steam line, were considered. S-RELAP5 break mass flow rate is computed using the Moody critical flow model modified h that only steam flows out the break.

 .5.1.5.1.2 Power Decalibration er decalibration is caused by density-induced changes in the reactor vessel downcomer dowing of the power-range ex-core detectors during heatup or cooldown transients. The lear power levels indicated by those instruments are lower than the actual reactor power levels n the coolant entering the reactor vessel is cooler than the normal temperature for full power ration (and higher when the vessel inlet coolant is warmer than the normal full power perature). This effect is included in the modeling of any power-dependent reactor trips ited in the analysis of full power cooldown events and low power events. The Variable rpower trip, the Thermal Margin/Low Pressure (TM/LP) trip function, and the Local Power sity (LPD) trip all depend on the indicated nuclear power level.
 .5.1.5.1.3 Harsh Containment Conditions sh containment conditions can be caused by the release of steam within the reactor tainment. Under such conditions, only those trips which have been qualified for harsh ironments are credited, and increased uncertainties are included in the setpoints of all ironmentally qualified trips which are credited.
 .5.1.5.1.4 Boron Injection on injection into the primary system acts to mitigate the return to power. Injection of boron is deled from the HPSI system. The HPSI system is conservatively modeled to take suction from Refueling Water Storage Tank (RWST) at 35°F with a boron concentration of 1720 ppm.

ially, the line volume between the check valves isolating the system pumps and the cold leg ction location is assumed to be filled with unborated water. The time required to flush this orated water from the safety injection lines is included as an integral part of the S-RELAP5 SS calculation. In the pre-scram SLB event, the analysis is terminated shortly after reactor trip, efore injection of borated water is not a factor in the analysis.

 .5.1.5.1.5 Single Failure Assumption rder to simulate the asymmetric thermal-hydraulic and reactivity feedback effects that occur ng the pre-scram SLB event, the core is divided into an affected sector (one-half of the core) an unaffected sector (one-half of the core). The single failure assumed in this analysis is the of one channel of Nuclear Instrumentation (NI) which provides power indication to the 14.1-10                                   Rev. 35

ed around the reactor vessel in positions that result in one detector seeing the flux only from affected region, one seeing the flux only from the unaffected region, and two detectors seeing rly equal flux from both regions. If one of these latter two is out of service, and the other is med to be a single failure, the remaining two channels will be required to cause an RPS trip h power or TM/LP). Since the power in the affected region will always be higher than in the ffected region, it is sufficient to model the NI channel reading the unaffected region only.

 .5.1.5.1.6 Feedwater breaks smaller than 0.4 square feet, the Main Feed Water (MFW) control valves are assumed odulate to match MFW flow to steam flow. For larger breaks, the MFW valves are servatively assumed to open fully at the initiation of the event and remain open. The MFW ps are assumed to be operating at rated speed. The MFW flow increases as the secondary sure decreases at the lowest possible fluid temperature until the feedwater regulator valve es. Fluid temperature is determined by assuming heating of the feedwater ceases at the same e the break is initiated. The MFW flow is terminated 14 seconds after receiving the isolation al.
 .5.1.5.1.7 Trips and Delays uation signals and delays are given in Table 14.1.5.1-4. Biases to account for uncertainties are uded in the trip setpoints as shown. In the pre-scram SLB event, the analysis is terminated rtly after reactor trip, therefore injection of borated water is not a factor in the analysis.
 .5.1.5.1.8 Neutronics core kinetics input for this calculation consisted of the minimum required control rod tdown worth at EOC, and EOC values associated with the reactivity feedback curves, delayed tron fraction, delayed neutron fraction distribution and related time constants, and prompt tron generation time. The S-RELAP5 default fission product and actinide decay constants e utilized for this calculation.

core reactivity is derived from input of several functions. These include effects from control worth, moderator density changes, boron concentration, and Doppler effects. The reactivity is ghted between the core sectors. The S-RELAP5 analyses for cases with off site power ilable were performed with an MTC ranging from -8 pcm/°F to -32 pcm/°F. In all cases, the t limiting scenarios were those performed with an MTC of -32 pcm/°F. The analyses for cases h a loss of off site power were performed with an MTC of +4.0 pcm/°F. A summary of the lear input and assumptions is given in Table 14.1.5.1-5.

 .5.1.5.1.9 Decay Heat presence of radioisotope decay heat at the initiation of the SLB event will reduce the rate and extent of cooldown of the primary system. The initial decay heat is calculated on the basis of 14.1-11                                     Rev. 35

SLB cases.

 .5.1.5.1.10 Nodalization NSSS transient calculations presented in this report utilized the nodalization model described eference 14.1-1. The nodalization treats all major NSSS components and subcomponents as rete elements, with the exception of the secondary side of the steam generators. In addition, omponents with long axial dimensions are divided into subcells adequate to minimize erical diffusion and smearing of gradients.

rder to simulate the asymmetric thermal-hydraulic and reactivity feedback effects that occur ng the pre-scram SLB event, the core is divided into an affected sector (one-half of the core) an unaffected sector (one-half of the core).

 .5.1.5.1.11 Interloop Mixing ing an actual SLB transient, some mixing between the parallel channels within the reactor sure vessel will occur in the downcomer, the lower plenum, the core, and the upper plenum to lateral momentum imbalances, and turbulence or eddy mixing. The mixing will act to uce the positive reactivity feedback effects due to a reduced rate and magnitude of cooldown he affected loop and associated core sector.

his analysis, no credit is taken for turbulent or eddy mixing of coolant between loops or the llel flow channels within the reactor pressure vessel. However, interloop mixing is calculated ccur due to flow in interloop junctions in the upper and lower plenums. Mixing in the lower um was effectively reduced to zero by using an extremely high loss coefficient between the cted and intact sectors.

 .5.1.5.2    Minimum Departure From Nucleate Boiling and Linear Heat Generation Rate Analysis PRISM (Reference 14.1-3) core neutronics code is used to calculate the core radial power ributions for XCOBRA-IIIC (Reference 14.1-4) during the asymmetric transients with off site er available only. The PRISM model is a three-dimensional representation of the entire core, h four radial nodes and 24 axial nodes for each fuel assembly.

ed on the overall core conditions calculated by S-RELAP5 for the symmetric cases (or ELAP5 and PRISM for the asymmetric cases with off site power available) at the peak heat time-point, the XCOBRA-IIIC fuel assembly thermal-hydraulic code is used to calculate the and enthalpy distributions for the entire core and the DNB performance for the DNB-ting assembly. The XCOBRA-IIIC model consists of a thermal-hydraulic model of the core resenting each assembly by a single channel) linked to a detailed thermal-hydraulic model he limiting assembly (representing each subchannel by a single channel). The limiting mbly DNBR calculations are performed using the HTP DNB correlation (Reference 14.1-5). 14.1-12 Rev. 35

sient. The increase in peaking is determined by PRISM.

 .5.1.6 Analysis Results ummary of the calculated peak reactor power is presented in Table 14.1.5.1-6 for the limiting NBR and LHGR cases. The MDNBR values for the limiting cases are also in this Table. For s where off site power was available for operation of the primary coolant system pumps, the NBR and the maximum LHGR occurred at the time of the maximum power condition. For s where off site power is lost and the primary system pumps coast down, the maximum GR and the MDNBR occur when the worst combination of core power, flow, inlet perature, and pressure are present.

scenario which resulted in the highest power level and the largest LHGR is the HFP 0.20 ft2 mmetric break inside containment with off site power available for operation of the primary lant pumps. This case is presented in detail. scenario which resulted in the limiting MDNBR is the HFP case with a loss of off site power is also presented in detail.

 .5.1.6.1   Hot Full Power 0.20 ft2 Break Inside Containment Upstream of a Check Valve with Off site Power Available S-RELAP5 simulation of the NSSS during the HFP asymmetric break transient with off site er available is illustrated in Figures 14.1.5.1-1 through 14.1.5.1-6. A tabulation of the uence of events is presented in Table 14.1.5.1-7. The S-RELAP5 computation was terminated seconds after break initiation. This is well beyond the time of MDNBR or peak LHGR. The eral response of the reactor was the same for all the asymmetric break sizes but the sequence vents was quicker as the break size increased.
 .5.1.6.1.1 Secondary System Parameters n break initiation the break flow increased sharply and then began to decline in response to ng secondary side pressure. When the turbine trip occurred, the break flow increased due to a l pressure increase. The main steam line flow rate from each generator initially increased (see ure 14.1.5.1-6) in response to the break and the assumed instantaneous full opening of the ine control valves. The increased steam flow creates a mismatch between the core heat eration rate and the steam generator heat removal rate. This power mismatch causes the mary-to-secondary heat transfer rate to increase, which in turn causes the primary system to l down (see Figure 14.1.5.1-2). When the reactor scram occurred, the turbine valves closed steam flow declined sharply.

14.1-13 Rev. 35

roximately five seconds after the break occurred, the core inlet temperature began to decline. h a negative MTC (see Figure 14.1.5.1-3), the primary system cooldown caused the reactor er level to increase. The core power continues to increase until reactor scram on high tainment pressure occurs. This terminated the power excursion. The pressurizer pressure and l began to decline as the volume of water in the primary system shrank. The core inlet mass rate increased due to the increasing density of the primary system fluid while the reactor lant pumps speed remained constant.

 .5.1.6.1.3 Departure From Nucleate Boiling Ratio and Linear Heat Generation Rate Results MDNBR value for this scenario was calculated to be below the 95/95 HTP correlation limit.

amount of failed fuel is determined by comparing the core power distribution to the power at ch DNB occurs. This results in a predicted failure of 3.7% of the fuel rods in the core. peak LHR for the LHR-limiting case (0.20 ft2 break inside containment and upstream of a ck valve) is calculated to be less than the FMCLHR limit. Therefore, it is apparent that terline melt is not predicted to occur. Thus, no fuel failures are predicted to occur due to ation of the centerline melt criteria.

 .5.1.6.2    Hot Full Power 3.51 ft2 Inside Containment Asymmetric Break Concurrent with a Loss of Off site Power ANF-RELAP NSSS simulation of the most limiting pre-scram SLB scenario from an NBR standpoint (i.e., HFP 3.51 ft2 inside containment asymmetric break concurrent with a of off site power) is illustrated in Figures 14.1.5.1-7 through 14.1.5.1-11. A tabulation of the uence of events is presented in Table 14.1.5.1-8. The ANF-RELAP computation was inated 60 seconds after break initiation. This is well beyond the time of MDNBR or peak GR.

transient is initiated by the opening of the break. The RCPs tripped shortly after transient ation. The sharp reduction in the reactor coolant flow causes this pre-trip pumps off ulation to become a heat up transient very similar to a Loss of Coolant Flow event. Typically, Steam Line Break calculation is a cooldown event. Because this case is a heat up event the t positive BOC neutronics conditions are used, and the maximum inside containment mmetric break size is used. The maximum break size causes the biggest decrease in primary sure. Maximizing the primary system pressure decrease causes the maximum decrease in derator density and the maximum positive moderator feedback. The RCP trip causes the RCS to decrease rapidly throughout this transient. The decreasing RCS flow causes the transient e of the fluid in the core to increase and the fluid temperature begins to rise. The increasing d temperature causes positive moderator feedback, which in turn causes an increase in core er. However, the decreasing RCS flow causes the heat transfer to the fluid to decrease. The ease in core power is offset by the decrease in heat transfer from the fuel rods, such that, the 14.1-14 Rev. 35

 .5.1.6.2.1 Departure From Nucleate Boiling Ratio and Linear Heat Generation Rate Results MDNBR value for the pre-scram 3.51 ft2 asymmetric break inside containment with a loss of site power was calculated to be below the 95/95 HTP correlation limit. The number of failed mblies is determined by comparing the core power distribution to the assembly power where B occurs. This results in a predicted failure of 3.7% of the fuel rods in the core.

peak LHR for this case is bounded by the 0.20 ft2 inside containment asymmetric break. refore, the LHGR for this case is below the FCMLHR limit and no fuel failures are predicted ccur due to violation of the centerline melt criteria.

 .5.1.7 Conclusions HFP 0.20 ft2 break inside containment and upstream of a check valve (asymmetric break) h off site power available was determined to be the most limiting in this analysis from an GR standpoint. In no scenario evaluated, however, was fuel failure calculated to occur as a lt of violating the FCMLHR limit.

h the HFP 0.20 ft2 asymmetric break inside containment with off site power available and the P 3.51 ft2 asymmetric break inside containment coincident with a loss of off site power were rmined to be the most limiting in this analysis from the standpoint of MDNBR. The MDNBR elow the 95/95 HTP correlation limit. This results in a predicted failure of 3.7% of the fuel s in the core. However, for any outside containment main steam line break scenario, no fuel ure due to violation of the DNBR limit was predicted to occur.

 .5.2 Post-Scram Analysis
 .5.2.1 Event Initiator s event is initiated by a rupture in the main steam piping downstream of the integral steam erator flow restrictors and upstream of the MSIVs which results in an uncontrolled steam ase from the secondary system.
 .5.2.2 Event Description increase in energy removal through the secondary system results in a severe overcooling of primary system. In the presence of a negative MTC, this cooldown causes a decrease in the tdown margin (following reactor scram) such that a return to power might be possible owing a steam line rupture. This is a potential problem because of the high power peaking ors which exist, assuming the most reactive control rod to be stuck in its fully withdrawn ition.

14.1-15 Rev. 35

ctor protection is provided by the low steam generator pressure and water level trips, LPD , TM/LP trip, high containment pressure trip, and SIAS. Reactor protection for the Steam tem Piping Failures Inside and Outside of Containment event is summarized in Table 14.1.5.2-

 .5.2.4 Disposition and Justification ated power conditions, the stored energy in the primary coolant is maximized, the available mal margin is minimized, and the pre-trip power level is maximized. These conditions result he greatest potential for cooldown and provide the greatest challenge to the SAFDLS.

iating this event from rated power also results in the highest post-trip power since it imizes the concentration of delayed neutrons providing for the greatest power rise for a given itive reactivity insertion. Thus, this event initiated from rated power conditions will bound all r cases initiated from at power operation modes. the zero power and subcritical plant states (Modes 2-6), there is a potential for a return to er at reduced pressure conditions. The most limiting steam line break (SLB) event at zero er is one which is initiated at the highest temperature, thereby providing the greatest capacity cooldown. This occurs in Modes 2 and 3. Thus, the event initiated from Modes 2 and 3 will nd those initiated from Modes 4-6. Further, the limiting initial conditions will occur when the is just critical. These conditions will maximize the available positive reactivity and produce quickest and largest return to power. Thus, the SLB initiated from critical conditions in Mode ill bound the results of the event initiated form subcritical Mode 3 conditions. Technical Specifications only require a minimum of one RCP to be operating in Mode 3. One p operation provides the limiting minimum initial core flow case. Minimizing core flow imizes the clad to coolant heat transfer coefficient and degrades the ability to remove heat erated within the fuel pins. Conversely, however, a maximum loop flow will maximize the ary to secondary heat transfer coefficient, thus providing for the greatest cooldown. Higher p flow will sweep the cooler fluid into the core faster, maximizing the rate of positive tivity addition and the peak power level. worst combination of conditions is achieved for the four-pump loss of off site power case. In situation, the initial loop flow is maximized resulting in the greatest initial cooldown, while final loop flow is minimized providing the greatest challenge to the DNB SAFDL. Since the ral circulation flow which is established at the end of the transient will be the same regardless hether one or four pumps were initially operating the results of the four-pump loss of off site er case will bound those of the one-pump case. Thus, only four-pump operation need be lyzed for the Mode 2 case. event is analyzed to support the Technical Specification EOC MTC limit. This event must be lyzed both with and without a coincident loss of off site power. Typically there are two single ures which are considered for the off site power available case. The first is failure of a High ssure Safety Injection (HPSI) pump to start. The second is failure of an MSIV to close, 14.1-16 Rev. 35

m generator to reach the break. This is not deemed credible. Thus, the single failure to be sidered with off site power available is failure of a HPSI pump to start. For the loss-of-off site er case, the limiting single failure is the failure of a diesel generator to start. This is assumed esult in the loss of one HPSI pump. The disposition of events for the Steam System Piping ures Inside and Outside of Containment event is summarized in Table 14.1.5.2-2.

 .5.2.5 Definition of Events Analyzed post-scram SLB is initiated by a rupture in the main steam piping downstream of the integral m generator flow restrictors and upstream of the MSIVs which results in an uncontrolled m release from the secondary system. The effects of harsh containment conditions (where licable) are included in the following analyses:
1. HFP and HZP breaks outside containment with off site power available
2. HFP and HZP breaks outside containment with a loss of off site power
3. HFP and HZP breaks inside containment with off site power available
4. HFP and HZP breaks inside containment with a loss of off site power event is analyzed to support the Technical Specification EOC MTC limit. This event must be lyzed both with and without a coincident loss of off site power.

single failure assumed in this analysis results in the disabling of one of the two HPSI pumps uired to be in service during normal operation. In addition to the single failure, there is no it taken for the charging pump system. This assumption results in an additional delay in the e required for boron to reach the core. The delay is amplified when combined with the mption of a stagnant upper head which serves to maintain the primary system pressure due to hing of the hot fluid in the upper head. increase in energy removal through the secondary system results in a severe overcooling of primary system. In the presence of a negative MTC, this cooldown results in a large decrease he shutdown margin and a return to power. This return to power is exacerbated because of the h power peaking factors which exist, with the most reactive control rod stuck in its full hdrawn position. outlined in Reference 14.1-1, three computerized calculations are required prior to the final ulation of the Minimum Departure From Nucleate Boiling Ratio (MDNBR) values and the imum Linear Heat Generation Rate (LHGR) values utilized in the determination of fuel ure. The NSSS response is computed using the AREVA S-RELAP5 code (Reference 14.1-1), detailed core and hot assembly power distributions and the reactivity at the time of peak post-m power are calculated using the PRISM code (Reference 14.1-3), and the detailed core and assembly flow and enthalpy distributions are calculated using the XCOBRA-IIIC code 14.1-17 Rev. 35

 .5.2.5.1    Analysis of Results S-RELAP5 analysis provides the NSSS boundary conditions for the PRISM and the OBRA-IIIC calculations. This section presents a description of the treatment of factors which have a significant impact on NSSS response and resultant MDNBR and LHGR values. The t specific parameters used in this analysis are listed in Tables 14.1.5.2-3 to 14.1.5.2-5.

servatisms are included in parameters or factors known to have significant effects on the SS performance and resulting MDNBR and LHGR values.

 .5.2.5.1.1 Break Location, Size, and Flow Model post-scram SLB event is initiated by a double-ended guillotine break of a main steam line nstream of the integral steam generator flow restrictors and upstream of the MSIVs. The flow hoked at the integral steam generator flow restrictor, which has an area of 3.51 ft2. On the m generator side of the break, steam flows out of the break throughout the entire transient. On MSIV side of the break, break flow terminates after the MSIVs are fully closed. As an added servatism, the main steam check valves are not credited in the analysis. The event occurs current with the most reactive control rod stuck out of the core. The break flow areas for the cted and intact steam generators are listed in Table 14.1.5.2-3.

S-RELAP5 break mass flow rate is computed using the Moody critical flow model modified h that only steam flows out the break.

 .5.2.5.1.2 Boron Injection on injection into the primary system acts to mitigate the return to power. Injection of boron is deled from the HPSI system. The HPSI system is conservatively modeled to take suction from Refueling Water Storage Tank (RWST) at 35°F with a boron concentration of 1720 ppm.

ially, the line volume between the check valves isolating the system pumps and the cold leg ction location is assumed to be filled with unborated water. The time required to flush this orated water from the safety injection lines is included as an integral part of the S-RELAP5 SS calculation. The characteristics of the HPSI system are listed in Table 14.1.5.2-3. The very curve for the HPSI system used in this analysis is given in Figure 14.1.5.2-1.

 .5.2.5.1.3 Single Failure Assumption single failure assumed in the engineered safeguards system results in the disabling of one of two HPSI pumps required to be in service during normal operation. In addition to the single ure, there is no credit taken for the charging pump system. This assumption results in an itional delay in the time required for boron to reach the reactor core. The delay is further lified when combined with the assumption of a stagnant upper head which serves to maintain primary system pressure due to flashing of the hot fluid in the upper head.

14.1-18 Rev. 35

the HFP scenarios, normal MFW flow is assumed to be delivered to both Steam Generators. MFW flow increases as the secondary pressure decreases at the lowest possible fluid perature until the feedwater regulating valve closes. Fluid temperature is determined by ming heating of the feedwater ceases at the same time the break is initiated. The MFW flow is inated 14 seconds after receiving the isolation signal. In addition, a sensitivity study was ormed on the effect of MFW on the post-scram SLB results. The study concluded that the t-scram SLB results are insensitive to the MFW flow. the HFP scenarios, the AFW flow is assumed to be zero at break initiation. After 180 seconds, W is delivered at the maximum capacity of the AFW system with flow restrictors installed on AFW delivery lines. For the HZP scenarios, the AFW flow is increased to the maximum acity immediately at break initiation. For all scenarios, all of the AFW flow is directed to the cted steam generator to maximize the cooldown rate. AFW flow remains directed to the cted steam generator until the S-RELAP5 computation was terminated 600 seconds after k initiation. By 600 seconds, core power is decreasing. After this point in time, the operators terminate AFW flow to the affected steam generator, which will cause the primary system peratures to increase, further decreasing the core power. The timing of the isolation of AFW to affected steam generator is not a critical parameter.

 .5.2.5.1.5 Trips and Delays s for the HPSI, main feedwater valves, and MSIVs are given in Table 14.1.5.2-4. Biases to ount for uncertainties are included in the trip setpoints as shown. For the steam and feedwater es, the delay times given are between the time the trip setpoint is reached and the time full e closure is reached. For the HPSI system, the delay time given is from the time the setpoint is hed until the pumps have accelerated to rated speed. Additional delay time required to sweep lines of unborated water is accounted for by setting the boron concentration of the injected to zero until the volume of the injection lines has been cleared.
 .5.2.5.1.6 Neutronics core kinetics input for this calculation consisted of the minimum required control rod tdown worth at the EOC, and EOC values associated with the reactivity feedback curves, yed neutron fraction, delayed neutron fraction distribution and related time constants, and mpt neutron generation time. The S-RELAP5 default fission product and actinide decay stants were utilized for this calculation.

core reactivity is derived from input of several functions. These include effects from control worth, moderator density changes, boron concentration, and Doppler effects. The reactivity is ghted between the core sectors. Different reactivity functions were utilized where necessary the HZP and the HFP cases. The S-RELAP5 analyses were performed with an MTC of pcm/°F. A summary of the nuclear input and assumptions is given in Table 14.1.5.2-5. 14.1-19 Rev. 35

presence of radioisotope decay heat at the initiation of the SLB event will reduce the rate and extent of cooldown of the primary system. For the HFP scenarios, the initial decay heat is ulated on the basis of infinite irradiation time at a power of 2700 MW prior to transient ation. For the HZP scenarios, the initial decay heat is calculated on the basis of infinite diation time at a power of 1 W prior to transient initiation. For both scenarios, decay heat erated from return to power is calculated. This treatment of decay heat serves to maximize the ed energy in the HFP cases and to minimize it in the HZP cases. This treatment provides ting stored energy conditions for the SLB cases.

 .5.2.5.1.8 Nodalization NSSS transient calculations utilized the nodalization model described in Reference 14.1-1.

nodalization treats all major NSSS components and subcomponents as discrete elements, h the exception of the secondary side of the steam generators. In addition, all components with g axial dimensions are divided into subcells adequate to minimize numerical diffusion and aring of gradients. rder to simulate the asymmetric thermal hydraulic and reactivity feedback effects that occur ng an SLB transient, the core is nodalized into three radial sectors. One sector corresponds to region immediately surrounding the assembly where the most reactive control rod is assumed k out of the core. This sector is termed the 'stuck rod' sector. The remainder of the region of core which is directly affected by the loop containing the break is the second sector and is ed the 'affected' sector. The remainder of the core and the other loop is termed either the ffected' or the 'intact' sector or loop.

 .5.2.5.1.9 Interloop Mixing ing an actual SLB transient, some mixing between the parallel channels within the reactor sure vessel will occur in the downcomer, the lower plenum, the core, and the upper plenum to lateral momentum imbalances, and turbulence or eddy mixing. The mixing will act to uce the positive reactivity feedback effects due to a reduced rate and magnitude of cooldown he affected loop and associated core sector.

his analysis, no credit is taken for turbulent or eddy mixing of coolant between loops or the llel flow channels within the reactor pressure vessel (RPV). However, interloop mixing is ulated to occur due to flow in interloop junctions in the upper and lower plenums. Mixing in lower plenum was reduced to a minimum by using an extremely high loss coefficient between affected and intact sectors.

 .5.2.5.1.10 Harsh Containment Conditions sh containment conditions can be caused by the release of steam within the reactor tainment. Under such conditions, only those trips which have been qualified for harsh 14.1-20                                   Rev. 35
 .5.2.5.2    Minimum Departure From Nucleate Boiling and Linear Heat Generation Rate Analysis NBR calculations require determination of the power, enthalpy, and flow distributions within highest power assembly of the stuck rod core sector. Similarly, determination of the maximum GR also requires characterization of the power distribution. The power distribution within the
 , including the highest powered assembly within the stuck rod core sector, is calculated with SM (Reference 14.1-3). Flow and enthalpy distributions within the core, including the highest ered assembly within the stuck rod core sector, are calculated with XCOBRA-IIIC ference 14.1-4). In order to obtain compatible flows, moderator densities, and powers within high power assemblies, iteration between PRISM and XCOBRA-IIIC is conducted.

this calculation, the modified Barnett correlation was found to be suitable for the MDNBR ulation. The modified Barnett correlation is based upon closed channels and primarily orm power distribution data. The correlation is based on assembly inlet (or upstream) fluid ditions rather than on local fluid conditions as is the case with subchannel based correlations.

 .5.2.6 Analysis Results ummary of calculated results important to this analysis is presented in Table 14.1.5.2-6 for the ting MDNBR and LHGR scenarios. The MDNBR values are listed together with the esponding core power values at the time of MDNBR which corresponds to the maximum t-scram power level. The outside containment cases, regardless of whether or not off site er was or was not available, were found to be the most limiting. For cases where off site er was available for operation of the primary coolant system pumps, the MDNBR and the imum LHGR occurred at the time of the maximum power condition. For cases where off site er is lost and the primary system pumps coast down, the maximum LHGR and the MDNBR ur when the worst combination of core power, flow, inlet temperature, and pressure are ent. These conditions occurred at the time of peak power in this analysis.

scenario which resulted in the highest post-scram power level and the largest LHGR is that ated from HZP with the break occurring outside containment and with off site power available operation of the primary coolant pumps. This case is presented in detail. NSSS responses for the scenarios with loss of off site power for operation of the primary em coolant pumps are different from those scenarios where off site power is available ughout the transient due to the pump coastdown and subsequent natural circulation of the ary coolant. Post-scram maximum power levels attained during the transient are significantly er. Lower power levels result from lower positive moderator feedback. The positive derator feedback is reduced due to the coolant density reductions that occur axially upwards in core at low core flow rates, even for low core power levels. Lower power levels cause NBR values to increase, but lowering flow rates cause MDNBR values to decrease. Overall, 14.1-21 Rev. 35

he two loss-of-off site-power scenarios analyzed, the HZP break occurring outside tainment case resulted in lower MDNBR values. The general response of the HFP and HZP s with loss of off site power is comparable. Because the two scenarios are quite similar in s of their general response, only the limiting MDNBR case (i.e., HZP break outside tainment and without off site power) is presented in detail.

 .5.2.6.1     Hot Zero Power Outside Containment with Off site Power Available S-RELAP5 simulation of the NSSS during the HZP transient with off site power available is strated in Figures 14.1.5.2-2 through 14.1.5.2-9. A tabulation of the sequence of events is ented in Table 14.1.5.2-7. The S-RELAP5 computation was terminated 600 seconds after k initiation. This is well beyond the time of MDNBR or peak LHGR.
 .5.2.6.1.1 Secondary System Thermal Hydraulic Parameters m flow out the break is the source of the NSSS cooldown. Break flow for the steam erators is plotted in Figure 14.1.5.2-2. Secondary pressure for the steam generators is plotted igure 14.1.5.2-3. After break initiation, the pressure in the affected steam generator decreased ediately and then stabilized around 180 seconds. The mass inventory in the affected steam erator decreased rapidly for the first 100 seconds of the transient and slowly thereafter. The ffected steam generator mass inventory remained essentially constant throughout the transient.

intact steam generator blows down for a short period until the MSIVs completely close roximately 15 seconds after the break is initiated. The pressure recovers as the intact steam erator equilibrates with the primary system.

 .5.2.6.1.2 Primary System Thermal Hydraulic Parameters primary system coolant temperature and pressurizer pressure and level responses resulting m the break flow are illustrated in Figures 14.1.5.2-4 through 14.1.5.2-6. The primary system sure decays rapidly as the coolant contracts due to cooldown and the pressurizer empties. The IVs close at 15 seconds, ending the blowdown of the intact steam generators and reducing the of energy removal from the primary fluid. The pressurizer emptied at approximately 40 onds and system pressure (which increased slowly for the duration of the transient) was eafter established by the saturation temperature of the primary coolant in the upper head of the tor vessel.
 .5.2.6.1.3 Reactivity and Core Power reactivity transient calculated by S-RELAP5 is illustrated in Figure 14.1.5.2-8. Initially, the is assumed to be at zero power. All control rods, except the most reactive one, are assumed to nserted into the core following the reactor trip signal. The reactivity transient then proceeds.

total core reactivity, initially at 0.00$ decreased initially due to reactor scram worth, then 14.1-22 Rev. 35

tivities balancing the scram reactivity balancing the scram reactivity, when boron began ring the core, causing the power to decrease. SI flow to the RCS began 41 seconds after break initiation and 25 seconds after the HPSI ation signal. Twenty-five seconds was the assumed time for the HPSI pumps to reach rated ed. ure 14.1.5.2-9 shows the transient reactor power. The reactor power initially declined due to rtion of the control rods. The severe cooldown resulted in power increasing after 80 seconds. aximum power level of 296 MW or 11% of rated power occurred at 336 seconds.

 .5.2.6.1.4 PRISM and XCOBRA-IIIC Results PRISM calculation is made initially on the basis of S-RELAP5 input. Each assembly within three channels is assumed to have a uniform flow corresponding to the sector flows calculated h S-RELAP5. Due to high power peaking in the region of the stuck control rod, large derator density reductions are calculated to occur in the top portions of several assemblies in region of the core in the PRISM calculation and are responsible for the significant reduction eactivity observed when PRISM is compared to S-RELAP5. An XCOBRA-IIIC analysis is conducted to define the flow and enthalpy distribution within the high power assembly.

omparison of the overall change in reactivity from the event initiation to the time of maximum GR between S-RELAP5 and PRISM shows the S-RELAP5 power calculation is conservative.

 .5.2.6.1.5 Departure From Nucleate Boiling Ratio and Linear Heat Generation Rate Results the MDNBR portion of the calculation, the radial power distribution was modified to servatively account for local rod power distribution effects within the hot assembly. This was e by raising the power of the hot assembly to bound the peak rod power.

the bases of these conservative assumptions, the MDNBR value was calculated to be greater the modified Barnett 95/95 DNBR correlation limit. Therefore, no fuel rods would be ected to fail during this transient scenario from an MDNBR stand point. analysis of the peak LHGR also comes from the PRISM and XCOBRA-IIIC analysis. The k LHGR is calculated from the S-RELAP5 total core power and the PRISM radial and axial king. The peak LHGR was calculated to be less than the FCMLHR limit for the HZP outside tainment break with off site power available event. As such, no fuel pins are predicted to fail to the violation of the FCMLHR limit.

 .5.2.6.2   Hot Zero Power Outside Containment with Loss of Off Site Power S-RELAP5 NSSS simulation of the most limiting SLB scenario from an MDNBR standpoint
, HZP outside containment break with a loss of off site power) is illustrated in 14.1-23                                  Rev. 35

onds after initiation of the break. The S-RELAP5 computation was terminated 600 seconds r break initiation. This is well beyond the time of MDNBR and maximum LHGR.

 .5.2.6.2.1 Secondary System Thermal Hydraulic Parameters m flow out the break is the source of the NSSS cooldown. Steam flow for the affected steam erator is plotted in Figure 14.1.5.2-10. Secondary pressure for the steam generators is plotted igure 14.1.5.2-11. The affected steam generator blows down through the break throughout the sient. The pressure and mass flow rate dropped rapidly at first and then proceeded downward slower decay rate until natural circulation flow was established by approximately 220 onds.

intact steam generators blow down for a short period until the MSIVs completely close roximately 14 seconds after the break is initiated. The pressure recovers as the intact steam erator equilibrates with the primary system. Subsequently, the intact steam generator pressure ains essentially constant as the primary intact coolant loop approaches natural circulation ditions.

 .5.2.6.2.2 Primary System Thermal Hydraulic Parameters primary system core coolant temperature and pressurizer pressure and level responses lting from the break flow are illustrated in Figures 14.1.5.2-12 through 14.1.5.2-14. The ary system pressure decays rapidly as the coolant contracts due to the cooldown and the surizer empties. Continued pressure reduction in the primary system causes the relatively hot nant liquid in the head of the RPV vessel to flash. The flashing in the upper head, coupled h near equilibration of other NSSS parameters, retards the pressure decay from that point ward.

omparison of intact and affected core sector inlet temperatures throughout the transient cates significant differences due to the limited cross flow allowed between loops. The core or flows all show the same trend due to the coastdown of the primary coolant pumps. That is, lows decrease rapidly until natural circulation conditions are achieved in the two flow loops.

 .5.2.6.2.3 Reactivity and Core Power reactivity transient calculated by S-RELAP5 is illustrated in Figure 14.1.5.2-15. Initially, the is assumed to be at zero power. The total core reactivity, initially at 0.00$ decreased initially to reactor scram worth, then steadily increased due to moderator and Doppler feedback ciated with the primary system cooldown. The rise in reactor power was arrested when boron an entering the core at 337 seconds. Power then declined slowly due to an increasing boron centration in the primary system.

HPSI actuation signal was received at 21 seconds. After a 25 second delay, during which the SI pumps reached rated speed, HPSI flow to the RCS began, at 46 seconds. 14.1-24 Rev. 35

d power (167 MW) at 356 seconds.

 .5.2.6.2.4 PRISM and XCOBRA-IIIC Results PRISM calculation is initially made on the basis of S-RELAP5 predicted core power, flow, sure, and inlet temperatures. The PRISM calculations provide the radial and axial power ributions for use in the XCOBRA-IIIC code. Due to the high power peaking in the region of stuck control rod, and the low core average natural circulation flow rates, large moderator sity decreases are calculated in several assemblies in this region in the PRISM calculation and responsible for the significant reduction in reactivity observed when PRISM is compared to ELAP5. An XCOBRA-IIIC analysis is also conducted to define the flow and enthalpy ribution within the high power assembly.

omparison of the overall change in reactivity from the event initiation to the time of minimum BR between S-RELAP5 and PRISM shows the S-RELAP5 power calculation is conservative.

 .5.2.6.2.5 Departure From Nucleate Boiling Ratio and Linear Heat Generation Rate Results MDNBR of the hot fuel assembly is calculated to be greater than the modified Barnett 95/95 BR correlation limit. Therefore, no fuel rods are expected to fail from an MDNBR standpoint.

before, the analysis of the peak LHGR comes from the PRISM and the XCOBRA-IIIC lysis. The peak LHGR was less than the FCMLHR limit. Therefore, it is apparent that terline melt is not predicted to occur. Thus, no fuel failures are predicted to occur due to ation of the centerline melt criteria.

 .5.2.7 Conclusions HFP and HZP scenarios, with off site power maintained for operation of the primary coolant ps, resulted in a return to higher power levels than the scenarios where off site power is lost.

wever, these scenarios provide substantially greater margin to the MDNBR limit because of higher coolant flow rate. In no scenario evaluated, however, was fuel failure calculated to ur as a result of penetration of the MDNBR safety limit. Even though the scenarios with off power available have substantially greater margin to the MDNBR limit because of a higher lant flow rate, the higher power levels in combination with the highly skewed power ribution due to the assumed stuck rod cluster resulted in them having the least margin to the centerline melt limit. HZP outside containment break scenario concurrent with a loss of off site power was rmined to be the most limiting in this analysis from an MDNBR standpoint. The MDNBR of hot fuel assembly is calculated to be above the modified Barnett 95/95 DNBR correlation

t. Therefore, no fuel rods are expected to fail from an MDNBR standpoint.

14.1-25 Rev. 35

highest return to power and highest calculated LHGR. No fuel pins are calculated to fail due to violation of the FCMLHR limit. 1.5.3 Radiological Consequences of a Main Steam Line Break main steam line break is postulated to occur in a main steam line outside the containment. radiological consequences of a main steam line break inside containment is bounded by the n steam line break outside containment. The plant is assumed to be operating with Technical cification coolant concentrations. A 150 gpd primary to secondary leak is assumed to occur in h steam generator. main steam line break (MSLB) accident begins with a break in one of the main steam lines ing from a steam generator (affected generator) to the turbine coincident with a loss of off site er. As a result, the condenser is unavailable and cool down of the primary system is through release of steam to the environment from the intact generator. In order to maximize doses, k scenarios are assessed in the following structures: 1) Turbine Building, 2) Containment and nclosure Building. The analysis for the MSLB in Containment assumes 3.7% fuel failure. The lyses for the MSLB in the Turbine Building and Enclosure Building assume no fuel failure but ress the pre-accident (60Ci/gm Dose Equivalent (DEQ) I-131) and concurrent iodine (500 es the iodine appearance rates at 1Ci/gm DEQ I-131) spike scenarios as well as Technical cification levels of activity consistent with Dose Equivalent I-131 and Xe-133 limits. operator action is credited in the event of MSLB in the Enclosure or Turbine Building. There sufficient radioactivity in the release to alarm control room radiation monitors with sequent isolation of the control room. This action requires an operator to isolate the control m within 4 hours of MSLB and align Control Room Emergency Ventilation for filtered rculation within an additional 1 hour. noble gases and iodines in the primary coolant that leak into the faulted steam generator ng the transient are released directly to the environment without holdup or decontamination ept for MSLB in containment. For the MSLB in containment, releases from the faulted steam erator are released to the environment at Technical Specification containment leak rates. An ne partition factor of 0.01 is used for the releases from the unaffected steam generator. Off-site er is assumed to be lost, thus making the condenser unavailable. The steam releases from the n steam line break are from the turbine building blowout panels as the atmospheric dispersion or is greater for this release point than the enclosure building blowout panels. The steam ases from the intact steam generator are from the MSSVs/ADVs. radiological consequences of a main steam line break to the EAB, LPZ and Millstone 2 trol Room are reported in Table 14.1.5.3-2. The assumptions used to perform this evaluation summarized in Table 14.1.5.3-1. resulting doses to the EAB, LPZ and Control Room do not exceed the guidelines provided in CFR 50.67 and Regulatory Guide 1.183. 14.1-26 Rev. 35

-1 SRP Chapter 15 Non-LOCA Methodology for Pressurized Water Reactors, EMF-2310(P)(A), Revision 1, Framatome-ANP, May 2004. -2 ANF-RELAP Methodology for Pressurized Water Reactors: Analysis of Non-LOCA Chapter 15 Events, ANF-89-151(P)(A), Advanced Nuclear Fuels Corporation, May 1992. -3 Reactor Analysis Systems for PWRs, Volume 1 - Methodology Description, Volume 2 - Benchmarking Results, EMF-96-029(P)(A), Siemens Power Corporation, January 1997. -4 XCOBRA-IIIC: A Computer Code to Determine the Distribution of Coolant During Steady-State and Transient Core Operation, XN-NF-75-21(A), Revision 2, Exxon Nuclear Company, January 1986. -5 HTP: Departure from Nucleate Boiling Correlation for High Thermal Performance Fuel, EMF-92-153(P)(A), Revision 1, Siemens Power Corporation, January 2005. 14.1-27 Rev. 35

ABLE 14.1.1-1 AVAILABLE REACTOR PROTECTION FOR THE DECREASE IN FEEDWATER TEMPERATURE EVENT Reactor Operational Modes Reactor Protection 1 Variable Overpower Trip Thermal Margin/Low Pressure Trip Local Power Density Trip Low Steam Generator Pressure Trip 2 Variable Overpower Trip Low Steam Generator Pressure Trip 3 Variable Overpower Trip 4-6 Not a credible event for these reactor operating conditions since there is no extraction steam to the feedwater heaters 14.1-28 Rev. 35

BLE 14.1.1-2 DISPOSITION OF EVENTS FOR THE DECREASE IN FEEDWATER TEMPERATURE EVENT Reactor Operational Modes Disposition 1 Bounded by Event 14.1.3, Increase in Steam Flow Event 2, 3 Bounded by the above 4-6 No analysis required; not a credible event 14.1-29 Rev. 35

ABLE 14.1.2-1 AVAILABLE REACTOR PROTECTION FOR THE INCREASE IN FEEDWATER FLOW EVENT eactor Operational Modes Reactor Protection 1 Variable Overpower Trip Local Power Density Trip Thermal Margin/Low Pressure Trip Low Steam Generator Pressure Trip Safety Injection Actuation Signal Reactor Trip on Turbine Trip due to High Steam Generator Water Level

  • 2 Low Steam Generator Pressure Trip Variable Overpower Trip Safety Injection Actuation Signal 3 Variable Overpower Trip Safety Injection Actuation Signal 4 Technical Specification Requirements on Shutdown Margin Inherent Negative Doppler Feedback **

5, 6 No analysis required; no significant consequences Additional protection is provided by this control-grade reactor trip. Provides defense in depth. 14.1-30 Rev. 35

BLE 14.1.2-2 DISPOSITION OF EVENTS FOR THE INCREASE IN FEEDWATER FLOW EVENT Reactor Operational Modes Disposition 1 Bounded by Event 14.1.3 (Increase in Steam Flow) 2 Bounded by the Mode 3 Case 3-6 Bounded by Event 14.1.3 14.1-31 Rev. 35

ABLE 14.1.3-1 AVAILABLE REACTOR PROTECTION FOR THE INCREASE IN STEAM FLOW EVENT Reactor Operational Modes Reactor Protection 1 Low Steam Generator Pressure Trip Low Steam Generator Water Level Trip Thermal Margin/Low Pressure Trip Local Power Density Trip Variable Overpower Trip Safety Injection Actuation Signal 2 Low Steam Generator Pressure Trip Low Steam Generator Water Level Trip Variable Overpower Trip Safety Injection Actuation Signal 3 Variable Overpower Trip Safety Injection Actuation Signal 4 Technical Specification Requirements on Shutdown Margin Inherent Negative Doppler Feedback 5, 6 No Analysis Required; No Significant Consequences Provides defense in depth. 14.1-32 Rev. 35

BLE 14.1.3-2 DISPOSITION OF EVENTS FOR THE INCREASE IN STEAM FLOW EVENT Reactor Operational Modes Disposition 1 Analyze 2, 3, 4 Bounded by the above 5, 6 No analysis required; no significant consequences 14.1-33 Rev. 35

TABLE 14.1.3-3 INITIAL CONDITIONS FOR THE INCREASE IN STEAM FLOW EVENT Parameter Initial Value actor power 2754 MWt (102% of rated) ld leg temperature 549°F tal RCS flow rate 360,000 gpm ssurizer pressure 2250 psia ssurizer level 65% of span am generator pressure 875 psia am generator fluid mass (feedwater and steam) 167,237 lbm per steam generator 14.1-34 Rev. 35

BLE 14.1.3-4 EVENT

SUMMARY

FOR THE INCREASE IN STEAM FLOW EVENT Time Event (seconds) am dump to condenser valves and turbine bypass valves open fully 0.0 rbine control valves open fully 0.0 icated thermal power reaches Variable Overpower trip ceiling (111.7%) 22.1 rbine trips on reactor scram signal 23.0 actor power reaches maximum value 23.5 ram CEA insertion begins 23.5 DNBR occurs 23.7 14.1-35 Rev. 35

TABLE 14.1.3-5 PEAK REACTOR POWER LEVELS FOR INCREASE IN STEAM FLOW EVENT Peak Reactor Power Case Amount of Steam Release (% of Rated) rbine control valves fully open ~11% excess 117.8% ermediate case ~20% excess 117.8% am dump to condenser valves, ~41% excess 117.7% bine bypass valves, and turbine ntrol valves fully open 14.1-36 Rev. 35

MPS-2 FSAR FIGURE 14.1.3-1 NORMALIZED POWER AND HEAT FLUX FOR THE INCREASE IN STEAM FLOW EVENT Normalized Power or Heat Flux (fraction of rated) Reactor power Indicated thermal power Indicated nuclear power Core-average heat flux Time (s) Rev. 21

MPS-2 FSAR FIGURE 14.1.3-2 REACTIVITY FEEDBACK FOR THE INCREASE IN STEAM FLOW EVENT Reactivity (dollars) Total Moderator Doppler Scram Time (s) Rev. 21

MPS-2 FSAR FIGURE 14.1.3-3 REACTOR COOLANT TEMPERATURES FOR THE INCREASE IN STEAM FLOW EVENT Coolant Temperature (F) Loop 1 hot leg Loop 2 hot leg Loop 1 cold legs Loop 2 cold legs Time (s) Rev. 21

MPS-2 FSAR FIGURE 14.1.3-4 CORE INLET MASS FLOW RATE FOR THE INCREASE IN STEAM FLOW EVENT Core Inlet Mass Flow Rate (lbm/s) Time (s) Rev. 21

MPS-2 FSAR FIGURE 14.1.3-5 PRESSURIZER PRESSURE FOR THE INCREASE IN STEAM FLOW EVENT 2400 2200 0 u; 2000

 ~

CD L.

J

(/J 1800 (/J CD L. 0. L. CD 1600 N i:

J

(/J en CD. L. 1400 0. 1200 1000 4 6 8 10 12 14 16 18 20 22 24 0 2 Time (s) Rev. 21

MPS-2 FSAR FIGURE 14.1.3-6 STEAM GENERATOR PRESSURES FOR THE INCREASE IN STEAM FLOW EVENT Steam Generator Pressure (psia) Steam Generator 1 Steam Generator 2 Time (s) Rev. 21

MPS-2 FSAR FIGURE 14.1.3-7 STEAM MASS FLOW RATES FOR THE INCREASE IN STEAM FLOW EVENT Turbine Excess Steam Mass Flow Rate (lbm/s) Time (s) Rev. 21

MPS-2 FSAR FIGURE 14.1.3-8 MAIN FEEDWATER FLOW FOR THE INCREASE IN STEAM FLOW EVENT Steam Generator 1 Steam Generator 2 MFW Mass Flow Rate (lbm/s) Time (s) Rev. 21

MPS-2 FSAR FIGURE 14.1.3-9 MAIN FEEDWATER TEMPERATURE FOR THE INCREASE IN STEAM FLOW EVENT MFW Temperature (F) Time (s) Rev. 21

ABLE 14.1.4-1 AVAILABLE REACTOR PROTECTION FOR THE INADVERTENT OPENING OF A STEAM GENERATOR RELIEF OR SAFETY VALVES eactor Operational Modes Reactor Protection 1 Low Steam Generator Pressure Trip Low Steam Generator Water Level Trip Variable Overpower Trip Local Power Density Trip Thermal Margin/Low Pressure Trip Safety Injection Actuation Signal 2 Low Steam Generator Pressure Trip Low Steam Generator Water Level Trip Variable Overpower Trip Safety Injection Actuation Signal 3, 4 Technical Specification Requirements on Shutdown Margin, Inherent Negative Doppler Feedback

  • 5, 6 No Analysis Required; Not a Credible Event Provides defense in depth.

14.1-46 Rev. 35

BLE 14.1.4-2 DISPOSITION OF EVENTS FOR THE INADVERTENT OPENING OF A STEAM GENERATOR RELIEF OR SAFETY VALVE EVENT Reactor Operational Modes Disposition 1-4 Bounded by analyses presented for Event 14.1.3 5, 6 Not a credible event; no analysis required 14.1-47 Rev. 35

ABLE 14.1.5.1-1 AVAILABLE REACTOR PROTECTION FOR STEAM SYSTEM PIPING FAILURES INSIDE AND OUTSIDE OF CONTAINMENT PRE-SCRAM ANALYSIS Reactor Operating Conditions Reactor Protection 1 Low Steam Generator Pressure Trip Low Steam Generator Water Level Trip Low Reactor Coolant Flow Variable Overpower Trip Local Power Density Trip Thermal Margin/Low Pressure Trip High Containment Pressure Trip Safety Injection Actuation Signal 2 Low Steam Generator Pressure Trip Low Steam Generator Water Level Trip Low Reactor Coolant Flow Variable Overpower Trip High Containment Pressure Trip Safety Injection Actuation Signal 3-6 Technical Specification Requirements on Shutdown Margin, Inherent Negative Doppler Feedback 14.1-48 Rev. 35

TABLE 14.1.5.1-2 DISPOSITION OF EVENTS FOR STEAM SYSTEM PIPING FAILURES INSIDE AND OUTSIDE CONTAINMENT PRE-SCRAM ANALYSIS Reactor Operating Conditions Disposition 1 Analyze 2 Analyze 3-6 Bounded by the above 14.1-49 Rev. 35

BLE 14.1.5.1-3 S-RELAP5 THERMAL-HYDRAULIC INPUT (PRE-SCRAM STEAM LINE BREAK) Initial Condition Thermal-Hydraulic Input HFP actor Power (MW) 2754 ssurizer Pressure (psia) 2250 ssurizer Level (%) 65 ld Leg Coolant Temperature (°F) 549 tal Primary Flow Rate (lbm/sec) 37,640 condary Pressure (psia) 876 re Bypass Flow Rate (lbm/sec) per Loop 724 in Feedwater Temperature (°F) 432 am Generator Mass Inventory (lbm) 167,237 14.1-50 Rev. 35

BLE 14.1.5.1-4 ACTUATION SIGNALS AND DELAYS (PRE-SCRAM STEAM LINE BREAK) Non-Harsh Containment Harsh Containment Reactor Trip Condition Setpoint Condition Setpoint Delay riable Overpower (ceiling) 111.6% of rated (NI Not credited 0.9 s power); 114% of rated (thermal power) w Reactor Coolant Flow Credited 85% flow 0.65 s gh Containment Pressure Not applicable 5.83 psig 0.9 s w Steam Generator Pressure 658 psia 550 psia 0.9 s

/LP (floor)                   1728 psia             1700 psia        0.9 s
/LP (function)            Evaluated from function   Not credited     0.9 s given in Technical Specification 14.1-51                          Rev. 35

TABLE 14.1.5.1-5 S-RELAP5 NEUTRONICS INPUT AND ASSUMPTIONS (PRE-SCRAM STEAM LINE BREAK) Point Kinetics Input Value ective Delayed Neutron Fraction 0.005245 derator Temperature Coefficient (pcm/°F) Off Site Power Available (Technical Specification most -8 to -32 negative limit) Loss of Off Site Power (Technical Specification most +4 positive limit above 70% RTP) P Scram Worth (pcm) 6425 utdown Margin Requirement (pcm) 3600 ppler Coefficient Off Site Power Available Nominal EOC Loss of Off Site Power 0.80 x least negative value at BOC ion Product and Actinide Decay Constants ault values in ANF-RELAP utilized 14.1-52 Rev. 35

TABLE 14.1.5.1-6 MDNBR AND PEAK REACTOR POWER LEVEL

SUMMARY

(PRE-SCRAM STEAM LINE BREAK) Type of Size of Peak Reactor Location of Break Cooldown Break MDNBR Power (% of rated) tside containment, Symmetric 3.25 ft2 ** 129.83% wnstream of check valves 3.50 ft2 ** 130.84% 3.75 ft2 ** 128.19% tside containment, upstream of Asymmetric 1.80 ft2 ** 124.62% eck valve 2.00 ft2 ** 126.05% 2.20 ft2 ** 126.94% 2.40 ft2 ** 122.46% ide containment, upstream of Asymmetric 0.05 ft2 ** 134.38% eck valve 0.10 ft2 ** 136.02% 0.20 ft2 ** 138.81%

  • 0.40 ft2 ** 125.87%

ide containment, upstream of Asymmetric 3.51 ft2 0.88 106.86% eck valve with loss of off site wer The peak LHRs for all pre-scram breaks are bounded by the peak LHR for the 0.20 ft2 break inside containment and upstream of a check valve. The MDNBRs for these pre-scram breaks are bounded by the MDNBR for the 3.51 ft2 break inside containment, upstream of a check valve with the loss of off site power. 14.1-53 Rev. 35

BLE 14.1.5.1-7 LHGR-LIMITING PRE-SCRAM STEAM LINE BREAK SEQUENCE F EVENTS: HFP 0.20FT2 ASYMMETRIC BREAK INSIDE CONTAINMENT WITH OFFSITE POWER AVAILABLE Time (sec.) Event 0 Break upstream of main steam line check valves opens 0 Turbine control valves open fully 115.2 High containment pressure trip setpoint reached 116.4 Turbine trips on reactor scram signal 116.4 Scram CEA insertion begins 116.4 Reactor power reaches maximum value 116.5 MDNBR occurs 14.1-54 Rev. 35

TABLE 14.1.5.1-8 MDNBR-LIMITING PRE-SCRAM STEAM LINE BREAK SEQUENCE OF EVENTS:HFP 3.51FT2 ASYMMETRIC BREAK INSIDE CONTAINMENT WITH LOSS OF OFFSITE POWER Time (seconds) Event 0 Break occurs 0 RCPs trip 0 Peak LHGR (kW/ft) 2 Scram signal on low flow trip 3 Scram CEA Insertion begins 3 Maximum Power (Fraction of RTP) 4 MDNBR 14.1-55 Rev. 35

MPS-2 FSAR FIGURE 14.1.5.1-1 NORMALIZED CORE POWER (0.20 FT2 ASYMMETRIC BREAK INSIDE CONTAINMENT) Normalized Power or Heat Flux (fraction of rated) Reactor Power Indicated thermal power Indicated nuclear power Core-average heat flux Time (s) Rev. 21

MPS-2 FSAR FIGURE 14.1.5.1-2 CORE INLET TEMPERATURES (0.20 FT2 ASYMMETRIC BREAK INSIDE CONTAINMENT) Coolant Temperature (F) Loop 1 hot leg Loop 2 hot leg Loop 1 cold legs Loop 2 cold legs Time (s) Rev. 21

MPS-2 FSAR FIGURE 14.1.5.1-3 REACTIVITY FEEDBACK (0.20 FT2 ASYMMMETRIC BREAK INSIDE CONTAINMENT) Reactivity (dollars) Total Moderator Doppler Scram Time (s) Rev. 21

MPS-2 FSAR FIGURE 14.1.5.1-4 PRESSURIZER PRESSURE (0.20 FT2 ASYMMETRIC BREAK INSIDE CONTAINMENT) Pressurizer Pressure (psia) Time (s) Rev. 21

MPS-2 FSAR FIGURE 14.1.5.1-5 STEAM GENERATOR PRESSURES (0.20 FT2 ASYMMETRIC BREAK INSIDE CONTAINMENT) SG 1 SG 2 Steam Generator Pressure (psia) Time (s) Rev. 21

MPS-2 FSAR FIGURE 14.1.5.1-6 STEAM MASS FLOW RATES (0.20 FT2 ASYMMETRIC BREAK INSIDE CONTAINMENT) Turbine Break Steam Mass Flow Rate (lbm/s) Time (s) Rev. 21

MPS-2 FSAR FIGURE 14.1.5.1-7 NORMALIZED POWER AND HEAT FLUX (ASYMMETRIC 3.51 FT2 BREAK INSIDE CONTAINMENT WITH LOSS OF OFFSITE POWER) 1.3 Reactor Power Normalized Power or Heat Flux (fraction of rated) 1.2 Core-average heat flux 1.1 1.0

                                                     .9
                                                      .8
                                                      .7
                                                      .6
                                                      .5
                                                      .4
                                                      .3
                                                      .2
                                                      .1
                                                      .0 0   1   2   3   4   5   6   7   8   9   10 11 12 13 14 15 16 17 18 19 20 Time (s)

MARCH 1999

MPS-2 FSAR FIGURE 14.1.5.1-8 REACTOR COOLANT TEMPERATURES (ASYMMETRIC 3.51 FT2 BREAK INSIDE CONTAINMENT WITH LOSS OF OFFSITE POWER) 650 640 630 620 Coolant Temperature (F) 610 600 590 580 570 560 550 540 530 Loop 1 hot leg Loop 2 hot leg 520 Loop 1 cold legs 510 Loop 2 cold legs 500 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 Time (s) MARCH 1999

MPS-2 FSAR FIGURE 14.1.5.1-9 NORMALIZED REACTOR COOLANT SYSTEM FLOW RATE (ASYMMETRIC 3.51 FT2 BREAK INSIDE CONTAINMENT WITH LOSS OF OFFSITE POWER) 100 90 Normalized Flow Rate (percent) 80 70 60 50 40 30 20 10 0 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 Time (s) MARCH 1999

MPS-2 FSAR FIGURE 14.1.5.1-10 PRESSURIZER PRESSURE (ASYMMETRIC 3.51 FT2 BREAK INSIDE CONTAINMENT WITH LOSS OF OFFSITE POWER) 2300 2250 2200 Pressurizer Pressure (psia) 2150 2100 2050 2000 1950 1900 1850 1800 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 Time (s) MARCH 1999

MPS-2 FSAR FIGURE 14.1.5.1-11 STEAM GENERATOR PRESSURES (ASYMMETRIC 3.51 FT2 BREAK INSIDE CONTAINMENT WITH LOSS OF OFFSITE POWER) 1200 1100 Steam Generator Pressure (psia) 1000 900 800 700 600 500 400 300 200 SG 1 100 SG 1 0 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 Time (s) MARCH 1999

ABLE 14.1.5.2-1 AVAILABLE REACTOR PROTECTION FOR STEAM SYSTEM PIPING FAILURES INSIDE AND OUTSIDE OF CONTAINMENT, POST-SCRAM ANALYSIS Reactor Operating Conditions Reactor Protection 1 Low Steam Generator Pressure Trip Low Steam Generator Water Level Trip Local Power Density Trip Thermal Margin/Low Pressure Trip High Containment Pressure Trip Safety Injection Actuation Signal 2 Low Steam Generator Pressure Trip Low Steam Generator Water Level Trip High Containment Pressure Trip Safety Injection Actuation Signal 3-6 Technical Specification Requirements on Shutdown Margin, Inherent Negative Doppler Feedback 14.1-67 Rev. 35

TABLE 14.1.5.2-2 DISPOSITION OF EVENTS FOR STEAM SYSTEM PIPING AILURES INSIDE AND OUTSIDE OF CONTAINMENT, POST-SCRAM ANALYSIS Reactor Operating Conditions Disposition 1 Analyze 2 Analyze 3-6 Bounded by the above 14.1-68 Rev. 35

Initial Condition Thermal-Hydraulic Input HFP HZP Core Power (MW) 2700 1E-6 Primary Pressure (psia) 2250 2250 Pressurizer Level (%) 65 40 Cold Leg Temperature (°F) 549 532 Primary Flow Rate per Loop (lbm/sec) 18,820 19,241 Secondary Pressure (psia) 881 892 Steam Generator Mass Inventory (lbm) 167,237 253,989 Total Steam Flow (lbm/sec) per Steam Generator 1634 4 Break Characteristics imum Flow Area Affected Steam Generator (ft2) 3.51 Unaffected Steam Generator (ft2) 3.51 ation of Pipe Break Downstream of steam generator integral flow restrictor and upstream of MSIV Injection Systems HFP HZP Total HPSI Pumps 3 3 Active HPSI Pumps 2 2 Single Failure (No credit for mounted spare) 1 HPSI pump 1 HPSI pump Active Charging Pumps 0 0 Refueling Water Storage Tank Boron 1720 1720 Concentration (ppm) HPSI Delivery Curve Figure 14.1.5.2-1 Figure 14.1.5.2-1 14.1-69 Rev. 35

BLE 14.1.5.2-4 ACTUATION SIGNALS AND DELAYS (POST-SCRAM STEAM LINE BREAK) Parameter Setpoints Inside Containment Outside Containment Low Steam Generator Pressure Trip 550 psia 658 psia Low Pressurizer Pressure SIAS 1500 psia 1578 psia Low Steam Generator Pressure MSI 370 psia 478 psia IV Closure Required Actuation Signal (3) Above Delay - 6.9 seconds SI Actuation Required Actuation Signal (2) Above Delay - 25.0 seconds n Feedwater Valve Closure Required Actuation Signal (3) Above Delay - 14.0 seconds ctor Scram Required Actuation Signal (1) Above Delay - 0.9 second instrument delay, 3.0 second insertion time 14.1-70 Rev. 35

TABLE 14.1.5.2-5 S-RELAP5 NEUTRONICS INPUT AND ASSUMPTIONS (POST-SCRAM STEAM LINE BREAK) Point Kinetics Input Value ective Delayed Neutron Fraction 0.005245 derator Temperature Coefficient (pcm/°F) -32.0 P Scram Worth (pcm) 6425.0 utdown Margin Requirement (pcm) 3600.0 ck Rod Location Within half-core section cooled by affected loop. ion Product and Actinide Decay Constants Default values in ANF-RELAP utilized. 14.1-71 Rev. 35

TABLE 14.1.5.2-6 POST-SCRAM STEAM LINE BREAK ANALYSIS

SUMMARY

Maximum Fuel nitial Offsite Post-Scram Maximum Failure ower Power Break Return to LHGR (% of evel Available Location Power (MW) MDNBR (kW/ft) Core) HFP No outside 163 2.002 14.97 0.0 containment HFP Yes outside 278 2.417 22.77 0.0 containment HZP No outside 167 1.944 15.51 0.0 containment HZP Yes outside 296 2.374 23.10 0.0 containment 14.1-72 Rev. 35

TABLE 14.1.5.2-7 LHGR-LIMITING SEQUENCE OF EVENTS - HZP OFFSITE POWER AVAILABLE Time (s) Event

0. Reactor at HZP 0.+ Double-ended guillotine break. Shutdown reactivity inserted. AFW increased to maximum flow, all directed to affected steam generator.

7.6 MSIV closure trip signal 14.4 MSIVs closed 17.2 SI signal 42.2 SI pumps at rated speed (25 s delay) 336. Peak post-scram power reached (296 MW) 337.4 SI lines cleared. Boron begins to enter primary system 600. Calculation terminated; power decreasing 14.1-73 Rev. 35

TABLE 14.1.5.2-8 MDNBR-LIMITING POST-SCRAM STEAM LINE BREAK ANALYSIS

SUMMARY

ime (s) Event

0. Reactor at HZP 0.+ Double-ended guillotine break. Loss of offsite power. Shutdown reactivity inserted.

Full AFW flow started, all directed to the affected steam generator. 7.3 MSIV closure trip signal 14.1 MSIVs closed 21.4 SI signal 46.4 SI pumps at rated speed (25 s delay) 336.6 SI lines cleared. Boron begins to enter primary system 356. Peak post-scram power reached (167 MW) 600. Calculation terminated. Power decreasing. 14.1-74 Rev. 35

MPS-2 FSAR FIGURE 14.1.5.2-1 ONE PUMP HIGH PRESSURE SAFETY INJECTION SYSTEM DELIVERY vs. PRIMARY PRESSURE (POST-SCRAM STEAM LINE BREAK) Primary System Backpressure (psia) Flow to Primary System (lbm/sec) Rev. 21

MPS-2 FSAR FIGURE 14.1.5.2-2 STEAM GENERATOR BREAK FLOW (HZP POST-SCRAM STEAM LINE OUTSIDE CONTAINMENT BREAK WITH OFFSITE POWER AVAILABLE) Affected SG Unaffected SG Mass Flow (lbm/sec) Time (sec) Rev. 21

MPS-2 FSAR FIGURE 14.1.5.2-3 GENERATOR'S SECONDARY PRESSURES (HZP POST-SCRAM STEAM LNE OUTSIDE CONTAINMENT BREAK WITH OFFSITE POWER AVAILABLE) Unaffected SG Affected SG Pressure (psia) Time (sec) Rev. 21

MPS-2 FSAR FIGURE 14.1.5.2-4 CORE INLET TEMPERATURE (HZP POST-SCRAM STEAM LINE OUTSIDE CONTANMENT BREAK WITH OFFSITE POWER AVAILABLE) Affected Region Unaffected Region Temperature (F) Time (sec) Rev. 21

MPS-2 FSAR FIGURE 14.1.5.2-5 PRESSURIZER PRESSURE (HZP POST-SCRAM STEAM LINE OUTSIDE CONTAINMENT BREAK WITH OFFSITE POWER AVAILABLE) Pressure (psia) Time (sec) Rev. 21

MPS-2 FSAR FIGURE 14.1.5.2-6 PRESSURIZER LEVEL (HZP POST-SCRAM STEAM LINE OUTSIDE CONTAINMENT BREAK WITH OFFSITE POWER AVAILABLE) Level (% of Span) Time (sec) Rev. 21

MPS-2 FSAR FIGURE 14.1.5.2-7 STEAM GENERATOR'S SECONDARY MASS (HZP POST-SCRAM STEAM LINE OUTSIDE CONTAINMENT BREAK WITH OFFSITE POWER AVAILABLE) Unaffected SG Affected SG Mass (lbm) Time (sec) Rev. 21

MPS-2 FSAR FIGURE 14.1.5.2-8 REACTIVITY COMPONENTS (HZP POST-SCRAM STEAM LINE OUTSIDE CONTAINMENT BREAK WITH OFFSITE POWER AVAILABLE) Reactivity ($) Total Boron Moderator Doppler Scram Time (sec) Rev. 21

MPS-2 FSAR FIGURE 14.1.5.2-9 REACTIVITY POWER (HZP POST-SCRAM STEAM LINE OUTSIDE CONTAINMENT BREAK WITH OFFSITE POWER AVAILABLE) Reactor Power (MW) Time (sec) Rev. 21

MPS-2 FSAR FIGURE 14.1.5.2-10 STEAM GENERATOR BREAK FLOW (HZP POST-SCRAM STEAM LINE OUTSIDE CONTAINMENT BREAK WITH LOSS OF OFFSITE POWER) Affected SG Unaffected SG Mass Flow (lbm/sec) Time (sec) Rev. 21

MPS-2 FSAR FIGURE 14.1.5.2-11 STEAM GENERATOR'S SECONDARY PRESSURES (HZP POST-SCRAM STEAM LINE OUTSIDE CONTAINMENT BREAK WITH LOSS OF OFFSITE POWER) Unaffected SG Affected SG Pressure (psia) Time (sec) Rev. 21

MPS-2 FSAR FIGURE 14.1.5.2-12 CORE INLET TEMPERATURES (HZP POST-SCRAM STEAM LINE OUTSIDE CONTAINMENT BREAK WITH LOSS OF OFFSITE POWER) Affected Region Unaffected Region Temperature (F) Time (sec) Rev. 21

MPS-2 FSAR FIGURE 14.1.5.2-13 PRESSURIZER PRESSURE (HZP POST-SCRAM STEAM LINE OUTSIDE CONTAINMENT BREAK WITH LOSS OF OFFSITE POWER) Pressure (psia) Time (sec) Rev. 21

MPS-2 FSAR FIGURE 14.1.5.2-14 PRESSURIZER LEVEL (HZP POST-SCRAM STEAM LINE OUTSIDE CONTAINMENT BREAK WITH LOSS OF OFFSITE POWER) Level (% of Span) Time (sec) Rev. 21

MPS-2 FSAR FIGURE 14.1.5.2-15 REACTIVITY COMPONENTS (HZP POST-SCRAM STEAM LINE OUTSIDE CONTAINMENT BREAK WITH LOSS OF OFFSITE POWER) Reactivity ($) Total Boron Moderator Doppler Scram Time (sec) Rev. 21

MPS-2 FSAR FIGURE 14.1.5.2-16 REACTOR POWER (HZP POST-SCRAM STEAM LINE OUTSIDE CONTAINMENT BREAK WITH LOSS OF OFFSITE POWER) 500.0 450.0 400.0 r--. 3: 350.0

2 I.... 300.0 Q) 0 250.0 CL
 \...

0 200.0

-+J u

0 Q) 150.0 a::: 100.0 50.0

            .0
               .0        100.0    200.0        300.0 400.0     500.0     600.0 Time (sec)

Rev. 21

e Power Level (MWt) 2754 mary to Secondary Leak Rate per Steam Generator 150 gpd mary Coolant Iodine Concentration 1 Ci/gpm DE I-131 ondary Coolant Iodine Concentration 0.1 Ci/gm DE I-131 mary Coolant Noble Gas Concentration 1100 Ci/gm DE Xe-133 accident Spike Iodine Concentration 60 Ci/gm DE I-131 ed Fuel Percentage (Containment release only) 3.7% king Factor 1.69 ctor Coolant Mass 428,400 lbm ct Steam Generator Minimum Mass 91,092 lbm Boundary Breathing Rate (m3/sec) 0 - 8 hour 3.5E-04 8 - 24 hour .8E-04 24 - 720 hour 2.3E-04 Boundary Dispersion Factors (sec/m3) EAB: 0 - 2 hour 3.66E-04 LPZ: 0 - 4 hour 4.80E-05 4 - 8 hour 2.31E-05 8 - 24 hour 1.60E-05 24 - 96 hour 7.25E-06 96 - 720 hour 2.32E-06 trol Room Breathing Rate 3.5E-04 m3/sec trol Room Isolation Time - 4 hours MSLB outside containment (Operator action in the event the break is small enough to not automatically isolate the control room. 14.1-91 Rev. 35

trol Room Intake Prior to Isolation 800 cfm trol Room Inleakage After Isolation 200 cfm trol Room Emergency Filtered Recirculation Rate 2,250 cfm (t = 1 hour after isolation) trol Room Intake Dispersion Factors (sec/m3) tainment Enclosure Building - Ground Release 0 - 2 hour 3.00E-03 2 - 8 hour 1.87E-03 8 - 24 hour 6.64E-04 24 - 96 hour 5.83E-04 96 - 720 hour 4.97E-04 ospheric Dump Valves (ADVs) & Enclosure lding Blowout Panels 0 - 2 hour 7.40E-03 2 - 8 hour 5.71E-03 8 - 24 hour 2.13E-03 24 - 96 hour 1.74E-03 96 - 720 hour 1.43E-03 bine Building Blowout Panels 0 - 2 hour 1.22E-02 2 - 8 hour 8.67E-03 8 - 24 hour 3.77E-03 24 - 96 hour 2.92E-03 96 - 720 hour 2.23E-03 n Steam Safety Valves (MSSVs) 0 - 2 hour 3.03E-03 2 - 8 hour 2.30E-03 8 - 24 hour 8.46E-04 14.1-92 Rev. 35

96 - 720 hour 5.49E-04 trol Room Free Volume 35,650 ft3 trol Room Filter Efficiency (paticulate/elemental/organic) 90 / 90 / 70 % (1) e Conversion Factors Federal Guidance Reports 11 and 12 70% is a conservative analysis assumption for some iodine species. Technical Specifications can support assumptions for filter efficiencies of 90% for all iodine species. 14.1-93 Rev. 35

TABLE 14.1.5.3-2

SUMMARY

OF MILLSTONE 2 MSLB ACCIDENT DOSES EAB, LPZ, Control Room, MSLB rem - TEDE rem - TEDE rem - TEDE SLB in Enclosure Building pre-accident spike 9.1E-02 2.8E-02 2.6E+00 concurrent spike 1.6E-01 5.4E-02 3.8E+00 SLB in Turbine Building pre-accident spike 9.1E-02 2.9E-02 4.0E+00 concurrent spike 1.6E-01 5.4E-02 4.7E+00 SLB in Containment 3.7% fuel failure 1.4E-01 4.2E-02 2.0E+00 14.1-94 Rev. 35

14.1-95 Rev. 35 2.1 LOSS OF EXTERNAL LOAD 2.1.1 Event Initiator ajor loss of load can be initiated as the result of a loss of external electrical load or a turbine . Turbine stop valve closure is assumed as the initiator of this event because this is the fastest rejection which can be postulated which will challenge the plant overpressure and Specified eptable Fuel Design Limits (SAFDL) protection features. The assumed fast valve closure time 2 sec.) and the assumed unavailability of the steam dump system allow this event to bound the cts of Event 14.2.2 (Turbine Trip Steam dump system available) and the simultaneous ure of both Main Steam Isolation Valves (MSIV) case of Event 14.2.4 (Closure of the MSIVs Valve closure time > 0.02 seconds).

 .1.2 Event Description a full load reduction at power, the primary to secondary heat transfer would be severely inished because of the increase in secondary side temperature. Initially, in response to the load uction and diminished energy removal through the secondary system, the primary system peratures begin to increase. The increasing primary system average temperature causes an rge into the pressurizer due to the expanding primary fluid. The primary system pressure eases as the pressurizer steam space is compressed by the insurging liquid. Primary system rpressure protection is afforded by the pressurizer power-operated relief valves (PORV) and primary safety valves. Eventually, the secondary system pressure reaches the opening setpoint he secondary side safety valves and steam discharge occurs to limit the secondary side sure rise. Energy removal through the steam generator and pressurizer safety valves mitigates consequences of the load reduction. However, in analyzing the overpressurization aspects of event, no credit is taken for the PORVs on the primary system or the turbine bypass system steam dump system on the secondary system.
 .1.3 Reactor Protection ctor protection is provided by the high pressurizer pressure, variable overpower, thermal gin/low pressure (TM/LP), and low steam generator water level trips. If the turbine is tripped e initiation of this event, a direct reactor trip signal would be generated and the effects of this nt would be mitigated. However, no credit is taken for a direct reactor trip on turbine trip.

itionally, reactor protection is provided by the primary and secondary safety valves. Because he potential for increasing the primary system temperatures, with small increases in pressure, event can challenge the SAFDLs as well as the overpressure criteria mentioned above. ctor protection for the Loss of External Load event is summarized in Table 14.2.1-1. 2.1.4 Disposition and Justification s event is only credible for rated power and power operating conditions because there is no on the turbine at other reactor conditions. The consequences of this event for rated power 14.2-1 Rev. 35

match which occurs upon loss of load. ee separate cases of this event are analyzed for Millstone Unit 2 from rated power conditions. o cases maximize the primary and secondary system overpressurization criteria, respectively, the third case addresses the fuel design limits. There is no single failure considered which ld worsen the results. The disposition of events for the Loss of External Load event is marized in Table 14.2.1-2. 2.1.5 Definition of Events Analyzed objectives in analyzing this event are to demonstrate that: the primary pressure relief capacity ufficient to limit the pressure to less than 110% (2750 psia) of the design pressure, the ondary side pressure relief capacity is capable of limiting the pressure to less than 110% (1100

) of design pressure and the minimum Departure From Nucleate Boiling Ratio (DNBR) ains above the safety limit. No credit is taken for direct reactor trip on turbine trip, the turbine ass system or the steam dump system.

ee cases are analyzed for this event: Two cases addressing the primary and secondary system rpressurization criteria, respectively, and one case addressing the fuel design limits. In the ary overpressurization and fuel design limits cases the input parameters are biased to imize the increase in reactor power during the transient. The secondary overpressurization minimizes the increase in reactor power to reduce the pressurizer insurge and delay the tor trip. For the system overpressurization cases, the parameters and equipment operational es which significantly affect the results of the analysis are selected to maximize the system rpressure, while for the fuel design limit case the parameters and equipment states are selected educe the system pressurization and thereby provide a conservative estimate of the minimum BR during the transient. 2.1.6 Analysis Results s analysis was performed with AREVA Non-LOCA methodology, Reference 14.2-7, which s the ANF-RELAP code to calculate the system response, and the XCOBRA-IIIC code to ulate the minimum DNBR. cases are initiated with a ramp closure of the turbine control valve in 0.02 seconds. The surization of the secondary side results in decreased primary-to-secondary heat transfer, and a stantial rise in primary system temperature. This results in an insurge into the pressurizer, pressing the steam space and pressurizing the primary system. The reactor trips on high sure. The primary and secondary system safety valve setpoints were modeled with a +/- 3% t allowance, and the flow characteristics were modeled with a 3% allowance for accumulation. he primary system overpressurization case, the capacity of the pressurizer safety valves limit pressurizer pressure to a maximum of 2576 psia. The maximum Reactor Coolant System sure at the bottom of the reactor vessel is 2717 psia. 14.2-2 Rev. 35

RV setpoint. The peak secondary system pressure is 1086 psia. The minimum DNBR for this nt is bounded by the minimum DNBR of the Section 14.3.1 loss of forced reactor coolant flow nt. Because of this, the DNB SAFDL is not violated. This event does not challenge the MLHR limit. Therefore, LHR is not evaluated. responses of key system variables are given in Figures 14.2.1-1 to 14.2.1-5 for the primary em overpressurization case, Figures 14.2.1-6 to 14.2.1-11 for the secondary system rpressurization case, and Figures 14.2.1-12 to 14.2.1-17 for the minimum DNBR case. The uence of events for each of these cases is given in Tables 14.2.1-3, 14.2.1-4 and 14.2.1-5, ectively. primary and secondary side pressure relief valves have sufficient capacity to limit the ective system pressure to less than 110% (2750 and 1100 psia) of their design pressure. he event that one or more of the main steam safety valves are inoperable, analyses have onstrated that the design pressure limits are met provided thermal power is limited to the es given in Technical Specification 3.7.1.1. 2.1.7 Conclusion calculated minimum DNBR for the Loss of Load event is above the heat flux correlation ty limit, so the Departure From Nucleate Boiling (DNB) SAFDL is not exceeded in this event. peak pellet LHR is less than the FCMLHR limit. The maximum primary and secondary em pressures remain below 110% of design pressure. Thus, the Loss of External Load event been demonstrated to meet all required acceptance criteria.

 .2 TURBINE TRIP
 .2.1 Event Initiator s event is initiated by a turbine trip which results in closure of the turbine stop valves and a d reduction in energy removal through the steam generators.
 .2.2 Event Description reactor protection system is designed to generate a reactor trip signal automatically when the ine is tripped. Following reactor trip, there would be a rapid decrease in the energy being erated in the primary system. This would mitigate the consequences of the turbine trip event.

mary and secondary system overpressurization protection is provided by the code safety valves oth the primary and secondary systems and the secondary atmospheric dump valves. Also, if condenser was available, the steam bypass system would be activated to reduce the secondary em pressure. 14.2-3 Rev. 35

ctor protection is provided by the high pressurizer pressure trip, variable overpower trip, TM/ trip, low steam generator water level trip, and a nonsafety grade reactor trip on turbine trip. itional protection is also provided by the primary and secondary side safety valves. Reactor ection for the Turbine Trip event is summarized in Table 14.2.2-1.

 .2.4 Disposition and Justification s event is only credible for rated power and power operating conditions since the turbine will er be in tripped condition or there will be no load on the steam generators for other reactor ration conditions. The consequences of this event for rated power operation bound the event sequences for other operating conditions because of the higher initial stored energy in the ary system, maximum power to load mismatch potential, and the reduced SAFDL margin for d power operation. Because of the limiting assumptions used in the analysis of the sequences of the Loss of External Load (Event 14.2.1), the consequences of the Turbine Trip nt are bounded by the consequences of Event 14.2.1, which is analyzed for Millstone Unit 2.

major assumptions used in Event 14.2.1 are the conservatively rapid turbine stop valve ure time, the failure to trip the reactor on turbine trip, and the assumed unavailability of the ospheric steam dump system. The disposition of events for the Turbine Trip event is marized in Table 14.2.2-2.

 .3 LOSS OF CONDENSER VACUUM s event is not in the current licensing basis for Millstone Unit 2 and therefore is not analyzed.
 .4 CLOSURE OF THE MAIN STEAM ISOLATION VALVES
 .4.1 Event Initiator event postulated is the loss of control air to the MSIV operator. Either one or both MSIVs inadvertently close. The valves are swinging disc-type check valves, installed in a reversed ition and held open against steam flow by a pneumatically operating cylinder assembly. The es are spring-loaded to the closed position.
 .4.2 Event Description inadvertent MSIV closure is primarily of concern in boiling water reactors as indicated in the ndard Review Plan (Reference 14.2-1), but closure of the MSIVs in a pressurized water reactor ld cause a drastic reduction in the load on the reactor. As such, the consequences of a dual IV closure are similar to the consequences of Event 14.2.1. Although the valve closure time the MSIVs is less than 6 seconds, this is much longer than the turbine stop valve closure time med in Event 14.2.1 (0.02 seconds); as such, the transient events will proceed somewhat wer and be less severe than in the case of Event 14.2.1.

14.2-4 Rev. 35

oint of the steam line safety valves. The primary to secondary heat transfer will be inished, resulting in a heatup of the associated primary side loop. In response to the drop in m flow through the turbine control valves, the steam flow out of the unisolated steam erator will increase. Depressurization of the steam generator will result, causing an increase in t removal from the primary. The associated primary side loop will thus experience a ldown. The side of the core subjected to the cooldown will experience a power rise in the ence of a negative moderator temperature coefficient (MTC). 2.4.3 Reactor Protection ctor protection for the dual MSIV closure is provided by the high pressurizer pressure trip, able overpower trip, TM/LP trip, and low steam generator water level trip. Additional ection is provided by the primary and secondary side safety valves. ctor protection for the single MSIV closure is provided by the low steam generator level and steam generator pressure trips. Due to the location of the excore detectors and the mmetries associated with this event, the variable overpower and TM/LP trip may not get the uired 2 out of 4 channels tripped. Further, since one loop will be cooling down and one will be ting up, the pressure may be either increasing or decreasing. Thus, this event cannot take credit the variable high power, TM/LP or high pressure trips. Additional protection continues to be vided by the primary and secondary side safety valves. Reactor Protection for these events are marized in Table 14.2.4-1. 2.4.4 Disposition and Justification simultaneous closure of both MSIVs, the event will progress very similarly to Event 14.2.1. such, the limiting case is obtained when the event is initiated from rated full power conditions. to the decreased heat load and steam flow rates in Modes 2-4 the consequences of the event bounded by the Mode 1 analysis. The absence of any secondary side heat removal in Modes 5 6 eliminates the need to consider the event in these Modes. The turbine stop valve closure e employed in Section 14.2.1 analysis (0.02 sec) is much smaller than the MSIV closure time. s, the consequences of Event 14.2.1 will bound those of the dual MSIV closure event. asymmetric conditions resulting from the closure of only one of the two MSIVs are similar to predicted for a steam line break (SLB). That is, the primary coolant loop associated with the ed MSIV experiences a heatup due to the loss of heat sink and the primary coolant loop ciated with the open MSIV experiences a cooldown due to the perceived load increase. The perature increase seen by the hot loop will be limited by the actuation of the steam generator ty valves. The temperature decrease seen by the cooling loop will continue until such time as actor trip is generated. ce the loop experiencing the cooldown will see the larger temperature change, the limiting ditions for the event are at end of cycle (EOC). The EOC MTC is larger in absolute magnitude the beginning of cycle (BOC) MTC. When the larger MTC is coupled with the larger 14.2-5 Rev. 35

ce the asymmetries associated with the event preclude taking credit for the high pressure or able overpower trip, the single MSIV closure cannot be bounded without analysis by the loss oad, Event 14.2.1. limiting single MSIV closure case is that which is initiated from rated power in Mode 1. As the simultaneous closure of both MSIVs, the reduced heat load and steam flow rates reduce consequences in all other Modes. For the single MSIV closure case there is also no potential a post-trip return to power since the remaining MSIV and the turbine stop valves provide undant means for terminating the remaining steam flow from the unaffected steam generator. refore, there is no possibility of a sustained cooldown, thus preventing the addition of icient reactivity to the core to return to criticality after reactor trip. There is no single failure sidered which could worsen the results. The disposition of events for the Closure of the IVs events is summarized in Table 14.2.4-2.

 .4.5 Definition of Events Analyzed discussed above, the limiting case is obtained when the event is initiated from rated full-power ditions. For simultaneous closure of both MSIVs, the event will progress very similar to Event
 .1. The turbine stop valve closure time employed in the Event 14.2.1 analysis (0.02 seconds) uch smaller than the MSIV closure time. Thus, the consequences of Event 14.2.1 will bound e of the dual MSIV closure event.

objective in analyzing this event is to demonstrate that: the secondary side safety relief acity is capable of limiting the pressure to less than 110% (1100 psia) of the design pressure, the minimum DNBR remains above the safety limit, and that the peak LHR is below the terline melt limit. asymmetric conditions resulting from the closure of only one of the two MSIVs is similar to predicted for an SLB. That is, the primary coolant loop associated with the closed MSIV eriences a heatup due to the loss of heat sink and the primary coolant loop associated with the n MSIV experiences a cooldown due to the perceived load increase. The temperature increase n by the hot loop will be limited by the actuation of the steam generator safety valves. The perature decrease seen by the cooling loop will continue until such time as a reactor trip is erated. ce the loop experiencing the cooldown will see the larger temperature change, the limiting ditions for the event are at EOC. The EOC MTC is larger in absolute magnitude than the BOC C. When the larger MTC is coupled with the larger temperature change in the cooling loop, a er overall increase in core power will be predicted. This larger increase in core power will duce the limiting DNB conditions for the event. 14.2-6 Rev. 35

analysis used EOC initial conditions and assumed that one of the two MSIVs close thereby ating a heatup on the side of the core associated with the closed MSIV. The side of the core ciated with the open MSIV experiences a cooldown corresponding to an approximate bling of the steam flow. This is because the entire steam demand is shifted to a single steam erator. As the pressure in the steam generator drops, the amount of steam which is supplied by steam generator is also modeled to decrease. Two sensitivity cases were analyzed. In the first e, at transient initiation, the flow area of the steam control valve was set to a constant value al to the steady state flow area, thereby simulating the manual mode of operation of the turbine trol valves. In this case, the steam flow from a single steam generator with the open MSIV dly increases following the initiation of the transient to pick up the load from the isolated m generator, nearly doubling the initial steam flow through the single MSIV. In the second

 , steam flow is modeled as a function of steam generator pressure based on EOC coastdown
. This case simulates the automatic mode of operation of the turbine control valves. The ual mode case, having constant turbine control valve flow area, yielded lower steam flow s through the transient.

non-LOCA methodology (Reference 14.2-2) is used to analyze the MSIV closure event. Due he event asymmetry and the fact that the event proceeds much like a Steam Line Break (SLB) nt prior to reactor trip, portions of the pre-scram SLB methodology (Reference 14.2-2) and S-RELAP5 input deck were used to perform this analysis. The input deck was modified to the MSSV inlet piping and individual MSSVs to the steam line having the closed MSIV. For single MSIV closure event, the neutronics input required to predict the radial power ribution between the cold and the hot side of the core was, however, redeveloped based on nt specific PRISM (Reference 14.2-3) calculations. Core radial power distributions from full-er EOC PRISM cases with differences between the hot-region inlet temperature and the cold-on inlet temperature are used to determine the power split between the halves of the core as a ction of the difference in inlet temperatures. Since the temperature differences used in the SM cases meets or exceeds the inlet temperature difference calculated by S-RELAP5 during transient calculation, the power splits used in S-RELAP5 are bounding. The PRISM ulations for the single MSIV closure event are similar to those used in the pre-scram SLB nt. The pre-scram SLB and single MSIV closure event analyses both require power ribution data assuming that an all rods out power distribution is appropriate. limiting results were obtained from the manual mode turbine control valve case which has er steam flow rates. The results of the limiting EOC analysis are given in the event summary, le 14.2.4-3, and in Figures 14.2.4-1 through 14.2.4-5. As indicated in the event summary e the secondary safety valves open early in the transient limiting the temperature rise on the side of the core associated with the closed MSIV. The reactor trips on low steam generator sure which terminates the power rise. peak LHR and Minimum Departure from Nucleate Boiling Ratio (MDNBR) are predicted to ur on the cold side of the core. The peak LHR occurs at the time of the reactor trip, and the NBR occurs at the time of the peak heat flux. The peak LHR is less than the FCMLHR limit. minimum DNBR for this event is bounded by the minimum DNBR of the Section 14.3.1 loss 14.2-7 Rev. 35

secondary side safety valve setpoints were modeled assuming a +3% tolerance on the ning set pressures, and that the valves would relieve the rated flow at an inlet pressure of 1098

. The calculated maximum secondary side pressure is 1093 psia, which is less than 110%

00 psia) of design pressure.

 .4.7 Conclusion calculated minimum DNBR for the single MSIV closure event is above the HTP critical heat correlation safety limit, so the DNB SAFDL is not exceeded in this event. The peak LHR is than the FCMLHR limit. The maximum secondary side pressure is below 110% of design sure. Thus, the single MSIV closure event has been demonstrated to meet all required eptance criteria.

2.5 STEAM PRESSURE REGULATOR FAILURE lstone Unit 2 does not have any steam line pressure regulators, so this event is not credible for plant. No analysis needs to be considered for this event.

 .6 LOSS OF NONEMERGENCY AC POWER TO THE STATION AUXILIARIES s event is not in the current licensing basis for Millstone Unit 2 and therefore is not analyzed.
 .7 LOSS OF NORMAL FEEDWATER FLOW
 .7.1 Event Initiator Loss of Normal Feedwater Flow transient is initiated by a trip of the main feedwater pumps malfunction in the feedwater control valves.
 .7.2 Event Description loss of main feedwater flow will increase the secondary-side temperature and reduce the m generator heat removal capability because the main feedwater system is supplying cooled water to the steam generators. The rise in the secondary-side temperature leads to a rise he primary system coolant temperature. As the primary system temperatures increase, the lant expands into the pressurizer which increases the pressure by compressing the steam ume.

temperatures of the secondary sides and primary loops are controlled by the opening and ing of the main steam safety valves and/or the steam dump valves. The long term cooling of primary system is assured by the secondary-side water inventory supplied by the Auxiliary dwater System (AFWS). Two motor-driven auxiliary feedwater (AF) pumps are automatically ted upon a steam generator low liquid level signal. If a loss of offsite power occurs, the motor-14.2-8 Rev. 35

2.7.3 Reactor Protection tem overpressure protection is provided by the primary and secondary system safety valves. A tor trip occurs on low steam generator level with additional reactor protection provided by the h pressurizer pressure trip, variable overpower trip, and the TM/LP trip. Reactor protection for Loss of Normal Feedwater Flow event is summarized in Table 14.2.7-1. 2.7.4 Disposition and Justification s event is only credible for rated power and power operating conditions because the main water system is not required to provide feedwater to the steam generators for other reactor rating conditions. The consequences of this event for rated power operation bound the sequences for other conditions because of the higher initial stored energy in the primary em, the minimum steam generator inventory, and the greater impact of the loss of feedwater on the secondary system. near-term pressurization and DNB aspects of this event are bounded by those Events 14.2.1 14.3.1, respectively. In Event 14.2.1, reactor trip is delayed until a high pressurizer pressure al is received. This results in a higher power level at trip, greater pressurization and greater llenge to the SAFDLs than in Event 14.2.7. Long-term pressurization, if it occurs, is very dual and is arrested by opening of the pressurizer code safety valves. vent 14.3.1, the RCPs are tripped as the initiating event. Reactor trip occurs on low coolant

 , and the core flow rate at the time of trip is significantly lower than in the Loss of Normal dwater Flow event where the RCPs are tripped coincident with the reactor trip. The core er to flow ratio is much higher for Event 14.3.1, thereby producing a more limiting minimum BR.

Loss of Normal Feedwater event is analyzed to assess the maximum expected pressurizer l swell and the long-term adequacy of the AFWS to restore and maintain steam generator ntory and prevent steam generator dryout. The maximum level swell is examined to assure the pressurizer does not become water solid. Each case was analyzed using 102% power, imum allowed positive reactivity feedback, and maximum permitted pressurizer level. The power initial condition maximizes the core decay heat that must be removed in the post-scram od. A primary concern in simulating this event is to demonstrate adequate long-term cooling ability. The single active failure assumptions reduce heat removal capacity by significantly ting the amount of AFW flow supplied to the steam generators. o single failures considered in the Loss of Normal Feedwater Flow event are the failure of a or-driven AFW pump to start and the failure of the steam-driven AFW pump to start. Also sidered is a loss of offsite power coincident with reactor trip (Reference 14.2-7). The osition of events for the Loss of Normal Feedwater Flow event is summarized in le 14.2.7-2. 14.2-9 Rev. 35

Millstone Unit 2 AFW system consists of two independent motor-driven pumps which are med to start automatically within 240 seconds of AFW system actuation on low-low steam erator level. There is also a steam-driven AFW pump which may be started by operator action. e operator action is credited 10 minutes following reactor trip in the safety analysis.) The ng configuration allows each pump to supply both steam generators simultaneously. There are potential single active failures in this configuration: One is the failure of the steam-driven W pump to start, and the other is the failure of one of the two motor-driven AFW pumps to

t. Because of the differences in pump capacity and actuation times, it is not immediately ious which single active failure is the most limiting.

ther uncertainty when analyzing the Loss of Normal Feedwater Flow event is the effect of the P trip. If the RCPs remain on, the pump heat imposes a significant heat load on the system. If RCPs are tripped, primary to secondary heat removal capability is degraded due to sole ance on natural circulation. loss of offsite power option (RCP trip), combined with the two single active failure sibilities produces a total of four base cases. The four cases collectively demonstrate pliance with both the pressurizer overfill criterion and the steam generator secondary water ntory criterion when the MSSVs are the sole secondary steam release path. The biases and al conditions for the cases are identical and are selected to maximize pressurizer level ease and to minimize steam generator level recovery. A fifth case considers the effects of m dump system operation and determines whether the SG inventory boiloff required to cool RCS to no load temperature is offset by increased AFW flow at lower SG pressures. initiating event for each case is an instantaneous loss of main feedwater. analysis is performed with the ANF-RELAP code (Reference 14.2-6). The ANF-RELAP e includes relevant aspects of the mass and energy balance of the primary and secondary ems. itional conservative conditions are applied for analysis of each case to present the greatest llenge to the event acceptance criteria. In accordance with AREVA methodology, symmetric plugging is modeled for the four base cases utilizing the MSSVs as the sole secondary steam ase path. No steam generator tube plugging is applied for the fifth case. This conservatively imizes post-trip steam generator liquid inventories by producing slightly higher steam erator pressures and consequently lower AFW flows when steam generator pressure is trolled by the steam dumps instead of by the MSSVs.

 .7.5.1 Analysis Results Loss of Normal Feedwater Flow event is initiated from 102% power with each steam erator at nominal liquid levels. A total instantaneous loss of all Main Feedwater flow initiates event. When loss of offsite power assumptions are applied, the loss of offsite power and tor coolant pump trip is assumed to occur at scram. The reactor trips on the steam generator 14.2-10                                    Rev. 35

luated with an analytical low level trip setpoint of 43%. cases where off site power is available and the RCPs maintain forced coolant flow through primary system produce the lowest steam generator inventories. The loss of one of the two or-driven AFW pumps combined with post-trip RCS temperature control using the steam ps resulted in a minimum steam generator liquid mass inventory of 5,540 lbs per steam erator at 654 seconds. The steam generator level steadily recovers from this minimum level, ensuring continued heat removal. An event summary is presented in Table 14.2.7-3. The sient responses are presented in Figures 14.2.7-1 through 14.2.7-6. The transient execution e was 1800 seconds. cases where offsite power is assumed to be lost coincident with the reactor trip, and primary econdary heat transfer is achieved via natural circulation, generated the highest pressurizer ls. The loss of one of the two motor-driven AFW pumps produced the maximum pressurizer l of 76.3% at 43 seconds. Sufficient steam volume remains to preclude the expulsion of liquid m the pressurizer safety valves. An event summary is presented in Table 14.2.7-4. The sient responses are presented in Figures 14.2.7-6 through 14.2.7-10. The transient execution e was 2400 seconds.

 .7.6 Conclusions ss of normal feedwater event does not result in the violation of SAFDLs, peak pressurizer sure does not exceed 110% of the design rating and primary liquid is not expelled through the surizer safety valves. Adequate cooling water is supplied by the AFWS to allow a safe and erly plant shutdown and to prevent steam generator dryout. Thus, the loss of normal feedwater nt has been demonstrated to meet all required acceptance criteria.

2.8 FEEDWATER SYSTEM PIPE BREAKS INSIDE AND OUTSIDE CONTAINMENT s event is not in the current licensing basis for Millstone Unit 2 and, therefore, is not analyzed.

 .9 REFERENCES
 -1    Standard Review Plan for the Review of Safety Analysis Reports for Nuclear Power Plants, NUREG-0800, U.S. Nuclear Regulatory Commission, July 1981.
 -2    SRP Chapter 15 Non-LOCA Methodology for Pressurized Water Reactors, EMF-2310(P)(A), Revision 1, Framatome-ANP, May 2004.
 -3    Reactor Analysis Systems for PWRs, Volume 1 - Methodology Description, Volume 2 -

Benchmarking Results, EMF-96-029 (P)(A), Siemens Power Corporation, January 1997.

 -4    Technical Specifications for Millstone Unit 2, Docket Number 50-336.

14.2-11 Rev. 35

1990. -6 ANF-RELAP Methodology for Pressurized Water Reactors: Analysis of Non-LOCA Chapter 15 Events, ANF-89-151(P)(A), Siemens Power Corporation, May 1992. -7 Millstone Unit No. 2 Loss of Normal Feedwater Flow Transient with Reduced Auxiliary Feedwater Flow, EMF-98-015, Rev. 1, Siemens Power Corporation, December 1998. -8 HTP: Departure from Nucleate Boiling Correlation for High Thermal Performance Fuel, EMF-92-153(P)(A), Revision 1, Siemens Power Corporation, January 2005. 14.2-12 Rev. 35

TABLE 14.2.1-1 AVAILABLE REACTOR FOR THE LOSS OF EXTERNAL LOAD EVENT Reactor Operational Mode Reactor Protection 1 High Pressurizer Pressure Trip Variable Overpower Trip Thermal Margin/Low Pressure Trip Low Steam Generator Water Level Trip 2 High Pressurizer Pressure Trip Variable Overpower Trip Low Steam Generator Water Level Trip 3-6 No Analysis Required; Not a Credible Event 14.2-13 Rev. 35

BLE 14.2.1-2 DISPOSITION OF EVENTS FOR THE LOSS OF EXTERNAL LOAD EVENT Reactor Operational Mode Reactor Protection 1 Analyze 2 Bounded by the above, no analysis required 3-6 No Analysis Required; Not a Credible Event 14.2-14 Rev. 35

BLE 14.2.1-3 EVENT

SUMMARY

FOR THE LOSS OF EXTERNAL LOAD EVENT (PRIMARY OVERPRESSURIZATION CASE) Event Time (seconds) rbine Trip 0.0 st MSSV opens 4.3 mary pressure reaches high pressure trip setpoint 4.9 actor scrams 6.3 ssurizer safety valves open 7.4 ak pressurizer pressure 7.4 mary system peak pressure 7.9 st MSSV opens 9.5 d of calculation 20.0 14.2-15 Rev. 35

BLE 14.2.1-4 EVENT

SUMMARY

FOR THE LOSS OF EXTERNAL LOAD EVENT (SECONDARY OVERPRESSURIZATION CASE) Event Time (seconds) rbine Trip 0.0 st MSSV opens 2.9 ssurizer spray begins 3.2 ssurizer PORV opens 4.9 mary pressure reaches high-pressure trip setpoint 4.9 ssurizer safety valves open 5.0 ak pressurizer pressure 5.0 actor scrams 6.3 st MSSV opens 10.1 am Generator secondary peak pressure 10.2 d of calculation 20.0 14.2-16 Rev. 35

BLE 14.2.1-5 EVENT

SUMMARY

FOR THE LOSS OF EXTERNAL LOAD EVENT (MINIMUM DEPARTURE FROM NUCLEATE BOILING RATIO CASE) Event Time (seconds) rbine Trip 0.0 ssurizer spray on 3.4 st MSSV opens 4.3 mary pressure reaches high pressure trip setpoint 4.9 ssurizer PORVs open 4.9 mary safety valves open 4.9 ak pressurizer pressure 5.0 actor scrams 6.3 d of calculation 20.0 14.2-17 Rev. 35

MNPS -2 FSAR 3000.0 25 00.0 ~ ~ 2- 2000.0 L Q)

 ~

0 15 00.0 0-

 ~-

0 +-' u 1000.0 0 (l) 0::: 500.0

           .0 o 2 .5      5       7.5        10      12.5  15    17.5     20 Time (sec)

FIGURE 14.2 .1-1 REACTOR POWER LEVEL FOR LOSS OF EXTERNAL LOAD (PRIMARY OVERPRESSURIZATION CASE) SEPTEMBER 1998

MNPS-2 t-:SAR 200000

 ~

N 1~

  • 150000
 +....1
 '+-

1.- _C

J
 + -'

OJ 100000 X

J LL

+ -' 0 50000 Q) I o o 5 10 15 20 Time (sec) FIGURE 14.2.1 -2 CORE AVERAGE HEAT FLUX FOR LOSS OF EXTERNAL LOAD (PRIMARY OVERPRESSURIZATION CASE) SEPTEMBER 1998

MNPS -2 FSAR 62 0.0 I I I I I I I T il Hot Leg Temper at ure Cold Leg Temper ature ~ LL 60 0.0 <c:> (f) CD L

=J

+-' 0 L Q) 58 0.0 0... E Q) ~ o, '" 0 ...- '" 0 5 60. 0 ./ -l ".- 54 0.0 o 2 .5 5 7.5 10 12.5 15 17.5 20 Time (sec) FIGURE 14.2 .1 -3 REACTOR COOLANT SYSTEM TEMPERATURES FOR LOSS OF EX TERNAL LOAD (PRIMARY OVERPRESSURIZATION CASE) SEPTEMBER 19 9 8

MNPS-2 ~SAR 2 80 0.0 Pr essur izer I

                                                       '" "                              Reactor Vessel Bottom I          -,
                                                /               '\
                                              /                    \
                                            /                        '\

2 600.0 / -,

                                        /                                  -,

~ , .-0

                                 /
                                      "                                         \

if) / \ 0.. / -, Q) L 2 40 0.0

J '"

(f) en .... .... (1) L 0... 22 00.0 2 000.0 o 2. 5 5 7.5 10 12.5 15 17.5 20 Time (sec) FIGURE 14.2 .1 -4 PRIMARY SYSTEM PRESSURES FOR LOSS OF EXTERNAL LOAD (PRIMARY OVERPRESSURIZATION CASE) SEPTEMBER 1998

MNPS-I.. rSAR 2.0

            .0

~

t£}:

'--../

-+-'

.- - 2. 0

-+-'

U 0 Q) 0:::

       - 4.0
       -6.0 o 2.5      5      7.5        10     12.5     15  17.5     20 Time (sec)

FIGURE 14.2.1-5 TOTAL REACTIVITY FOR LOSS OF EXTERNAL LOAD (PRIMARY OVERPRESSURIZATION CASE) SEPTEMBER 1998

MNPS-L ,-SAR

     .3000.0 2 5 00.0

-s 2 2000.0 I-Q)

 ~

0 15 00.0 0-I- 0

-f-J U   100 0.0 0

Q) o: 500.0 I I ! I I I

           .0 o 2.5       5      7.5       10      12.5    15     17.5      20 Time (sec)

FIGURE 14.2.1 -6 REACTOR POWER LEVEL FOR LOSS OF EXTERNAL LOAD (SECONDARY OVERPRESSURIZATION CASE) SEPTEMBER 1998

MNPS-2 fSAR 200000

~,

C"'l

  ,'t-150000
+--'
'+-

I L

..c
J
+-'

OJ 100000

J LL

+-' 0 50000 Q)

r:::

o _ -J.. 1_ -..1----L_ ..L--.L.._ L . - - .* l - - - - lL _ - ' - - - - - ' -_ _- l . . - _ .lI _ - - ' - - - - l . _ - l - - - L _ . . . . l . . - - - - L _ . . L - _ o 5 10 15 20 Time (sec) FIGURE 14.2 .1 -7 CORE AVERAGE HEAT FLUX FOR LOSS OF EXTERNAL LOAD (SECONDARY OVERPRESSURIZATION CASE) SEPTEMBER 1998

MNPS-2 FSAR 62 0.0 Hot Leg Temperature Cold Leg Temper ature LL

----...- 60 0.0 (f)

Q) L

J
 +-'

0 L Q) 5 80.0 0... E Q) f-0... ...". - -- -- ---_...-- ------- -- - - 0 0 560.0 ---1 540.0 o 2.5 5 7.5 10 12.5 15 17.5 20 Time (sec ) FIGURE 14.2.1 -8 REACTOR COOLANT SYSTEM TEMPERATURES FOR LOSS OF EXTERNAL LOAD (SECONDARY OVERPRESSURIZATION CASE) SEPTEMBER 1998

MNPS-2 FSAR 2800.0 2 600. 0 ~ 0

.-(f)

Q

 <D     2 4 00 .0
 \-

(f) en

<D L

0-22 0 0. 0 2 00 0.0 o 2.5 5 7.5 10 12.5 15 17.5 20 Time (sec ) FIGURE 14.2 .1-9 PRESSURIZER PRESSURE FOR LOSS OF EXTER",AL LOAD (SECONDARY OVERPRESSURIZATION CASE) SEPTEMBER 1998

MNPS -2 r-"AR

2. 0 I j I I I I I I
           .0

~ ffl <c:>

>     -2.0

+-' u 0 1) et:

     -- 4 .0
    **-6.0                        I   I    I   I   I   I   I   I   ! I   I I I  I I o 2.5      5      7.5                  10                12.5    15       20 Time (sec)

FIGURE 14.2.1 -10 TOTAL REACTIVITY FOR LOSS OF EXTERNAL LOAD (SECONDARY OVERPRESSURIZATION CASE) SEPTEMBER 1998

MNPS-2 FSAR 1100.0 1050.0 1000.0 .-0(J) Q <c::> ([) 950.0

 .-:J (J)

(J) Q) I.- CL 900.0 850.0 Steam Generator 1

                                                         -  - -- Steam Generator 2 800.0 0     2. 5     5     7.5         10      12.5     15    17.5      20
                                            --urne   (sec)

FIGURE 14.2.1-11 MAXIMUM SECONDARY SYSTEM PRESSURES FOR LOSS OF EXTERNAL LOAD (SECONDARY OVERPRESSURIZATION CASE) SEPTEMBER 1998

MNPS -2 t-~AR

         .3000 .0 2 500.0 3:

~ _.e::.. 2000 .0 '-.-/ L Q)

  ~

0 1500.0 0-L 0 +-' u 0 1000.0 Q) 0:: 50 0 .0

               .0        I  I  I I  I  I
                  .0       2.5     5.0    7.5       10.0         12.5  15.0 17.5   20.0 Time (sec)

FIGURE 14.2.1-12 REACTOR POWER LEVEL FOR LOSS OF EXTERNAL LOAD (MDNBR CASE) SEPTEMBER 1998

MNPS~2 FSAR 1.2 1.0 x

J u, .8
-t-'

0 Q) I u .6 Q) N 0 E L

        .4 0

Z

       .2
      ..0
             .0   2.5     6.0      7.6       10.0      12.5  16.0     17.5     20.0 Time (sec)

FIGURE 14.2.1-13 NORMALIZED HEAT FLUX FOR LOSS OF EXTERNAL LOAD (MDNBR CASE) SEPTEMBER 1998

MNPS -2 t'~AR 620.0 -,- , I I I I I ,.---,., l I ,,---.--.-, I I r I 0 0 Core Inlet 0 0 Hot Leg-1

                                                                                  ~   Hot Leg -2 600.0 LL Q)

I-

J
-+-'

0 580.0 I-Q) Q-E C[) f--- 560.0 540.0 I I I ( I (

                   .0         2.5     5.0   7.5         10.0        12.5   15.0         17.5       20 .0 Time (s ec)

FIGURE 14.2.1-14 REACTOR COOLANT SYSTEM TEMPERATURE FOR LOSS OF EXTERNAL LOAD (M DNBR CASE) SEPTEMBER 1998

MNPS -2 FSAR 2500 .0 2400.0 2 300.0 Q) I-

J (f)

(f) 2200. 0 Q) I-0.. 21 0 0.0 2 0 00 .0 5.0 7.5 10.0 ' 2.6 15 .0 17.5 20.0

              .0      2.5 Tim e (se c )

FIGURE 14 .2.1-1 5 PRESSURIZER PRESSURE FOR LOSS OF EXTE RNAL LOAD (MDN BR CASE ) SEPTEMBER 1998

MNPS-2 t-SAR 2.0

             *0 W

'--../

- ~ -'
         - 2. 0
-+-~

u 0 Q) ex:

        - 4.0
        - 6.0                 !  I  I                            I   I  1 I   !  I
                .0   2.5        5.0     7.5       10.0      12.5   15.0     17.5      20 .0 Ti me (sec)

FIGURE 14.2.1-16 T l1TAL REACTIVITY FOR LOSS OF EXTERNAL LOAD (MDNBR CASE) SEPTEMBER 1998

MNPS -2 rvAR 110 0.0 10 50.0 ..----.. 10 0 0. 0 0

.-(f) 0..

Q) L-950.0

J (f)

(f) Q) L-

o. 900.0 850.0 0 0 Steam Generat or -1 0 0 St eam Gener at or-2 800.0
                  .0     2 .5    5.0    7.5       10.0      12.5     15.0    17.5       20.0 Time (s ec )

FIGURE 14.2.1 -17 MAXIMUM SECONDARY SYSTEM PRESSURE FOR LOSS OF EXTERNAL LOAD (M DNBR CASE) SEPTEMBER 19 98

ABLE 14.2.2-1 AVAILABLE REACTOR PROTECTION FOR THE TURBINE TRIP EVENT Reactor Operational Mode Reactor Protection 1 High Pressurizer Pressure Trip Nonsafety Grade Reactor Trip on Turbine Trip Variable Overpower Trip Thermal Margin/Low Pressure Trip Low Steam Generator Water Level Trip 2 High Pressurizer Pressure Trip Variable Overpower Trip Low Steam Generator Water Level Trip 3-6 No Analysis Required; Not a Credible Event 14.2-35 Rev. 35

TABLE 14.2.2-2 DISPOSITION OF EVENTS FOR THE TURBINE TRIP EVENT Reactor Operational Mode Disposition 1 Bounded by Event 14.2.1 for the rated power operating condition (number 1). 2 Same as above. 3-6 No Analysis Required; Not a Credible Event. 14.2-36 Rev. 35

ABLE 14.2.4-1 AVAILABLE REACTOR PROTECTION FOR THE CLOSURE OF THE MAIN STEAM ISOLATION VALVES EVENTS Reactor Operational Mode Reactor Protection 1 High Pressurizer Pressure Trip Variable Overpower Trip Thermal Margin/Low Pressure Trip Low Steam Generator Water Level Trip Low Steam Generator Pressure 2 High Pressurizer Pressure Trip Variable Overpower Trip Low Steam Generator Water Level Trip Low Steam Generator Pressure 3 Variable Overpower Trip 4-6 No Analysis Required 14.2-37 Rev. 35

ABLE 14.2.4-2 DISPOSITION OF EVENTS FOR THE CLOSURE OF THE MAIN STEAM ISOLATION VALVES EVENTS Reactor Operational Mode Disposition 1 Dual MSIV closure; Bounded by Event 14.2.1 Single MSIV Closure; Analyze 2-6 Bounded by Mode 1 14.2-38 Rev. 35

ABLE 14.2.4-3 EVENT

SUMMARY

FOR THE MAIN STEAM ISOLATION VALVE CLOSURE EVENT (LOWER STEAM FLOW CASE) Event Time (seconds) actor at full power 0.0 e MSIV closes instantaneously, flow though other MSIV increases 0.0 st MSSV opens 4.1 ak pressurizer pressure 12.4 ak Steam Generator secondary pressure 12.5 st MSSV opens 12.7 am Generator pressure reaches low-pressure trip setpoint 24.0 actor scrams (beginning of CEA insertion) 25.4 ak reactor power 25.4 d of calculation 40.0 14.2-39 Rev. 35

MPS-2 FSAR FIGURE 14.2.4-1 REACTOR POWER LEVEL FOR MSIV CLOSURE (LOWER STEAM FLOW CASE) Reactor Power (MW) Time (s) Rev. 21

MPS-2 FSAR FIGURE 14.2.4-2 REACTOR COOLANT SYSTEM TEMPERATURES FOR MSIV CLOSURE (LOWER STEAM FLOW CASE) Open MSIV Loop Tcold Open MSIV Loop Tcold Closed MSIV Loop Thot Closed MSIV Loop Thot Loop Temperature (F) Time (s) Rev. 21

MPS-2 FSAR FIGURE 14.2.4-3 PRESSURIZER PRESSURE FOR MSIV CLOSURE (LOWER STEAM FLOW CASE) Pressure (psia) Time (s) Rev. 21

MPS-2 FSAR FIGURE 14.2.4-4 ISOLATED STEAM GENERATOR PRESSURE AT BOTTOM OF BOILER REGION FOR MSIV CLOSURE (LOWER STEAM FLOW CASE) Pressure (psia) Time (s) Rev. 21

MPS-2 FSAR FIGURE 14.2.4-5 OPEN MSIV STEAM GENERATOR STEAM DOME PRESSURE FOR MSIV CLOSURE (LOWER STEAM FLOW CASE) Pressure (psia) Time (s) Rev. 21

TABLE 14.2.7-1 AVAILABLE REACTOR PROTECTION FOR THE LOSS OF NORMAL FEEDWATER FLOW EVENT Reactor Operating Conditions Reactor Protection 1 Low Steam Generator Water Level Trip High Pressurizer Pressure Trip Thermal Margin/Low Pressure Trip Variable Overpower Trip 2 High Pressurizer Pressure Trip Variable Overpower Trip Low Steam Generator Water Level Trip 3 Variable Overpower Trip 4-6 No Analysis Required; Not a Credible Event 14.2-45 Rev. 35

TABLE 14.2.7-2 DISPOSITION OF EVENTS FOR THE LOSS OF NORMAL FEEDWATER FLOW EVENT Reactor Operating Conditions Disposition 1 Analyze to assess maximum pressurizer level swell and long term adequacy of AFW. Pressurization and DNB aspects bounded by Event 14.2.1. 2, 3 Bounded by the above, no analysis required. 4-6 No Analysis Required; Not a Credible Event 14.2-46 Rev. 35

ABLE 14.2.7-3 SEQUENCE OF EVENTS FOR MINIMUM STEAM GENERATOR NVENTORY CASE: ONE MOTOR-DRIVEN AFW PUMP FAILS TO START WITH OFFSITE POWER AND STEAM DUMPS Time (seconds) Event 0 Total loss of main feedwater 20 Pressurizer spray actuates 27.9 Reactor trip signal on low steam generator water level 28.8 Control rods begin to drop 29.9 Main turbine trip 32 Maximum pressurizer level, 73% 38 Peak steam generator pressure (993 psia) 48 AFW actuation signal on low-low Steam Generator water level 53 Charging flow initiated in response to pressurizer level program 58 Steam generator blowdown isolated 288 Train A motor-driven AFW pump starts 628 Steam-driven AFW pump starts 654 Minimum Steam generator liquid inventory occurs 1800 End of calculation 14.2-47 Rev. 35

ABLE 14.2.7-4 SEQUENCE OF EVENTS FOR MAXIMUM PRESSURIZER LEVEL ASE: LOSS OF OFFSITE POWER, ONE MOTOR-DRIVEN AFW PUMP FAILS TO START Time (seconds) Event 0 Total loss of main feedwater 34.1 Reactor trip signal on low steam generator water level 35.1 Control rods begin to drop; RCPs tripped 36.1 Main turbine trip 39 Pressurizer PORV cycles open/closed 41 AFW actuation signal on low-low steam generator water level signal 43 Maximum pressurizer level, 76% 46 Peak steam generator pressure (1055 psia) 51 steam generator blowdown isolated 73 Charging flow on (pressurizer level below program) 170 Charging flow off (pressurizer level at program setpoint) 281 Train A motor-driven AFW pump starts 635 Steam-driven AFW pump starts 730 Minimum steam generator liquid inventory occurs 757 Maximum post-trip RCS average temperature (571°F) 2400 End of calculation 14.2-48 Rev. 35

MNP~*2 FSAR 650 I I I

                                                 ,                  I                ,

I I I o---a I RCS Hot Leg Temp. I I I I I 0---0 625 RCS Cold Leg Temp.

                              - - - - - - - ~ - - - - - - - -:- - - - - - - - ~ - ~                      RCS Averag e Temp.

I I I I I I I I I I r.x.: I I I I I oil 600 Q)

                             ------ - ~I --- ----~I --- -- - --~- ---- --~ -------~- ----_.-

I I I I I I

              ~                                                                                       ,                   I I                  I                I                 I I                                                                         I Q)                                               ,                 I                 I                   I I                  I                 I I-. 575
                ~
                                ----- - ~-    I                  r
                                                   - -- - - - , - - - - - - - - r --- ----~I - -- --- -~ -- - - - -- -

I I I

                                              ,                  I                 I                 I                   ,
              ....   14 Q)
                         -------,-------,--------r-------,-------,-------

I 1 I I 1

            ....:l                        I                                              1               1

_______ J J1 ~ J___ __JI _ 12

            ..-0. .                       1                            I                 I               1                 1 1                           ,                  ,               1                 I
l I I I 1 I 0' 10 ------,-------,--- -----r------- ------,-------
.3 I , , . 1

_____ JI J, 1~___ __JI JI _

           '0        8 Q)                         ,                           I                 I                I                 ,

Vl I I I 1 1

0. I I I , 1
               <<3    6   ----      --,-------,--------                                       -------~-------,-------

I , I , 1 0 I , I

                                                                                         ~

U 4 ______ J__ _ JI J1 _ I , I 1 0 I ' I 1 CI) 1 I I 1 2 - - - - - - - , - - ........lIl...iOi. . -, - - - - - - - - r - - - - - - - ~ - - - - - - - - - - - - - I I I I I I I I I 1 0 0 400 800 1200 1600 2000 2400 Time (sec) FIGURE 14.2.7-5 STEAM GENERATOR COLLAPSED LIQUID LEVEL FOR MINIMUM STEAM GENERATOR INVENTORY CASE: OFFSITE POWER AVAILABLE, "B" MOTOR-DRIVEN AFW PUMP FAILS TO START APRIL 1999

MNF-- .....2 FSAR 650 I I I 1 I I I I I I I I I I I I I I I I I I I

                          - - - _ - - _ ..I _ __ - __ _   _ _ _____    ~  _______   J - - - __ - - _1 __ _ _ _ _ _ _

625 I I I I I I I I I I I I I I I I I I I I

            ~                                                                            I              I
,.l I I I tlb 600 Q)
                            -- - - - - ~I -- - ---    I
                                                              - --- - - - Ir- - --- - - ~I -- - - -- -~1 - -- - - - - -

I I I I I Cl

                          ,V_I                                                           I              I oJ r-. 575 ~ ------{~ -----~- -------~ -. -----~ -------J. --                                          --

I I I

            ....;j                        I               I                I            i               I I
                                              ----        I                I             I
             ~                            I                                              I              I Q)                                    -...,  I               ,..

I I I

0. 550 1 ------~ --------r-*---

I I E I I I Q) I I I I I Eo-< I I I I I I

                                                         ~

_ ___ _ __ JI __ ___ __ I_ _ ___ _ __L 525 ~ RCS Hol Leg Temp. I t t -- I I t 0---0 RCS Cold Leg Temp. I I I I t I A---A RCS Average Temp. I I I 500 o 400 BOO 1200 1600 2000 2400 Time (sec) FIGURE 14.2.7-6 REACTOR COOLANT SYSTEM LOOP TEMPERATURES FOR MAXIMUM PRESSURIZER LEVEL CASE: LOSS OF OFFSITE POWER , ONE MOTOR*DRIVEN AFW PUMP FAILS TO START APRIL 1999

MNPS-2 FSAR 1200 I  :  :  : ~ SOl 1150 ---- ---, ------ -,-------- r---- -- -,-- I , , I 0---0 SO 2 ____ ___ I ~ ~I IL ~I ~ _ 1100 I I t I , I , , I , I I , I ,

      ......... 1050 tU              ------,--- ----,--------r-------,-- -----,-------
       °tii                  ___ _ _ __ __ _ _  ~                    ~       J        _
0. 1000 I I I I Q) I I I I l-o I I I I I
         ;:l     950   -------, -------,--- -----r-------, -------, -------

I I I I I gJ __ __ __ _ I~ ~I IL ~I JI _ Q) l-o 900 I , I , I 0.. I I I I I I I I I I o 850 -- -----,----- --,--------r-------,---- ---,-------- en , I I , ,

                                   ~

_ _ _ __ __ I JI IL ~I JI _ 800 I , I , I I I I I , I I I I t 750 -------, -------,--------r-------,- ------,- ------ I I I , , I I I I I 700 0 400 800 1200 1800 2000 2400 Time (sec) FIGURE 14.2.7-7 STEAM GENERATOR DOME PRESSURE FOR MAXIMUM PRESSURIZER LEVEL CASE: LOSS OF OFFSITE POWER, ONE MOTOR-DRIVEN AFW PUMP FAILS TO START APRIL 1999

MNPS-2 FSAR 100 I I I I I I I I I I I

               - -----~-------~--- -----~-------f-------~-------

I I I 1 __ ~,r: \ J " " -_ _-+I---I"J--...J I

 ......                         I Q)                 I         I         I                       I a;    50   --      ~ -------~--------~-------l-------J-------
 ...:l                I         I         I       I               I I         I         I       I               I I         I         I       I               I I         I         I       1               I I         I         I       I               I I         I         I       I               I I         I         I       I               I
              -------i-------~--------~-------~-------~- ------

I I I I I I I I I I I I I I I I I I I I

:  : I~ CNTR~VAR-6201 o

o 400 BOO 1200 1600 2000 2400 Time (sec) FIGURE 14.2.7-8 PRESSURIZER LEVEL FOR MAXIMUM PRESSURIZER LEVEL CASE: LOSS OF OFFSITE POWER, ONE MOTOR-DRIVEN AFW PUMP FAILS TO START APRIL 1999

MNPS-2 FSAR 150000 I Q---{] SG 1 I I I I I 0---<) SG2

                      --            4               ~                  ~              ~----            -------

125000 I I I I I I I I I I I I

       ---6                          I               I                I               I I               I                I               I              I
       ~     100000   - - - - - - - ~ - - - - - - - ~ - - - - - - - - ~ - - - - - - - ~ - -- -- - - ~ -- - -- - -
       '-'"                          I               I                I               1              I I               I                I               I              I
         !/l                         I               I                I               I              I
         !/l                         I               I                 I              I              I ttl                         I               I                 I              I              I
E 75000 -------,-------,--------r----- --,-------,---- ---

I I I I I I I I I I I I I I I

s::l I I I I I 0<

3 50000 ------~-------~--------~-------~-------~------- I I I I I I I I I I o I I I I en JI JI Il _ I 25000 I I o o 400 aoo 1200 1600 2000 2400 Time (sec) FIGURE 14.2.7-9 STEAM GENERATOR LIQUID MASS INVENTORY FOR MAXIMUM PRESSURIZER LEVEL CASE: LOSS OF OFFSITE POWER, ONE MOTOR-DRIVEN AFW PUMP FAILS TO START APRIL 1999

MNPS-2 FSAR 20 I I t I I t t I 18 -------~-------,--------r-------~--

        .-.....                    I           I        I         I
         -+J          _______ JI              ~

I IL JI ~ _

        ........ 16                I           I          ,       I           I
         ......                    I           ,        I         I           I QJ                      I           I         I        I           I
           >     1.   -------~-------,--------r-------~-------,-------

I I I I I

         .3            ______ JI              ~

I IL JI ~ I _

         '"d     12                ,           I        I         ,           I
         *S                        I           I        I         ,           I
                                   ,           I        I         I           ,
         ..0'
            .. 10      -----~-------,--------r-------,-------,----~~~~
         ~                         I           I         I        I           I

_ _ _ _ _ _ JI ~ I , L J, _

         't:l     8                I           I        I         I  ---.....-,

QJ Vl I I I I I p., I I I Cd 6 I I I 0 ~ I

                                         ....-=-         ,L      JI          ~

I _ o , I , I t.'J

  • I I

I I I I I I en 2 -------~-------,--------r-------,-------,------- I I I I t I I t I I 0 0 400 800 1200 1600 2000 2400 Time (sec) FIGURE 14.2.7-10 STEAM GENERATOR COLLAPSED LIQUID LEVEL FOR MAXIMUM PRESSURIZER LEVEL CASE: LOSS OF OFFSITE POWER, ONE MOTOR-DRIVEN AFW PUMP FAILS TO START APRIL 1999

3.1 LOSS OF FORCED REACTOR COOLANT FLOW 3.1.1 Event Initiator loss of forced reactor coolant flow in the primary system may result from a mechanical or trical failure in a main reactor coolant pump (RCP) or in the power supply to these pumps. ced coolant flow may be completely or partially lost. The limiting event initiator is that which lts in the trip of all four RCPs.

 .1.2 Event Description Loss of Forced Reactor Coolant Flow transient is initiated by a loss of the electrical power plied to or a mechanical failure in a reactor coolant system (RCS) pump. These failures may lt in a complete or partial loss of forced coolant flow. The immediate result of the loss of ed coolant flow is an increase in the coolant temperature as it flows through the reactor core.

r to reactor trip, the combination of decreased flow and increased temperature poses a llenge to Departure From Nucleate Boiling (DNB) limits.

 .1.3 Reactor Protection ctor protection is provided by the following reactor trips:
1. Low reactor coolant flow;
2. Thermal margin/low pressure (TM/LP); and
3. High pressurizer pressure trip.

ctor protection for the Loss of Forced Reactor Coolant Flow event is summarized in le 14.3.1-1.

 .1.4 Disposition and Justification power sources for the main RCPs are the most likely initiator for a loss of flow event olving more than one pump. A mechanical or electrical fault in one of the pumps will only lt in a single pump loss of forced coolant flow transient. The normal power supplies for the ps are from two buses which receive power from the main generator. Two pumps, in opposite ps, are powered from each bus. If there is a generator trip, the pumps are automatically sferred to a bus supplied from the external power lines. A generator trip with the failure of this sfer could result in a loss of power to all four pumps.

he case of four pump operation, two situations must be considered: two pump coastdown and a l loss of forced coolant flow. Considering first the total loss of flow cases, the consequences of postulated event are bounded by rated power operation. 14.3-1 Rev. 35

the two pump loss of flow cases, the magnitude of the coastdown is less severe than the four p coastdown, and the consequences of this event are bounded by the four pump loss of flow nt. For the two pump flow coastdown cases, there is always some degree of forced reactor lant flow. These events are, therefore, not as challenging as the four pump coastdown events. omparison of the governing parameters indicates that these events are bounded by the four p loss of flow event from full rated power conditions. ummary, the four pump loss of flow event is the bounding event for the 14.3.1 events in all des of operation. The only active system challenged is the reactor protection system (RPS) ch is redundant and single failure proof. disposition of events for the Loss of Forced Reactor Coolant Flow event is summarized in le 14.3.1-2.

 .1.5 Definition of Events Analyzed s event is analyzed from full-power initial conditions. The core thermal margins are imized at full power conditions resulting in this being the bounding mode of operation for this nt. One case is analyzed for this event to assess the challenge to the DNB Specified eptable Fuel Design Limit (SAFDL).

loss of coolant flow immediately results in a loss of system heat rejection capacity. This ses the primary system coolant temperature to increase. The objective of selecting input and ing is to minimize Departure From Nucleate Boiling Ratio (DNBR). The event analysis is, efore, biased to minimize pressure which minimizes DNBR. The steam bypass and the ospheric dump valves are both assumed not to operate, which again most challenges the DNB FDL. 3.1.6 Analysis Results transient is initiated by tripping all four primary coolant pumps. As the pumps coast down, core flow is reduced, causing a reactor scram on low flow. No credit was taken for the RCP er speed trip. As the flow coasts down, primary temperatures increase. This increase in perature causes a subsequent power rise due to moderator reactivity feedback. The primary llenge to DNB is from the decreasing flow rate and resulting increase in coolant temperatures. deterministic Minimum Departure from Nucleate Boiling Ratio (MDNBR) may violate the design limit for this event. Because of this, the DNBR consequences of this event were luated using AREVA statistical setpoint methodology (Reference 14.3-1). The event imum Departure From Nucleate Boiling Ratio (MDNBR) was shown to be greater than mal margin limits. This event does not challenge the FCMLHR limit. Therefore, LHR is not luated 14.3-2 Rev. 35

 .1.7 Conclusion statistical setpoint analysis demonstrates that the MDNBR limit is not penetrated by the Loss orced Reactor Coolant Flow event. Maximum peak pellet LHR for this event is below the MLHR limit.
 .2 FLOW CONTROLLER MALFUNCTION re are no flow control devices on the primary RCS of Millstone Unit 2. This event is therefore credible and need not be analyzed.
 .3 REACTOR COOLANT PUMP ROTOR SEIZURE
 .3.1 Event Initiator s event is initiated by an instantaneous seizure of an RCP rotor.
 .3.2 Event Description RCP seizure causes an immediate reduction in RCS flow rate. As in the Loss of Forced lant Flow event (Event 14.3.1), the impact of losing an RCS pump is a decrease in the active rate in the reactor core and an increase in core temperatures. Prior to reactor trip, the bination of decreased flow and increased temperature poses a challenge to DNB limits. A surization of the primary system will also occur due to the heatup of the primary coolant ch causes a rapid insurge into the pressurizer.
 .3.3 Reactor Protection ctor protection for the RCP rotor seizure event is provided by the low reactor coolant flow

, TM/LP trip, and the high pressurizer pressure trip. ctor protection for the Reactor Coolant Pump Rotor Seizure event is summarized in le 14.3.3-1.

 .3.4 Disposition and Justification s event is a concern for only rated power and power operating conditions because for other tor operating conditions there is sufficient thermal margin so there will not be a challenge to fuel design limits. The core heat flux to flow ratio is an excellent indicator of the potential B challenge for a loss of flow event. The highest ratios for this event are predicted to occur ng the first few seconds of the transient from full-power rated operating conditions. The sequences of this event will therefore be bounded by a pump rotor seizure event initiated from

-power rated conditions. There is no single failure considered which could worsen the results. 14.3-3 Rev. 35

3.3.5 Definition of Events Analyzed case is analyzed for this event to maximize the challenge to the DNB limit. The bounding rating mode for this event is full-power initial conditions. 3.3.6 Analysis Results locked rotor analysis assumes the locked pump loss coefficient given by the homologous p curves at zero pump speed. The sequence of events is given in Table 14.3.3-3 and the onses of key system variables are given in Figures 14.3.3-1 to 14.3.3-7 for the deterministic

 . This event does not challenge the FCMLHR limit. Therefore, LHR is not evaluated.

DNBR consequences of the Loss of Flow event (14.3.1) were evaluated using AREVA istical setpoint methodology (Reference 14.3-1), and the MDNBR was shown to be greater thermal design limits. Because the Rotor Seizure event (14.3.3) is inherently similar to and a deterministic MDNBR greater than the Loss of Flow event, penetration of thermal design ts is precluded for the Rotor Seizure event, as well.

 .3.7 Conclusion MDNBR limits are not exceeded by this event. The peak LHR is less than the FCMLHR t.
 .4 REACTOR COOLANT PUMP SHAFT BREAK s event is not in the current licensing basis for Millstone Unit 2 and is, therefore, not analyzed.
 .5 REFERENCES
 -1    Statistical Setpoint/Transient Methodology for Combustion Engineering Type Reactors, EMF-1961(P)(A), Revision 0, Siemens Power Corporation, July 2000.

14.3-4 Rev. 35

BLE 14.3.1-1 AVAILABLE REACTOR PROTECTION FOR THE LOSS OF FORCED REACTOR COOLANT FLOW EVENT eactor Operational Mode Reactor Protection 4 pump operation) Low Reactor Coolant Flow Trip Thermal Margin/Low Pressure Trip High Pressurizer Pressure Trip 4 pump operation) High Pressurizer Pressure Trip Technical Specification requirements on number of operating pumps (less than 4 pump High Pressurizer Pressure Trip eration) Technical Specification requirements on number of operating pumps 14.3-5 Rev. 35

BLE 14.3.1-2 DISPOSITION OF EVENTS FOR THE LOSS OF FORCED REACTOR COOLANT FLOW EVENT Reactor Operational Mode Disposition 1 Analyze 2-6 Bounded by the above, no analysis required 14.3-6 Rev. 35

TABLE 14.3.1-3 EVENT

SUMMARY

FOR THE LOSS OF FORCED REACTOR COOLANT FLOW Event Time (seconds) tiate Four Pump Coastdown 0.00 tdown Flow Valve Open 0.00 actor Scram Signal 1.29 d Insertion Begins 2.44 ak Power 2.45 DNBR 3.7 ak Core Average Temperature 4.10 ak Pressurizer Pressure 5.53 am Line Safety Valves Open 7.35 14.3-7 Rev. 35

MNPS*2 FSAR 3000 2500

 .....>  2000
r L

[_. PL L Q) 1500

 ;J 0

0.... 1000 - SOD -------- ~-----

                                                                 - - - - - - -J 2      3      i       5            6     7    8 9      10 TLma (sac)

FIGURE 14 .3 .1-1 REACTOR POWER LEVEL FOR LOSS OF FORCED REACTOR COO LANT FLOW OCTOBER 1998

MNPS-2 FSAR X

J u,

t2S000 a (I)

r:

(I) OJ a (I) a: 50000 '---.L-_l--'------lL--L--.1------J.--l--1.--l---l-~--l._....L__.l_...L___l _.J o 2 3 ~ 5 6 7 6 9 TLme (sec) FIGURE 14.3.1-2 CORE AVERAGE HEAT FLUX FOR LOSS OF FORCED REACTOR COOLANT FLOW OCTOBER 1998

MNPS-2 t-~AR 620 , I I I _---to-- '.. "",.--____, I I

                                                                                                          ---    -~- ~~~
                                                                                                                          ~--

610 f- ---~--=- u, 600 f- - TAVGI en (Il

                                                                                                    -------          TILl 0     '-
                                                                                                    - _ . - reID
                                                                                                    **.***********   TCLI S90 -                                                                                                                        -

aJ L

J
~        580   l:-----------                                                                                                          -

L Ql 0.. E 570 f- - Ql 560 - 550 I I , t I , , I l I s~o a 2 3 .-i 5 6 7 8 9 10 TLme (sec) FIGURE 14.3.1-3 REACTOR COOLANT SYSTEM TEMPERATURES FOR LOSS OF FORCED REACTOR COOLANT FLOW OCTOBER 1998

MNPS-2 r~AR 0

 .J        2JSO (l)

CL a> L 230J

J 0)
0) c=

OJ C-O- 22SO L eIl N .J C-

J (l) 2200 tn OJ C-n, 2150 2 3 ~ 5 B 7 B 9 TLme (sec)

FIGURE 14.3 .1 -4 PRESSURIZER PRESSURE FOR LOSS OF FORCED REACTOR COOLANT FLOW OCTOBER 1998

MNPS-2 FSAR I / , t-----------.n.I~J~~--._.

                                                             --._-._._-._.~

0

           -1     f-(J')

L 0 0 0 I-

           -3
I)
 ....>     -~    f-

.j .j

....>     -s    I-U 0

m 0::: -6 I-

         -7    I-
         -0    f-I
         -9 0             2 FIGURE 14.3.1-5 REACTIVITIES FOR LOSS OF FORCED REACTOR COOLANT FLOW OCTOBER 1998

MNPS-2 FSAR 35000 lJ a> HLPCR (f) I

<,      30000 E
..D
-J
l 0 25000

-J (I..- (f) U 0::: 20000 15000 10000 l..--.-.L._l--.L._L---L._.L---L._L--L-_.l--L-_..l.----'-_...l----1._J-.-L_-L.----ll--' o 2 3 ~ 5 6 7 8 9 10 TLme (sec) FIGURE 14.3.1-6 PRIMARY COOLANT FLOW RATE FOR LOSS OF FORCED REACTOR COOLANT FLOW OCTOBER 1998

MNPS-2 .-::>AR 1050 0

.J (I) 0..

1£XXJ II> L 9SO

J 0')

0') II> root L 900 CL II> E 0 eso 0 (!, en Em 750 700 o i 5 6 7 o 9 10 TLme (sec) FIGURE 14.3.1-7 SECONDARY PRESSURE FOR LOSS OF FORCED REACTOR COOLANT FLOW OCTOBER 1998

TABLE 14.3.3-1 AVAILABLE REACTOR PROTECTION FOR THE REACTOR COOLANT PUMP ROTOR SEIZURE EVENT Reactor Operational Mode Reactor Protection 1 Low Reactor Coolant Flow Trip Thermal Margin/Low Pressure Trip High Pressurizer Pressure Trip 2 High Pressurizer Pressure Trip Available Thermal Margin (1) 3-6 Available Thermal Margin (1) Defense In Depth 14.3-15 Rev. 35

BLE 14.3.3-2 DISPOSITION OF EVENTS FOR THE REACTOR COOLANT PUMP ROTOR SEIZURE EVENT Reactor Operational Mode Disposition 1 Analyze 2 Bounded by the above 3-6 No analysis required 14.3-16 Rev. 35

BLE 14.3.3-3 EVENT

SUMMARY

FOR THE REACTOR COOLANT PUMP ROTOR SEIZURE Event Time (seconds) actor Coolant Pump Rotor Seizes 0.00 actor Scram Signal 0.08 d Insertion Begins 1.23 ak Power 1.23 DNBR 1.7 ak Core Average Temperature 1.90 ak Pressurizer Pressure 3.68 am Line Safety Valves Open 5.30 ak Steam Dome Pressure 6.35 14.3-17 Rev. 35

MNr . "l FSAR -.) 2000 3:: 1: L PL CD 1500

J 0

G.- 1000 1 2 J i 5 6 7 B 9 TLme (sec) FIGURE 14.3.3-1 REACTOR POWER LEVEL FOR REACTOR COOLANT PUMP ROTOR SEIZURE APRIL 1993

MNP' FSAR 200000 N

~

4-I 175000 L ..c: "- :::J

~

m 150000 x

J LL 125(11)0 0

CD I CD 100000 m 0 L CD a: 75000 50000 '-----L._.L--L._...L-.--"-_-'---l_-L------'-_---L._..L-..--L_~___I_ __.l__L.___L___' o 5 10 TLme (sec) FIGURE 14.3.3-2 CORE AVERAGE HEAT FLUX FOR REACTOR COOLANT PUMP ROTOR SEIZURE APRIL 1993

MNP~ FSAR TAVGl

                                 ., ... -- ~' ~~--------- ----'- -~ .......                                                                nt..l 610                     , "                                            ' .... ~ ~                                                     rcro
                       .-~' -                                                          .. " ....                                           T~I
                                                                                                          ..... ~
                                                                                                                     ~

L.. 600 ~"- m ~- Q) o ----- -- -~ 590 (I) L

J
+>   580 o

L (I)

 ~   570 Q) f--t 560
               ._.--.                        :::::::::::::::::::==::==::::::::::=::::=:

550 510 L---'--...L.-----L--L---'---L-.L----L-..L.--.J.--..£--'-----'--..L---L---l--.L---.L-~--:to o 2 3 ~ 5 6 7 B 9 TLme (sec) FIGURE 14.3.3-3 REACTOR COOLANT SYSTEM TEMPERATURES FOR REACTOR COOLANT PUMP ROTOR SEIZURE APRIL 1993

MNPS" FSAR 0 .J 2350 C/) o, PPR CD L

J 2300 C/)

C/) ID C-o, 2250 L CD N .J L

J C/) 2200 C/)

CD L n, 2150 2 3 i 5 6 7 8 9 10 TLme (sec) FIGURE 14.3.3-4 PRESSURIZER PRESSURE FOR REACTOR COOLANT PUMP ROTOR SEIZURE APRIL 1993

MNP ') FSAR 9 . I I I

                                                                    .        I 7 I-                                                                                   -

,... (I( (I) 5 I-

                                                                ------- DKOor L                                                              - - - - OKHOO 0

.J .J 3 I- - a 0 - J) 1 - -------- ---------- .J J-------_~--"'. . . =.::::==--:~.:-:.:.:::=..::: :.:.::.:.:=-=~=-=~~.--- --.-- .J

 > -1   I-                                                                                    -

o 0 (I) -3 ... - 0:::

     -5  I-
     -7  t-                                                                                    -
                .      I              I                       I                I
     -9 0              2                                      6                8                10 TLme (sec)

FIGURE 14.3.3-5 REACTIVITIES FOR REACTOR COOLANT PUMP ROTOR SEIZURE APRIL 1993

MNpf'? FSAR 35000 3iOOO 33000 0 CD ..a..PCR (I) I <, 32000 E ..0 ..J 31000

J 0

..J u, 30000 Ul U a::: 29000 28000 27000 26000 0 1 2 3 i 5 6 7 8 9 10 TLme (sec) FIGURE 14.3.3-6 PRIMARY COOLANT FLOW RATE FOR REACTOR COOLANT PUMP ROTOR SEIZURE APRIL 1993

MNPS-2 r~AR

                                                              . ~. _   . _   . -. - .

1050 ./

                                                   /
                                              /"            _ . _ . _ - POOl 1000                                 -:                            P00 2
                                      /

950 Q) /

 \...
J /

(f) (f)

 ~     900                    /

Q.

                           /
~      850               /

Cl ~....----- / o (J) -' 800 750 700 o 2 J 456 7 8 9 to Time (s ec ) FIGURE 14.3.3 -7 SECONDARY PRESSURE FOR REACTOR COOLANT PUMP ROTOR SEIZURE OCTOBER 1998

 .1 UNCONTROLLED CONTROL ROD/BANK WITHDRAWAL FROM A SUBCRITICAL OR LOW-POWER STARTUP CONDITION
 .1.1 Event Initiator iated by the uncontrolled withdrawal of the control rod/banks in sequence, this event results in insertion of positive reactivity and consequently a power excursion. This event could be sed by a malfunction in the reactor control or rod control systems. The consequences of the k withdrawal from operating Modes 2-6 are considered in this event category; the sequences at rated power and power operating initial conditions are considered in Event 4.2.

control rods are wired together into preselected bank configurations. These circuits prevent control rods from being withdrawn in other than their respective banks. Power is supplied to banks in such a way that no more than two banks can be withdrawn at the same time and in r proper withdrawal sequence. 4.1.2 Event Description s event is initiated by the uncontrolled withdrawal of control rod banks in sequence. This hdrawal adds positive reactivity to the core which leads to a power excursion. As the control ks are withdrawn, the positive reactivity insertion causes a significant core power increase as reactor approaches prompt criticality. Low coolant flow rates in the core, combined with a d surge of power and pronounced radial and axial power peaking, represent a challenge both he Departure from Nucleate Boiling (DNB) and fuel centerline melt acceptance criteria. The B acceptance criteria may also be challenged by the reduced Reactor Coolant System (RCS) sure for a Mode 3 initial condition. Doppler reactivity feedback from the negative Doppler fficient limit the power excursion until the transient can be terminated by the Reactor tection System (RPS).

 .1.3 Reactor Protection power transient is eventually terminated (as well as the control rod withdrawal) by the RPS ne of the following signals:
1. Variable overpower trip or
2. High pressurizer pressure trip.

ctor protection for the Uncontrolled Control Rod Bank Withdrawal from a Subcritical or Low er Startup Condition event is summarized in Table 14.4.1-1. 14.4-1 Rev. 35

Technical Specifications (Reference 14.4-1) for Millstone Unit 2 require that the control rod es be deenergized in Modes 4-6 whenever the RCS boron concentration is less than the eling requirement. A rod withdrawal from these modes is therefore not considered a credible nt. ing Mode 3 operations, the control rod drive mechanism may be energized provided the hnical Specification requirements that 4 RCPs are operating, the RCS temperature is greater 500°F, the RCS pressure is greater than 2000 psia, and the variable overpower trip is rable are met. Consequently, a rod withdrawal from these operating conditions within this de is possible. The greatest power rise for this event is obtained when it is initiated from the est power. refore, the event initiated from a Mode 3 condition at 2000 psia will bound all other low-er or subcritical cases. The only active system challenged in this event is the RPS, which is undant and single failure proof. disposition of events for the Uncontrolled Control Rod/Bank Withdrawal from a Subcritical ow-Power Startup Condition event is summarized in Table 14.4.1-2.

 .1.5 Definition of Events Analyzed discussed in Section 14.4.1.4, the event was analyzed from a Mode 3 initial condition at 2000
. These conditions will bound all other low power or subcritical cases. Axial and radial power ributions for various control rod configurations, ranging from the critical configurations to all trol rods fully withdrawn, were considered. Conservative system conditions were used in the lysis to bound potential initial conditions for the transient. Four coolant pumps were sidered to be in operation, consistent with the Technical Specification minimum for Mode 3 ration with Control Element Assembly (CEA) drives energized and shutdown requirements

. Credit is taken only for the variable overpower trip and high pressurizer pressure trips; other s are not modeled. The variable overpower trip setpoint was conservatively set to the zero-er initial condition of 14.6% of rated thermal power plus 12.62% to account for uncertainties he variable overpower trip minimum setpoint, power calibration, calorimetric power, and for er decalibration allowance. The variable overpower trip delay was set to a conservatively e value of 0.7 seconds. Biased beginning-of-cycle (BOC) kinetics values were assumed to imize the reactor power during the transient.

 .1.6 Analysis Results event is initiated from Mode 3 with both shutdown CEA banks fully withdrawn and all ulating CEA banks fully inserted. The resultant power excursion results in a fuel temperature ease and negative Doppler reactivity feedback which limits the peak power. The transient is inated when control rods are inserted upon a variable high power trip. The responses of key em parameters are plotted in Figures 14.4.1-1 to 14.4.1-5. The sequence of events is given in le 14.4.1-3.

14.4-2 Rev. 35

ulated MDNBR was well above the 95/95 acceptance criterion for the HTP DNB correlation

t. This ensures that, with 95% probability and 95% confidence, DNB is not expected to occur.

peak fuel centerline temperature is calculated to be well below the melting point. Thus, no failures are predicted to occur. 4.1.7 Conclusion fuel failures are predicted for this event. Therefore, the event meets the applicable acceptance eria. 4.2 UNCONTROLLED CONTROL ROD/BANK WITHDRAWAL AT POWER 4.2.1 Event Initiator s event is initiated by an uncontrolled control rod/bank withdrawal from power operating ditions. 4.2.2 Event Description s event is initiated by an uncontrolled withdrawal of a control bank, causing a positive tivity addition to the reactor core. This positive reactivity addition causes an increase in the power and primary coolant system temperatures. Due to the increasing power and peratures, the DNB limits are challenged.

 .2.3 Reactor Protection challenge to the fuel design limits is terminated by the automatic action of the RPS which inates the bank withdrawal and inserts negative reactivity to terminate the power transient.

automatic action of the RPS is initiated as the result of one of the following signals:

1. Variable overpower trip;
2. Local power density (LPD) trip;
3. Thermal margin/low pressure (TM/LP) trip; or
4. High pressurizer pressure trip.

ctor protection for the Uncontrolled Control Rod/Bank Withdrawal at Power event is marized in Table 14.4.2-1. 14.4-3 Rev. 35

s event is designed to address the safety challenge posed by an uncontrolled control rod/bank hdrawal transient from power conditions. This event addresses all the power operating ditions and the rated power operating conditions. It is performed to test the adequacy of the able overpower and TM/LP trip setpoints in mitigating the challenge to the Specified eptable Fuel Design Limits (SAFDL). od withdrawal initiated from lower powers will provide less of a challenge to the SAFDLs due ncreased initial thermal margin, a lesser amount of setpoint overshoot, and a decreased able overpower trip setpoint resulting in a greater thermal margin at trip. The event initiated m full power will then bound those initiated from lower power conditions. This event will efore be analyzed at full power for conditions ranging from BOC to end of cycle (EOC) for a ctrum of reactivity insertion rates. The only active system challenged by this event is the RPS, ch is redundant and single failure proof. disposition of events for the Uncontrolled Control Rod/Bank Withdrawal at Power event is marized in Table 14.4.2-2. 4.2.5 Definition of Events Analyzed analysis evaluates the consequences of an uncontrolled control rod bank withdrawal from d power. A spectrum of reactivity insertion rates were evaluated in order to bound events ging from a slow dilution of the primary system boron concentration to the fastest allowed trol bank withdrawals. Specifically, the analysis encompasses reactivity insertion rates from 4

  -6 to 4 x 10 -4 delta rho/sec.
 .2.6 Analysis Results uncontrolled control bank withdrawal transients were analyzed for full-power conditions 0% of rated). The limiting uncontrolled control rod bank withdrawal at 100% power occurred h EOC kinetics at an insertion rate of 4 x 10 -6 delta rho/sec. The MDNBR was calculated to be ve the CHF correlation limit. This transient tripped on a TM/LP signal. The maximum peak et linear heat rate (LHR) occurs in a 100% power case which uses BOC kinetics. The variable h power and LPD trips ensure the maximum peak pellet LHR is less than the FCMLHR limit.

sequence of events for the Uncontrolled Bank Withdrawal transient is given in Table 14.4.2-he transient response of key system variables are given in Figures 14.4.2-1 to 14.4.2-6. 4.2.7 Conclusion ctivity insertion transient calculations demonstrate that the DNBR limit will not be penetrated ng any credible reactivity insertion transient at full power. The maximum peak pellet linear t generation rate for this event is less than the FCMLHR limit. Applicable acceptance criteria therefore met, and the adequate functioning of the TM/LP trip is demonstrated. 14.4-4 Rev. 35

control rod misoperation event encompasses a number of transients resulting from different nt initiators. The specific events addressed under this event category include the following:

1. Dropped control rod or control rod bank;
2. Dropped part-length control rod;
3. Malpositioning of the part-length control rod group;
4. Statically misaligned control rod/control rod bank;
5. Single control rod withdrawal;
6. Reactivity control device removal error during refueling; and
7. Variations in reactivity load to be compensated by burnup or on-line refueling.
 .3.1 Dropped Control Rod/Bank
 .3.1.1 Event Initiator Dropped Control Rod/Bank event is initiated by a de-energized control rod drive mechanism y a malfunction associated with a control rod bank.
 .3.1.2 Event Description ropped Control Rod/Bank event is initiated by a deenergized control element drive hanism (CEDM) or another failure in the control rod system. The reactor power initially ps in response to the insertion of negative reactivity. However, the local peaking increases due he local effect on the power distribution. The reactor core will attempt to return to a new ilibrium at the original power level as a result of moderator and Doppler reactivity feedback.

ause of the increased peaking and the potential return to the initial power level, the Dropped trol Rod/Bank event poses a challenge to the DNB margin.

 .3.1.3 Reactor Protection e amount of reactivity is large enough to cause a significant reduction in core power, a reactor could be generated by the variable overpower trip prior to returning to full power. Reactor ection for the Dropped Control Rod/Bank event is summarized in Table 14.4.3.1-1.

14.4-5 Rev. 35

ce the control rod drive mechanisms are deenergized in Modes 4-6 and reactor power is ted to zero percent with keff < 0.99 in Mode 3, there will be no consequences of this event for e modes. mately, the consequences of this event are a return to power at elevated peaking conditions. s, the worst case is obtained when the combination of final power level, increased peaking, core inlet temperature are maximized. This occurs for cases initiated from full-power. The -power case thus bounds all other power operation conditions. a single dropped control rod, a reactor trip is not expected. Thus, a DNB evaluation assuming turn to full-power at maximum dropped rod peaking will be performed to demonstrate that the FDLs are not violated. return to power for a dropped control bank is limited by the capacity of the turbine control

e. In response to a decrease in the secondary side steam flow resulting from a drop in core er, the turbine valve will throttle open in an attempt to maintain a constant load demand. If the tivity worth of the dropped control bank is sufficiently large, the turbine valve will not have ugh excess capacity for the reactor to return to full power. The lower power level could be et by the higher peaking factor associated with a dropped control bank. It should be noted that operator will have multiple indications that a dropped rod/bank has occurred via CEA iation alarms and rod bottoming signals. The only active system challenged in this event is the S, which is redundant and single failure proof.

disposition of events for the Control Rod Misoperation (Dropped Control Rod/Bank) event is marized in Table 14.4.3.1-2.

 .3.1.5 Definition of Events Analyzed analysis evaluates the consequences of this event from rated power conditions. A spectrum of pped control rod/bank cases were analyzed at full power with increased radial peaking and hnical Specifications minimum primary coolant flow. Radial peaking augmentation factors for Dropped Control Rod/Bank event are calculated at full power for different exposure ditions. Bounding radial peaking augmentation factors were used in the analysis. In addition, nding values of control rod and bank worth were used.
 .3.1.6 Analysis Results sequence of events for the limiting dropped control rod/bank case is given in Table 14.4.3.1-he transient response or key system parameters are given in Figures 14.4.3.1-1 to 14.4.3.1-4.

limiting case, both from the standpoint of MDNBR and peak LHR, was with a 1079.8 pcm pped control bank. n transient initiation, the core power decreased in response to the negative reactivity insertion lting from the dropped control bank. The primary coolant and fuel temperatures decreased in 14.4-6 Rev. 35

lted in the VHP trip being reset to a power level of 89.3% of RTP. The measured DT power eeded the VHP trip setpoint at 88.3 seconds. Throughout the transient, the core inlet mass flow increased and the primary-side pressure decreased due to changes in the moderator density lting from the primary-side cooldown. peak pellet LHR is calculated to be less than the FCMLHR limit. The minimum DNBR for event is bounded by the minimum DNBR of the Section 14.3.1 loss of forced reactor coolant event, which is greater than thermal margin limits.

 .3.1.7 Conclusion of the cases analyzed were above the HTP 95/95 mixed core DNB safety limit and below the k LHGR limit. Therefore, no fuel failure is predicted to occur. Applicable acceptance criteria the Dropped Control Rod/Bank event are therefore met for Millstone Unit 2.
 .3.2 Dropped Part-Length Control Rod part-length control rods have been removed from the Millstone Unit 2 core. Therefore, this nt is not applicable.
 .3.3 Malpositioning of the Part-Length Control Rod Group part-length control rods have been removed from the Millstone Unit 2 core. Therefore, this nt is not applicable.
 .3.4 Statically Misaligned Control Rod/Bank s event is not in the current licensing basis for Millstone Unit 2 and therefore is not analyzed.
 .3.5 Single Control Rod Withdrawal
 .3.5.1 Event Initiator s event is initiated by the inadvertent withdrawal of a single CEA from the core. No single trical or mechanical failure in the Rod Control System could cause the accidental withdrawal single CEA from the inserted CEA bank during full power operation. Procedures are ilable to permit the operator to withdraw a single CEA in the control bank since this feature is essary in order to retrieve an assembly should one be accidentally dropped. The event can ur only as the result of multiple wiring failures or multiple operator actions in disregard of ilable event indication.

he extremely unlikely event of simultaneous electrical failures which could result in single A withdrawal, the rod position indicators and deviation alarms would indicate the relative itions of the assemblies in the bank. Withdrawal of a single CEA by operator action, whether 14.4-7 Rev. 35

 .3.5.2 Event Description withdrawal of a single full-length CEA is initiated by the inadvertent withdrawal of a single trol rod. The ensuing reactivity insertion causes core power to increase. In the event that the ondary steam dump control system does not respond to the increased power production, ondary system temperature and pressure will increase, causing a corresponding increase in ary coolant temperature. This increase in primary coolant temperature occurs slowly enough the pressurizer pressure control system, if available, is capable of suppressing the primary sure increase. The degradation of coolant conditions coupled with the power increase is ntially the same as expected for slow CEA bank withdrawals at power and may approach B conditions in the hot channel.

single CEA withdrawal is distinguished from the withdrawal of a CEA bank by a severe al power redistribution. High radial power peaking is localized in the region of the single hdrawn CEA and may, in severe cases, surpass the design limits. Thus, assemblies in the ediate vicinity of the withdrawn CEA may experience boiling transition for a short time od. Some fuel damage might occur.

 .3.5.3 Reactor Protection challenge to the fuel design limits is terminated by the automatic action of the RPS which inates the CEA withdrawal and inserts negative reactivity to terminate the power transient.

automatic action of the RPS is initiated as the result of one of the following signals:

1. Variable overpower trip;
2. LPD trip;
3. TM/LP trip; or
4. High pressurizer pressure trip.

ctor protection for the Single Control Rod Withdrawal event is summarized in Table 14.4.3.5-

 .3.5.4 Disposition and Justification overall system response to the withdrawal of a single CEA will be identical to the response to ow withdrawal of a CEA bank. The only difference will be that the core will experience lized peaking in the vicinity of the withdrawn CEA that is not present if an entire bank is hdrawn. Therefore, the disposition of the single CEA withdrawal will be identical to that of the A bank withdrawal.

14.4-8 Rev. 35

disposition of events for the Single Control Rod Withdrawal event is summarized in le 14.4.3.5-2.

 .3.5.5 Definition of Events Analyzed s event was analyzed at rated power conditions. Radial peaking augmentation factors to ount for localized peaking redistribution were utilized in the assessment of the challenge to NBR limits.
 .3.5.6 Analysis Results ial peaking augmentation factors for single control rod withdrawal events are calculated at power for different exposure conditions. Bounding radial peaking augmentation factors were d in the analysis. In addition, bounding value of control rod worth was used.

deterministic MDNBR for the single rod withdrawal event is greater than the HTP correlation

t. The peak pellet LHR is calculated to be less than the FCMLHR limit. Therefore, no fuel ure is predicted to occur.
 .3.5.7 Conclusion maximum peak LHR for the single rod withdrawal event is such that fuel centerline melt is expected. In addition, the minimum DNBR is greater than the limit. Thus, no fuel failure is dicted to occur.
 .3.6 Reactivity Control Device Removal Error During Refueling lstone Unit 2 has no reactivity control devices which are used during refueling and could vertently be removed. Boron dilution during refueling is considered in Event 14.4.6.

refore, this event is not applicable.

 .3.7 Variations in Reactivity Load to be Compensated by Burnup or On-Line Refueling s event considered the anticipated variations in the reactivity load of the reactor, to be pensated by means of action such as buildup and burnup of xenon poisoning, fuel burnup, on-refueling, fuel followers, temperature moderator and void coefficients.

visions for xenon changes and fuel burnup are described in Chapter 3. On line refueling will be performed on Millstone Unit 2. The core design does not include fuel followers. The safety lyses are based upon the most adverse combination of temperature, moderator and void fficients. Therefore, this event has no significant consequences and is not analyzed. 14.4-9 Rev. 35

4.4.1 Event Initiator s event is initiated by the startup of an inactive reactor coolant pump (RCP). 4.4.2 Event Description h primary coolant loop is equipped with two single-suction centrifugal pumps, one per cold which are located between the steam generator outlet and the reactor vessel inlet nozzles. A reversing mechanism is provided to prevent reverse rotation of the pump rotor. This feature limits backflow through the pump under nonoperating conditions. Note that there is no kflow in the hot leg (or steam generator) associated with the side of the plant that has the tive RCP. The inadvertent actuation of an inactive pump would therefore lead to a decrease in derator temperature and, with a negative moderator coefficient, an increase in core reactivity h a potential increase in core power level. 4.4.3 Reactor Protection ctor protection for this event is afforded by Technical Specification requirements on shutdown gin and RCP operation. Reactor protection for the Startup of an Inactive Loop event is marized in Table 14.4.4-1.

 .4.4 Disposition and Justification s event is not credible in operating Modes 1 and 2 because Technical Specifications require all RCPs to be operating. It is not credible in Mode 6 due to administrative procedures requiring the pumps be prevented from starting.

hnical specification requirements on shutdown margin in Modes 3-5 are such that any tivity insertion due to an inactive loop start is not great enough to reach criticality. Thus, the sequences of this event in Modes 3-5 are minimal and no analysis is required. The disposition vents for the Startup of an Inactive Loop event is summarized in Table 14.4.4-2.

 .5 FLOW CONTROLLER MALFUNCTION lstone Unit 2 does not have any flow control devices on the primary reactor coolant loops so event is not credible and does not need to be analyzed.

14.4-10 Rev. 35

REACTOR COOLANT

 .6.1 Event Initiator ilution of the primary system boron concentration can occur as a result of adding primary de water into the RCS via the Chemical and Volume Control System (CVCS). The greatest tion rate occurs for operation of the CVCS charging pumps. The three available charging ps can inject water into the primary system at a maximum rate of 147 gpm. For Modes 4, 5, 6, only two charging pumps are to be operable for a maximum rate of 98 gpm.
 .6.2 Event Description oron dilution event can occur when demineralized water is added to the primary coolant em via the CVCS resulting in decreasing boron concentration in the primary system coolant.

s dilution of the primary system coolant boron concentration results in the addition of positive tivity to the core. This event can lead to an erosion of shutdown margin for subcritical initial ditions, or a slow power excursion for at-power conditions. A boron dilution at rated or power rating conditions behaves in a manner similar to a slow uncontrolled rod withdrawal transient ent 14.4.2).

 .6.3 Reactor Protection ctor protection for the boron dilution event during operating Modes 3-6 is provided by hnical Specification shutdown margin requirements, administrative procedures, and sufficient e for the operator to take the appropriate action in the unlikely event that a boron dilution uld occur. Reactor protection for the reactor critical, power operation, and rated power rating conditions is provided by various trips and operator response time. Reactor protection the CVCS Malfunction that Results in a Decrease in the Boron Concentration in the Reactor lant event is summarized in Table 14.4.6-1.

4.6.4 Disposition and Justification boron dilutions in reactor Modes 1-6, the challenge to the SAFDLs is very similar to that of w control rod withdrawals and can be bounded by the consequences of control rod withdrawal nts as analyzed for events 14.4.2 and 14.4.1. A spectrum of control rod withdrawal reactivity ition rates is considered for Events 14.4.2 and 14.4.1, so the range of reactivity addition rates be established to encompass the predicted reactivity addition rates for boron dilution events odes 1-6. re must be 15 minutes (modes 3, 4, and 5) or 30 minutes (mode 6) from the onset of the tion prior to a complete erosion of shutdown margin. The disposition of events for the CVCS function that Results in a Decrease in the Boron Concentration in the Reactor Coolant event is marized in Table 14.4.6-2. 14.4-11 Rev. 35

boron dilution analysis evaluates the time to criticality caused by the dilution of the primary em boron and the subsequent loss of shutdown margin. This analysis determines the shutdown ling system flow rate needed to meet the time criteria for Refueling (Mode 6), Cold Shutdown de 5), and Hot Shutdown (Mode 4). The systems that would be involved in the boron dilution nt, depending upon the mode of operation are the RCS, the shutdown cooling system and the CS. major differences between the operating modes are the system parameters which affect the at which boron dilution occurs and the boron mixing model used once the demineralized er is injected into the lower plenum of the reactor vessel. Parameters such as charging pump acity and primary system water volume affect the dilution rate. The mixing model used ends on whether the RCPs are operating and whether the shutdown cooling system discharge both cold legs when the dilution is postulated to occur. the six modes of operation, two mixing models are used: (1) instantaneous mixing and (2) tion front. In the instantaneous mixing mode, the diluting water is assumed to uniformly mix h the entire RCS volume immediately upon injection into the primary system. Instantaneous ing is assumed to occur if one or more RCPs are in operation. ilution front model is used to simulate operation of the shutdown cooling system when the n RCPs are not running. It is conservatively assumed that the diluted water from the shutdown ling system discharged to the lower plenum will not immediately mix with the entire reactor lant due to the relatively low flow rate. Rather, the boron dilution occurs locally at the rging/shutdown cooling mixing location. The diluted mixture then flows through the RCS em until it reaches the mixing location where further dilution occurs. Thus, the RCS boron centration can be viewed as a series of dilution fronts traveling through the RCS. metric and asymmetric variations of the dilution front model are considered. In the metric flow variation, diluting water is assumed to be injected into two or more cold legs ted on opposite loops of the RCS. In the asymmetric flow variation, diluting water is assumed e injected into only one cold leg. The asymmetric flow variation is more limiting than the metric flow variation. In either case, the time to criticality is reduced if the shutdown cooling em flow is reduced. boron dilution analysis also included calculations to determine the maximum and minimum tivity insertion rates during Startup (Mode 2) and Full Power Operation (Mode 1). These es are used to confirm that the reactivity insertion rates used for the uncontrolled rod hdrawal analysis remain bounding of the boron dilution event. 4.6.6 Analysis Results le 14.4.6-3 presents the minimum shutdown cooling flow for Modes 4 through 6 required to id complete erosion of shutdown margin within the required time. These results are based on asymmetric dilution front model. For Modes 4 to 6, the reactor vessel is filled to the mid-plane 14.4-12 Rev. 35

tivity shutdown margin of 3.6% was used for all operating modes except Mode 6. For Mode Keff of 0.95 was used. The results indicate that as long as the shutdown cooling system flow ains above the flowrates provided in Table 14.4.6-3, the operator response time criteria will be sfied. le 14.4.6-4 presents the results of the calculated time to loss of shutdown margin for Modes 1 ugh 5 based on the use of the instantaneous mixing assumption. For Modes 1 to 3, three rging pumps were assumed to be operable and only two charging pumps were considered to be rable for Modes 4 and 5. For Modes 1 and 2 where the reactor is initially critical, the results onstrate that there is at least 15 minutes until loss of shutdown margin. The results indicate the time to loss of shutdown margin criteria is satisfied for all cases in which an RCS pump is peration. 4.6.7 Conclusions results of the boron dilution analysis show that there is not a complete erosion of shutdown gin within 15 minutes for modes 1 through 5 and 30 minutes in mode 6. The operator can ate reboration to recover the shutdown margin. For all modes there is adequate time for the rator to manually terminate the source of dilution flow. 4.7 INADVERTENT LOADING AND OPERATION OF A FUEL ASSEMBLY IN AN IMPROPER POSITION s event is not in the current licensing basis for Millstone Unit 2 and therefore is not analyzed. 4.8 SPECTRUM OF CONTROL ROD EJECTION ACCIDENTS 4.8.1 Event Initiator s accident is initiated by a failure in the control rod drive pressure housing which could result he rapid ejection of a control rod. 4.8.2 Event Description s event is initiated by a failure in the CEDM pressure housing causing a rapid ejection of the cted control rod. This results in a rapid loss of negative reactivity causing a nuclear power sient. In addition to the power transient, the ejected rod results in a highly perturbed power ribution which, coupled with the power transient, could possibly lead to localized fuel age. Also, the rapid nuclear power excursion can result in a significant short-term heatup of coolant with a resultant RCS pressure increase, although on the long-term the RCS will ressurize due to the break in the reactor coolant pressure boundary. 14.4-13 Rev. 35

ctor protection for the Spectrum of Control Rod Ejection Accidents is summarized in le 14.4.8-1. Doppler feedback inherent in the fuel also limits the nuclear power excursion.

 .8.4 Disposition and Justification s event is not a concern in Modes 4-6 as all control rods are required to be fully inserted per hnical Specifications and no one CEA possesses enough reactivity worth to overcome the imum allowed shutdown margin. The fuel energy content is maximized by starting from rated er initial conditions, so the consequences of this event are bounding for power operating al conditions. However, because of the complex interaction of the ejected rod worth and ted peaking factor (which are maximized at hot zero power (HZP) operating conditions, and pler feedback effects, it is difficult to bound the consequences of the event for either rated er or HZP operating conditions without additional analysis. Therefore, the consequences of event are analyzed for both rated power and HZP operating conditions. Separate evaluations performed for deposited enthalpy, DNBR, and RCS pressurization for each of these initial rating conditions.

ddition to the rod ejection, this event is characterized by a small-break loss-of-coolant dent (SBLOCA) as the failure of the pressure housing is assumed to result in a breach of the ary coolant pressure boundary. The short-term aspects of the event are dominated by the rod tion, while the long-term aspects are dominated by the SBLOCA. The limiting SBLOCA is luated in Event 14.6.5 and is typically a cold leg break. In the rod ejection, the break is more racteristic of a hot leg break and therefore will be bounded by the SBLOCA. Also in the rod tion, a much earlier reactor trip occurs, resulting in lower powers and temperatures than in nt 14.6.5. It is concluded that the long-term aspects of the rod ejection are bounded by those of nt 14.6.5 for small breaks. Thus, only the short-term rod ejection consequences need be luated. Note also that the limiting 14.6.5 event occurs for rated power operating conditions. disposition of events for the Spectrum of Control Rod Ejection Accidents is summarized in le 14.4.8-2. 4.8.5 Definition of Events Analyzed to the complex interaction of the ejected rod worth, ejected peaking factor and Doppler back effects, it is difficult to bound the consequences of the event for either rated power or P operating conditions without analysis. Therefore, each of these conditions were evaluated at h BOC and EOC for deposited enthalpy, DNBR and pressurization concerns. the evaluation of the DNBR and pressurization consequences, concurrent loss of offsite er is assumed. No credit is taken for the variable overpower trip in the analysis of the surization consequences of a control rod ejection. 14.4-14 Rev. 35

hot full-power (HFP) control rod ejection event was determined to deposit more energy into primary system than the event initiated from HZP. Therefore, in terms of the event acceptance eria, the HFP event poses a greater challenge than the HZP event. For this analysis, the event assumed to initiate from HFP at 102% of rated full power. assess the acceptability of the outcome of an HFP rod ejection event, two cases were mined. The first case determines the maximum pressurization potential of the primary system ng this event. The second case evaluates the MDNBR. For both the maximum pressurization minimum DNB case, BOC and EOC kinetics were employed to establish the respective ting cases. limiting minimum DNB case is calculated to occur for BOC kinetics. Core boundary ditions used to evaluate the DNBR conservatively account for depressurization due to the tulated breach in the CEDM housing. As a consequence of this event, less than 11.5% of the rods are calculated to fail due to penetration of DNBR limits. The responses of key system meters are shown in Figures 14.4.8-1 to 14.4.8-6. maximum pressurization case occurs for EOC kinetics. The peak RCS pressure at the bottom he reactor vessel remains below 110% of the pressure vessel design limit. The peak RCS sure is conservatively calculated to be 2748 psia. Key system parameters for the overpressure are plotted in Figures 14.4.8-7 to 14.4.8-12. sequence of events for the Control Rod Ejection transient is given in Tables 14.4.8-3 and

 .8-4.

deposited enthalpy portion of the rod ejection accident has been evaluated with the cedures developed in the Generic Rod Ejection Analysis (Reference 14.4-3). The ejected rod ths and hot pellet peaking factors were calculated using the PRISM code. No credit was taken the power flattening effects of Doppler or moderator feedback in the calculation of ejected rod ths or resultant peaking factors. The calculations performed used a full-core three-ensional PRISM model. The pellet energy deposition resulting from an ejected rod was servatively evaluated explicitly for BOC and EOC conditions. The rod ejection accident was nd to result in an energy deposition of less than the 280 cal/g limit as stated in Regulatory de 1.77. The significant parameters for the analyses, along with the results, are summarized in le 14.4.8-5. 4.8.7 Conclusion maximum RCS pressure does not exceed 110% of the design pressure. Less than 11.5% of the will experience fuel failure due to penetration of DNBR limits. Deposited enthalpy is less the limit of 280 cal/g. 14.4-15 Rev. 35

o release paths are considered independently for the rod ejection accident. Each release path is sidered independently as the only one available. The actual doses for the accident would be a posite of doses resulting from portions of the release going out the two different pathways. nario 1: A postulated CREA that releases the failed fuel activity into the RCS, which is released in its entirety, into containment via the ruptured control rod drive mechanism housing, is mixed in the free volume of the containment and then released at containment Technical Specification leak rate. Releases occur via the Millstone stack and the Unit 2 containment. nario 2: A postulated CREA that releases the failed fuel activity into the RCS which is then transmitted to the secondary side via steam generator tube leakage. The condenser is assumed to be unavailable due to a loss of offsite power. Releases occur from both steam generators via the MSSVs and the ADVs. nario 1: Containment Release sequent to a CREA, the following activity is assumed to be instantaneously released and to ogeneously mix in the free volume of containment.

  • 11.5% of the core gap activity and consideration of a peaking factor of 1.69 activity transport model assumes full enclosure building bypass until the enclosure building achieved drawdown to negative pressure. After the drawdown time, 1.4% of the activity that s from the containment completely bypasses the enclosure building for the first 24 hours of accident, after which it is reduced by 50% for the duration of the accident.

nario 2: Release via the MSSVs/ADVs owing a CREA, the activity available for release via the secondary system consists of:

  • Steam Generator tube leakage containing 11.5% of the core activity and consideration of a peaking factor of 1.69 assumed that offsite power is lost, therefore the main condenser is not available and releases via the MSSVs/ADVs from both steam generators. The releases from both steam generators tinue for about 16 hours until shutdown cooling commences. During this 16 hour period, tor coolant is assumed to leak into the steam generators at a maximum leak rate of 150 gpd steam generator. The noble gases that enter the steam generators are released directly to the ironment without holdup while the iodine activity is released from the steam generators in portion to the steaming rate and the partition factor.

14.4-16 Rev. 35

delines of 10 CFR 50.67 and Regulatory Guide 1.183. 4.9 SPECTRUM OF ROD DROP ACCIDENTS (BOILING WATER REACTOR) lstone Unit 2 is not a Boiling Water Reactor and as such this event is not applicable.

.10 REFERENCES
-1   Technical Specifications for Millstone Unit 2, Docket Number 50-336.
-2   Standard Review Plan for the Review of Safety Analysis Reports for Nuclear Power Plants, NUREG-0800, U.S. Nuclear Regulatory Commission, July 1981.
-3   A Generic Analysis of the Control Rod Ejection Transient for Pressurized Water Reactors, XN-NF-78-44(A), Exxon Nuclear Company, October 1983.

14.4-17 Rev. 35

BLE 14.4.1-1 AVAILABLE REACTOR PROTECTION FOR THE UNCONTROLLED NTROL ROD/BANK WITHDRAWAL FROM A SUBCRITICAL OR LOW-POWER STARTUP CONDITION EVENT eactor Operational Mode Reactor Protection 1 Considered as FSAR Event 14.4.2 2 Variable Overpower Trip High Pressurizer Pressure Trip Rod Withdrawal Prohibit on Variable Overpower Pretrip Alarm (1) 3 Variable Overpower Trip Rod Withdrawal Prohibit on Variable Overpower Pretrip Alarm (1) 4-6 Not a Credible Event; No Analysis Required Control grade. Provides defense in depth. Not credited in any safety analysis event. 14.4-18 Rev. 35

BLE 14.4.1-2 DISPOSITION OF EVENTS FOR THE UNCONTROLLED CONTROL OD/BANK WITHDRAWAL FROM A SUBCRITICAL OR LOW-POWER STARTUP CONDITION EVENT eactor Operational Mode Disposition 1 Considered as FSAR Event 14.4.2 2 Bounded by Mode 3 3 Analyze at 2,000 psia 4-6 Not a Credible Event; No Analysis Required 14.4-19 Rev. 35

TABLE 14.4.1-3 EVENT

SUMMARY

FOR THE UNCONTROLLED BANK WITHDRAWAL FROM LOW-POWER EVENT Event Time (seconds) nk Withdrawal Begins 0.00 riable Overpower Trip Setpoint Reached 37.6 ram Reactivity Insertion Begins 38.8 ak Nuclear Power (155.7% of rated) 39.0 ak Core Heat Flux (58.0% of rated) 40.0 ak Reactor Vessel Upper Plenum Coolant 41.2 mperature 14.4-20 Rev. 35

MPS-2 FSAR FIGURE 14.4.1-1 REACTOR POWER LEVEL FOR LOW POWER BANK WITHDRAWAL 5000 4000 Reactor Power (MW) 3000 2000 1000 0 0 10 20 30 40 50 Time (sec) Rev. 21

MPS-2 FSAR FIGURE 14.4.1-2 CORE AVERAGE HEAT FLUX FOR LOW POWER BANK WITHDRAWAL Core-Average Fuel Rod Heat Flux (BTU/hr-ft2) 120,000 100,000 80,000 60,000 40,000 20,000 0 0 10 20 30 40 50 Time (sec) Rev. 21

MPS-2 FSAR FIGURE 14.4.1-3 REACTOR COOLANT TEMPERATURES FOR LOW POWER BANK WITHDRAWAL 570 Vessel Upper Plenum Average Core Core Inlet Reactor Coo ant Temperature (F) 560 550 540 530 0 10 20 30 40 50 Time (sec) Rev. 21

MPS-2 FSAR FIGURE 14.4.1-4 PRESSURIZER PRESSURE FOR LOW POWER BANK WITHDRAWAL 2300 Pressurizer Pressure (psia) 2200 2100 2000 1900 0 10 20 30 40 50 Time (sec) Rev. 21

MPS-2 FSAR FIGURE 14.4.1-5 REACTIVITIES FOR LOW POWER BANK WITHDRAWAL 1.0 0.0 Total Moderator

                  -1.0                   Doppler Reactivity ($)
                  -2.0
                  -3.0
                  -4.0
                  -5.0 0   10     20                30   40   50 Time (sec)

Rev. 21

BLE 14.4.2-1 AVAILABLE REACTOR PROTECTION FOR THE UNCONTROLLED CONTROL ROD/BANK WITHDRAWAL AT POWER EVENT eactor Operational Mode Reactor Protection 1 Variable Overpower Trip Local Power Density Trip Thermal Margin/Low Pressure Trip High Pressurizer Pressure Trip Rod Withdrawal Prohibit Action on Variable Overpower or TM/LP Pretrip Alarm (1) 2-6 Not considered in this section Control grade. Provides defense in depth. Not credited in any safety analysis event. 14.4-26 Rev. 35

BLE 14.4.2-2 DISPOSITION OF EVENTS FOR THE UNCONTROLLED CONTROL ROD/BANK WITHDRAWAL AT POWER EVENT eactor Operational Mode Disposition 1 Analyze at rated power 2-6 No analysis required; not considered in this section 14.4-27 Rev. 35

TABLE 14.4.2-3 EVENT

SUMMARY

FOR THE UNCONTROLLED ROD/BANK WITHDRAWAL EVENT FOR THE LIMITING 100% POWER CASE Event Time (seconds) rt Rod Withdrawal 0.00 tdown Flow Valve Open 0.00

 /LP Trip Signal                               511.96 rbine Stop Valve Closed                         512.88 ak Power Level                                  513.32 DNBR                                            513.36 ak Core Average Temperature                     513.39 am Line Safety Valves Open                     514.94 ak Steam Dome Pressure                          516.96 14.4-28                     Rev. 35

MAY, 1990 MNPS-2 FSAR J.L 1000 . 1100 . 1000 - I... u

~   1100     -
£
     * "Aft
                                                                                \

o , r T . o 100 110 100 100 100 110 400 ... 100 I&CI Tlme..eo FIGURE 14.4.2-1 REACTOR CORE POWER FOR AN UNCONTROLLED BANK WITHDRAWAL AT POWER

MAY. 1990 MNPS-2 FSAR

                                                                     ~

110000 - 116000 - I-

'10000I-10000I-I-
                                                                          \

o I , I ~ I I , I I o too tao 100 160 100 ItA 400 .. 100 6U 1lme,.eo FIGURE 14.4.2-2 CORE AVERAGE HEAT FLUX FOR AN UNCONTROLLED BANK WIT HDRAWAL AT POWER

MAY, 1990 MNPS-2 FSAR 820 110 --~

            --            -  -        -                           ~

I I

                                                                          \

I 580

  • me-
                                                      ---=   *
                                                                         ~"

&60

          ,                                               I
                             - ...-- , - - --,                 ,       I o
  • 100
  • 110
  • 200 250 300 160 .00 .eo 600 560 Time. sec FIGURE 14.4.2-3 REACT OR COOLAW SYSTEM TEMPERATURES FOR AN UNCONTROLLED BANK WITHDRAWAL AT POWER

MAY, 1990 MNPS-2 FSAR at'II 1111 1 ~ au alOO 5.. ~ ZO'IG I.. 1000 at 1016 1000 ltr11 0 10 lOG 110 aGO 160 100 110 too ... 100 16@ nm...eo FIGURE 14.4 .2-4 PRESSURIZER PRESSURE FOR AN UNCONTROLLED BANK WITHDRAWAL AT POWER

MNPS-2 FSAR MAY, 1990 u o .. ----.--...-----.------.~.

IBBrc
                                                              *     .~
                                                                         ~

-7.6

-10     *     ,       ,        *       *      *    *      *      *
  • o 10 100 110 100 .. 100 .. 400 460 100 160 11me.8eo FIGURE 14.4.2-5 REACTIVITIES FOR AN UNCONTROLLED BANK WITHDRAWAL AT POWER

MAY. 1990 MNPS-2 FSAR JDQL lHO - t ~ 1200 -

 ~.
**  1100 -

t u , a 0 . ~ 1000 () I'll 100 - aoo * , * , , , * * * , o 10 100 110 .. lao lOG laO 40G ao 100 160

                                          'ftme. seo FIGURE14.4.2-6     SECONDARY PRESSURE FOR AN UNCONTROLLED BANK WIT HDRAWAL AT POWER

TABLE 14.4.3.1-1 AVAILABLE REACTOR PROTECTION FOR THE DROPPED CONTROL ROD/BANK EVENT eactor Operational Mode Reactor Protection 1 Variable Overpower Trip Thermal Margin/Low Pressure Trip Local Power Density Trip Available Thermal Margin (1) 2 Variable Overpower Trip Available Thermal Margin (1) 3-6 No Significant Consequences for these Reactor Operational Modes Provides defense in depth 14.4-35 Rev. 35

BLE 14.4.3.1-2 DISPOSITION OF EVENTS FOR THE DROPPED CONTROL ROD/ BANK EVENT Reactor Operational Mode Disposition 1 Analyze at rated power 2 Bounded by the above; no analysis required Available Thermal (1) 3-6 No analysis required Provides defense in depth. 14.4-36 Rev. 35

ABLE 14.4.3.1-3 EVENT

SUMMARY

FOR THE LIMITING DROPPED CONTROL ROD/BANK CASE Event Time (seconds) nimum Worth Control Rod Drops into Core 0.0 Power Exceeds VHP Trip Setpoint 88.3 DNBR and Peak LHGR occur 89.5 alysis Trerminated 100.0 14.4-37 Rev. 35

MPS-2 FSAR FIGURE 14.4.3.1-1 REACTOR POWER LEVEL FOR THE LIMITING DROPPED CONTROL ROD/BANK CASE 1.5 Normalized Power or Heat Flux (fraction of rated) 1.25 1.0

                                                      .75
                                                       .5 Reactor Power
                                                      .25                                Indicated thermal power Indicated nuclear power Core-average heat flux
                                                       .0
                                                            .0   10.0   20.0   30.0    40.0      50.0      60.0    70.0   80.0   90.0   100.0 Time (sec)

JUNE 2000

MPS-2 FSAR FIGURE 14.4.3.1-2 REACTOR COOLANT SYSTEM TEMPERATURES FOR THE LIMITING DROPPED CONTROL ROD/BANK CASE 625 600 Average Coolant Temperature (F) 575 550 525 500 475 T-hot T-avg T-cold 450

                                        .0   10.0   20.0   30.0    40.0      50.0      60.0   70.0   80.0   90.0   100.0 Time (sec)

JUNE 2000

MPS-2 FSAR FIGURE 14.4.3.1-3 PRESSURIZER PRESURE FOR THE LIMITING DROPPED CONTROL ROD/BANK CASE 2300 2200 Pressurizer Pressure (psia) 2100 2000 1900 1800

                                     .0   10.0   20.0   30.0    40.0     50.0       60.0   70.0   80.0   90.0   100.0 Time (sec)

JUNE 2000

MPS-2 FSAR FIGURE 14.4.3.1-4 SECONDARY PRESSURE FOR THE LIMITING DROPPED ROD/BANK CASE 880 SG-1 860 SG-2 840 Steam Generator Pressure (psia) 820 800 780 760 740 720 700 680 660 640

                                        .0   10.0   20.0   30.0   40.0     50.0       60.0   70.0   80.0   90.0   100.0 Time (sec)

JUNE 2000

TABLE 14.4.3.5-1 AVAILABLE REACTOR PROTECTION FOR THE SINGLE CONTROL ROD WITHDRAWAL EVENT Reactor Operational Mode Reactor Protection 1 Variable Overpower Trip Local Power Density Trip Thermal Margin/Low Pressure Trip High Pressurizer Pressure Trip Rod Withdrawal Prohibit Action on Variable Overpower or TM/LP Pretrip Alarm (1) 2 Variable Overpower Trip High Pressurizer Pressure Trip Rod Withdrawal Prohibit on Variable Overpower Pretrip Alarm (1) 3 Variable Overpower Trip Rod Withdrawal Prohibit on Variable Overpower Pretrip Alarm (1) 4-6 Not a Credible Event; No Analysis Required Control grade. Provides defense in depth. Not credited in any safety analysis event. 14.4-42 Rev. 35

ABLE 14.4.3.5-2 DISPOSITION OF EVENTS FOR THE SINGLE CONTROL ROD WITHDRAWAL EVENT eactor Operational Mode Disposition 1 Analyze at rated power 2 Bounded by Mode 1 (1) 3 Bounded by the above 4-6 Not a Credible Event; No Analysis Required Mode 2 operation is dispositioned as bounded by Mode 1 operation because the MDNBR for an Uncontrolled Bank Withdrawal from Startup (Event 14.4.1) was found to be bounded by the MDNBR for an Uncontrolled Bank Withdrawal at Power (Event 14.4.2). 14.4-43 Rev. 35

TABLE 14.4.4-1 AVAILABLE REACTOR PROTECTION Reactor Operational Mode Reactor Protection 1, 2, 6 Not Applicable 3-5 Technical Specification Requirements on Shutdown Margin and Reactor Coolant Pump Operation 14.4-44 Rev. 35

BLE 14.4.4-2 DISPOSITION OF EVENTS FOR THE STARTUP OF AN INACTIVE LOOP EVENT Reactor Operational Mode Disposition 1, 2, 6 Not Applicable 3-5 No analysis required; minimal consequences 14.4-45 Rev. 35

TABLE 14.4.6-1 AVAILABLE REACTOR PROTECTION FOR CHEMICAL AND LUME CONTROL SYSTEM MALFUNCTION THAT RESULTS IN A DECREASE IN THE BORON CONCENTRATION IN THE REACTOR COOLANT EVENT Reactor Operational Mode Reactor Protection 1 Local Power Density Trip Variable Overpower Trip Thermal Margin / Low Pressure Trip High Pressurizer Pressure Trip 2 Variable Overpower Trip High Pressurizer Pressure Trip 3-6 Technical Specification Shutdown Margin Requirements Administrative Procedures Operator Response Time 14.4-46 Rev. 35

ABLE 14.4.6-2 DISPOSITION OF EVENTS FOR THE CHEMICAL AND VOLUME ONTROL SYSTEM MALFUNCTION THAT RESULTS IN A DECREASE IN THE BORON CONCENTRATION IN THE REACTOR COOLANT EVENT Reactor Operational Mode Disposition 1-9 Analyze for loss of shutdown margin 14.4-47 Rev. 35

ABLE 14.4.6-3

SUMMARY

OF RESULTS FOR THE BORON DILUTION EVENT ASYMMETRIC DILUTION FRONT MODEL Minimum Analytical Required SDC Flow to Satisfy Licensing Criteria for Licensing Criteria for Operator Response Time Operator Response Time eactor Operational Mode (gpm) (1) (minutes) Mode 6 688 30 Mode 5 808 15 Mode 4 848 15 The required flows do not include SDC flow measurement uncertainties. Plant procedures maintain the SDC flowrate at a higher value to account for instrument uncertainty. 14.4-48 Rev. 35

ABLE 14.4.6-4

SUMMARY

OF RESULTS FOR THE BORON DILUTION EVENT INSTANTANEOUS MIXING MODE Licensing Criteria for Time Calculated Time to Loss of to Loss of Shutdown Margin eactor Operational Mode Shutdown Margin (minutes) (minutes) Mode 5 100 15 Mode 4 125 15 Mode 3 73 15 Mode 1 and 2 71 15 14.4-49 Rev. 35

ABLE 14.4.8-1 AVAILABLE REACTOR PROTECTION FOR THE SPECTRUM OF CONTROL ROD EJECTION ACCIDENTS eactor Operational Mode Reactor Protection 1 Variable Overpower Trip Thermal Margin / Low Pressure Trip High Pressurizer Pressure Trip 2 Variable Overpower Trip High Pressurizer Pressure Trip 3 Variable Overpower Trip 4-6 No Reactor Protection Required; Ejected Rod Worth Less than the Technical Specification Minimum Shutdown Margin. No Significant Consequence for this Operating Condition. 14.4-50 Rev. 35

ABLE 14.4.8-2 DISPOSITION OF EVENTS FOR THE SPECTRUM OF CONTROL ROD EJECTION ACCIDENTS eactor Operational Mode Disposition 1 Analyze for short-term response. Long-term bounded by Event 14.6.5 2, 3 Analyze 4-6 No analysis required 14.4-51 Rev. 35

TABLE 14.4.8-3 EVENT

SUMMARY

FOR A CONTROL ROD EJECTION (MAXIMUM PRESSURIZATION CASE) Event Time (seconds) ntrol Rod Ejects 0.00 arging Pumps On 0.00 ssurizer Heaters On 0.00 actor Scram Signal 4.23 ak Power 5.61 ssurizer Safety Valves Open 5.94 ak Core Average Temperature 6.53 ak Pressurizer Pressure 7.08 am Line Safety Valves Open 9.60 ak Steam Dome Pressure 11.63 14.4-52 Rev. 35

TABLE 14.4.8-4 EVENT

SUMMARY

FOR A CONTROL ROD EJECTION MINIMUM DEPARTURE FROM NUCLEATE BOILING RATIO CASE Event Time (seconds) ntrol Rod Ejects 0.00 tdown Valve Open 0.00 actor Scram Signal 3.23 ak Power 4.10 ak Core Average Heat Flux 4.22 DNBR 4.22 ak Core Average Temperature 4.64 ak Pressurizer Pressure 5.42 am Line Safety Valves Open 7.53 ak Steam Dome Pressure 9.56 14.4-53 Rev. 35

TABLE 14.4.8-5 DNBOUNDING BEGINNING OF CYCLE/END OF CYCLE EJECTED ROD ANALYSIS HFP HZP Contribution(a) to Energy Contribution (a) to Energy Value Deposition Value Deposition A. 103.4 cal/g --- 18.0 cal/g --- B. Generic Initial Fuel 40.8 cal/g --- 16.7 cal/g --- Enthalpy C. Delta Initial Fuel 62.6 cal/g 62.6 cal/g 1.3 cal/g 1.3 cal/g Enthalpy D. Maximum Control Rod 200 pcm 131.4 cal/g 800 pcm 109.1 cal/g Worth E. Doppler Coefficient -0.8 pcm/°F 1.17 -0.8 pcm/°F 1.31 F. Delayed Neutron 0.0045 1.06 0.0045 1.28 Fraction G. Power Peaking Factor 6.0 --- 14.0 --- Total Fuel Enthalpy 240.6 cal/g (b) 185.1 cal/g (b) (a) The contribution to the total pellet energy deposition is a function of initial fuel enthalpy, maximum control rod worth, Doppler coefficient, and delayed neutron fraction. The energy deposition contribution values and factors are derived from data calculated the generic analysis of the control rod ejection transient document XN-NF-78-44. (b) Total pellet energy deposition (cal/g) is calculated by the equation: Total (cal/g) = (C+D) * (E) * (F). 14.4-54 Rev

TABLE 14.4.8-6 CREA RADIOLOGICAL ANALYSIS ASSUMPTIONS re Power Level 2754 Mwt led Fuel Percentage 11.5% aking Factor 1.69 rcentage of Core Activity in Gap 10% Noble Gas 10% Halogens mposition of Iodine in the Core Gap (particulate/elemental/organic) 95 / 4.85 / 0.15 % - containment 0 / 97 / 3 % - secondary side actor Coolant Mass 430,000 lbs am Generator Minimum Mass 100,000 lbs/SG ntainment Free Volume 1.899E6 ft3 ntainment Leak Rate 0-1 day: 0.5% vol per day 1-30 days: 0.25% vol per day closure Building Bypass Fraction 1.4% me Before Enclosure Building Filtration System (EBFS) is 170 seconds lly Functional closure Building Filter Efficiency 70 / 70 / 70% (1) (particulate/elemental/organic) e Boundary Meteorology X/Qs ound Level Release B: 0 - 2 hr 3.66E-04 Z: 0 - 4 hr 4.80E-05 4 - 8 hr 2.31E-05 14.4-55 Rev. 35

8 - 24 hr 1.60E-05 24 - 96 hr 7.25E-06 96 - 720 hr 2.32E-06 llstone Stack Release (includes fumigation): B: 0 - 2 hr 1.00E-04 Z: 0- 4 hr 2.69E-05 4 - 8 hr 1.07E-05 8 - 24 hr 6.72E-06 24 - 96 hr 2.46E-06 96 - 720 hr 5.83E-07 ntrol Room Breathing Rate 3.5E-04 m3/sec ntrol Room Isolation Time post-accident 105 seconds ntrol Room Intake Prior to Isolation 800 cfm ntrol Room Inleakage During Isolation 200 cfm ntrol Room Emergency Filtered Recirculation Rate (from 1 2,250 cfm ur after isolation) ntrol Room Intake Dispersion Factors (sec/m3) Ground Millstone Stack ADV 0 - 2 hr 3.00E-3 2.51E-4 7.40E-3 2 - 4 hr 1.87E-3 2.51E-4 5.71E-3 4 - 8 hr 1.87E-3 1.96E-5 5.71E-3 8 - 24 hr 6.64E-4 5.46E-6 2.13E-3 24 - 96 hr 5.83E-4 3.43E-7 1.74E-3 96 - 720 hr 4.97E-4 6.44E-9 1.43E-3 ntrol Room Free Volume 35,656 ft3 ntrol Room Filter Efficiency (particulate/elemental/organic) 70 / 70 / 70 %(1) se Conversion Factors Federal Guideline Reports 11 and 12 70% is a conservative analysis assumption for some iodine species. Technical Specifications can support assumptions for filter efficiencies of 90% for all iodine species. 14.4-56 Rev. 35

TABLE 14.4.8-7 RADIOLOGICAL CONSEQUENCES OF A CREA EAB, LPZ, Control Room, CREA rem-TEDE rem-TEDE rem-TEDE ntainment Release 5.4 E -01 4.7 E -01 1.6E+00 condary Side Release 7.9 E -01 1.8 E -01 3.9E+00 14.4-57 Rev. 35

MNPS-" t:'S.AR 3500 r---r----.---,----~--~--_r_--~--_y_--__r--___,

                                                                                - - - PL 3000

.....> 2500 ~ L: fu2000

J o

0-

0) 1500 L

o U 1000 500 Ol-_ _. . J -_ _ ~ _ _. . L __ __ L __ ___'___ __ _ ' __ __..L._ ____L_ ____I__ _ _.I o 2 6 8 10 TLme (sec) FIGURE 14.4.8-1 CORE POWER FOR A CEA EJECTION (MDNBR sa) CASE) APRIL 1993

MNP~'" FSAR ..... 20000 I W OOA 0 J: 17500 N

+>

4-,L .c 15000 f---4 (I) x 12500

J l.a..
 ..-)

0 10000 m

r:

m m 0 7500 L m a: 5000 0 2 i 6 8 10 TLme (sec) FIGURE 14.4.8-2 CORE AVERAGE HEAT FLUX FOR A CEA EJECTION (MDNSR CASE) APRIL 1993

MNpr ') FSAR 6iO I 1--y-----r--~--_,_--.,----_,_--~---r=1I=====1 I I TAVGI

                                                                                                                                                                   .............. lew 630 -                                                                                                                                                         - --- --- TCLI
                                                                                                                                                                   - - . - THLI 620 f-
                                                                                                                                   .--.- .- ._ '---- - - - - u..                                                                                                _.---.         ----

m

  . 610 f-                                                         ---.---

Q) - o f -. _ . -" ---.--. 600 t- - Q) L

l 590 I- -
+>

0 L Q) -:: f-n, 580 E Q) f-4 570 '- - 560 I- -

          ~- - - - - - - - - - - -_ -                                                                          * *_
                                        ~ U LI :.: : : : : : : : : : : :.::: : : : " . : :.::.: :::::.::'" ~ ~ _  * * *_ . _

550 __ _ ___ _ _ ___ _ _ ___ ~~~ ~ ~~~ ~~~~~u ---n ~ ~~~ I I . , I 510 o 2 . 6 8 to TLme (sec) FIGURE 14 .4 .8-3 PRIMARY SYSTEM TEMPERATURES FOR A CEA EJECTION (MDNBR CASE) APRIL 1993

MNF ., FSAR 2100 0 .J (f) 2350 0.- m L

J (f)

(f) 2300 m L (L L

 <D N

.J 2250 t.

J (f)

(f) m L (L 2200 2150 o 2 6 B 10 TLme (sec) FIGURE 14.4.8-4 PZR PRESSURE FOR A CEA EJECTION (MDNBR CASE) APRIL 1993

MNPf FSAR 1

                                       **** :::: : : :: :: :: :.: :::ru   r.:':. :-.:-'. ~.:"'. :"'
o. ..

0 ------- ------ -- (J) L -1 0 ~

                                                                                                       \  ------- Ot<oor I

0 0 -2

n
~

.J -3 .J

~

0 -i 0

 <D 0:::
     -5
     -6
     -7 ~-.L---:---L---I-----l-----L--.Lc:==:::L==                                                                 ==~=l o           2                4                                  6                                    B                     10 TLme (sec)

FIGURE 14.4.8-5 REACTIVITiES FOR A CEA EJECTION (MDNBR CASE) APRIL 1993

11 00 r------r------r---~--__r_--__r_--___r--____,..__--_r__--~--_, ~ 0 1050 .J en o, m L

J en 1000 en m

L n, m E 950 0 0 ~ if) 900 B50 o 2 6 8 10 TLme (sec) FIGURE 14.4.8-6 SECONDARY PRESSURE FOR A CEA EJECTION (MDNBR CASE) APRIL 1993

MNP" "" FSAR 3500 r----r-----r----r---.,-----r---~--___r--~--~--___, 3000

                                                         - - - PL

...> 2500 3:- L:

 ~  2000 o

0-

 ~   1500 o

o 1000 500 o o 2 6 B 10 TLme (sec) FIGURE 14.4.8-7 CORE POWER FOR A CEA EJECTION (OVERPRESSURE) APRIL 1993

MNP"" ') FSAR ....... 20000 I W 0

e: 18000 N

4-I L

..c 16000
J
~

m x 11000

J lL-
 -..J 0   12000 Q)
c

([) [P-o 10000 L ([) cr: 8000 0 2 6 8 10

                                     " TLme (sec)

FIGURE 14 .4.8-8 CORE AVERAGE HEAT FLUX FOR A CEA EJECTION (OVERPRESSURE) APRIL 1993

MNPS-2 FSAR 6~O

                                                                 ~~~                              ~ -

630 - 620 u_ TAYGI t ... ..... ..*... TCt.. J tn 810 ------- Trt.1 Q) 0 - _ . - Tela 800 ID L

J

-.J 590 - 0 L etJ 580 CL E ID ~ 570 560 550

                 .-- ._ . ~ l~ :~ :::::= :.-==:::: :.~:.=:.= :.= :.~ :.~ : .~ :

SiC 0 2 G e 10 TLme (sec) FIGURE 14.4.8 -9 PRIMARY SYSTEM TEMPERATURES FOR A CEA EJECTION (OVERPRESSURE) OCTOBER 1998

MNf ') FSAR 2700 2650 0 .J (/) 2600 0.. PPR Q) 2550 L (/) 2500 (/) ([J L 2450 (L L CD 2100 N .J L 2350 (/) (/) CD 2300 L (L 2250 2200 2150 0 2 i 6 o JO TLme (sec) FIGURE 14.4.8-10 PZR PRESSURE FOR A CEA EJECTION (OVERPRESSURE) APRIL 1993

MNPS FSAR 0 - - - n .* *.* * . *. **** . .* ** **.*** ** * *** * * * ** ** *** ** **** *. ** **** ~ ~ ~ ~~.~~ ~ ~~~.~ ~~.:.~~~. : ::::::::::~:> ..- _..:;..:::..:.- --------

     -1 (J)

L 0 -J -2 I.. . .*.*.. ~

                                                                                      -------            DKHOO I

0 0

     -3
7)

...., -1 .J .J ...., -5 u 0 (J) 0::: -6

      -7
      -8
      -9 0                               2                                             1                             6                              8                                     10 TLme (sec)

FIGURE 14.4.8-11 REACTIVITIES FOR A CEA EJECTION (OVERPRESSURE) APRIL 1993

MNP~ FSAR lI00..----r---...-----,.----r---.___---r---....---r---,----r--,-----, ~ 0 1050 .J 0) ......a. CD L

J 1000 0) 0)

CD L 0.... CD E 9SO 0 0 (!) (J') 900 85O'--_--L_ _..L.-_-&._ _...L-_---JL---_-L_ _.L--_---L_ _...L-_---I_ _- . L . _ - - I o 2 6 8 10 12 TLme (sec) FIGURE 14.4.8-12 SECONDARY PRESSURE FOR A CEA EJECTION (OVERPRESSURE) APRIL 1993

.1 INADVERTENT OPERATION OF THE EMERGENCY CORE COOLING SYSTEM THAT INCREASES REACTOR COOLANT INVENTORY s event is not in the current licensing basis for Millstone Unit 2 and therefore is not analyzed.
.2 CHEMICAL VOLUME AND CONTROL SYSTEM MALFUNCTION THAT INCREASES REACTOR COOLANT INVENTORY s event is not in the current licensing basis for Millstone Unit 2 and therefore is not analyzed.

potential consequences of diluting the primary system boron concentration are addressed in nt 14.4.6. 1 Rev. 21

MPS2 UFSAR 14.6 DECREASES IN REACTOR COOLANT INVENTORY 14.6.1 INADVERTENT OPENING OF A PRESSURIZED WATER REACTOR PRESSURIZER PRESSURE RELIEF VALVE 14.6.1.1 Event Initiator The event is postulated to occur as a result of the inadvertent opening of one or more pressurizer pressure relief or safety valves due to an electrical or mechanical failure. The limiting event is obtained by assuming the inadvertent opening of a pressurizer safety valve which bounds the capacity of two pressurizer power-operated relief valves (PORVs). 14.6.1.2 Event Description The opening of the pressurizer pressure relief valve or safety valve results in a blowdown of primary coolant as steam through the faulted valves. Primary system pressure drops rapidly until the pressurizer liquid is depleted, and then quite rapidly to a pressure determined by the saturation curve at the temperature of the coolant in the upper vessel head. Reactor scram will occur on thermal margin/low pressure (TM/LP) before the pressurizer liquid is depleted, terminating the challenge to Specified Acceptable Fuel Design Limits (SAFDLs). In this initial stage, pressurizer heaters would actuate in an attempt to maintain pressure, but would be turned off on a low-level signal before the heater elements were uncovered. 14.6.1.3 Reactor Protection The TM/LP trip provides initial protection against loss of thermal margin and possible fuel damage. Reactor protection for the Inadvertent Opening of a Pressurized Water Reactor (PWR) Pressurizer Pressure Relief Valve event is summarized in Table 14.6.1-1. 14.6.1.4 Disposition and Justification The event proceeds as a depressurization of the primary coolant system with a loss of inventory. The core power and primary loop temperatures are relatively unaffected by the pressure drop. Thus, a short term challenge to the SAFDLs exists due to the depressurization prior to scram. There is also a long term concern in that if primary inventory cannot be restored and maintained, core uncovery may result. The greatest challenge to core uncovery exists at rated power conditions when the core power and primary coolant stored energy are maximized. The greatest challenge to the SAFDLs occurs for the event initiated at rated power where the margin to Departure from Nucleate Boiling (DNB) is minimized. An evaluation of the SAFDL challenge is also made for 5% power operating conditions in Mode 2 when the TM/LP trip may be bypassed. In this mode, the primary system may depressurize below the TM/LP setpoint pressure without an automatic reactor trip occurring. The Safety 14.6-1 Rev. 35

MPS2 UFSAR Injection System (SIS) will, however, be available to inject boron and provide for inventory makeup. The disposition of events for the Inadvertent Opening of a PWR Pressurizer Pressure Relief Valve event is summarized in Table 14.6.1-2. 14.6.1.5 Definition of Events Analyzed As discussed above, this event is analyzed for Minimum Departure from Nucleate Boiling Ratio (MDNBR) for both Modes 1 (full power), and 2 (startup). The startup power case is analyzed because the TM/LP trip can be manually bypassed below 5% power. The system response for the full power case was evaluated by using PTSPWR2 (Reference 14.6-1). The full power event MDNBR was calculated using XCOBRA-IIIC (Reference 14.6-2). The system response for the startup case was determined by conservative problem constraints. The maximum power was limited to 7% of the rated power. Above this power the assumed TM/ LP trip bypass is automatically removed. The system pressure is conservatively assumed to be at the core inlet saturation pressure. The core inlet temperature is assumed to be at a level consistent with a maximum power rise of 7% and a conservative time delay before the SIS terminates the event. XCOBRA-IIIC was used with these system responses to predict the hot channel mass flux required for the critical heat flux calculation. The thermal margin was conservatively determined by the Modified Barnett critical heat flux correlation (Reference 14.6-3), with the system pressure reduced to the 725 psia upper limit of the Modified Barnett correlation. 14.6.1.6 Analysis Results The sequence of events for the full power analysis are given in Table 14.6.1-3. Figures 14.6.1-1 to 14.6.1-6 show the transient response for key system variables. The MDNBR for this event initiated from full power is above the CHF correlation limit. This event does not challenge the FCMLHR limit. Therefore, LHR is not evaluated. The startup mode case resulted in a minimum critical heat flux ratio of above 10, as calculated by the Modified Barnett correlation. The peak pellet LHR is less than the full power value. Thus, the startup mode is bounded by the full power mode. The High Pressure Safety Injection (HPSI) system has been shown to have sufficient capacity to compensate for the loss of primary coolant mass through the inadvertent opening of the pressurizer pressure relief valves. Analysis has shown that core uncovery does not occur during this event. 14.6.1.7 Conclusions The results of the analysis demonstrate that the event acceptance criteria are met since the MDNBR predicted for the full power case is greater than the DNBR safety limit and the minimum Critical Heat Flux Ratio (CHFR) predicted for the startup mode case is greater than the Modified 14.6-2 Rev. 35

MPS2 UFSAR Barnett Critical Heat Flux (CHF) limit. The correlation limits assure with 95% probability and 95% confidence, that DNB is not expected to occur; therefore, no fuel is expected to fail. The FCMLHR limit is not violated in this event. 14.6.2 RADIOLOGICAL CONSEQUENCES OF THE FAILURE OF SMALL LINES CARRYING PRIMARY COOLANT OUTSIDE OF CONTAINMENT Millstone Unit 2 does not have any instrument lines connected to the reactor coolant system (RCS) which penetrate the containment. A break in either the letdown line or a RCS sample line is not in the current licensing basis for Millstone Unit 2. Therefore, this event is not analyzed. 14.6.3 RADIOLOGICAL CONSEQUENCES OF STEAM GENERATOR TUBE FAILURE 14.6.3.1 Event Initiator The event is initiated by a loss of integrity in a single tube in a steam generator, resulting in a flow of primary side reactor coolant water into the secondary side. 14.6.3.2 Event Description Experience with nuclear steam generators indicates that the probability of complete severance of a tube is small. The more probable modes of failure are those involving the occurrence of pinholes or small cracks in the tubes, and of cracks in the seal welds between the tubes and tube sheet. A leaking steam generator tube would allow transport of primary coolant into the main steam system. Radioactivity contained in the primary coolant would mix with shell side water in the affected steam generator. Some of this radioactivity would be transported by steam to the turbine and then to the condenser. Noncondensible radioactive materials would then be passed to the atmosphere through the condenser air ejector discharge via the Plant stack. The vent path is via the Millstone stack until actuation of an Enclosure Building Filtration Actuation Signal (EBFAS). Actuation of EBFAS automatically isolates the vent path to the Millstone stack after which the Operators manually align the vent path to the Unit 2 stack. The radioactive products would be sensed by the condenser air ejector radiation monitor or the stack radiation monitor. These monitors have audible alarms that will be annunciated in the control room to alert the operator to abnormal activity levels so that corrective action could be taken. The behavior of the systems will vary depending upon the size of the steam generator tube failure. For small leaks the chemical and volume control charging pumps will be able to maintain the necessary primary coolant inventory and an automatic reactor trip will not occur. The gaseous fission products will be released from the main steam system at the air ejector discharge and will be discharged via the Plant stack. Nonvolatile fission products will tend to concentrate in the water of the steam generators. 14.6-3 Rev. 35

MPS2 UFSAR For leaks larger than the capacity of the charging pumps, the pressurizer water level and pressure will decrease and a reactor trip will occur. Upon reactor trip, the turbine will trip and the steam system atmospheric dump valves, steam generator safety valves and the turbine bypass valves will open. In this case it is possible that in addition to the noble fission gases a substantial amount of the radioiodines contained in the secondary system may also be released to the atmosphere through the steam generator safety valves and atmospheric dump valves. The amount of radioactivity released increases with break size. For this analysis, a double-ended break of one tube was assumed. The selection of one double-ended break as an upper limit is conservatively based upon the experience obtained with other steam generators. 14.6.3.3 Reactor Protection The leak rate through the double-ended rupture of one tube is greater than the maximum flow available from the charging pumps. Therefore, the Primary Coolant system pressure will decrease and a low pressurizer pressure trip or TM/LP trip will occur. The thermal margin trip has a low pressure floor below which trip will always occur. Following the reactor trip the Primary Coolant System is cooled by exhausting steam through the atmospheric dump valves, steam generator safety valves, and turbine bypass valves. The radioactivity exhausted through the atmospheric dump valves and steam generator safety valves passes directly to the atmosphere. The radioactivity exhausted through the turbine bypass valves flows to the condenser where the gaseous products remaining are vented to the atmosphere through the condenser air ejector and Plant stack. Due to loss of offsite power, a release pathway via the condenser is not credited. Reactor protection for the Radiological Consequences of Steam Generator Tube Failure event is summarized in Table 14.6.3-1. 14.6.3.4 Disposition and Justification The radiological consequences of a steam generator tube rupture (SGTR) accident are maximized at rated power operation due to the stored energy in the primary coolant which must be removed by the steam generators in order to bring the primary and secondary systems into pressure equilibrium, thereby terminating the primary to secondary leak. The challenge to the SAFDLs exists due to the depressurization prior to scram. As such, this challenge is very similar to that which exists due to the inadvertent opening of a pressurizer relief valve (Event 14.6.1). Since the depressurization rates associated with Event 14.6.1 are substantially larger than those encountered for this event, the corresponding pressure undershoot will also be greater. Event 14.6.1 will thus be characterized by lower pressures at the time of MDNBR than those obtained for this event. Therefore, the DNB aspects of this event will be bounded by those of Event 14.6.1. The disposition of events for the Radiological Consequences of Steam Generator Tube Failure event is summarized in Table 14.6.3-2. 14.6-4 Rev. 35

MPS2 UFSAR 14.6.3.5 Definition of Events Analyzed The analysis of the SGTR event was performed with assumptions regarding system operation that were chosen to maximize the radiological doses. The analysis assumed that a loss of offsite power and a reactor trip occur at the initiation of a double ended rupture of a steam generator tube. This causes a loss of forced circulation which results in a higher hotleg temperature and a larger portion of the break flow flashing. The loss of offsite power also results in the loss of the ability to steam via the condenser. Plant cooldown occurs via atmospheric dump valves or main steam safety valves, which is the release pathway for primary and secondary activity. This results in a slower cooldown and RCS depressurization, and a reduced capability to cool down the plant via the unaffected steam generator. All of these effects result in higher doses. The plant simulation includes modeling of the RCS, the steam generators, the main steam and feedwater systems, the charging and letdown systems, and the HPSI System. The pressurizer was modeled as a non-equilibrium volume. Single failure is not postulated in conjunction with the SGTR event. The following assumptions are made to ensure a conservative estimate of the radiological consequences:

1. The initial core power is 2754 MWt;
2. The initial reactor pressure is 2300 psia including instrument uncertainty;
3. The initial main steam pressure is 933 psia including instrument uncertainty;
4. The initial inlet temperature is 551.25°F including instrument uncertainty;
5. A double-ended rupture of one steam generator tube occurs instantaneously;
6. On reactor trip and turbine trip, loss of offsite power is assumed along with loss of instrument air and the condenser. The ADVs may be operated by local manual action due to loss of instrument air;
7. Following the reactor trip, the MSSVs lift for removal of decay heat from the RCS;
8. The analysis assumed the lowest allowed opening setpoint (-3% drift) for the ruptured steam generator MSSVs and the highest allowed opening setpoint (+3%

drift) for the intact steam generator MSSVs;

9. The reseat pressure of the MSSVs on the ruptured steam generator is 12% below the opening pressure, Reference 14.6-5, while the reseat pressure of the MSSVs on the intact steam generator is nominal 6% below the opening pressure. This maximizes the releases to the atmosphere from the ruptured steam generator;
10. All three charging pumps are assumed to be operable, which will lead to a larger primary to secondary break flow. Letdown is conservatively isolated at the time of tube rupture; 14.6-5 Rev. 35

MPS2 UFSAR

11. SIAS is initiated on low pressurizer pressure which starts two HPSI pumps to deliver maximum flow;
12. AFW auto-initiates, accounting for system delay, and delivers a minimum flow.

The operator actions assumed in this analysis are consistent with the EOPs. The major post-trip analysis assumptions regarding operator actions are:

1. Commence Cooldown to Hotleg Temperature Less Than 515°F Once the event is diagnosed, the operators will cool the RCS at a maximum controllable rate until the hotleg temperature of both loops reaches 515°F, for ruptured steam generator isolation. This temperature assumed in the analysis conservatively includes instrument uncertainties to delay the time till the ruptured steam generator can be isolated. The analysis assumes a loss of offsite power leading to a loss of the condenser. Therefore, the cooldown is performed using the ADVs. Since the analysis assumes a loss of offsite power, a loss of instrument air is postulated, requiring a local manual control of the ADVs for this cooldown. The analysis assumes that, to account for local manual operator action, the cooldown starts 30 minutes from the time of reactor trip.
2. Reduce and Control RCS Pressure The analysis conservatively does not depressurize the RCS till after the hotleg temperature is less than 515°F and the ruptured steam generator is isolated. In the EOPs, the RCS depressurization begins just after the cooldown to 515°F commences. It is more conservative for dose consequences to delay the RCS depressurization since this will provide a larger primary to secondary break flow rate.
3. Determine and Isolate the Most Affected Steam Generator The operator isolates the most affected steam generator once the hotleg temperature of the loops have reached the isolation temperature of hotleg less than 515°F.
4. Cooldown and Depressurize RCS to SDC Entry Condition Cooldown and depressurization to SDC entry would minimize the primary to secondary break flow. The analysis assumes that cooldown to SDC entry is achieved by steaming just the intact steam generator per the EOPs. This is performed for 16 hours from the time of the tube rupture. The analysis conservatively assumes that the hotleg temperatures of the two loops fail to stay coupled, impeding the depressurization to SDC condition. There are five options available in the EOPs to cool and depressurize the isolated steam generator: 1) if RCPs are operating, use at least one RCP and perform a backflow into the RCS; 2) 14.6-6 Rev. 35

MPS2 UFSAR if time permits, allow ambient cooling; 3) if the condenser is available, steam to the condenser; 4) feed and bleed via steam generator blowdown; 5) steam to the atmosphere using ADV and feeding. Since the last option would lead to a larger offsite dose, it is the method modeled. Given a loss of offsite power/loss of instrument air condition, the feed and bleed via the steam generator blowdown or ambient cooling may be chosen to limit the offsite dose as well as dose to the operator for performing a local manual operation of the ruptured steam generator ADV.

5. Maintain Isolated Steam Generator Level Less Than 90%

The EOPs prevent the ruptured steam generator from overfill by maintaining the pressurizer pressure within 50 psi of the isolated steam generator pressure or backflow into the RCS, in order to minimize the primary to secondary break flow. Alternatively, the steam generator blowdown may be used to restore level less than 90% narrow range level. For offsite dose purposes, this is not explicitly modeled. However, the model assumes primary side depressurization, facilitated by steaming of the isolated steam generator. Therefore, the pressurizer pressure can be maintained within 50 psi of the isolated steam generator, avoiding ruptured steam generator overfill. 14.6.3.6 Analysis Results 14.6.3.6.1 Thermal-Hydraulic Calculation The portion of the SGTR analysis, till the time the hotleg temperatures reach less than 515°F, was performed using RETRAN-02 MOD 3 (Reference 14.6-4) computer code. The sequence of results for this transient is presented in Table 14.6.3-3. Figures 14.6.3-1 through 14.6.3-9 present the dynamic behavior of important NSSS parameters during this event. Following a double-ended break of a steam generator tube rupture, reactor coolant flows from the primary side into the secondary side of the ruptured steam generator (see Figure 14.6.3-5). A portion of this break flow is released as flashed steam (see Figure 14.6.3-6). The model has the reactor tripping at the time of tube break. This is conservative, since any pre-trip mass releases would be via the condenser air ejector where a partition factor would greatly reduce the iodine releases (this is further discussed in Section 14.6.3.6.2). Therefore, to conservatively maximize the direct atmospheric releases, the earliest possible trip is limiting. A loss of offsite power at the time of trip leads to a loss of forced flow and a momentary spike in the coldleg temperature as shown in Figure 14.6.3-1. The pressurizer level decreases as the reactor coolant shrinks post-trip. Also, the break flow is greater than the capacity of the charging pumps. As a result, the pressurizer level decreases as shown in Figure 14.6.3-2. The pressurizer pressure also drops as shown in Figure 14.6.3-3. While all three charging pumps and pressurizer heaters attempt to maintain level and pressure, letdown is conservatively isolated at the time of tube break. The pressurizer heaters are turned off as the pressurizer level decreases towards heater uncovery. 14.6-7 Rev. 35

MPS2 UFSAR As the steam bypass to the condenser is assumed to be unavailable, the post-trip steaming is accomplished via the ADVs and the MSSVs. However, the ADVs require instrument air, which is postulated to be lost with the loss of offsite power. Therefore, no releases from the ADVs are modeled till 1,800 seconds from trip, when local manual operator action can be credited (see Figure 14.6.3-7). Hence, the post-trip steaming to remove decay heat is accomplished, during the initial 30 minutes, by the MSSVs (see Figure 14.6.3-8). The turbine valve closure, due to reactor trip, causes the steam generator pressure to rise, as shown in Figure 14.6.3-4, till the MSSV lift pressure is reached. The main feedwater flow is terminated at the time of trip and the AFW initiates on low steam generator level at 277 seconds accounting for system response time. As the colder AFW is delivered, a hotter volume of feedwater is swept in first. Two AFW pumps deliver a minimum flow rate (see Figure 14.6.3-9). The pressurizer level and pressure continue to decrease as the energy transfer to the secondary side shrinks the reactor coolant and the tube break flow continues to deplete the primary inventory. The decrease in pressure results in actuation of SIAS at 496 seconds. Once RCS pressure decreases below the HPSI shutoff head pressure, two HPSI pumps deliver maximum flow to slow the decrease in pressurizer pressure. The pressurizer pressure approaches an equilibrium pressure as the combined HPSI and charging flow rate matches the break flow rate. The hotleg temperature of 515°F is reached in 3,637 seconds post-trip. Specific analyses of the potential for fuel failure is not performed for the steam generator tube accident. The potential for fuel failure is bounded by the analysis for the inadvertent opening of the pressurizer relief valve (Event 14.6.1). The analyses for Event 14.6.1 show that fuel failure does not occur for that event, therefore, fuel failure does not occur following a steam generator tube rupture. 14.6.3.6.2 Radiological Calculation The intent of this radiological consequences analysis is to verify that offsite and control room doses do not exceed the guidelines of 10 CFR 50.67 and Regulatory Guide 1.183. The mass releases following a SGTR were determined for use in evaluating the offsite and control room radiation exposure. Figures 14.6.3-5 through 14.6.3-8 show the break mass flow rate and the steam mass flow rate predicted by the thermal hydraulic analysis. This includes the flashing of the break flow as it enters the secondary side of the steam generator. Table 14.6.3-4 summarizes the mass releases for the SGTR event. This includes 92,000 lbm additional mass releases from the ruptured steam generator after it is identified and isolated associated with facilitating cooldown and depressurization for SDC entry, as well as 2,719,000 lbm released from the intact steam generator for cooldown to SDC entry. The SGTR accident is a penetration of the barrier between the RCS and the main steam system. The integrity of this barrier is significant from the standpoint of radiological safety in that a leaking steam generator tube allows the transfer of reactor coolant into the main steam system. Radioactivity contained in the reactor coolant is transported directly to the atmosphere via the ADVs and MSSVs. 14.6-8 Rev. 35

MPS2 UFSAR The effects of iodine spiking were accounted for in the analysis. Two different spiking models were evaluated. The first assumes that the primary coolant concentration of I-131 (DEQ) is at the Technical Specification limit of 1.0 Ci/gm and the resulting tube rupture causes the iodine appearance rate to increase by a factor of 335 over the equilibrium appearance rate corresponding to the 1.0 Ci/gm (DEQ) I-131 coolant concentration. The duration of the spike is assumed to be 8 hours. The second spiking model assumes that a pre-accident iodine spike causes the primary coolant to reach an I-131 (DEQ) concentration of 60 Ci/gm at the time of the tube rupture. This concentration is assumed to last for the duration of the accident. The RADTRAD-NAI computer program is used to calculate the TEDE dose due to RCS and secondary side activity. RADTRAD-NAI is a multiple compartment activity transport code with the dose model consistent with the Regulatory Guideline 1.183 model. The offsite and control room doses calculated are presented in Table 14.6.3-6. The results are bounding for the assumptions on Table 14.6.3-5 and the thermal-hydraulic results presented in Table 14.6.3-4. 14.6.3.7 Conclusion The radiological release criterion for this analysis, as well as the calculated results, are presented in Table 14.6.3-6. The calculated results are less than the NRC criteria for both the cases evaluated. 14.6.4 RADIOLOGICAL CONSEQUENCES OF A MAIN STEAM LINE FAILURE OUTSIDE CONTAINMENT This event is only applicable to Boiling Water Reactors (BWR). As such, this event is not applicable to Millstone Unit 2. 14.6.5 LOSS OF COOLANT ACCIDENTS RESULTING FROM A SPECTRUM OF POSTULATED PIPING BREAKS WITHIN THE REACTOR COOLANT PRESSURE BOUNDARY This event is initiated by a breach in the primary coolant system pressure boundary. Basically, a range of break sizes from small leaks up to a complete double-ended severance of a primary coolant system pipe must be considered. Typically, these breaks are classified as large breaks or small breaks. Large-break loss-of-coolant accidents (LBLOCA) are discussed in Section 14.6.5.1. Small-break loss-of-coolant accidents (SBLOCA) are discussed in Section 14.6.5.2. 14.6.5.1 Large Break Loss of Coolant Accidents 14.6.5.1.1 Event Initiator This event is initiated by a large break in the primary coolant system pressure boundary. The size of breaks typically considered to be large breaks are from 0.5 ft2 up to a double-ended severance of a primary coolant system pipe. 14.6-9 Rev. 35

MPS2 UFSAR 14.6.5.1.2 Event Description The LBLOCA events are characterized by four sequential phases. They are:

1. blowdown
2. refill
3. reflood
4. long term cooling The blowdown phase immediately follows the initiation of a large break. Primary system water is discharged through the break into containment. The system pressure decreases rapidly during the initial subcooled blowdown. As the saturation pressure is approached, local boiling and flashing takes place in the core and the reactor goes subcritical via the negative moderator reactivity feedback. The blowdown flow becomes a water-vapor mixture. The depressurization rate is reduced when core pressure falls below the saturation pressure. The water level continues to decrease until a large amount of water from the safety injection tanks reaches the lower plenum.

The refill phase starts when the safety injection tank water begins to fill the lower plenum. At this time, the core is uncovered by water and the fuel rods are cooled primarily by thermal radiation. The reflood phase begins when the water level reaches the bottom of the core. The long term cooling phase starts after the core has quenched to the point where the zircaloy-water reaction is suppressed, or the water level covers the active fuel. During this phase, the water inventory is controlled by the safety injection pumps. The continuous operation of these pumps ensures the long term dissipation of the decay heat. 14.6.5.1.3 Reactor Protection No credit is taken for a reactor trip by the reactor protection system (RPS). The RPS is not necessary due to the rapid depletion of the moderator which shuts down the reactor core almost immediately, followed by ECCS injection which contains sufficient boron to maintain the reactor core in a subcritical configuration. Technical specification limits on hot rod power serve to limit the peak cladding temperature (PCT). Available Reactor Protection for the Large Break Loss of Coolant Accidents is summarized in Table 14.6.5.1-1. 14.6.5.1.4 Disposition and Justification Section 15.6.5 of Reference 14.6-7 indicates that the primary acceptance criteria for this event are to limit offsite doses, to limit fuel clad oxidation, and to keep PCTs below 2200°F. Offsite doses are maximized by assuming the highest concentration of radionuclides contained within the fuel 14.6-10 Rev. 35

MPS2 UFSAR pins at event initiation. This is accomplished by assuming steady state radionuclide concentrations characteristic of long term operation of the plant at full power. Fuel pin cladding temperatures and oxidation rates are maximized by initiating the event with the highest cladding temperatures and linear heat generation rates (LHGR). Thus, the most limiting results for this event are obtained with the plant operating at full power in Mode 1. These results will bound those from Modes 2-6. Disposition of Events for the Large Break Loss of Coolant Accidents is summarized in Table 14.6.5.1-2. 14.6.5.1.5 Definition of Events Analyzed The purpose of the LBLOCA analysis is to demonstrate that the criteria stated in 10 CFR 50.46(b) are met. The criteria are:

1. The calculated peak fuel element cladding temperature does not exceed the 2200°F limit.
2. The amount of fuel element cladding which reacts chemically with water or steam does not exceed 1% of the total amount of zircaloy in the core.
3. The cladding temperature transient is terminated at a time when the core geometry is still amenable to cooling. The hot fuel rod cladding oxidation limit of 17% is not exceeded during or after quenching.
4. The core temperature is reduced and decay heat is removed for an extended period of time, as required by the long-lived radioactivity remaining in the core.

14.6.5.1.5.1 Description of Large Break Loss of Coolant Accident Transient A LBLOCA is defined as the rupture of the RCS primary piping from 0.5 ft2 in area up to and including a double-ended guillotine break. The limiting break occurs on the pump discharge side of a cold leg pipe. Loss of offsite power is assumed to occur coincident with the LBLOCA. Primary coolant pump coastdown occurs coincident with the loss of offsite power. Following the break, depressurization of the RCS, including the pressurizer, occurs. A reactor trip signal occurs when the pressurizer low pressure trip setpoint is reached. Reactor trip and scram are conservatively neglected in the LBLOCA analysis. Early in the blowdown, the reactor core experiences flow reversal and stagnation which causes the fuel rods to pass through CHF. Following CHF, the fuel rods dissipate heat through the transition and film boiling modes of heat transfer. Rewet is precluded during blowdown by Appendix K of 10 CFR 50. An SIS signal is actuated when the appropriate setpoint (high containment pressure) is reached. Due to loss of offsite power, a time delay for startup of diesel generators and SIS pumps is assumed. Once the time delay criteria is met and the system pressure falls below the shutoff head of the HPSI pumps or Low Pressure Safety Injection (LPSI) pumps, SIS flow is injected into the cold legs. The single failure criterion is met by assuming that either one diesel or one LPSI pump 14.6-11 Rev. 35

MPS2 UFSAR fails. Loss of diesel results in the loss of one HPSI pump, one LPSI pump, one containment spray train and two CAR fans. The Loss-of-LPSI pump case assumes only the loss of one LPSI pump. When the system pressure falls below the safety injection tank pressure, flow from the safety injection tanks is injected into the cold legs. Flow from the ECCS is assumed to bypass the core and flow to the break until the end of bypass (EOBY) is predicted to occur (sustained downflow in the downcomer). Following EOBY, ECCS flow fills the downcomer and lower plenum until the liquid level reaches the bottom of the core (beginning of core recovery, or BOCREC time). During the refill period, heat is transferred from the fuel rods by radiation heat transfer. The reflood period begins at BOCREC time. ECCS fluid fills the downcomer and provides the driving head to move coolant through the core. As the mixture level moves up the core, steam is generated. Steam binding occurs as the steam flows through the intact and broken loop steam generators and pumps. The pumps are assumed to have a locked rotor (per Appendix K of 10 CFR 50) which tends to reduce the reflood rate. The fuel rods are eventually cooled and quenched by radiation and convective heat transfer as the quench front moves up the core. The reflood heat transfer rate is predicted through experimentally determined heat transfer and carry-over rate fraction correlations. 14.6.5.1.5.2 Description of Analytical Models The AREVA EXEM/PWR evaluation model (Reference 14.6-8 as modified by Reference 14.6-9) was used to perform the analysis. This evaluation model consists of the following computer codes:

1. RODEX2 for computation of initial fuel stored energy, fission gas release, and gap conductance;
2. RELAP4-EM for the system blowdown and accumulator/SIS flow split calculations;
3. CONTEMPT/LT-22 as modified in accordance with NRC Branch Technical Position CSB 6-1 for computation of containment back pressure;
4. REFLEX for computation of system reflood; and
5. TOODEE2 for the calculation of fuel rod heatup during the refill and reflood portions of the LOCA transient.

The quench time, quench velocity, and carryover rate fraction correlations in REFLEX, and the heat transfer correlations in TOODEE2 are based on AREVAs Fuel Cooling Test Facility data. The governing conservation equations for mass, energy, and momentum transfer are used along with appropriate correlations consistent with Appendix K of 10 CFR 50. The reactor core in RELAP4 is modeled with heat generation rates determined from reactor kinetics equations with reactivity feedback, and with actinide and decay heating as required by Appendix K. Appropriate conservatisms specified by Appendix K of 10 CFR 50 are incorporated in all of the models. 14.6-12 Rev. 35

MPS2 UFSAR 14.6.5.1.5.3 Plant Description and Summary of Analysis Parameters The Millstone Unit 2 nuclear power plant is a Combustion Engineering (CE) designed PWR which has two hot leg pipes, two U-tube steam generators, and four cold leg pipes with one RCP in each cold leg. The plant utilizes a large dry containment. The RCS is nodalized into control volumes representing reasonably homogeneous regions, interconnected by flow paths or junctions. The two cold legs connected to the intact loop steam generator were assumed to be symmetrical and were modeled as one intact cold leg with appropriately scaled input. The model considers four safety injection tanks, a pressurizer, and two steam generators with both primary and secondary sides of the steam generators modeled. The HPSI and LPSI pumps were modeled as fill junctions at the safety injection tank lines, with conservative flows given as a function of system back pressure. The pump performance curves were characteristic of pumps typically used in CE plants. The reactor core was modeled radially with an average core and a hot assembly as parallel flow channels, each with three axial nodes. A total steam generator tube plugging level of 500 tubes per steam generator (symmetric) was assumed. Values for system parameters used in the analysis are given in Table 14.6.5.1-3. 14.6.5.1.5.4 Base Calculations Calculations were performed for all combinations of the following parameters:

  • 0.4, 0.6, 0.8, 1.0 DECLG and 0.8 and 1.0 SECLS breaks
  • Beginning-of-cycle (BOC), middle-of-cycle (MOC), and end-of-cycle (EOC) axial power shapes
  • Loss-of-diesel and loss-of-LPSI single failures Thus, this analysis comprised a total of 36 complete (blowdown, refill, and reflood) calculations to determine the limiting break and plant scenario. All calculations were performed at a peak LHR of 15.1 kW/ft. The BOC, MOC, and EOC axial shapes were peaked at a relative core height of 0.5, 0.77, and 0.85, respectively. BOC stored energy (where maximum densification occurs) was conservatively used in all of the BOC and MOC axial shape calculations. MOC stored energy was used in the EOC axial shape calculations.

PCT results for each break and limiting single failure are shown in Table 14.6.5.1-4. The table shows that the 1.0 DECLG, EOC axial shape, and loss-of-diesel single failure produced the highest PCT, 1814°F. Additional calculated results for the limiting case are shown in Table 14.6.5.1-5. The sequence of events for the overall limiting case is shown in Table 14.6.5.1-

6. Graphical results of parameters depicting the 1.0 DECLG, EOC axial shape, and loss-of-diesel single failure are shown in Figures 14.6.5.1-1 through 14.6.5.1-19.

14.6-13 Rev. 35

MPS2 UFSAR 14.6.5.1.5.5 Exposure Study The calculations described in Section 14.6.5.1.5.4 support exposures out to EOC. Additional evaluation is required to consider exposures out to end of life (EOL) with a maximum assembly average exposure of 56,000 MWd/MTU. The AREVA methodology predicts maximum fuel stored energy to occur near BOC where maximum densification occurs. Closure of the fuel-cladding gap at higher exposures significantly reduces the fuel stored energy. Beyond exposures of about 30,000 MWd/MTU, the stored energy begins to increase due to fission gas release to the gap, but is still significantly less than the stored energy at MOC. In addition, the power level of rods at EOL is significantly lower than the peak powered rod, resulting in a significantly lower stored energy than the peak powered rod. The reduced stored energy at high exposures outweighs any adverse effects of higher rod internal pressure. Thus, the peak cladding temperature is lower at high exposures than the limiting case reported for the base calculations in Section 14.6.5.1.5.4. This was confirmed in a separate EOL calculation, with 1.0 DECLG and loss-of-diesel single failure, which resulted in a PCT of 1362°F. Therefore, a peak LHR of 15.1 kW/ft is supported for assembly average exposures up to 56,000 MWd/MTU. 14.6.5.1.5.6 Reduced Reactor Coolant System Temperature Operation End-of-cycle full power primary coolant temperature (Tave) coastdown with a maximum reduction in primary coolant temperature of 12°F was evaluated. From the base calculations, the EOC axial shape combined with MOC stored energy resulted in a PCT of 1814°F. At EOC conditions, the hot rod stored energy will be considerably less than at MOC due to closure of the gap from fuel swelling and clad creep effects. This difference in stored energy is considered to more than offset any adverse effects on PCT of a 12°F Tave reduction. This was confirmed in a separate Tave coastdown calculation. In that a calculation which was performed using the 1.0 DECLG break and loss-of-diesel single failure, the PCT was 1758°F. Thus, this analysis bounds operation at EOC with up to a 12°F Tave coastdown. A full power coastdown to an indicated cold leg temperature of 537°F at EOC is bounded by this evaluation. 14.6.5.1.6 Summary of Results The analysis identified a double-ended cold leg guillotine break with a discharge coefficient of 1.0 (1.0 DECLG) as the limiting break size. The initial reactor conditions for the limiting break corresponded to an MOC stored energy combined with an EOC axial power shape peaked at a relative core height of 0.85. The limiting scenario also included the loss of one diesel generator as a single failure. The PCT for the limiting case was calculated to be 1814°F. The transient response for this limiting case is shown in Figures 14.6.5.1-1 through 14.6.5.1-19. The summary of results is given in Table 14.6.5.1-5. Margin between the calculated PCT and the 2200°F limit of 10 CFR 50.46 is available to accommodate other permanent adjustments due to 10 CFR 50.59 Safety Evaluations and LOCA model assessments. These adjustments are summarized in the 30-day and annual 10 CFR 50.46 reporting of PCT Margin Utilization. The reporting process and attention to PCT margin assure that the PCT remains below the 2200°F limit of 10 CFR 50.46. 14.6-14 Rev. 35

MPS2 UFSAR The analysis supports full power operation at 2754 MWt (2700 MWt plus 2% uncertainty) with a total steam generator tube plugging of up to 500 tubes per steam generator. The analysis supports assembly average exposures of up to 56,000 MWd/MTU. The analysis also supports operation at full power, with an indicated RCS cold leg temperature of 537°F at EOC. The analysis demonstrates that the 10 CFR 50.46(b) criteria are satisfied for the Millstone Unit 2 reactor with an axial and exposure independent LHR of 15.1 kw/ft. 14.6.5.1.7 Post Analysis of Record Evaluations In addition to the analyses presented in this section, evaluations and reanalyses may be performed as need to address ECCS Evaluation Model errors and emergent issues, or to support plant changes. The issues or changes are evaluated, and the impact on the peak cladding temperature (PCT) is determined. The resultant increases or decreases in PCT are applied to the analysis of record PCT. The PCT, including all penalties and benefits, is presented in Table 14.6.5.1-7 for the large break LOCA. The current PCT is demonstrated to be less than the 10 CFR 50.46(b) requirement of 2200°F. 14.6.5.1.8 Conclusions The results of the LBLOCA analysis for Millstone Unit 2 showed the 1.0 DECLG break size to be the limiting break with current EXEM/PWR as modified by SEM/PWR models. The analysis supports operation of Millstone Unit 2 at a power level of 2700 MWt and a total steam generator tube plugging of up to 500 tubes per steam generator. The analysis supports a peak LHR of 15.1 kW/ft with an axial and exposure independent power peaking limit. The analysis supports assembly average exposures of up to 56,000 MWd/MTU. The analysis supports full power operation with an indicated RCS cold leg temperature of 537°F at EOC. A reduction in RCS cold leg temperature to less than 537°F is acceptable as long as there is a corresponding decrease in power level. Operation of Millstone Unit 2 with AREVA 14x14 fuel at or below the LHR limit assures that the NRC acceptance criteria (10 CFR 50.46(b)) for LOCA pipe breaks up to and including the double-ended severance of a reactor coolant pipe will be met with the ECCS for Millstone Unit 2. 14.6.5.2 Small Break Loss of Coolant Accident The following subsections present the ECCS Small Break Loss of Coolant Accidents (SBLOCA) performance analysis supporting AREVA Standard CE14 HTP Fuel Assembly with M5 fuel rod cladding. AREVA fuel with Zirc 4 cladding is evaluated as discussed in FSAR Section 14.6.5.2.7. 14.6.5.2.1 Event Initiator This event is initiated by a small break in the primary coolant system pressure boundary. The size of breaks typically considered to be small breaks are from a leak exceeding the makeup capacity of the charging system up to approximately 0.5 ft2. The most limiting break location is in the RCS cold leg pipe on the discharge side of the reactor coolant pump. 14.6-15 Rev. 35

MPS2 UFSAR 14.6.5.2.2 Event Description The principal PWR design feature for mitigating the consequences of an SBLOCA is the Emergency Core Cooling System (ECCS) which maintains the water inventory. Its major subsystems for restoring water inventory are the HPSI system, and the LPSI system and the safety injection tanks. An SBLOCA is characterized by slow RCS depressurization rates and mass transfer rates within the RCS relative to similar parameters calculated for LBLOCA. If the break area is large enough that the charging pumps cannot maintain the reactor coolant inventory and allow RCS pressure control, the RCS will depressurize. The depressurization produces a low pressurizer pressure (TM/LP) reactor trip and an SIAS. The rate of RCS depressurization following SIAS depends on the break area and the HPSI shutoff head. With a combination of a very small break and a sufficiently high HPSI shutoff head, the depressurization may be arrested. If the break area is sufficiently large to allow continued depressurization and net loss of coolant inventory even with the HPSI pumps in operation, the coolant level in the reactor vessel may recede below the top of the reactor core. If sufficient steam is produced in the RCS, natural circulation (the RCPs will have been tripped by this time to reduce coolant loss out the break) around the RCS loops will cease. Eventually, loss of reactor coolant inventory is arrested by ECCS flow exceeding flow out the break. In either case, the coolant level within the reactor vessel will rise, and the RCS will eventually be refilled (although leaking) to the cold leg elevation. 14.6.5.2.3 Reactor Protection Primary reactor protection for this event is provided by the low pressurizer pressure (TM/LP) trip and the SIAS on a low pressurizer pressure signal. Available Reactor Protection for the Small Break Loss of Coolant Accidents resulting from a Spectrum of Postulated Piping Breaks Within the Reactor Coolant Pressure Boundary event is summarized in Table 14.6.5.2-1. 14.6.5.2.4 Disposition and Justification Section 15.6.5 of Reference 14.6-11 indicates that the primary acceptance criteria for this event are to limit offsite doses, to limit fuel clad oxidation, and to keep PCT below 2200°F. Offsite doses are maximized by assuming the highest concentration of radionuclides contained within the fuel pins at event initiation. This is accomplished by assuming steady state radionuclide concentrations characteristic of long term operation of the plant at full power. Fuel pin cladding temperatures and oxidation rates are maximized by initiating the event with the highest cladding temperatures and LHGR. Thus, the most limiting results for this event are obtained with the plant operating at full power in Mode 1. These results will bound those from Modes 2-6. Disposition of Events for the Small Break Loss of Coolant Accidents Resulting from a Spectrum of Postulated Piping Breaks Within the Reactor Coolant Pressure Boundary event is summarized in Table 14.6.5.2-2. 14.6-16 Rev. 35

MPS2 UFSAR 14.6.5.2.5 Definition of Events Analyzed The purpose of the SBLOCA analysis is to demonstrate that the criteria stated in 10 CFR 50.46(b) are met. The criteria are:

1. The calculated maximum fuel element cladding temperature shall not exceed 2200°F.
2. The calculated total local oxidation of the cladding shall not exceed 0.17 times the total cladding thickness before oxidation.
3. The calculated total amount of hydrogen generated from the chemical reaction of the cladding with water or steam shall not exceed 0.01 times the hypothetical amount that would be generated if all of the metal in the cladding cylinders surrounding the fuel, excluding the cladding surrounding the plenum volume, were to react.
4. The core temperature is reduced and decay heat is removed for an extended period of time, as required by the long-lived radioactivity remaining in the core.

14.6.5.2.5.1 Description of Small Break Loss of Coolant Accident Transient The SBLOCA is generally defined as a break in the PWR pressure boundary which has an area of 0.5 ft2 (approximately 10% of cold leg pipe area) or less. This range of break areas encompasses small lines which penetrate the primary pressure boundary. Small breaks could involve pressurizer relief and safety valves, charging and letdown lines, drain lines, and instrumentation lines. The limiting break size is generally in the neighborhood of 2% of the cold leg pipe area. The most limiting break location is in the cold leg pipe at the discharge side of the pumps, particularly with primary pumps tripped early to conservatively model a possible loss-of-offsite power on reactor trip. This break location results in the largest amount of inventory loss and the largest fraction of ECCS fluid ejected out the break. This produces the greatest degree of core uncovery and the longest fuel rod heatup time. The SBLOCA transient is characterized by a slow depressurization of the primary system with a reactor trip occurring at a low primary pressure of 1700 psia (conservatively bounding the actual value of 1865 psia) in the Millstone Unit 2 plant. The SIAS occurs when the system has depressurized to 1500 psia. The capacity and shutoff head of the HPSI pumps are important parameters in the SBLOCA transient. The single failure criterion is satisfied by the loss of one diesel generator. In the Millstone Unit 2 SBLOCA analysis, only one train of safety injection is available. The SBLOCA transient can be categorized into three ranges of break sizes. The scenario is different for each range of break sizes. Very small breaks are characterized by inventory losses that are less than the makeup capacity of the HPSI pumps such that core uncovery is limited. The core level is eventually recovered and hot rod heatup is limited. Breaks with a large flow area are characterized by a sufficiently large primary system depressurization rate such that the safety 14.6-17 Rev. 35

MPS2 UFSAR injection tank pressure is reached in sufficient time to limit the core uncovery and hot rod heatup. The HPSI pumps have limited influence on those transients. Breaks with a flow area between the two extremes are generally the most limiting. In those transients, the rate of inventory loss from the primary system is large enough that the HPSI pumps cannot preclude significant core uncovery. The primary system depressurization rate is very slow, extending the time required to reach the safety injection tank pressure. This tends to maximize the heatup time of the hot rod and produces the maximum PCT. It also results in the longest time-at-temperature, which maximizes the local cladding oxidation. The limiting break is usually one with a large enough break area that there is a prolonged period of core uncovery. The time allowed for the operators to manually trip the primary coolant pumps following the SIAS is an important parameter. Allowing the pumps to continue to operate during an SBLOCA would delay break uncovery (transition to mostly steam flow), resulting in additional inventory loss from the system. Subsequent loss of the primary pumps (due to cavitation or an eventual manual trip) would result in more core uncovery, and higher PCTs. The base calculations for this analysis tripped the primary coolant pumps at the reactor scram signal, when a loss of offsite power is assumed to occur. A sensitivity study was also performed to determine an acceptable primary coolant pump trip delay time (Section 14.6.5.2.5.5). 14.6.5.2.5.2 Analytical Models The approved AREVA SBLOCA evaluation model (References 14.6-12 and 14.6-13) consists of two computer codes. The appropriate conservatisms prescribed by Appendix K of 10 CFR 50 are incorporated.

1. The RODEX2-2A code was used to determine the burnup dependent initial fuel conditions for the system calculations.
2. The S-RELAP5 code was used to model the transient thermal-hydraulic response of the reactor coolant and main steam systems. The S-RELAP5 code also models the thermal-hydraulic response of the hot rod during the transient. The governing conservation equations for mass, energy, and momentum transfer are used along with appropriate correlations consistent with Appendix K of 10 CFR 50.

14.6.5.2.5.3 Plant Description and Summary of Analysis Parameters The Millstone Unit 2 nuclear power plant is a CE designed PWR which has two hot leg pipes, two U-tube steam generators, and four cold leg pipes with one RCP in each cold leg. The plant utilizes a large dry containment. The RCS is nodalized in the S-RELAP5 model into control volumes representing reasonably homogeneous regions, interconnected by flow paths or junctions. Each of the cold legs is modeled separately. The model considers four safety injection tanks, a pressurizer, and two steam generators with both primary and secondary sides of the steam generators modeled. The HPSI pumps were modeled as sources injecting directly into all RCS cold legs, with conservative flows given as a function of system backpressure. The LPSI pumps were modeled as sources injecting directly into two of the RCS cold legs, with conservative flows given as a function of system backpressure. The pumped ECCS flow rates as a function of cold 14.6-18 Rev. 35

MPS2 UFSAR leg pressure are based on a detailed Safety Injection System flow calculation which includes pump degradation. No charging pump flow was credited in the analysis. The primary coolant pump performance curves are characteristic of the Millstone Unit 2 pumps. Symmetrical steam generator tube plugging of 500 tubes per steam generator was assumed. The analysis assumed loss of offsite power concurrent with reactor scram. The single failure criterion required by Appendix K was satisfied by assuming the loss of one Emergency Diesel Generator (EDG), which resulted in the disabling of one HPSI pump, one LPSI pump and one motor-driven auxiliary feedwater (AFW) pump. The swing HPSI pump was not credited, leaving only a single HPSI pump in operation. Initiation of the HPSI and LPSI systems were delayed by 25 and 25 seconds, respectively, beyond the time of SIAS representing the maximum Technical Specification delay time required for EDG startup, switching, and pump startup. The disabling of a motor-driven AFW pump would leave one motor-driven pump and the turbine-driven pump available. The initiation of the motor-driven pump was delayed 240 seconds beyond the time of the Auxiliary Feedwater Actuation Signal (AFAS) indicating low steam generator level (0% narrow range). The operator startup of the turbine-driven AFW pump was not credited in the analysis. The analysis assumes a constant minimum AFW flow rate of 72 gpm per SG. The sweep out of fluid at the main feedwater temperature in the piping between the AFW injection location and the steam generators is also accounted for. The analysis supports a total unrodded integrated radial peaking factor (FrT) of 1.854 (1.69 plus uncertainties), and a maximum Linear Heat Rate (LHR) of 15.1 kW/ft. All four RCPs were assumed to trip coincident with reactor scram, consistent with a loss of offsite power at the time of trip. Values for system parameters used in the analysis are given in Table 14.6.5.2-3. 14.6.5.2.5.4 Break Spectrum The Millstone Nuclear Plant Unit 2 break spectrum analysis for SBLOCA includes breaks of varying diameter up to 10% of the flow area for the cold leg. The spectrum includes a wide enough range of break sizes from 2.0 inch diameter to 9.49 inch diameters to establish a PCT trend. Additional break sizes are performed with a smaller break interval fine enough to identify the limiting break size and to capture different recovery phenomena. The limiting break size was determined to be 3.78-inch diameter (0.07793 ft2), resulting in a PCT of 1707°F. The PCT results for the break spectrum are presented in Table 14.6.5.2-6. Figure 14.6.5.2-1 shows the calculated PCTs for the break spectrum. The predicted event times for the break spectrum are provided in Table 14.6.5.2-5. For the break spectrum analysis, RCP trip is assumed to occur on reactor trip. The break opens at 0 seconds and initiates a subcooled depressurization of the primary system. The low pressurizer pressure trip setpoint is reached at 19 seconds and within 2 seconds the reactor is tripped. Offsite power is lost, coincident with the turbine trip, RCP trip, and MFW pump trip (Figure 14.6.5.2-2, Figure 14.6.5.2-11, Figure 14.6.5.2-12 and Table 14.6.5.2-5). The SIAS is issued at 27 seconds on low pressurizer pressure. As MFW to the SGs is ramped down, 14.6-19 Rev. 35

MPS2 UFSAR the pressure in the SGs increase for approximately 30 seconds until the MSSV inlet reaches the lowest opening pressure setpoint. This provides core heat removal in the early stages of the transient. The primary system depressurization continues at a relatively fast rate for the first 125 seconds as fluid rushes out of the break (Figure 14.6.5.2-3). The broken leg loop seal clears at 356 seconds (Figure 14.6.5.2-6), as demonstrated in the horizontal loop seals and break void fractions in Figure 14.6.5.2-6 and Figure 14.6.5.2-5, respectively. Prior to loop seal clearing in the broken leg, the core uncovers about 4 feet below the top of the active fuel (Figure 14.6.5.2-9, Figure 14.6.5.2-20 and Figure 14.6.5.2-21). As there is no loop flow, a large amount of steam is generated and accumulated in the core by the decay heat until enough pressure is built to blow the upflow leg of the loop seal in the broken leg around 356 seconds into the transient. This causes an abrupt level drop in the downcomer region (Figure 14.6.5.2-8) with a simultaneous core recovery (Figure 14.6.5.2-9). As the broken leg clears, the plant then enters a fairly slow boil-off phase where mass is lost out the break, and the primary system continues to empty. All intact loops remained plugged for the duration of the transient. As liquid drains out of the loop piping, the break flow transitions from liquid to two phase flow, and then to steam. The break flow becomes primarily steam around 366 seconds resulting in a reduced mass flow rate out of the break (Figure 14.6.5.2-4) and an increase in the depressurization rate of the primary system (Figure 14.6.5.2-3). The liquid level in the reactor vessel continues to drop until the reactor vessel reaches a minimum level at 1386 seconds (Figure 14.6.5.2-10). Although HPSI flow to the primary system cold legs began at approximately 62 seconds into the transient (Figure 14.6.5.2-16), it does not provide sufficient inventory at this time to offset the large amounts lost out the break at this time. As effective cooling is lost in the core, the fuel rods begin to heat up at approximately 800 seconds (Figure 14.6.5.2-22). The fuel continues to heat up until the maximum PCT of 1707°F is reached at 1824 seconds. Fuel rod rupture does occur for the hot rod, the calculated blockage factor indicates that the channel around the hot rod is not completely blocked and that all other channels in the core are also not completely blocked. Therefore, the hot rod and all other channels in the core are amenable to cooling. For this break size, the HPSI flow is eventually sufficient to compensate for the rate of inventory loss out of the break. At the time of the PCT, the primary system has de pressurized to a pressure slightly above the SIT pressure and the LPSI shut-off head. SIT injection begins at 4580 seconds (Figure 14.6.5.2-18), followed by LPSI injection 56 seconds later (Figure 14.6.5.2-17), resulting in no influence on PCT turnaround. The downcomer level (Figure 14.6.5.2-8) and the reactor vessel inventory (Figure 14.6.5.2-10) start slowly increasing at approximately 1400 seconds. The onset of SIT and LPSI injection helps the reactor vessel levels to step up, ensuring core recovery and long term core cooling. 14.6-20 Rev. 35

MPS2 UFSAR In conclusion, the limiting PCT break spectrum case is a 3.78-inch diameter cold leg break. The PCT of this case is 1707°F. The transient maximum local oxidation (MLO) is 3.6% and the maximum corewide oxidation (CWO) is less than 0.04%. The total maximum local oxidation is less than 6%, including a pre-transient oxidation of 2.3%. The hot rod resulted in rupture, but remained amenable to cooling. The results of the analysis demonstrate the adequacy of the ECCS to support the 10 CFR 50.46(b) (1-4) criteria (Reference 14.6-14). 14.6.5.2.5.5 Pump Trip Delay Results For plants such as Millstone Nuclear Plant Unit 2 that do not have an automatic RCP trip, a delayed RCP trip can potentially result in a more limiting condition than tripping the RCPs at reactor trip. Continued operation of the RCPs may result in earlier loop seal clearing with associated two-phase flow out the break, which would result in less inventory loss out the break early in the transient, but in the longer term could result in more overall inventory loss out the break. It has been postulated that tripping the pumps when the minimum RCS inventory occurs could cause a collapse of voids in the core, thus depressing the core level and provoking a deeper core uncovery, and a potentially higher PCT. Therefore, an RCP trip sensitivity for both the cold and hot leg breaks was performed with a delay time following the loss of subcooling margin in the cold leg to demonstrate 10 CFR 50.46(b)(1-4) criteria. This manual RCP trip study was performed consistent with the Combustion Engineering Owners Group guidelines described in Generic Letter 86-06 (Reference 14.6-10) where compliance with 10 CFR 50.46 is demonstrated when operator action to trip the RCPs is taken within 2 minutes after the RCP trip criterion is reached using the 10 CFR 50 Appendix K method. Also, consistent with Generic Letter 86-06, additional delayed RCP trip sensitivity studies were performed to determine the maximum delay time for operator action under a more realistic scenario. Best-estimate assumptions were applied using the same model as the Appendix K RCP analysis with relaxation in two areas: decay heat multiplier reduction from 1.2 to 1.0 and critical break flow model change from Moody to the Homogeneous Equilibrium Model. A range of RCP trip delay times was examined for both hot and cold leg break locations. The results of the delayed RCP trip sensitivity, with Appendix K assumptions demonstrated that all four RCPs can be tripped by the operator within 2 minutes after subcooling margin is lost in the cold leg pump suction in order to meet the 10 CFR 50.46(b)(1-4) criteria. In addition, relaxation of Appendix K assumptions demonstrated that longer delay times of up to 10 minutes could be accommodated and still meet the 10 CFR 50.46(b)(1-4) criteria. 14.6.5.2.5.6 Asymmetric Steam Generator Tube Plugging Small break LOCA analyses have no significant sensitivity to steam generator tube plugging. The analysis was performed using the maximum allowed total tube plugging of 1000 tubes applied symmetrically (500 tubes per steam generator), and supports the maximum allowed asymmetry between the generators of 500 tubes. By applying the maximum allowed tube plugging, the analysis is conservative because the initial primary system mass inventory is minimized. 14.6-21 Rev. 35

MPS2 UFSAR 14.6.5.2.5.7 Reduced Primary Temperature Operation An assessment to support an end of cycle full power coastdown to an indicated RCS cold leg temperature of 537°F has been performed. The result of the assessment confirmed that break spectrum analysis bounds operation during an end of cycle full power coastdown to an indicated RCS cold leg temperature of 537°F. 14.6.5.2.5.8 Attached Piping Break Sensitivity Study Although breaks in the attached piping are not typically PCT limiting, they do result in reduced ECCS flows available to mitigate the event. Therefore, an analysis of the limiting break size and location in attached piping was performed. For Millstone Nuclear Plant Unit 2, the limiting break location and size for an attached piping break is a double-ended guillotine break of a SIT line. The break is located in the SIT line connected to Loop 2B. For the double-ended guillotine break in the SIT line, the calculated PCT is 1239°F, which is bounded by the limiting PCT of the RCS break spectrum. The minimal HPSI and LPSI flow rates modeled were sufficient to prevent a subsequent heatup after the initial quench from the SIT discharge. 14.6.5.2.5.9 Safety Injection Low Fluid Temperature Sensitivity Study A sensitivity study was performed on which SIT and safety injection (SI) fluid temperatures were reduced to approximate nominal temperatures as opposed to the maximum temperatures used in the spectrum analysis. The result of the sensitivity study confirmed that the SIT and SI fluid temperatures used in the break spectrum are conservative. 14.6.5.2.6 Analysis Results The limiting PCT break spectrum case is a 3.78-inch diameter cold leg break. The PCT of this case is 1707°F. The transient maximum local oxidation (MLO) is 3.6% and the maximum core-wide oxidation (CWO) is less than 0.04%. The total maximum local oxidation is less than 6%, including a pretransient oxidation of 2.3%. The hot rod resulted in rupture, but remained amenable to cooling. The results of the analysis demonstrate the adequacy of the ECCS to support the 10 CFR 50.46(b) (1-4) criteria (Reference 14.6-14). The results of the delayed RCP trip sensitivity with Appendix K assumptions demonstrated that all four RCPs can be tripped by the operator within 2 minutes after subcooling margin is lost in the cold leg pump suction in order to meet the 10 CFR 50.46(b)(1-4) criteria. In addition, relaxation of Appendix K assumptions demonstrated that longer delay times of up to 10 minutes could be accommodated and still meet the 10 CFR 50.46(b)(1-4) criteria. The attached pipe break and the ECCS temperature sensitivity results are bounded by the RCS break spectrum. The sensitivity studies justify the applicability of the break spectrum as the licensing basis. 14.6-22 Rev. 35

MPS2 UFSAR Margin between the calculated PCT and the 2200°F limit of 10 CFR 50.46 is available to accommodate other permanent adjustments due to 10 CFR 50.59 Safety Evaluations and LOCA model assessments. These adjustments are summarized in the 30 day and annual 10 CFR 50.46 reporting of PCT Margin Utilization. The reporting process and attention to PCT margin assure that the PCT remains below the 2200°F limit of 10 CFR 50.46. The analysis supports full power operation at 2754 MWt (2700 MWt plus 2% uncertainty). A maximum LHR of 15.1 kW/ft and a radial peaking factor of 1.69 are supported by this analysis. The analysis demonstrates that the 10 CFR 50.46(b) criteria are satisfied for the Millstone Unit 2 reactor. 14.6.5.2.7 Post Analysis of Record Evaluations In addition to the analyses presented in this section, evaluations and reanalysis may be performed as needed to address ECCS Evaluation Model errors and emergent issues, or to support plant changes. The issues or changes are evaluated, and the impact on the peak cladding temperature (PCT) is determined. The resultant increases or decreases in PCT are applied to the analysis of record PCT. A separate study applying Zirc-4 fuel and the model updates in Supplement 1 (Reference 14.6-13) of the methodology was performed, which resulted in a +4 °F increase in PCT relative to M5. This assessment was performed to support operation utilizing AREVA fuel with Zirc-4 cladding. This increase in PCT is treated as a penalty to the Analysis of Record under the provisions of 10 CFR 50.46. The PCT, including this penalty, is presented in Table 14.6.5.2-7 for the Small Break LOCA. The current PCT is demonstrated to be less than the 10 CFR 50.46(b) requirement of 2200°F. 14.6.5.2.8 Conclusions The limiting PCT break spectrum case is a 3.78-inch diameter cold leg break. The PCT of this case is 1707°F. The transient maximum local oxidation (MLO) is 3.6% and the maximum core-wide oxidation (CWO) is less than 0.04%. The total maximum local oxidation is less than 6%, including a pretransient oxidation of 2.3%. The hot rod resulted in rupture, but remained amenable to cooling. The results of the analysis demonstrate the adequacy of the ECCS to support the 10 CFR 50.46(b) (1-4) criteria (Reference 14.6-14) given in Section 14.6.5.2.5. 14.6.5.3 Post-LOCA Long Term Cooling Following the short term mitigation actions for a large or small break LOCA (as discussed in Sections 14.6.5.1 and 14.6.5.2, respectively), long term cooling will continue to maintain the core at an acceptably low temperature. LOCA mitigation in the long term will be accomplished by the methods referred to as the Long Term Cooling (LTC) Plan. The LTC Plan (shown in Figure 14.6.5.3-1) consists of the events and actions that will assure acceptable long term core cooling and prevention of boric acid precipitation in the core region. The Post-LOCA Long Term Cooling Analysis, demonstrates that Post-LOCA Long Term core cooling and boric acid precipitation prevention can be accomplished for all LOCAs. 14.6-23 Rev. 35

MPS2 UFSAR 14.6.5.3.1 The Post-LOCA Long Term Cooling Plan Figure 14.6.5.3-1 shows the basic sequence of events for the initial automatic actions and the subsequent operator actions of the LTC Plan. The operators first action is to initiate a plant cooldown within 1 hour post-LOCA by releasing steam from the steam generators. The steam is released either through the turbine bypass system, if it is available, or through the atmospheric dump valves (ADVs). When pressurizer pressure is less than 600 psia and stable with a controlled cooldown in progress, the Safety Injection Tanks (SITs) are isolated to avoid injecting nitrogen, a non-condensable gas, into the reactor coolant system (RCS). At 8 to 10 hours post-LOCA, the operator will determine if the RCS is filled. If the RCS is filled, then natural circulation will prevent boric acid precipitation and simultaneous hot and cold leg injection will not be necessary. The operators will attempt to establish shutdown cooling (SDC) if entry conditions exist or can be established. If SDC cannot be established (whether due to single failure, SDC pressure/temperature limits unsatisfied, or RCS activity beyond appropriate limits), then steam generator (SG) cooling will be continued. If the RCS is not filled at 8 to 10 hours post-LOCA, then simultaneous hot and cold leg injection will be established to provide core flushing. The preferred method of simultaneous hot and cold leg injection is low pressure safety injection (LPSI) to the hot side (one LPSI pump to the SDC warm up line to the SDC suction line to a hot leg.) Cold side injection will be via a high pressure safety injection (HPSI) pump and the running LPSI pump. If the preferred method cannot be established, then the alternative method of simultaneous hot and cold leg injection, which is HPSI injection to the hot side (1 HPSI pump to a charging line to the pressurizer auxiliary spray line to a hot leg), will be established. Cold side injection will be via a LPSI pump. Either of these simultaneous hot and cold leg injection methods will provide flow that is adequate to cool the core and prevent boric acid precipitation. 14.6.5.3.2 Post-LOCA Long Term Cooling Equipment and Operator Actions The following discussion elaborates on equipment and operator actions that support the Long Term Cooling Plan. As stated above, simultaneous hot and cold leg injection will be required if the RCS is not filled at 8 to 10 hours post-LOCA. If simultaneous hot and cold leg injection is required, then either the preferred method (LPSI hot leg injection) or the alternative method (HPSI hot leg injection) can be established and operated despite various single failures to simultaneous hot and cold leg injection equipment. Either simultaneous hot and cold leg injection configuration will provide adequate delivery flows while ensuring acceptable HPSI and LPSI pump operation. Operator actions outside the control room will be required to realign manually operated valves. Additional operator actions outside the control room will be required in the event of either a facility Z1 or Z2 loss of power (when offsite power is unavailable). These additional operator actions are described as follows. 14.6-24 Rev. 35

MPS2 UFSAR For a failure of Facility Z1, the position of the LPSI injection valves 2-SI-615, 2-SI-625 must be known, whereas a failure of Facility Z2 requires the position of LPSI injection valves 2-SI-635 and 2-SI-645 to be known. This is required to correctly align for simultaneous hot and cold leg injection. These positions would be determined by operator actions outside the control room. For the failure of the emergency Facility Z1 AC power to the safety injection system (SIS), the operator actions to establish LPSI hot leg injection would include aligning an alternate power source to SDC suction line valve 2-SI-651. This manual aligning would require operator action outside the control room. For the failure of the emergency Facility Z2 DC power for the SIS train 2, the operator actions to establish HPSI hot leg injection would include aligning an alternate power source to charging valves 2-CH-517 and 2-CH-519. This manual aligning would require operator action outside the control room. In addition to the above operator actions directly concerned with the simultaneous hot and cold leg injection realignment and operation, the following operator actions outside the control room may be required to support the Long Term Cooling Plan. If the condenser is unavailable, the auxiliary feedwater (AFW) system and the ADVs will be used to cooldown the RCS. The cooldown will be initiated within 1 hour after the start of the LOCA. If, when initiating the cooldown, the ADVs are in a closed position, then they will be manually opened by operator action outside the control room. If the primary source of AFW to the steam generatorthe condensate storage tank (CST)becomes depleted beyond the capability of the Ecolochem system to replenish, then operator action outside the control room will be required to manually realign the Fire Protection System to supply AFW. 14.6.5.3.3 Assumptions Used in the Long Term Cooling Analysis The major assumptions used in performing the Long Term Cooling analysis are listed below:

1. No offsite power is available.
2. The worst single failure is the failure of an emergency diesel generator. As a consequence of the failure, one ECCS train, one containment spray pump and one motor-driven auxiliary feedwater pump are unavailable.
3. Plant cooldown begins at two hours post-LOCA. (The EOPs conservatively initiate the cooldown within one hour post-LOCA.)
4. The analysis assumes that a cooldown rate of 40°F/hr is maintained until the ADVs are fully open (i.e., until flow limiting of the ADVs causes the cooldown rate to decrease from 40°F/hr).
5. The SITs are isolated prior to establishing shutdown cooling.

14.6-25 Rev. 35

MPS2 UFSAR

6. The pressurizer is included in the mass that is cooled down in establishing shutdown cooling entry conditions.
7. A continuous supply of auxiliary feedwater is available for the duration of steam generator cooling. One turbine-driven and one motor-driven auxiliary feedwater pump are assumed to be in operation.
8. Initial boric acid concentrations and inventories and pump flow rates used in the boric acid precipitation analysis are selected to maximize the boric acid concentration in the core.
9. A boric acid precipitation limit of 27.6 wt% is used in the large break LOCA boric acid precipitation analysis. This is the precipitation limit in saturated water at 14.7 psia.

14.6.5.3.4 Method of Analysis The objective of the post-LOCA Long Term Cooling analysis is to demonstrate that the Long Term Cooling Plan provides conformance to 10 CFR 50.46 Criterion 5, Long Term Cooling, of the ECCS acceptance criteria (Reference 14.6-14). Conformance is demonstrated by showing that under the Long Term Cooling Plan the calculated core temperature is maintained at an acceptably low value and that the boric acid concentration in the core is maintained below its solubility limit. The Millstone 2 post-LOCA Long Term Cooling analysis was performed using the NRC accepted computer codes described in Reference 14.6-15. As described in Reference 14.6-15, the CELDA computer code is used to analyze the post-LOCA thermal-hydraulic response of the RCS for a spectrum of break sizes. The NATFLOW computer code is used to calculate RCS temperatures for the purpose of determining when the shutdown cooling entry temperature is achieved. The steam generator cooldown transient that is used as a boundary condition in CELDA and NATFLOW is calculated using the CEPAC computer code. The BORON computer code is used to calculate the boric acid concentration in the core following the LOCA. The Millstone 2 post-LOCA Long Term Cooling Plan was developed using the NRC accepted methods described in Reference 14.6-15 with the following modification. In Reference 14.6-15, RCS pressure is used as the basis for determining whether to branch to shut down cooling (or continue steam generator cooling if SDC is inoperable) or to branch to simultaneous hot and cold side injection. In the Millstone 2 Long Term Cooling Plan, RCS status (i.e., the RCS is or is not filled) is used as the basis. This approach is consistent with the ABB CE emergency procedure guidelines (Reference 14.6-16). 14.6.5.3.5 Parameters Used in the Long Term Cooling Analysis Significant core and system parameters used in the Long Term Cooling analysis are presented in Table 14.6.5.3-1. 14.6-26 Rev. 35

MPS2 UFSAR 14.6.5.3.6 Results of the Long Term Cooling Analysis Figure 14.6.5.3-1 shows the Millstone 2 Long Term Cooling Plan. At 8 to 10 hours post-LOCA, the operator determines whether or not the RCS is filled. Eight to 10 hours is used as the decision time because it provides the operator with ample time to initiate simultaneous hot and cold side injection prior to the earliest time that boric acid precipitation would occur. The left branch of the Long Term Cooling Plan applies to those break sizes for which the RCS is filled at 8 to 10 hours. For these breaks, SDC operation or continued steam generator cooling will provide heat removal and natural circulation will prevent further boric acid buildup in the core. The right branch of the Long Term Cooling Plan applies to those break sizes for which the RCS is not filled at 8 to 10 hours. For those break sizes, simultaneous hot and cold side injection is used to maintain core cooling and to provide for boric acid precipitation control. A double-ended guillotine break in the cold leg is the limiting break for boric acid precipitation control. A double-ended guillotine break is limiting because the low RCS pressure associated with such a large break minimizes the boric acid solubility limit in the core. The cold leg is the limiting break location because it requires the initiation of hot side injection in order to create a core flushing flow to control boric acid precipitation. For a cold leg break, the core flushing flow is the difference between the hot side injection flow rate and the core boiloff flow rate. As shown in Figure 14.6.5.3-3, the initiation of a hot side injection flow rate of at least 180 gpm at 13 hours post-LOCA provides a substantial and time-increasing core flushing. Figure 14.6.5.3-4 shows that with no core flushing flow, boric acid would begin to precipitate at approximately 15 hours post-LOCA. However, with a hot side injection flow rate of 180 gpm, initiated at 13 hours post-LOCA, the maximum boric acid concentration in the core is 25.4 wt% as compared to the precipitation limit of 27.6 wt%. The margin provided for the prevention of boric acid precipitation by a constant core flushing flow of 20 gpm is also shown in Figure 14.6.5.3-4. The time by which the entrainment of hot side injection by the steam flowing in the hot leg would cease was calculated to be less than 2 hours post-LOCA. Therefore, the initiation of hot side injection at 13 hours is well after the potential for the entrainment of the hot side injection has ended. In order for the most limiting configuration of simultaneous hot and cold leg injection to provide the required hot side injection flow rate of 180 gpm, the RCS pressure must be 86 psia or less. As shown in Figure 14.6.5.3-2, the RCS will not be filled at 8 hours for a break area as small as 0.01 ft2. Consequently, this is the smallest break for which simultaneous hot and cold side injection would be required. The 0.01 ft2 break was calculated to achieve a RCS pressure of 86 psia prior to 13 hours post-LOCA. Larger breaks will also reach 86 psia prior to 13 hours post-LOCA. This demonstrates that the simultaneous hot and cold leg injection configurations will provide sufficient hot side injection for all breaks for which it may be required. 14.6-27 Rev. 35

MPS2 UFSAR 14.6.5.3.7 Conclusions of the Long Term Cooling Analysis The Millstone 2 post-LOCA Long Term Cooling analysis demonstrates conformance to 10 CFR 50.46 Criterion 5 of the ECCS acceptance criteria (Reference 14.6-14) for a complete spectrum of break sizes and locations. For breaks that are small enough for the RCS to refill at 8 to 10 hours post-LOCA, shutdown cooling or steam generator cooling provides core cooling and boric acid precipitation control. For breaks that are too large for the RCS to refill at 8 to 10 hours, initiating simultaneous hot and cold side injection provides core cooling and boric acid precipitation control. A simultaneous hot and cold side injection flow rate of 180 gpm (i.e., a flow rate of 180 gpm to both the hot side and cold side of the RCS) initiated by 13 hours post-LOCA maintains the boric acid concentration in the core below the solubility limit. 14.

6.6 REFERENCES

14.6-1 Description of the Exxon Nuclear Plant Transient Simulation Model for Pressurized Water Reactors (PTS-PWR), XN-NF-74-5(A), Rev. 2 and Supplements 3-6, Exxon Nuclear Company, Richland, WA 99352, October 1986. 14.6-2 XCOBRA-IIIC: A Computer Code to Determine the Distribution of Coolant During Steady-State and Transient Core Operation, XN-NF-75-21(A), Revision 2, Exxon Nuclear Company. 14.6-3 E. Daniel Hughes, A Correlation of Rod Bundle Critical Heat Flux for Water in the Pressure Range 150 to 725 psia, IN-1412 (TID-4500), Idaho Nuclear Corporation, July 1970. 14.6-4 RETRAN-02 A Program For Transient Thermal Hydraulic Analysis of Complex Fluid Flow Systems, EPRI NP-1850-CNN, dated October 1984. 14.6-5 W. G. Counsil letter to J. R. Miller, Docket No. 50-336, dated December 12, 1983. 14.6-6 Technical Specifications for Millstone Unit 2, Docket No. 50-336, Updated through Amendment Number 116. 14.6-7 Letter, Dennis M. Crutchfield (USNRC Asst. Director division of PWR Licensing-B) to Gary M. Ward (ENC Manager, Reload Licensing), Safety Evaluation of Exxon Nuclear Company's Large Break ECCS Evaluation Model EXEM/PWR and Acceptance for Referencing of Related Licensing Topical Reports, dated July 8, 1986. 14.6-8 EXEM PWR LBLOCA Evaluation Model as defined by the following references:

a. XN-NF-82-20(A), Revision 1, and Supplements 1 through 4, Exxon Nuclear Company Evaluation Model EXEM/PWR ECCS Model Updates, Exxon Nuclear Company, Inc., Richland, WA 99352. Revision 1 dated January 1990, Supplements 1 to 4 dated January 1990.

14.6-28 Rev. 35

MPS2 UFSAR

b. XN-NF-82-07(A), Revision 1, Exxon Nuclear Company ECCS Cladding Swelling and Rupture Model, Exxon Nuclear Company, Richland, WA 99352, November 1982.
c. XN-NF-81-58(A) Revision 2, and Supplements 1 through 4, RODEX2 Fuel Rod Thermal-Mechanical Response Evaluation Model, Exxon Nuclear Company, Richland, WA 99352. Revision 2 and Supplement 2 dated March 1984, Revision 2, Supplements 3 and 4 dated June 1990.
d. XN-NF-85-16(A), Volume 1 through Supplement 3; Volume 2, Revision 1 and Supplement 1, PWR 17x17 Fuel Cooling Test Program, Exxon Nuclear Company, Inc., Richland, WA 99352, February 1990.
e. XN-NF-85-105(A), Revision 0 and Supplement 1, Scaling of FCTF-Based Reflood Heat Transfer Correlation for Other Bundle Designs, Exxon Nuclear Company, Inc., Richland, WA 99352, January 1990.

14.6-9 EMF-2087(P), Revision 0, SEM/PWR-98: ECCS Evaluation Model for PWR LBLOCA Applications, Siemens Power Corporation, August 1998. 14.6-10 Nuclear Regulatory Commission Generic Letter 86-06,

Subject:

Implementation of TMI Action Item II.K.3.5, 'Automatic Trip of Reactor Coolant Pumps,' (Generic Letter No. 86-06), May 29, 1986 14.6-11 Standard Review Plan for the Review of Safety Analysis Reports for Nuclear Power Plants, NUREG-0800, U.S. Nuclear Regulatory Commission, July 1981. 14.6-12 EMF-2328(P)(A), Rev. 0, PWR Small Break LOCA Evaluation Model, S-RELAP5 Based, March 2001. 14.6-13 EMF-2328(P)(A), Rev. 0, Supplement 1 (P)(A), Rev. 0, PWR Small Break LOCA Evaluation Model, S-RELAP5 Based, September 2015. 14.6-14 Code of Federal Regulations, Title 10, Part 50, Section 50.46, Acceptance Criteria for Emergency Core Cooling Systems for Light Water Nuclear Power Reactors. 14.6-15 CENPD-254-P-A, Post-LOCA Long Term Cooling Evaluation Model, June 1980, (Proprietary). 14.6-16 CEN-152, Rev. 04, Combustion Engineering Emergency Procedure Guidelines, October, 1996. 14.6-29 Rev. 35

ABLE 14.6.1-1 AVAILABLE REACTOR PROTECTION FOR THE INADVERTENT OPENING OF A PRESSURIZED WATER REACTOR PRESSURIZER PRESSURE RELIEF VALVE EVENT Reactor Operational Mode Reactor Protection 1 Thermal Margin/Low Pressure Trip Safety Injection Actuation Signal 2, 3 Safety Injection Actuation Signal Available Thermal Margin

  • 4 Available Thermal Margin
  • 5, 6 No Significant Consequences for these Reactor Operating Conditions Defense in depth 14.6-30 Rev. 35

BLE 14.6.1-2 DISPOSITION OF EVENTS FOR THE INADVERTENT OPENING OF A PRESSURIZED WATER REACTOR PRESSURIZER RELIEF VALVE EVENT Reactor Operational Mode Disposition 1, 2 Analyze for DNBR 3-6 Bounded by the above 14.6-31 Rev. 35

TABLE 14.6.1-3 EVENT

SUMMARY

FOR AN INADVERTENT OPENING OF A PRESSURIZED WATER REACTOR PRESSURIZER PRESSURE RELIEF VALVE Event Time (seconds) tdown Valve Open 0.00 ssurizer Relief Valve Opens 0.01 actor Scram Signal 8.21 rbine Stop Valve Closed 9.13 ak Power 9.58 DNBR 9.69 ak Core Average Temperature 9.72 am Line Safety Valves Open 12.35 ak Steam Dome Pressure 13.60 14.6-32 Rev. 35

MNPS-2 FSAR MAY, 1990 SOOG

                                                                    --..e.L I.

20001-1000 . GOO,. o I I

  • T *
  • o u o ".6 10 ta.D 16 17.6 TIme. sec
                                                                            '0 FIGURE 14.6.1-1   REACTOR POWER LEVEL FOR AN INADVERTENT OPE NING OF A PRESSURIZED WATER REACTOR PRESS URIZER PRESSURE RELIEF VAL VE (RATED POWER)

MNPS-2 FSAR MAY. 1990 100000.....- - - - - - - - - - - - - - - - - - - - - - , - - -... l'IHOO ~ 110000 i e 125000 ~ 100000 IilD tI: Q)

<:lit 7&ODO e

u Jc 10000 aooo-t-----yo---...---......- - - . . - - -....- -.....- - -......- - - 4 o y.a 10 lIL6 lA 17.0 20

                                              'ftme.aec FIGURE 14.6.1- 2  CORE AVERAGE HEAT FLUX FOR AN INADVERTENT OPENING OF A PRESSURIZED WATER REACTOR PRE:.SS URIZER PRESS URE RELIEF VALVE (RATE D POWER)

MAY. 1990 MNPS-2 FSAR

    -              -            -           -~                                                     .-.

100 - " ~ ~ til

<<)

180 B t) ' ......... .a It Jot

.. HO                                                           / .---.~           ,,~"
        -                                                  /'                   , '

~ a ....~-.... ---_ .. .,.,"'- ," NO -

    &20            ,         ,         ,        -.        T              ,             ,

o 8.1 I '1.0 10 1&.1 11 11.. 20 TIme. sec FIGURE 14.6.1 ~3 REA CTOR COOLANT SYSTEM TEMPERATURES FOR AN INADVERT ENT OPE NING OF A PRESSURIZED WATER REACTOR PRESS URI ZER PRESSURE RELiE F VALVE (RATED POWER)

MAY, 1990 MNPS-2 FSAR 1800 JHL

     **ao

'? ..... 8.00 WI ~ t) aoao

 ~

Ul II) I) et... 100O t) -2 tHO

 ~

t.Q v.l 1100 E. JUO l'OO-t----,.--....,r----,...---r--.............- -......- -.....- - - 4 o 1.6 ao 12.6 16 17.1 20

                                          'ftme.88C FIGURE 14.6.1- 4    PRESSURIZER PRESSURE FOR A ~ ! INADVERTENT OPENING OF A PRESSURIZED WATER REACTOR PRESSURIZER PRESSURE RELIEF VALVE (RATED POWER)

MNPS-2 FSAR MAY. 1990 1 o .------..--.-.,....--...- -I -

IBE

-7 -I -

           ,          I                              I               I         I o      1.0               7.0         10          d.1             10        17.1       20
                                    'ftme.lI80 FIGURE 14.6.1- 5   REACT IVITIES FOR AN INADVERTENT OPENING OF A PRESSURIZED WATER REACTOR PRESSURIZ ER PRESSURE RELIEF VALVE (RATED POWER)

Mt\Y, 1990 MNPS-2 FSAR 1100 JJmL 1m 1060 1 loaD ~ 1000 ~ In G) I) 1'16 .t G taO a0 p () (/) 860+----r---.....--...,~--....- - _ . _ - -.....--__.--__t o u I 7.1 10 ta.a 16 17.6 20 TIme, 880 FIGURE 14.6.1-6 SECONDARY PRESSURE FOR AN INADVERTENT OPENING OF A PRESSURIZED WATER REACTOR PRESSURIZER PRESSURE RELIEF VALVE (RATED POWER)

BLE 14.6.3-1 AVAILABLE REACTOR PROTECTION FOR THE RADIOLOGICAL CONSEQUENCES OF STEAM GENERATOR TUBE RUPTURE EVENT Reactor Operational Modes Reactor Protection 1 Thermal Margin/Low Pressure Trip Safety Injection Actuation Signal 2, 3 Safety Injection Actuation Signal 4-6 No Significant Consequences for These Reactor Operational Modes 14.6-39 Rev. 35

TABLE 14.6.3-2 DISPOSITION OF EVENTS FOR THE RADIOLOGICAL CONSEQUENCES OF STEAM GENERATOR TUBE RUPTURE EVENT Reactor Operational Modes Disposition 1 Analyze radiological consequences Fuel performance bounded by Event 14.6.1 2-6 Bounded by the above 14.6-40 Rev. 35

ABLE 14.6.3-3 SEQUENCE OF EVENTS FOR THE STEAM GENERATOR TUBE RUPTURE EVENT TIME 1 SETPOINT OR seconds) EVENT VALUE 0 Tube Rupture Occurs 0 Reactor Trip 0 Concurrent Loss of Offsite Power; Loss of Forced Circulation 0 Loss of Instrument Air; Loss of ADV Auto-Actuation 1 Ruptured SG MSSVs Begin to Lift (MSSVs Cycle - See 970 psia Figure 14.6.3-8) 3 Intact SG MSSVs Begin to Lift (MSSVs Cycle - See 1030 psia Figure 14.6.3-8) 9 Maximum SG Pressure Reached

                    - Ruptured SG                                               1054 psia
                    - Intact SG                                                 1054 psia 37          AFW Actuation Condition Reached on Low SG Level                  10% Narrow Range 277         AFW Delivery Starts 530         Pressurizer Empties 1200        Intact SG MSSVs Close for Final Time                                940 psia 1220        SI Flow Begins to Enter RCS 1548        Ruptured SG MSSVs Close for Final Time                              880 psia 1800 2      RCS Cooldown to Thot < 515°F Initiated Using ADVs (Manual Local Action) 3090        Pressurizer Begins to Refill 3637        Thot < 515°F Achieved; Ruptured Steam Generator Isolated Time values are rounded to the nearest second.

This is an assumed analytical time and is not a required operator action time. 14.6-41 Rev. 35

BLE 14.6.3-4 MASS RELEASES FOR THE STEAM GENERATOR TUBE RUPTURE ACCIDENT fected Steam Generator - Break Flow Time Period, hours Total Break Flow Flashed Break Flow Liquid Break Flow From To lbm lbm lbm 0 1 150,000 5,000 145,000 After 1 hour (1) 51,600 1,200 50,400 fected Steam Generator - Total Steam Flow Through MSSVs and ADVs Time Period, hours Mass Flow From To lbm 0 1 1.700E+05 1 (1) 9.200E+04 act Steam Generator - Total Steam Flow Through MSSVs and ADVs Time Period, hours Steam Flow Rate, Steam Flow, lbm lbm/minute From To 0.00 1.00 2.000E+03 1.20E+05 1.00 1.11 7.330E+03 4.84E+04 1.11 1.71 5.147E+03 1.85E+05 1.71 2.33 4.200E+03 1.56E+05 2.33 2.74 3.840E+03 9.45E+04 2.74 3.18 3.810E+03 1.01E+05 3.18 3.72 3.780E+03 1.22E+05 3.72 6.50 2.743E+03 4.58E+05 6.50 17.61 2.151E+03 1.43E+06 until RCS is on shutdown cooling - release stops 14.6-42 Rev. 35

mary to Secondary Leak Rate for Intact Steam Generator 150 gpd mary Coolant Iodine Concentration 1Ci/gm DEQ I-131 ondary Coolant Iodine Concentration 0.1 Ci/gm DEQ I-131 mary Coolant Noble Gas Concentration 1100 Ci/gm DE Xe-133 accident Spike Iodine Concentration 60 Ci/gm DEQ I-131 current Iodine Spike equivalent to 335 times iodine appearance rate at 1 Ci/gm DEQ I-131 s of Offsite Power is assumed at time of tube rupture. m Generator Partition Factors Iodine 0.01 Noble Gases 1 Particulates 0.004 ctor Coolant Minimum Mass 423,00 lbm m Generator Minimum / Maximum Mass 80,000 / 280,000 lbm Boundary Breathing Rate (m3/sec) 0 - 8 hours 3.5 E -04 8 - 24 hours 1.8 E -04 24 - 720 hours 2.3 E -04 Boundary Dispersion Factors (sec/m3) EAB: 0 - 2 hours 3.66 E -04 LPZ: 4 - 8 hours 4.80 E -05 4 - 8 hours 2.31 E -05 8 - 24 hours 1.06 E -05 24 - 96 hours 7.25 E -06 96 - 720 hours 2.32 E -06 trol Room Breathing Rate 3.5 E -04 m3/sec trol Room Isolation Time after Event Initiation pre-accident spike - 20 seconds (includes time for damper closure and radiation monitor response) 14.6-43 Rev. 35

(includes time for damper closure and radiation monitor response) trol Room Intake Prior to Isolation 800 cfm trol Room Inleakage During Isolation 200 cfm trol Room Filtered Recirculation Rate (t=1 hour, 20 sec) 2,250 cfm Control Room Intake Dispersion Factors (sec/m3) MSSV ADV 2 hours 3.03 E -3 7.40 E -3 4 hours 2.30 E -3 5.71 E -3 8 hours 2.30 E -3 5.71 E -3 24 hours 8.46 E -4 2.13 E -3

- 96 hours                 6.73 E -4       1.74 E -3
- 720 hours                5.49 E -4       1.43 E -3 trol Room Free Volume                                                   35,656 ft3 trol Room Filter Efficiency (particulate/elemental/organic)             90 / 90 / 70 % (1) e Conversion Factors                                     Federal Guidance Reports 11 and 12 70% is a conservative analysis assumption for some iodine species. Technical Specifications can support assumptions for control room filter efficiencies of 90% for all iodine species.

14.6-44 Rev. 35

BLE 14.6.3-6

SUMMARY

- RADIOLOGICAL CONSEQUENCES OF THE STEAM GENERATOR TUBE RUPTURE EVENT SGTR          EAB, rem-TEDE  LPZ, rem-TEDE  Control Room, rem-TEDE ncurrent spike   1.2E+00        1.6 E -01      4.5E+00
-accident spike 1.4E+00        1.8 E -01      4.5E+00 14.6-45                       Rev. 35

MPS-2 FSAR FIGURE 14.6.3-1 STEAM GENERATOR TUBE RUPTURE WITH THE LOSS OF OFFSITE POWER RCS TEMPERATURE VS. TIME 610 600 590 580 THOT Loop 1 570 LL Q) Q) 560 0) Q)

s. 550 Q)
!.... 540
J

-+-'

~; 530 Q)

Q. E Q) 520 I- 510 Loop 1 - 49 0t-~~~~-t-~~~~-t~~~~~t--~~~~-t-~~~~-t-~~~~--+~-r-~~~~.--R-up_tu_re_d~~ 0 500 1000 1500 2000 2500 3000 3500 4000 Time (seconds) Rev. 21.2

MPS-2 FSAR FIGURE 14.6.3-2 STEAM GENERATOR TUBE RUPTURE WITH THE LOSS OF OFFSITE POWER PRESSURIZER LEVEL VS. TIME 100 90 80 70 60 50 40 30 Pressurizer Level (%) 20 10 0 0 500 1000 1500 2000 2500 3000 3500 4000 Time (seconds) Rev. 21.2

MPS-2 FSAR FIGURE 14.6.3-3 STEAM GENERATOR TUBE RUPTURE WITH THE LOSS OF OFFSITE POWER PRESSURIZER PRESSURE VS. TIME 2.50E+03 2.25E+03 2.00E+03 1.75E+03 Pressurizer Pressure (psia) 1.50E+03 1.25E+03 1.00E+03 0 500 1000 1500 2000 2500 3000 3500 4000 Time (seconds) Rev. 21.2

MPS-2 FSAR FIGURE 14.6.3-4 STEAM GENERATOR TUBE RUPTURE WITH THE LOSS OF OFFSITE POWER STEAM GENERATOR PRESSURE VS. TIME 1100 1000 900 800 Ruptured SG 700 Steam Generator Pressures (psia) Intact SG 600 0 500 1000 1500 2000 2500 3000 3500 4000 Time (seconds) Rev. 21.2

MPS-2 FSAR FIGURE 14.6.3-5 STEAM GENERATOR TUBE RUPTURE WITH THE LOSS OF OFFSITE POWER TOTAL BREAK FLOW RATE VS. TIME 65 60 55 50 45 40 35 Total Break Flow Rate (lbm/sec) 30 25 0 500 1000 1500 2000 2500 3000 3500 4000 Time (seconds) Rev. 21.2

MPS-2 FSAR FIGURE 14.6.3-6 STEAM GENERATOR TUBE RUPTURE WITH THE LOSS OF OFFSITE POWER FLASHED BREAK FLOW VS. TIME 6 5 4 3 Flashed Break Flow (lbm/sec) 2 1 0 0 500 1000 1500 2000 2500 3000 3500 4000 Time (seconds) Rev. 21.2

MPS-2 FSAR FIGURE 14.6.3-7 STEAM GENERATOR TUBE RUPTURE WITHTHE LOSS OF OFFSITE POWER ATMOSPHERIC DUMP VALVE FLOW RATE PER STEAM GENERATOR VS. TIME 260 240 220 200 180 160 140 120 100 80 ADV Flow Rate (lbm/sec) 60 Ruptured/Intact SGs 40 20 0 0 500 1000 1500 2000 2500 3000 3500 4000 Time (seconds) Rev. 21.2

MPS-2 FSAR FIGURE 14.6.3-8 STEAM GENERATOR TUBE RUPTURE WITH THE LOSS OF OFFSITE POWER MAIN STEAM SAFETY VALVE FLOW RATES PER STEAM GENERATOR VS TIME 1000 900 800 700 600 500 400 Intact SG 300 Ruptured SG MSSV Flow Rate (lbm/sec) 200 100 0 0 500 1000 1500 2000 2500 3000 3500 4000 Time (seconds) Rev. 21.2

MPS-2 FSAR FIGURE 14.6.3-9 STEAM GENERATOR TUBE RUPTURE WITH THE LOSS OF OFFSITE POWER AUXILIARY FEEDWATER VS. TIME 45 40 35 30 25 Intact SG 20 Ruptured SG 15 AFW Flowrate (lbm/sec) 10 5 0 0 500 1000 1500 2000 2500 3000 3500 4000 Time (seconds) Rev. 21.2

BLE 14.6.5.1-1 AVAILABLE REACTOR PROTECTION FOR THE LARGE BREAK LOSS OF COOLANT ACCIDENT Reactor Operating Conditions Reactor Protection 1, 2 No credit taken for reactor trip by the Reactor Protection System (RPS) ECCS - short and long-term cooling 3-6 No significant consequences for these reactor operating conditions 14.6-55 Rev. 35

ABLE 14.6.5.1-2 DISPOSITION OF EVENTS FOR THE LARGE BREAK LOSS OF COOLANT ACCIDENT Reactor Operating Conditions Disposition 1 Analyze 2-6 Bounded by the event initiated from Mode 1 14.6-56 Rev. 35

BLE 14.6.5.1-3 MILLSTONE UNIT 2 SYSTEM ANALYSIS PARAMETERS (LARGE BREAK LOSS OF COOLANT ACCIDENT ANALYSIS) mary Heat Output, MWt 2700

  • mary Coolant Flow Rate, lbm/hr 1.36 x 108 (360,000 gpm) mary Coolant System Volume, ft3 11,000 **

erating Pressure, psia 2250 et Coolant Temperature, °F 549 actor Vessel Volume, ft3 4538 ssurizer Total Volume, ft3 1500 ssurizer Liquid Total, ft3 800 T Total Volume, ft3 (one of four) 2019 T Liquid Volume, ft3 1150.5 T Pressure, psia 238.5 T Fluid Temperature, °F 106.8 tal Number of Tubes per Steam Generator 8523 am Generator Tube Plugging 5.9% mber of Tubes Plugged (Broken Loop) 500 mber of Tubes Plugged (Double Intact Loop) 500 am Generator Secondary Side Heat Transfer Area 87,130 roken Loop), ft2 am Generator Secondary Side Heat Transfer Area, 87,130 tact Loop), ft2 am Generator Secondary Flow Rate lbm/hr 6.04 x 106 am Generator Secondary Pressure (broken loop), psia 878.4 am Generator Secondary Pressure (intact loop), psia 878.4 am Generator Feedwater Temperature, °F 435 actor Coolant Pump Rated Head, feet 271.8 actor Coolant Pump Head, feet (DIL) 230.38 *** actor Coolant Pump Head, feet (SIL,BL) 233.00 *** 14.6-57 Rev. 35

actor Coolant Pump Rated Torque, ft-lbf 31,560 actor Coolant Pump Rated Speed, rpm 892 tial Reactor Coolant Pump Speed, rpm 866.8 *** actor Coolant Pump Moment of Inertia, lbm-ft2 100,000 ximum Containment Net Free Volume, ft3 1.938 x 106 ntainment Temperature, °F 101.6 S Fluid Temperature, °F 72.8 SI Delay Time, sec 25.0 SI Delay Time, sec 45.0 Primary Heat Output used in RELAP4-EM Model - 1.02 x 2700 = 2754 MWt. Includes pressurizer total volume and 5.9% SGTP Values used in RELAP4 for initialization. 14.6-58 Rev. 35

BLE 14.6.5.1-3 MILLSTONE UNIT 2 SYSTEM ANALYSIS PARAMETERS (LARGE BREAK LOSS OF COOLANT ACCIDENT ANALYSIS SIS Delivery Curves for Loss-of-Diesel Single Failure **** RCS DIL Pressure DIL HPSI SIL HPSI BL HPSI LPSI SIL LPSI BL LPSI (psia) (gpm) (gpm) (gpm) (gpm) (gpm) (gpm) 1144.34 0.00 0.00 0.00 0.00 0.00 0.00 1100.00 61.38 30.69 30.69 0.00 0.00 0.00 1050.00 87.75 43.88 43.88 0.00 0.00 0.00 1000.00 108.65 54.33 54.33 0.00 0.00 0.00 900.00 137.89 68.95 68.95 0.00 0.00 0.00 700.00 178.35 89.18 89.18 0.00 0.00 0.00 500.00 213.13 106.57 106.57 0.00 0.00 0.00 300.00 243.43 121.22 121.22 0.00 0.00 0.00 200.00 256.67 128.34 128.34 0.00 0.00 0.00 150.00 263.57 131.31 131.59 0.00 498.35 605.38 100.00 270.04 133.80 134.54 0.00 881.92 1039.03 50.00 276.41 136.26 137.44 0.00 1156.35 1349.84 14.70 280.83 138.01 139.47 0.00 1310.58 1524.66 0.00 280.83 138.01 139.47 0.00 1310.58 1524.66

  • SIS delivery to specific loops was chosen to ensure conservative results and thus does not reflect the actual plant cold leg/SIS train arrangement. For example, the larger of the two LPSI flows under loss-of-diesel conditions was directed to the broken loop. The model is insensitive to intact loop/SIS train assignments. Analysis delivery curves shown above differ from the latest calculated values. An evaluation has shown that the analysis values are bounding.

14.6-59 Rev. 35

BLE 14.6.5.1-3 MILLSTONE UNIT 2 SYSTEM ANALYSIS PARAMETERS (LARGE BREAK LOSS OF COOLANT ACCIDENT ANALYSIS SIS Delivery Curves for Loss-of-LPSI Single Failure ***** RCS ressure DIL HPSI SIL HPSI BL HPSI DIL LPSI SIL LPSI BL LPSI (psia) (gpm) (gpm) (gpm) (gpm) (gpm) (gpm) 1144.34 0.00 0.00 0.00 0.00 0.00 0.00 1100.00 122.57 61.13 61.32 0.00 0.00 0.00 1050.00 175.22 87.38 87.65 0.00 0.00 0.00 1000.00 216.94 108.18 108.52 0.00 0.00 0.00 900.00 275.49 137.38 137.81 0.00 0.00 0.00 700.00 356.28 177.66 178.22 0.00 0.00 0.00 500.00 425.78 212.31 212.98 0.00 0.00 0.00 300.00 484.25 241.46 242.23 0.00 0.00 0.00 200.00 512.67 255.64 256.45 0.00 0.00 0.00 150.00 525.61 262.00 263.11 653.94 274.77 398.76 100.00 537.51 267.76 269.33 1155.32 512.61 664.06 50.00 549.31 273.50 275.44 1475.25 663.50 834.75 14.70 557.54 277.50 279.70 1669.90 755.14 938.85 0.00 557.54 277.50 279.70 1669.90 755.14 938.85

    • SIS delivery to specific loops was chosen to ensure conservative results and thus does not reflect the actual plant cold leg/SIS train arrangement. For example, the largest of the LPSI flows under loss-of-LPSI conditions was directed to the broken loop. The model is insensitive to intact loop/SIS train assignments 14.6-60 Rev. 35

TABLE 14.6.5.1-4 MILLSTONE UNIT 2 LARGE BREAK LOSS OF COOLANT ACCIDENT ANALYSIS Summary of PCT Results Break Cd or Limiting Single eak Configuration Size Axial Shape Failure

  • PCT (°F)

BOC NOLPSI 1662 0.4 MOC NOLPSI 1711 EOC NOLPSI 1722 BOC NOLPSI 1664 0.6 MOC NOLPSI 1759 DECLG EOC NOLPSI 1770 BOC NOLPSI 1694 0.8 MOC NOLPSI 1784 EOC NOLPSI 1806 BOC NOLPSI 1688 1.0 MOC NOLPSI 1786 EOC NODIESEL 1814 BOC NOLPSI 1606 0.8 MOC NOLPSI 1669 SECLS EOC NODIESEL 1713 BOC NODIESEL 1630 1.0 MOC NODIESEL 1732 EOC NODIESEL 1759 NOLPSI denotes single failure of one LPSI pump, and NODIESEL denotes single failure of one diesel generator. 14.6-61 Rev. 35

TABLE 14.6.5.1-5 MILLSTONE UNIT 2 LARGE BREAK LOCA ANALYSIS Summary of Results for the Limiting Cases 1.0 DECLEG EOC Loss-of-Diesel t Rod Rupture Time (seconds) 43.002 Node Elevation (feet) 9.766 Flow Area Reduction (%) 46.37 ak Cladding Temperature Temperature (°F) 1814 Time (seconds) 135.784 Elevation (feet) 10.516 tal-Water Reaction

  • Local Maximum (%) 2.364 Elevation of Local Maximum (feet) 10.516 Hot Rod Total (%) 0.390 Core Maximum (%) <1 At 450 seconds 14.6-62 Rev. 35

TABLE 14.6.5.1-6 MILLSTONE UNIT 2 LARGE BREAK LOCA ANALYSIS Sequence of Events for the Overall Limiting Case (1.0 DECLG EOC Loss-of-Diesel) Event Time (s) alysis began 0.00 eak opened 0.05 AS issued 0.7 oken loop SIT injection began 9.5 uble intact loop SIT injection began 14.9 gle intact loop SIT injection began 14.9 fill began (EOBY) 18.1 flood began (BOCREC) 31.3 el rupture occurred 43.0 oken loop SIT emptied 48.1 uble intact loop SIT emptied 51.3 gle intact loop SIT emptied 51.9 T occurred 135.8 14.6-63 Rev. 35

Delta PCT CENSING BASIS PCT (°F) (°F) Analysis of Record Peak Clad Temperature (PCT) 1814 T ASSESSMENTS (Delta PCT) Corrected Corrosion Enhancement Factor -1 ICECON Coding Errors 0 Setting RFPAC Fuel Temperature at Start of Reflood -2 SISPNCH/ujun98 Code Error 0 Error in Flow Blockage Model in TOODEE2 0 Change in TOODEE2-Calculation of OMAX 0 Change in Gadolinia Modeling 0 PWR LBLOCA Split Break Modeling 0 TEOBY Calculation Error 0 . Inappropriate Heat Transfer in TOODEE2 0 . End-of-Bypass Prediction by TEOBY 0 . R4SS Overwrite of Junction Inetia 0 . Incorrect Junction Inertia Multipliers 1 . Errors Discovered During RODEX2 V&V 0 . Error in Broken Loop Steam Generator Tube Exit Junction 0 Inertia . RFPAC Refill and Reflood Calculation Code Errors 16 . Incorrect Pump Junction Area Used in RELA4 0 . Error in TOODEE2 Clad Thermal Expansion -1 . Accumulator Line Loss Error -1 . Inconsistent Loss Coefficients Used for Robinson LBLOCA 0 . Pump Head Adjustment for Pressure Balance Initialization -3 . ICECON Code Errors 0 . Containment Sump Modification and Replacement PZR 2 . Non-Conservative RODEX Fuel Pellet Temperature 20 . Array Index Issues in the RELAP4 Code 0 CENSING BASES PCT including all PCT ASSESSMENTS = 1845°F

MNP~-L FSAR 1.0 1 I I I I 0.75 - - L ill 3: 0 CL "D ill 0.5 - - N 0 E L 0 Z 0.25 - - 0.0

         "-----~--:-------:---~-----~

I I I I I 0.0 100.0 200 .0 300.0 400 .0 500 .0 600 .0 Time (sec) FIGURE 14.6.5.1-1 NORMALIZED POWER (1.0 DECLG EOC LOSS-OF-DIESEL) FEBRUARY 1999

MNPS*2 FSAR 6000.0 OIL SIL ,-... _.- .. BL o Q) If) 4000.0 E

-0
....,Q)                     "' -- ---

o 2000.0 I ' ----.~~~-.... _ ... 0::: ---.'t...:- I

  ~

o I I l.L J I If) I If) 0.0 --11.-----------------1 o 2

         -2000.0 0.0        20.0             40.0             60.0    80 .0 100.0      120.0 Time (sec)

FIGURE 14.6.5.1-2 SAFE fY INJECTION TANK DISCHARGE RATES (1.0 DECLG EOC LOSS-OF-DJESEL) FEBRUARY 1999

MNPS-2 FSAR 40.0 I I I I I r o Q) (f) 30 .0 ~ - E

 ..0 (l)
 +-'

o 20.0 ~ __ .__ ._J_ .~; =--",.J"W"S ~~ ':"OE':" ~ r'Ir' _. "=-';-_ .~ l,.'"'E"""II'. '1:"":": s-e-_. ~ ._........ :"~ J. ~.&

                                                                                                                                                    'C""':

n::

   ~

o LL e(f)n 10.0 '- - o

2 OIL
                                                                                                                         ----- --           SIL
                                                                                                                         --- .-             8L 0.0        I            I                 I                                    I                                   I 0.0 20.0         40.0             60.0                                80.0                                100.0                     120.0 Time (sec)

FIGURE 14.6.5.1-3 HIGH PRESSURE SAFETY INJECTION FLOW RATES (1.0 DECLG EOC LOSS-OF-DIESEL) FEBRUARY 1999

MNPS -2 FSAR 200 .0 I I r *_*_l--_._._.L._._._ ..l- ._. _ ._. I

                                          ; - ~-- - - --- -- -- --- - -- - ---- - - --- --- - - - - - - - -- - - -

r1 r ,........., I 150.0 ~ I - U I Q) CfJ <, r-' E 10 0.0 ~ -

.0                                  I II Q)
 ........                          I 0        50.0  ~                                                                                             -

a:::

    ~

I 0  ! l.J.... 0.0 CfJ CfJ 0

2 -50.0 I- DIL -

SIL BL

           -100.0         I    I                  I                    I                  I 0 .0 20.0 40 .0             60.0                80.0                100 .0                 120 .0 Time (sec)

FIGURE 14.6.5 .1-4 LOW PRESSURE SAFETY INJECTION FLOW RATES (1.0 DECLG EOC LOSS-OF-DIESEL) FEBRUARY 1999

MNPS-2 FSAR 3000.0 2500.0 ~ 0 2000.0 (I) 0-Q) L 1500.0

l (I)

(I) Q) L Q 1000.0 500.0 0.0 0.0 2.5 5.0 7.5 10.0 12.5 15.0 17.5 20.0 22.5 25.0 Time (sec) FIGURE 14.6.5.1 -5 UPPER PLENUM PRESSURE DURING SLOWDOWN (1.0 DECLG EOC LOSS-Of-DIESEL) FEBRUARY 1999

MNPS-2 FSAR

'<t o
  ...-   15.0 o

Q) (f) 10.0 E

.0 Q)
-+--'

0 5.0 CL

  ~

0 u, (f) (f) 0.0 0 2

         -5.0 0.0 2.5 5.0 7.5  10.0  12.5   15.0 17.5 20.0 22.5 25.0 Time (sec)

FIGURE 14.6.5.1-6 TOTAL BREAK FLOW RATE DURING BLOWDOWN (1.0 DECLG EOC LOSS-OF -DIESEL) FEBRUARY 1999

MNPS-2 FSAR 40000.0 () Q) if) 20000.0 E

  .D Q) 0            0.0 0::

3: 0 u, if) if) - 20000.0 0

2
            -40000 .0 0.0 2.5 5.0 7.5   10.0  12.5  15.0 17.5 20.0 22.5 25.0 Time (sec)

FIGURE 14.6.5.1-7 AVERAGE CORE INLET FLOW RATE DURING BLOWDOWN (1.0 DECLG EOC LOSS-OF-DIESEL) FEBRUARY 1999

MNP~ ... FSAR 200 .0 oQ) (f) 100.0 E

 ..0 Q) 0          0.0 0::::

3 0 1.L (f) (f) - 100.0 0

2
            -200.0 0.0 2.5 5.0 7.5  10.0  12.5   15.0 17.5 20.0 22.5 25.0 Time (sec)

FIGURE 14.6.5.1-8 HOT CHANNEL INLET FLOW RATE DURING SLOWDOWN (1.0 DECLG EOC LOSS-OF-DIESEL) FEBRUARY 1999

MNP~ -2 FSAR 0.8 -+-' 0

J a 0 .6 "D
J LL 0.4 0.0 2.5 5.0 7.5 10.0 12.5 15.0 17.5 20.0 22.5 25.0 Time (sec)

Fluid Quality Adjacent to peT -Node FIGURE 14.6.5.1 -9 PEAK Cl/,DDING TEMPERATURE NODE FLUID QUALITY DURING SLOWDOWN (1.0 DEClG EOC lOSS-OF-DIESEl) FEBRUARY 1999

MNPS-l fSAR I Avg Fuel

                                                                                             - -----. Clad Surface

_._. - Fluid 1500.0

                                                   " ... , ---\
     ---..                         ~~   ~

__ __ ' . r" \, LL

                             " ,/'                                \ '---                               .>: -_.-_..-- -,.

Q) l..  ! -------------------- j--- i\ i1~! Ii

J 0 1000.0  :

l.. , II ii' , '11.11

                                                                                                                       'i" i" Q) 0...             ,:'       (\                                                                   I , \ I ii! i 1\

I II \1' '1 /1 E Q)  : i \ ~ I " !I,!! ,, f-  !'I t; \f Ii ~ I J! I ~l

                      ~'-._.i \*-          ._ .r
                                                      .'\'                                                        ~ ' 1' 11 50 0.0                                       -- . -.-                                     ~                  I
                                                                         ' - '- '- .-.-.         ,,' r                     ill"
                                                                                          - .-.n... ;                      'j i 1..-

0.0 2.5 5.0 7.5 10.0 12.5 15.0 17.5 20.0 22.5 25.0 Ti me (sec) fIGURE 14.6.5.1-10 PEAK CLADDING TEMPERATURE NODE FUEL (AVERAGE) , CLADDING AND FLUID TEMPERATURES DURING BLOWDOWN (1.0 DECLG EOC LOSS-Of-DIESEL) FEBRUARY 1999

MNPS-Z FSAR

   --.. 300.0 ll..

I N

    -+-'

4-250.0 I L. J::

J
    -+-'  200.0 m
   '-../
    -+-'

C Q) 150 .0 () 4-4- Q)

    <3    100.0 c

o l.- f- 50.0

    -I-'

o Q) I O. a L...J......L-.L-.L-.L-.L-.L-L..l--L..J......L..l--L..J......L..J......L..J......L-L..JL-L..J.....J....J.....J....J.....J....J.....J....J.....J....J.....J....J.....J....J-LJ-LJ--'-L...J.....I....J.....I.-L...J 0.0 2.5 5.0 7.5 10.0 12.5 15.0 17.5 20.0 22.5 25.0 Time (sec) FIGURE 14.6.5.1-11 PEAK CLADDING TEMPERATURE NODE HEAT TRANSFER COEFFICIENT DURING BLOWDOWN (1.0 DECLG EOC LOSS-OF-DIESEL) FEBRUARY 1999

MNPS-2 FSAR v o

        ..- 60.0 50.0 N
     '+-

I 40.0 L..

    ..c
J CO
   "--"      30.0 X
J 1..L
   ......,   20.0 0

Q) I 10.0 0.0 0.0 2.5 5.0 7.5 10.0 12.5 15.0 17.5 20.0 22.5 25.0 Time (sec) FIGURE 14.6.5.1-12 PEAK CLADDING TEMPERATURE NODE HEAT FLUX DURIN G BLOWDOWN (1.0 DECLG EOC LOSS-OF-DIESEL) FEBRUARY 1999

MNPS-2 FSAR 50 .0 40 .0 0 C/) Q.. (l) L

        .30.0
)

C/) (f) (l) L o, 20.0 10.0 0.0 100.0 200.0 300 .0 400.0 500.0 600.0 Time (sec) FIGURE 14.6.5.1-13 CONTAINMENT PRESSURE (1.0 DECLG EOC LOSS -OF-DIESEL) FEBRUARY 1999

MNPS-2 FSAR 45.0 42 .5 0 40.0 (f) Q. Q) L 37.5

J (f)

(f) Q) L 0.. 35 .0 32 .5 30.0 0.0 100 .0 200.0 300.0 400.0 50 0.0 600.0 Time (sec) FIGURE 14.6.5.1-14 UPPER PLENUM PRESSURE (1.0 DECLG EOC LOSS*OF-DIESEL) FEBRUARY 1999

MNP~-L FSAR

       .30.0                      r                       I                         I                       I                 I               T 25.0   ~                                                                                                                                  -
-+-'

4- '-.../ 20.0 f-- r - Q) Q)

--.J   15.0   ~

Q) I....

J
+-'

x 10.0 ~ - L f-- 5.0 -

o. 0 1 1 I I I l-.-J1-J-L--'----'--'--'--..L.-..;L.--J.-..I..-.1........._-l.-...L.-.l.-Jl-..I.-..I..-.1...-L-l.-...L.-.l.-J~_.1.........I__L_..l._

0.0 100.0 200.0 .300.0 400.0 500.0 600.0 Time (sec) FIGURE 14.6.5.1-15 DOWNCOMER MIXTURE LEVEL (1.0 DECLG EOC LOSS-OF*DIESEL) FEBRUARY 1999

MNPS-2 fSAR ,,--.... 9.0 o Q) (f) c 7.5 Q) 0 o: 6.0 01 C

 -0 0        4.5 0

LL Q) o 3.0 U Q)

 ...... 1.5 o

Q)

 '+-
 '+-

W 0.0 0.0 100.0 200.0 300.0 400 .0 500.0 600.0 Time (sec) fiGURE 14.6.5.1-16 CORE EFFECTIVE FLOODING RATE (1.0 DECLG EOC LOSS-Of-DIESEL) FEBRUARY 1999

MNPS-2 FSAR 6.0

-+-'
"+-

Q) Q)

-l     4.0 Q)
  ~
l
-+-'

X

2 2.0 0.0 100 .0 200 .0 300.0 400.0 500.0 600 .0 Time (sec)

FIGURE 14.6.5.1-17 CORE MIXTURE LEVEL (1.0 DECLG EOC LOSS-OF-DIESEL) FEBRUARY 1999

MNPS-2 FSAR 12.0 10.0

'+-     8.0 Q)

Q) -l 6.0 ..c o c Q)

J 4.0 a

2.0 0.0 0.0 100 .0 200.0 300 .0 400.0 500 .0 600 .0 Time (sec) FIGURE 14.6.5.1-18 CORE QUENCH LEVEL (1.0 DECLG EOC LOSS ~OF -DIESEL) FEBR UARY 1999

MNPS-2 FSAR 2000 .0 1750.0 LL , 500.0 Q)

        ~                           ,
                                     \
J \
      +J                                 \                                          peT -node Clad Temp 1250.Q                      ,
J \

I. .

                                               ,,                                   Rupture-node Clad Temp Q)

Q ,, E ......... , Q) 1000.0 I-750 .0 500 .0 0.0 50.0 100.0 150.0 200.0 250.0 300.0 350.0 400.0 450.0 500.0 Time (sec) FIGURE 14.6.5.1-19 PEAK CLADDING TEMPERATURE NODE AND RUPTURED NODE CLADDING TEMPERATURES (1.0 DECLG EOC LOSS-OF-DIESEL) FEBRUARY 1999

BLE 14.6.5.2-1 AVAILABLE REACTOR PROTECTION FOR THE SMALL BREAK LOSS OF COOLANT ACCIDENT Reactor Operational Mode Reactor Protection 1 Thermal Margin/Low Pressure Trip Low Reactor Coolant Flow Trip Safety Injection Actuation Signal 2 Safety Injection Actuation Signal 3-6 No Significant Consequences for These Reactor Operating Conditions 14.6-84 Rev. 35

ABLE 14.6.5.2-2 DISPOSITION OF EVENTS FOR THE SMALL BREAK LOSS OF COOLANT ACCIDENT Reactor Operational Mode Disposition 1 Analyze 2-6 Bounded by the Event Initiated from Mode 1 14.6-85 Rev. 35

TABLE 14.6.5.2-3 MILLSTONE UNIT 2 SMALL BREAK LOSS OF COOLANT ACCIDENT SYSTEM ANALYSIS PARAMETERS actor Power, MWt 2754 ial Power Shape - ak LHR, kW/ft 15.1 dial Peaking Factor (1.69 plus uncertainties) 1.854 S Flow Rate, gpm 360,000 ssurizer Pressure, psia 2250 re Inlet Coolant Temperature, °F 549 T Pressure, psia 214.7 T Fluid Temperature, °F 120 T Water Volume, ft3 1135 ximum SG Tube Plugging Level per SG, % 5.87 Secondary Pressure, psia 880 FW Temperature, °F 435 W Flow Rate per SG, gpm 72 W Temperature, °F 70 w-Low SG Level Setpoint, % Narrow Range Span 0 W Delay, sec 240 SI and LPSI Fluid Temperature, °F 140 ssurizer Pressure - Low Reactor Trip Setpoint (RPS), psia 1700 actor Trip Delay Time on Low Pressurizer Pressure, sec 0.9 A Holding Coil Release Delay Time, sec 0.5 AS Activation Pressurizer Pressure Setpoint (Harsh Environment 1500 nditions), psia SI Pump Delay Time on SIAS, sec 25 SI Pump Delay Time on SIAS, sec 45 SSV Lift Pressure and Tolerance Nominal+ 3% 14.6-86 Rev. 35

ACCIDENT SYSTEM ANALYSIS PARAMETERS High Pressure Safety Injection Flow Rate for Cold Leg Breaks RCS Cold Leg Pressure (psia) Loop 1A (gpm) Loop 1B (gpm) Loop 2A (gpm) Loop 2B (gpm) .7 143 142 145 145 141 140 142 142 0 138 138 139 139 0 135 135 135 135 0 132 132 132 132 0 125 125 125 125 0 109 109 109 109 0 92 92 92 92 0 71 71 71 71 00 59 59 59 59 50 52 52 52 52 00 43 43 43 43 50 30 30 30 30 90 15 15 15 15 04 0 0 0 0 14.6-87 Rev. 35

ACCIDENT SYSTEM ANALYSIS PARAMETERS Low Pressure Safety Injection Flow Rates for Cold Leg Breaks CS Cold Leg Intact Loop 1A ressure (psia) (gpm) Intact Loop 1B (gpm) Intact Loop 2A (gpm) Intact Loop 2B (gpm) .7 1314 0 0 1369 1164 0 0 1214 0 904 0 0 945 0 519 0 0 546 0 0 0 0 0 14.6-88 Rev. 35

14.6-89 Rev. 35 Motor driven AFW Minimum RV mass Hot Rod rupture Time of core Non-condensable gas PCT (°F) Break opens SIAS issued PCT occurs AFAS Available HPSI available LPSI available Break uncovers uncovery occurs Low PZR Pressure HPSI Flow Begins LPSI Flow Begins Loop seal 1B clears Loop seal 2B clears occurs Loop seal 1A clears Loop seal 2A clears on, Initial Break diameter (in) Reactor Scram, RCP SIT injection begins 2.00 1135 0 66 67 79 90 104 124 242 -- 516 -- -- -- 1088 1100 -- 4116 -- 4330 3030 3.00 1384 0 29 30 38 54 63 83 98 -- 418 -- -- -- 512 552 -- 2082 -- 2451 280 3.60 1472 0 20 22 29 46 54 74 68 -- 414 -- -- -- 384 394 5700 1638 -- 2008 208 3.70 1501 0 19 21 27 46 52 72 64 -- 412 -- -- -- 364 374 5102 1572 -- 1978 200 3.75 1599 0 19 20 27 46 52 72 64 -- 422 -- -- -- 360 370 4660 1458 1772 1892 190 3.76 1651 0 19 20 27 46 52 72 62 -- 420 -- -- -- 358 368 4638 1428 1654 1857 188 3.78 1707 0 19 20 27 44 52 72 62 -- 408 -- -- -- 356 366 4580 1386 1563 1824 186 3.785 1690 0 19 20 27 44 52 72 62 -- 408 -- -- -- 356 366 4534 1396 1591 1852 186 3.79 1476 0 18 20 26 44 51 71 62 -- 408 944 -- 1950 1330 374 -- 1680 -- 2035 186 3.90 1606 0 18 19 25 44 50 70 58 -- 434 -- 366 2340 1176 350 1944 1598 1852 1954 192 4.02 1596 0 17 18 24 42 49 69 56 -- 430 700 -- 1250 1278 340 1738 1452 1684 1746 184 4.40 1581 0 14 16 22 40 47 67 48 -- 442 600 -- 1230 954 292 1278 1230 1270 1284 154 4.60 1621 0 13 15 20 40 45 65 46 -- 514 250 250 -- 478 272 1068 1072 1047 1076 144 4.80 1637 0 13 14 19 40 44 64 46 956 -- -- 480 -- 574 260 934 936 907 939 136 5.00 1615 0 12 13 19 38 44 64 44 848 -- -- 430 710 506 260 824 830 814 832 128 5.30 1591 0 11 13 18 38 43 63 44 714 -- -- 290 250 224 232 692 698 691 699 118 5.50 1398 0 11 12 17 38 42 62 42 692 -- 436 204 270 190 212 670 674 -- 674 112 6.00 1535 0 10 12 16 38 41 61 42 514 -- -- 186 180 180 188 496 500 -- 501 102 7.00 1510 0 9 11 14 38 39 59 40 360 -- 258 264 180 112 146 350 352 -- 353 74 8.00 1540 0 9 10 13 218 38 58 40 264 -- -- 98 116 86 114 254 256 -- 258 56 9.00 1399 0 9 10 13 -- 38 58 38 206 -- 70 86 66 102 84 198 202 -- 202 46 2 9.49 1437 0 9 10 12 -- 37 57 38 184 -- 74 76 60 84 78 178 180 -- 180 42 14.6-90 Rev

eak diameter (in) 2.00 3.00 3.60 3.70 3.75 3.76 eak Area (ft2) 0.02182 0.04909 0.07069 0.07467 0.07670 0.07711 ak Clad Temperature 1135 1384 1472 1501 1599 1651

)

me of PCT (sec) 4330 2451 2008 1978 1892 1857 me of Rupture (sec) -- -- -- -- 1772 1654 ansient MLO (%) 0.0864 0.4249 0.6647 0.7543 1.9966 2.6718 tal MLO (%) 2.337 2.676 2.915 3.005 4.247 4.922 re Wide Oxidation (%) 0.0042 0.0111 0.0124 0.0139 0.0254 0.0314 T Elevation (ft) 10.52 10.52 10.77 10.77 11.02 11.02 eak diameter (in) 3.78 3.785 3.79 3.90 4.02 4.40 eak Area (ft2) 0.07793 0.07814 0.07834 0.08296 0.08814 0.10559 ak Clad Temperature 1707 1690 1476 1606 1596 1581

)

me of PCT (sec) 1824 1852 2035 1954 1746 1284 me of Rupture (sec) 1563 1591 -- 1852 1684 1270 ansient MLO (%) 3.5273 3.2474 0.6689 1.3276 1.1046 0.6721 tal MLO (%) 5.778 5.498 2.920 3.578 3.355 2.923 re Wide Oxidation (%) 0.0396 0.0368 0.0119 0.0175 0.0162 0.0110 T Elevation (ft) 11.02 11.02 10.77 11.02 11.02 10.52 eak diameter (in) 4.60 4.80 5.00 5.30 5.50 6.00 eak Area (ft2) 0.11541 0.12566 0.13635 0.15321 0.16499 0.19635 ak Clad Temperature 1621 1637 1615 1591 1398 1535

)

me of PCT (sec) 1076 939 832 699 674 501 me of Rupture (sec) 1047 907 814 691 -- -- ansient MLO (%) 0.7874 0.8477 0.6456 0.5072 0.1142 0.2316 tal MLO (%) 3.038 3.098 2.896 2.758 2.365 2.482 re Wide Oxidation (%) 0.0108 0.0111 0.0091 0.0075 0.0020 0.0051 T Elevation (ft) 10.77 10.77 10.52 10.52 10.27 10. 27 14.6-91 Rev. 35

eak diameter (in) 7.00 8.00 9.00 9.49 eak Area (ft2) 0.26725 0.34907 0.44179 0.49120 ak Clad Temperature (°F) 1510 1540 1399 1437 me of PCT (sec) 353 258 202 180 me of Rupture (sec) -- -- -- -- ansient MLO (%) 0.2133 0.2247 0.1097 0.1367 tal MLO (%) 2.464 2.475 2.360 2.387 re Wide Oxidation (%) 0.0049 0.0046 0.0018 0.0024 T Elevation (ft) 10. 27 10. 27 10.02 10.02 14.6-92 Rev. 35

Delta PCT PCT (°F) CENSING BASIS (°F) Analysis of Record Peak Clad Temperature (PCT) 1707 T ASSESSMENTS (Delta PCT) Zirc-4 Cladding Assessment 4 CENSING BASES PCT including all PCT ASSESSMENTS = 1711°F 14.6-93 Rev. 35

MPS2 UFSAR FIGURE 14.6.5.2-1 PEAK CLADDING TEMPERATURE VERSUS BREAK SIZE (SBLOCA BREAK SPECTRUM) 14.6-94 Rev. 35

MPS2 UFSAR FIGURE 14.6.5.2-3 PRIMARY AND SECONDARY SYSTEM PRESSURES - 3.78-INCH BREAK 14.6-96 Rev. 35

MPS2 UFSAR FIGURE 14.6.5.2-6 LOOP SEAL VOID FRACTION - 3.78-INCH BREAK 14.6-99 Rev. 35

MPS2 UFSAR FIGURE 14.6.5.2-9 INNER AND OUTER CORE COLLAPSED LIQUID LEVEL - 3.78-INCH BREAK 14.6-102 Rev. 35

MPS2 UFSAR FIGURE 14.6.5.2-11 RCS LOOP MASS FLOW RATES - 3.78-INCH BREAK 14.6-104 Rev. 35

MPS2 UFSAR FIGURE 14.6.5.2-12 STEAM GENERATOR MAIN FEEDWATER MASS FLOW RATES

                           - 3.78-INCH BREAK 14.6-105               Rev. 35

MPS2 UFSAR FIGURE 14.6.5.2-13 STEAM GENERATOR AUXILIARY FEEDWATER MASS FLOW RATES - 3.78-INCH BREAK 14.6-106 Rev. 35

MPS2 UFSAR FIGURE 14.6.5.2-14 STEAM GENERATOR TOTAL MASS - 3.78-INCH BREAK 14.6-107 Rev. 35

MPS2 UFSAR FIGURE 14.6.5.2-15 (STEAM GENERATOR NARROW RANGE LEVEL % - 3.78-INCH BREAK 14.6-108 Rev. 35

MPS2 UFSAR FIGURE 14.6.5.2-16 (HIGH PRESSURE SAFETY INJECTION MASS FLOW RATES - 3.78-INCH BREAK 14.6-109 Rev. 35

MPS2 UFSAR FIGURE 14.6.5.2-17 (LOW PRESSURE SAFETY INJECTION MASS FLOW RATES - 3.78-INCH BREAK 14.6-110 Rev. 35

MPS2 UFSAR FIGURE 14.6.5.2-18 (SAFETY INJECTION TANK MASS FLOW RATES - 3.78-INCH BREAK 14.6-111 Rev. 35

MPS2 UFSAR FIGURE 14.6.5.2-19 (INTEGRATED BREAK FLOW AND ECCS FLOW - 3.78-INCH BREAK 14.6-112 Rev. 35

MPS-2 FSAR FIGURE 14.6.5.2-23 Deleted by PKG FSC 00-MP2-023

MPS-2 FSAR FIGURE 14.6.5.2-24 Deleted by PKG FSC 00-MP2-023

BLE 14.6.5.3-1 CORE AND SYSTEM PARAMETERS USED IN THE LTC ANALYSIS Parameter Value actor power level, MWt (102% of nominal) 2754 mber of plugged tubes per SG 1000

 / RCS cooldown rate, °F/hr (maximum)                              40 mospheric dump valve capacity at 1000 psia, lbm/hr/valve (minimum) 879,028 ric acid concentration, ppm (maximum) actor coolant system                                               2640 fueling water tank                                                 2640 fety injection tanks                                               2640 ric acid storage tanks                                            6139 tial inventory (maximum) actor coolant system, lbm                                          543,710 fueling water tank, gal                                            448,520 fety injection tanks, ft3/tank                                     1190 ric acid storage tank, ft3/tank                                   865.7 14.6-118                        Rev. 35

MNPS-2 FSAR LOCA I LOCA Lo ss- o f-C oo lan t Acc ident SIAS SIAS Sa fety Inject ion Actuation S igna l HPSI H igh Pressu re Safety Injection Pump I LPSI Low Pressure Safety Injec tio n Pum p HPSI and LPSI AF AS A uxiliary Feed wat e r Actuation S ign al Actuated (Cold Side) SIT Sa fety Injection Tank Auto SOC Shutdown Coo ling System SG Steam Generator I T ime After Stan of LOCA AFAS I "Manual " ind ica tes non-automatic functions Aux. Feed Flow Actuated Auto Yes No I Activate A ct ivate Atmospheric Turbine Bypass Stea m Du mp Manual Man ual I I I Isolate or Vent the SITs Manual Yes~ NO

                                                           .-------.;..:...<        Filled        >-~------,
                                                                                       ?

Es tablish SOC Condition s 8 hr :: t:: 10 hr In itiat e S imu ltaneous Hot and Cold Leg Inje ctio n Manual Ma nual Yes ~ Operabl e No

                                                           ?                              I Ac tuate                       ,                    Ma intain SG SOC                          t I

Heat Remova l M anual  : Existing I I ------~ Sec ure Steam Generator s Manual FIGURE 14.6.5.3-1 LONG TERM COOLING PLAN MARCH 1999

MNPS-2 FSAR \._--- ._---- - - ' --_._- ----- - --_. --- - - . . _- 16 , - -"'- --, 14 12 10

  .c fII e     8 E

i= 6 4 2 O-i--------------'-------'------------:.-----' o 0.005 0.01 0.015 0.0 2 0.025 0.03 0.035 Break Area. ft2 FIGURE 14.6.5.3-2 REACTOR COOLANT SYSTEM REFILL TIME VS. BREAK AREA MARCH 1999

MNPS-2 FSAR 350 , - - _ 300 250 200 ~ G-e 0-iQ 0:::

~

0 u.: 150 100 50 Flushing Flow:: Hot Side Injection Flow Rate - Core Boiloff O.l.--- -'-- -'-- -'-----!_-'-- -'--_----.J o 2 4 6 8 10 12 14 16 18 20 Time, hrs

                                  , _ _ Core Boiloff                 Hot Side Injection Flow Rate L - - - ._ _ . .           ~  __ . _ ..                       ~

FIGURE 14.6.5.3-3 CORE FLUSH BY HOT SIDE INJECTION FOR A DOUBLE-ENDED GUILLOTINE COLD LEG BREAK MARCH 1999

MNPS-2 FSAR 35 - - - - - --- _.. - .. _ - - - - -- --_ . . __. __ .- - - -- -- ------- -- --- - - --~ I I 30 25 - ~ 20 . .... c .~ ~ cCl o

§ 15 o

10 5 a .~ _ __'___ __'___ ____'___ __=___ __'__ -l a 2 4 6 8 10 12 14 16 18 20 Time, hrs __ - No Flushing Flow-- - -*- -- - - _____ 180 GPM HSI at 13 hrs

                  . . _. .   .20 GPM Flush ing at 13 hrs                      _. So lubility Lim it FIGURE 14.6.5.3-4 INNER VESSEL BORIC ACID CONCENTRATION VS TIME FOR A DOUBLE-ENDED GUILLOTINE COLD LEG BREAK MARCH 1999

7.1 WASTE GAS SYSTEM FAILURE s section has been moved to Section 11.1.4.4. 7.2 RADIOACTIVE LIQUID WASTE SYSTEM LEAK OR FAILURE (RELEASE TO ATMOSPHERE) s event is not in the current licensing basis for Millstone Unit 2 and therefore is not analyzed.

 .3 POSTULATED RADIOACTIVE RELEASES DUE TO LIQUID CONTAINING TANK FAILURES s event is not in the current licensing basis for Millstone Unit 2 and therefore is not analyzed.
 .4 RADIOLOGICAL CONSEQUENCES OF FUEL HANDLING ACCIDENT
 .4.1 General likelihood of a fuel handling accident is minimized by administrative controls and physical tations imposed upon fuel handling operations. All refueling operations are conducted in ordance with prescribed procedures under direct surveillance of a qualified supervisor. Also, ore any refueling operations begin, verification of complete control element assembly (CEA) rtion is obtained by tripping each CEA individually to obtain indication of assembly drop and ngagement from the drive shaft. Boron concentration in the coolant is raised to the refueling centration of 1720 ppm boron, or more per Technical Specifications and is verified by mical analysis. At the required boron concentration, the core will be more than 5 percent critical, even with all CEA's withdrawn.

er the vessel head is removed, the CEA drive shafts are removed from their respective mblies. A load cell is used to indicate that the drive shaft is free of the CEA as the lifting force pplied. maximum elevation to which the fuel assemblies can be raised is limited by the use of oders and limit switches in the fuel handling hoists and manipulators to ensure that the imum depth of water above the active fuel required for shielding is always present. This straint applies in fuel handling areas inside containment and in the spent fuel pool area. plementing the physical limits on fuel withdrawal, radiation monitors located at the fuel dling areas provide both audible and visual warning of high radiation levels in the event of a water level in the refueling cavity or fuel pool. Fuel pool structural integrity is assured by gning the pool and the spent fuel storage racks as Seismic Class I structures. design of the spent fuel storage racks and handling facilities in both the containment and fuel age area is such that fuel will always be in a subcritical geometrical array. The spent fuel pool tains a minimum of 2100 ppm of boron, and the refueling pool water contains a minimum of 14.7-1 Rev. 35

he spent fuel pool cooling system. At no time during the transfer from the reactor core to the nt fuel storage rack is the spent fuel removed from the water. l failure during refueling as a result of inadvertent criticality or overheating is not possible. possibility of damage to a fuel assembly as a consequence of mishandling is minimized by nsive personnel training, detailed procedures, and equipment design. Equipment design and inistrative controls preclude the handling of heavy objects such as shipping casks over the nt fuel storage racks with the exception of the consolidated fuel storage box or any object nded by the consolidated fuel storage box drop analysis and any single failure proof lift by the nt fuel cask crane in accordance with the guidelines of NUREG-0612. Inadvertent ngagement of a fuel assembly or consolidated fuel storage box from the fuel handling hine is prevented by mechanical interlocks. Consequently, the possibility of dropping either and damaging of a fuel assembly is remote. uld a fuel assembly be dropped or otherwise damaged during handling, radioactive release ld occur in either the containment or the auxiliary building. If ventilation is available and ndary integrity is set the ventilation exhaust air from both of these areas is monitored before ase to the atmosphere (see Section 7.5.6.3). The radiation monitors immediately indicate the eased activity level and alarm. The affected area would then be evacuated. ced ventilation is not required while handling irradiated fuel in containment or in the fuel ding, nor is containment or fuel handling area boundary integrity required. This allows any etration to the containment (including the equipment hatch and personnel access door) or fuel dling area boundaries (e.g., including roll-up doors) to be open during fuel movement. able radiological monitoring is recommended per the Millstone Effluent Control Program n boundary integrity is not set to ensure releases to the environment are monitored. re is no requirement for automatic isolation of containment purge to mitigate a release through containment purge system during fuel movement. 7.4.2 Method of Analysis the purpose of defining the upper limit on fuel damage as the result of a fuel handling dent, it is assumed that the fuel assembly or consolidated fuel storage box is dropped during dling by the spent fuel platform crane. Interlocks, procedural and administrative controls e such events unlikely. However, if assemblies are damaged to the extent that a number of rods fail, the accumulated fission gases and iodines in the fuel element gap could be released he surrounding water. Release of the particulate fission products is considered negligible due he surrounding water. fuel assemblies and consolidated fuel storage box are stored within the spent fuel rack at the om of the spent fuel pool. The top of the rack extends above the top of the stored fuel. A pped fuel assembly or consolidated fuel storage box could not strike more than one fuel mbly in the storage rack. Impact can occur only between the ends of the involved 14.7-2 Rev. 35

fuel assembly indicate that a fuel assembly in the storage rack is capable of absorbing the tic energy of the fuel assembly, heavy dummy fuel assembly, or consolidated fuel storage box p with no fuel rod failures. The worst fuel handling incident that could occur in the spent fuel l is the dropping of a fuel assembly to the fuel pool floor. It is assumed all of the fuel rods hin one fuel assembly will fail as a result of a fuel handling incident within containment or the nt fuel pool area. X/Q values have been chosen in the following manner: Site meteorological data has been mined for the years 1974 - 1981 for off site X/Qs and 1997 - 2001 for control room X/Qs. For h release point and dose calculation time period in question, the year with the largest (most servative) 95% maximum off site X/Q value has been chosen. Control room X/Qs are eloped consistent with Regulatory Guide 1.194 except for those from the Millstone Stack, ed on Regulatory Guide 1.145. each accident, the results indicate that the radiological consequences are within the criteria tified by Regulatory Guide 1.183 and 10 CFR 50.67. The limiting criteria are 6.3 rem TEDE EAB and LPZ and 5 rem TEDE for the control room.

 .4.2.1 Fuel Handling Accident in the Spent Fuel Pool s accident has been re-analyzed using the methods and assumptions contained in Regulatory de 1.183. A complete list of assumptions is provided in Table 14.7.4-1. The results of this lysis are within the limits as defined by 10 CFR 50.67 and within the criteria identified in ulatory Guide 1.183. This analysis does not require automatic initiation, isolation or lignment of main exhaust or AES ventilation system from radiation monitor response, nor s it require fuel handling area integrity.
 .4.2.2 Fuel Handling Accident in Containment s accident has been re-analyzed using the methods and assumptions contained in Regulatory de 1.183. A complete list of assumptions is provided in Table 14.7.4-2. The results of this lysis are within the limits as defined by 10 CFR 50.67 and within the criteria identified in ulatory Guide 1.183. This analysis does not require automatic isolation of purge from ation monitor response, nor does it require containment integrity because it is assumed that tainment penetrations such as the equipment hatch are open.

7.4.3 Results of Analysis

 .4.3.1 Fuel Handling Accident in the Spent Fuel Pool DE, rem EAB                                   1.5E+00 LPZ                                   2.0 E -01 14.7-3                                   Rev. 35
 .4.3.2 Fuel Handling Accident in Containment DE, rem EAB                                    1.5E+00 LPZ                                    2.0 E -01 Control Room                           3.1E+00
 .4.4 Conclusions es at the exclusion area boundary (EAB), low population zone (LPZ) and the control room are hin the requirements of 10 CFR 50.67 and within the guidelines identified in Regulatory Guide
3. Therefore, a fuel handling accident in the containment or spent fuel buildings will not ent any undue hazard to the health and safety of the public, nor will it compromise control m operations.
 .5 SPENT FUEL CASK DROP ACCIDENTS spent fuel cask crane is designed with special features to the structure that will ensure a single ure does not result in the loss of the capability of the system to safely retain the load. The raded spent fuel cask crane is designed to meet the single failure proof requirements of REG-0554 and NUREG-0612. When a spent fuel cask is rigged to the crane and handled in ordance with single failure criteria, a cask drop is not credible and need not be postulated.

refore, there will be no radiological consequences. wever, the cask drop accident that is postulated to have radiological consequences involves the kely scenerio where the spent fuel cask is disengaged from the crane in the cask laydown area tips over into the spent fuel pool damaging impacted fuel assemblies. This postulated cask p accident is referred to as a cask tip accident to avoid confusion. 7.5.1 Spent Fuel Cask Tip Accident spent fuel cask is assumed to initially be in the cask laydown area, where it is postulated to tip the spent fuel pool. All the fuel assemblies in the spent fuel pool that are within the distance L he center of the spent fuel cask laydown area are assumed to fail, where the distance L is the or dimension of the spent fuel cask. As described below, fuel assemblies in the spent fuel pool hin this distance L are assumed to have decayed for the minimum period specified in Technical cification 3.9.16. 7.5.2 Method of Analysis spent fuel cask tip accident is based upon the assumptions listed in Table 14.7.5-1. 14.7-4 Rev. 35

es. Each consolidated fuel canister or storage box contains the inventory of two individual fuel mblies. The analysis conservatively included the inventory of 33 more fuel assemblies than ally fit in the potential impact area. X/Q values have been chosen in the following manner: Site meteorological data has been mined for years 1974 - 1981 for offsite X/Qs and 1997 - 2001 for control room X/Qs. For h release point and dose calculation time period in question, the year with the largest (i.e., most servative) 95% maximum offsite X/Q value has been chosen. Control room X/Qs are eloped consistent with Regulatory Guide 1.194 except for those from the Millstone Stack, ed on Regulatory Guide 1.145. analysis considered two scenarios, an unisolated control room and an isolated control room: For an unisolated control room, the control room unfiltered ventilation flow consists of 800 cfm normal intake and 200 cfm inleakage for the entire accident. For an isolated control room, isolation occurs at 20 seconds, including radiation monitor and damper closure response times: 0-20 seconds Prior to isolation, the control room unfiltered ventilation flow of 1,000 cfm, which consists of 800 cfm normal intake and 200 cfm inleakage. 20 seconds - 1 hour 20 seconds During isolation but prior to recirculation, there is only 200 cfm unfiltered inleakage. 1 hour 20 seconds - 720 hours During isolation and recirculation, 200 cfm unfiltered inleakage and 2,250 cfm filtered recirculation continue for the remainder of the accident. unisolated control room provides the limiting dose consequences. 7.5.3 Results of Analysis DE, rem EAB 5.0E-01 LPZ 1.0E-01 Control Room 8.0E-01 results indicate, that the radiological consequences are within the criteria identified by ulatory Guide 1.183 and 10 CFR 50.67. The limiting criteria are 6.3 rem TEDE for EAB and

 , and 5 rem TEDE for the control room.

14.7-5 Rev. 35

spent fuel cask tip accident has been analyzed using the methods and assumptions contained egulatory Guide 1.183. The dose consequences at the exclusion area boundary (EAB), low ulation zone (LPZ), and the control room are within the limits as defined by 10 CFR 50.67 are within the criteria identified in Regulatory Guide 1.183. Therefore, a cask tip accident in spent fuel pool will not present any undue hazard to the heatlth and safety of the public, nor it compromise control room operations. 14.7-6 Rev. 35

14.7-7 Rev. 35 ABLE 14.7.4-1 ASSUMPTION FOR FUEL HANDLING ACCIDENT IN THE SPENT FUEL POOL Reactor Core Power Level: 2754 Mwt Iodine Pool Decontamination Factor: 200 Activity Released from Rods a) Iodines: 10% b) Noble Gases (Except Kr-85): 10% c) Kr-85: 30% c) I-131: 12% Chemical Form of Iodines Above Pool a) organic: 43% b) elemental: 57% 1 Assembly Assumed to Rupture Peaking Factor: 1.83 Decay time: 100 hours Duration of Release: 2 hours All Activity Bypasses EBFS Filters and is Released at Ground Level Using Worst Case X/Qs: Ground Level X/Qs (sec/m3) a) EAB: 3.66 E -04 b) LPZ: 4.80 E -05 c) Control Room: 3.00 E -03 Dose Conversion Factors Federal Guidance Report 11 & 12 ) Breathing Rate (m3/sec): 3.5 E -04 ) Control Room Breathing Rate 3.5 E -04 m3/sec ) Control Room Isolation Time post-accident (includes radiation 20 seconds monitor and damper closure response times) ) Control Room Intake Prior to Isolation 800 cfm ) Control Room Inleakage During Isolation 200 cfm 14.7-8 Rev. 35

) Control Room Emergency Filtered Reciculation Rate (from 1 2,250 cfm hour after isolation) ) Control Room Free Volume 35,656 ft3 ) Control Room Filter Efficiency (particulate/elemental/organic) 90 / 90 / 70 % (1) ) Dose Conversion Factors Federal Guidance Reports 11 and 12 70% is a conservative analysis assumption for some iodine species. Technical Specifications can support assumptions for control room filter efficiencies of 90% for all iodine species. 14.7-9 Rev. 35

TABLE 14.7.4-2 ASSUMPTION FOR FUEL HANDLING ACCIDENT IN CONTAINMENT Reactor Core Power Level: 2754 MWt Iodine Pool Decontamination Factor: 200 Activity Released from Rods a) Iodines: 10% b) Noble Gases (except Kr-85): 10% c) Kr-85: 30% d) I-131: 12% Chemical Form of Iodines Above Pool a) organic: 43% b) elemental: 57% 1 Assembly Assumed to Rupture Peaking Factor: 1.83 Decay Time: 100 hours Duration of Release 2 hours All Activity Bypasses EBFS Filters and is Released at Ground Level Using Worst Case X/Qs: Ground Level X/Qs (sec/m3) a) EAB: 3.66 E -04 b) LPZ: 4.80 E -05 c) Control Room: 3.00 E -03 Dose Conversion Factors Federal Guidance Report 11 & 12 Breathing Rate (m3/sec): 3.5 E -04 Control Room Breathing Rate 3.5 E -04 m3/sec Control Room Isolation Time post-accident (includes radiation 20 seconds monitor and damper closure response times) Control Room Intake Prior to Isolation 800 cfm Control Room Inleakage During Isolation 200 cfm 14.7-10 Rev. 35

Control Room Emergency Filtered Reciculation Rate (from 1 hour 2,250 cfm after isolation) Control Room Free Volume 35,656 ft3 Control Room Filter Efficiency (particulate/elemental/organic) 90 / 90 / 70 % (1) Dose Conversion Factors Federal Guidance Reports 11 and 12 70% is a conservative analysis assumption for some iodine species. Technical Specifications can support assumptions for control room filter efficiencies of 90% for all iodine species. 14.7-11 Rev. 35

Table deleted by FSARCR 02-MP2-015 14.7-12 Rev. 35

TABLE 14.7.5-1 ASSUMPTIONS FOR SPENT FUEL CASK TIP ACCIDENT PARAMETER VALUE actor Core Power Level 2,754 Mwt ine Pool Decontamination Factor 200 ction of Activity Released from Rods Iodines (Except I-131) 10% I-131 12% Noble Gases (Except Kr-85) 10% Kr-85 30% emical Form of Iodines Above Pool Organic 43% Elemental 57% ptured Assemblies & Decay Time 217 assemblies with 90 days decay and 1,376 assemblies with 5 years decay aking Factor 1 ration of Release 2 hours ound Level X/Qs EAB 3.66E-04 sec/m3 LPZ 4.80E-05 sec/m3 Control Room 3.00E-03 sec/m3 se Conversion Factors Federal Guidance Reports 11 & 12 eathing Rates EAB & LPZ (0-8 hours) 3.5E-04 m3/sec Control Room ((0-720 hours) 3.5E-04 m3/sec ntrol Room Isolation Time post-accident 20 seconds cludes radiation monitor and damper sure response times) ntrol Room Unfiltered Normal Intake 800 cfm ntrol Room Unfiltered Inleakage 200 cfm 14.7-13 Rev. 35

PARAMETER VALUE ntrol Room Emergency Filtered 2250 cfm circulation Rate (from 1 hour after isolation) ntrol Room Free Volume 35,656 ft3 ntrol Room Filter Efficiency (Particulate/ 90% / 90% / 70% (1) mental/Organic) (1) 70% is a conservative analysis assumption for some iodine species. Technical Specifications can support assumptions for control room filter efficiencies of 90% for all iodine species. 14.7-14 Rev. 35

 .1 FAILURES OF EQUIPMENT WHICH PROVIDES JOINT CONTROL/SAFETY FUNCTIONS lstone Unit 2 has no instrumentation which serves a combined function of process control and nitiation of emergency safety systems.
 .2 CONTAINMENT ANALYSIS
 .2.1 Main Steam Line Break Analysis
.2.1.1 Event Initiator he event of a Main Steam Line Break (MSLB), the release of steam into containment will lt in a rise in both temperature and pressure. The break is assumed to occur in the piping ween the steam generator and the containment wall penetration. Mass and energy releases are ted by the flow restrictor in the steam generator outlet nozzle.
.2.1.2 Protective Systems ineered Safety Features (ESF) systems which will operate to terminate the mass and energy ase to containment and suppress containment atmosphere temperature and pressure are the n Steam Isolation Signal (MSIS), Safety Injection Actuation Signal (SIAS), and Containment ay Actuation Signal (CSAS).

SIS will actuate on receipt of a containment high pressure signal to shut the following valves trip the main feedwater pumps: Steam Generator 1 & 2 Isolation Valves (HV-4217 & HV-4221 or MS-64A&B) Steam Generator 1&2 Isolation Valves Bypass (HV-4218 & HV-4222 or MS-65A&B) Steam Generator 1 & 2 Feedwater Isolation Valves (HV-5419 & HV-5420 or FW-5A&B) Steam Generator 1 & 2 Feedwater Regulating Valves (FV-5268 & FV-5269 or FW-51A&B) Main Steam Leg Low PT. Drains (HV-4193 & HV-4209 or MS-265B & MS-266B) 14.8-1 Rev. 35

(FV-5215 & FV-5216 or FW-41A&B) Feedwater Block Valve to Steam Generators 1 & 2 (HV-5263 & HV-5264 or FW-42A&B) Steam Generator 1 & 2 Feed Pump Discharge Valves (HV-5245 & HV-5247) IAS will activate the Containment Air Recirculation (CAR) fans and give a start signal to the rgency diesel generators (EDGs).

 .2.1.3 Method of Analysis omplete MSLB spectrum study has been performed to determine the limiting cases for peak tainment pressure. The NRC approved methodology (References 14.8-2 and 14.8-3) ciated with the Westinghouse SGN-III computer program was used to determine the mass and rgy releases to containment. Using these mass and energy releases, the NRC approved minion GOTHIC methodology (Reference 14.8-4) was used to determine the containment sure-temperature consequences of the MSLB.This methodology includes consideration for following: (a) inclusion of the steam line and feed line volumes into the overall determination lowdown volume available; (b) determination of temperature/pressure expansion factor for SGs and RCS to maximize the volume available for blowdown; (c) increase in feedwater flow he affected SG due to the increasing pressure imbalance between the affected and intact SG; inclusion of SG shell metal heat transfer as part of the energy release; and lastly, (e) a plete determination of the effects of different component single failures during the accident.
 .2.1.4 Major Assumptions major assumptions are as follows:
1. Offsite power is assumed to be available for most of the cases. This increases the primary to secondary heat transfer since the reactor coolant pumps (RCPs) are operating. To verify this assumption, loss of offsite power cases were included as part of the single failure analysis.
2. For determination of peak containment pressure, the initial containment pressure/

temperature is conservatively assumed to be at the Technical Specification maximum of 15.7 psia and 120°F.

3. Consistent with the NRC Standard Review Plan (SRP) Section 6.2.1.4, break spectrum studies were used to address moisture carryover.

14.8-2 Rev. 35

is operating.

5. Credit is taken for the main steam non-return valves to prevent blowdown of the unaffected SG into the containment.
6. The maximum RCS flow rate was conservatively assumed to maximize the heat transfer from the primary to secondary side.
7. Cases initiated from 0% power assumed auxiliary feedwater (AFW) is the sole source of steam generator inventory control and AFW flow to the affected steam generator is maximized from the beginning (time = 0) of the analysis. All other cases initiate AFW at a conservative minimum time of 180 seconds.
8. Reactor coolant pump heat was included.
9. All actuation signals are redundant and safety grade. In some cases credit is taken for actuation of nonsafety grade components initiated by the safety grade signals.
10. Relative humidity is assumed at 25 percent.
11. A 0.75" auxiliary steam line located between the two steam generators remains unisolated during the events. This causes the intact steam generator to continue to blowdown even after the MSIS.
12. A cavitating venturi installed in each AFW discharge line will limit AFW flow to a steam generator to 550 gpm.
13. Operator action to isolate the AFW is assumed to occur no greater than 30 minutes following MSIS.
14. The feedwater flow rate for the cases that initiate with the main feedwater system operating conservatively account the following:
  • Feedwater flow increases as the affected steam generator pressure decreases.
  • Upon receipt of a MSIS, the feedwater pumps are assumed to coast down at a rate based on plant operating experience.
  • If the ruptured steam generator pressure decreases to the discharge pressure of the still running condensate and heater drain pumps, flow will again begin to increase.

14.8-3 Rev. 35

.2.1.5 Initial Conditions and Input Data ial conditions and input data are given in Tables 14.8.2-1, 14.8.2-2 and 14.8.2-5. Table 14.8.2-ves an accounting of the amounts of steel assumed to be inside of containment. Since this erial acts as a heat sink to reduce containment temperature and pressure, minimum amounts used.
.2.1.6 Results rder to determine the limiting conditions, four different spectrum studies were performed.

se are as follows:

1. Power level and break size.
2. Feed system single failures.
3. Containment heat removal systems single failures.
4. Spectrum study for peak containment temperature.
 .2.1.6.1   Power Level and Break Size omprehensive sensitivity study was performed to determine the limiting break size for each er level. A sensitivity study was needed because of the interaction of power level with SG ntory and moisture carryover. The limiting break size at a given power level is the largest k size that would result in a pure steam blowdown, since a pure steam blowdown results in greatest amount of energy being transferred to the containment atmosphere in a short period of
e. The limiting results for each power level show that a maximum break size of 3.51 ft2 is ting for 25 percent, 50 percent, 75 percent, and 100 percent power. At 0 percent power the ting size break is 1.89 ft2.
 .2.1.6.2   Feed System Single Failures omprehensive feedwater system isolation single failure study was performed. For each single ure, a range of steady state initial power levels was analyzed, using the insights from the er Level/Break Size sensitivity study.
1. Feed Pump Failure to Trip - The failure of a feed pump to trip on MSIS results in additional feed water being pumped preferentially into the affected SG until the Feedwater Regulating Valves (FRV) and isolation valves shut. For 25 percent and 50 percent power levels, only one feedwater pump was assumed to be running when the accident commences. This event was not applicable to 0% cases (see 14.8-4 Rev. 35
2. Inadvertent Initiation of AFW Feedwater to Affected Steam Generator - Maximum AFW flow was assumed to be inadvertently initiated at the start of the event for cases initiated from greater than 0% power.
3. Feedwater Bypass Valve Fails Open - This failure is only credible when the FW bypass valve is initially open (cases initiated from 25% power and below). The failed open feedwater bypass results in additional feedwater being pumped preferentially into the affected SG until the FW pump discharge valves shut. In addition, even with the feed pump discharge valves shut, flashing in the feedwater lines continues to add energy into the affected steam generator. This effect has been taken into account.
4. Failure of Vital Bus Cabinet VA-10 or VA This failure could prevent closure of the FRVs and results in the loss of one train of the Containment Heat Removal Systems. Feedwater addition to the affected SG will continue until closure of the main feed pump discharge valves.
.2.1.6.3     Containment Heat Removal Systems Single Failures omprehensive containment heat removal systems single failure study was performed. For each le failure, a range of steady state initial power levels was analyzed, using the insights from the er Level/Break Size sensitivity study.
1. Failure of Two CAR fans to start - This failure is bounded by Section 14.8.2.1.6.2, item 4 described above.
2. Failure of one spray train to start - This failure is bounded by Section 14.8.2.1.6.2, item 4 described above.
3. Failure of the Vital Bus Transfer Mechanism - This failure results in a loss of the normal off-site power supply for the vital buses. Thus initiation of the containment sprays and CAR fans is delayed until the EDGs are powering the vital buses and auto sequencing has occurred. Since the FRVs have a backup DC power source, they are unaffected by this failure and will isolate the affected SG. The RCPs and certain other nonvital loads are also unaffected by this failure, which contributes to the severity of this accident by providing more rapid heat transfer from the primary to the affected SG.
4. Loss of Offsite Power with a Loss of One EDG - A loss of offsite power will result in loss of power to the RCPs, the condensate pumps and feedwater heater drain pumps. While only one train of containment heat removal systems is available, the loss of power to these pumps results in a greatly degraded heat transfer in the affected SG and less limiting results. Feedwater isolation will be unaffected since the FRVs are powered by DC backup power supplies.

14.8-5 Rev. 35

affected SG. With this failure, there is the potential for failure of the FRV and the other isolation valves to close. However, with the loss of the condensate and feedwater heater drain pumps, feedwater addition to the affected SG is terminated. The effect of continued energy addition to the affected SG from flashing in the feedwater lines has been taken into account.

.2.1.6.4    Maximum Containment Temperature Spectrum Study limiting peak pressure cases were re-run with the following modified assumptions to imize resultant containment temperature.
1. The initial containment pressure was reduced to 14.27 psia. This results in the maximum delay in containment spray actuation.
2. The relative humidity was increased to 100 percent.
3. The MSLB mass and energy releases model the Steam Generator steam super heating as it passes the uncovered portion of the Steam Generator tubes before exiting the break to address IE Information Notice 84-90. The containment wall re-evaporation is modeled using the GOTHIC built-in models for calculating the vaporization of the liquid in containment as described in Reference 14.8-4.
.2.1.7 Conclusions results of a MSLB initiated from 102 percent reactor power with coincident loss of offsite er and the failure of the Vital Bus Cabinet VA-10 or VA-20 produces the limiting containment k pressure of 53.8 psig. The peak containment atmospheric temperature for this case is

.1°F. With the loss of offsite power and this single failure, the condensate and feedwater heater n pumps are lost, and pumped feedwater addition to the affected steam generator is quickly inated. However, since the FRV and the other feedwater isolation valves fail to close on the IS, a significant volume of feedwater system remains connected to the affected steam erator. As the affected steam generator depressurizes, feedwater in this system flashes and s significant mass and energy to the affected steam generator. The portion of the feedwater flashes to steam is conservatively assumed to be directly added to the containment osphere separate from the mass and energy releases from the steam generator. The portion of feedwater liquid that reaches the affected steam generator as it depressurizes increases the cted steam generator liquid mass, which increases the steam generator mass and energy ases to containment. The plant response for the limiting peak pressure case is shown in ures 14.8.2-1 through 14.8.2-9 and the sequence of events is given in Table 14.8.2-4. results of MSLB initiated from 102 percent reactor power with offsite power available and failure of Vital Bus Cabinet VA-10 or VA-20produces the limiting containment peak ospheric temperature of 360.9°F. 14.8-6 Rev. 35

od of time and does not raise the containment structure above 289°F. 8.2.2 Loss of Coolant Accident Analysis

 .2.2.1 Events Analyzed nty four separate cases of Loss of Coolant Accidents (LOCAs) were analyzed with variations break locations, break sizes, single failures and availability of offsite power. Break locations lyzed are the reactor coolant pump suction leg, pump discharge leg and hot leg. Break sizes ude double-ended guillotine and slot breaks (9.82 sq. ft. area for the reactor coolant pump ion and discharge leg breaks, 19.24 sq. ft. area for a hot leg break) and smaller break sizes for leg breaks (10 sq. ft. and 2 sq. ft.) and reactor coolant pump suction and discharge leg breaks
q. ft. and 2 sq. ft.). Single failures considered are failure of an emergency diesel generator G) (for a loss of power (LOP) case) which fails 1 train of containment heat removal systems for no LOP), failure of either 1 spray system or 2 CAR fans. With an LOP 1 train of ECCS will rate, with no LOP both trains will operate.
 .2.2.2 Method of Analysis NRC approved, Westinghouse containment analysis methodology was used for the elopment of the short term mass and energy releases following a LOCA (Reference 14.8-3).

ss and energy input are provided through the End-of-Blowdown (EOB) from CEFLASH-4A, from the EOB to End-of-Post Reflood (EOPR) from FLOOD3. The long term boil-off phase s and energy input was calculated using the Dominion GOTHIC code (Reference 14.8-4). The tainment pressure and temperature response for the entire LOCA transient was calculated g the Dominion GOTHIC computer code.

 .2.2.3 Input and Assumptions
a. Containment input data, such as: heat sink area, spray flow rate, CAR fan cooler heat removal rate, spray water temperature, containment volume and initial containment temperature are the same as, or more conservative than, that used for the MSLB in Section 14.8.2.1.
b. Initial containment pressure is 15.7 psia for the peak pressure case and 14.27 psia for the peak temperature case.
c. Initial containment humidity is assumed to be 25 percent as in the MSLB for the peak pressure case and 100 percent for the peak temperature case.

14.8-7 Rev. 35

gallons.

e. Both the HPSI and the LPSI pumps operate prior to SRAS. Following SRAS, the LPSI are automatically stopped.
f. The Reactor Building Closed Cooling Water (RBCCW) is modeled with assumed flows before and after SRAS. The RBCCW is cooled by Service Water at 80°F.
g. The heat removal from the CAR fan cooler is modeled in the Dominion GOTHIC Code using a fan cooler model benchmarked to the post-LOCA specification data.

The specification identifies that one CAR fan is capable of removing a minimum of 80 million BTU/hr based on a containment air inlet temperature of 289°F and a fan flowrate of 34,800 cfm, along with a cooling water inlet temperature of 130°F and a flowrate of 2000 gpm.

h. A minimum spray flow of 1300 gpm is credited prior to SRAS, and 1350 gpm following SRAS.
.2.2.4 Results limiting LOCA with respect to maximum containment pressure was determined to be the uare foot discharge leg break with the LOP, the failure of two CAR fans and one spray train, minimum ECCS. The maximum calculated containment pressure for this case is 52.5 psig.

s pressure is rounded up to 53 psig to establish the Pa value stated in the containment leakage testing program Technical Specification. The limiting LOCA with respect to maximum tainment temperature was determined to be the 10 square foot hot leg break with the LOP, the ure of two CAR fans and one spray train and minimum ECCS. The maximum calculated tainment temperature for this case is 279.2°F. The maximum containment pressure and perature of these limiting LOCAs are bounded by the MSLB results provided in tion 14.8.2.1.

.2.2.5 Conclusion maximum containment pressure and temperature of the LOCA are less than the containment gn pressure and temperature of 54 psig and 289°F.
.3 DELETED
.4 RADIOLOGICAL CONSEQUENCES OF THE DESIGN BASIS ACCIDENT
.4.1 General OCA would increase the pressure in the containment resulting in a containment isolation and ation of the ECCS and containment spray systems. A SIAS signal automatically starts the 14.8-8                                   Rev. 35

ilable for release is consistent with the requirements of Regulatory Guide 1.183 ference 14.8-5). A SIAS also isolates the control room by closing the fresh air dampers within econds. Within 1 hour after control room isolation, the control room emergency ventilation EV) is properly aligned. CREV recirculates air within the control room through a charcoal r at 2,500 cfm (+/-10%) to remove iodines from the control room envelope. radiological consequences of a Design Basis LOCA at Millstone 2 were previously analyzed a low and high wind speed condition based on guidance from Regulatory Guide 1.4 ference 14.8-6) and SRP 6.5.3 (Reference 14.8-7). The low wind speed case was found to nd the high wind speed case. Therefore, the low wind speed case is the design basis for a CA and the high wind speed case is no longer analyzed. 8.4.2 Release Pathways release pathways to the environment subsequent to a LOCA are leakages from containment the enclosure building, which are collected and processed by EBFS and leakages from tainment and the RWST which bypass EBFS. tainment Leakage containment is assumed to leak at the design leak rate for 24 hours after the accident. After 24 rs, since the pressure has been decreased significantly, Regulatory Guide 1.183 allows for the rate to be reduced to one-half the design leakage rate. containment leakage for the first 110 seconds is assumed to bypass EBFS and is released ctly out the MP-2 containment. This is due to the fact that it takes 110 seconds for EBFS to ieve the required negative pressure in the enclosure building, thereby ensuring that leakage be into the enclosure building rather than out. FS collects most of the containment leakage and processes it through HEPA and charcoal rs and releases it up the Millstone stack. All containment leakage is collected and filtered by FS except for the small amount that is assumed to bypass EBFS and is released directly out the -2 containment. dit is taken for iodine removal due to containment sprays. The sprays are effective after 75 onds post-LOCA. The effectiveness of the sprays in removing elemental iodine ends at 3.03 rs and in removing particulate iodine at 3.23 hours. Credit is taken for iodine retention in the tainment sump based on post-LOCA sump pH 7.0 as discussed in Section 6.2.4.1. System Leakage Pathway t-accident radioactive releases from the ESF system are derived from fluid leakages assumed ng recirculation of the containment sump water through systems located outside containment. nuclide inventory assumed to be available for release from this pathway consists of 40% of core iodines. The quantity of leakage is based on the assumption that the ESF equipment leaks wice the maximum expected operational leak rate and that 10 percent of the iodine nuclides 14.8-9 Rev. 35

k. ST Backleakage Pathway t-accident radioactive releases from the ECCS system are a result of ECCS subsystems taining recirculated sump fluid backleaking to the RWST. The backflow rate to the RWST, as sult of isolation valve leakage, is predefined and time dependent. Due to this time dependency, contaminated sump fluid from backleakage does not enter into the RWST until 6.45 hours t-LOCA. Since the RWST is vented to atmosphere, the release is a result of the breathing rate he RWST due to solar heating. 8.4.3 Control Room Habitability radiation design objective of the control room is to limit the dose to personnel inside the trol room to 5 rem TEDE, during a DBA. The potential radiation dose to a control room rator is evaluated for the LOCA. The analysis is based on the assumptions and meteorological meters (X/Q values) given in a Tables 14.8.4-3 and 14.8.4-4. control room is designed to be continuously occupied for the duration of the accident, 30

s. Two basic sources of radiation have been evaluated: leakage of airborne activity into the trol room from sources described in Section 14.8.2 and direct dose from sources outside the trol room. The control room shielding serves to protect the operators from direct radiation due he passing cloud of radioactive effluent assumed to have leaked from containment, enclosure ding and the RWST. The control room walls also provide shielding protection for radiation nating from the CREV filters and containment shine.

IAS from Millstone 2 initiates control room isolation within 20 seconds by securing the fresh ntake dampers. Within 1 hour, CREV is in operation recirculating air in the control room elope through charcoal filters to remove radioactive iodines from the atmosphere. The ulated TEDE dose from a Millstone 2 LOCA is presented in Table 14.8.4-5 and is below the eral Design Criteria 19 and 10 CFR 50.67 limits. calculated TEDE dose from a Millstone 3 LOCA to the Millstone 2 control room is below the eral Design Criteria 19 limits and bounded by the dose consequences from the Millstone 2 CA. No credit has been taken for control room isolation or CREV operation. 8.4.4 Offsite Dose Computation radiological offsite dose consequences resulting from a postulated Millstone 2 LOCA are orted in Table 14.8.4-2. The offsite dose analysis shows that the consequences to the EAB hest 2 hour) and LPZ (0-30 day) are less than the limit of 25 rem TEDE as specified in CFR 50.67. The assumptions used to perform the radiological analysis are summarized in le 14.8.4-1. 14.8-10 Rev. 35

lysis shows that the offsite and control room radiological consequences are within CFR 50.67 criteria.

8.5 REFERENCES

8-1 Deleted.

-2    Preliminary Safety Analysis Report (PSAR) to CESSAR, Appendix 6B, Description of the SGN-111 Digital Computer Code Used In Developing Main Steam Line Break Mass/Energy Release Data For Containment Analysis.
-3    NRC Safety Evaluation Report - Standard Reference System CESSAR System 80, Combustion Engineering, Inc., December 1975.
-4    Dominion Topical Report DOM-NAF-3, Revision 0.0-P-A, GOTHIC Methodology For Analyzing the Response to Postulated Pipe Ruptures Inside Containment, September 2006.
-5    Regulatory Guide 1.183, Alternative Radiological Source Terms for Evaluating Design Basis Accidents at Nuclear Power Reactors, July 2000.
-6    Regulatory Guide 1.4, Assumptions used for Evaluating the Potential Radiological Consequences of a Loss of Coolant Accident for Pressurized Water Reactors Rev. 2, June 1974.
-7    SRP 6.5.3, Fission Product Control Systems and Structures.

14.8-11 Rev. 35

TABLE 14.8.2-1 CONTAINMENT DESIGN PARAMETERS Internal dimensions (feet) linder wall diameter 130.0 linder wall height 132.4 rved dome height 43.3 t free internal volume 1,899,000 cubic feet 14.8-12 Rev. 35

Reactor Coolant System Core thermal power level (MWt/% of rated power) 2754/102 Reactor coolant Pump heat (MWt) 17.1 Coolant pressure (psig) 2300 Inlet coolant temperature (ºF) 551.25 Internal coolant volume (cubic feet) (excludes the pressurizer) 10,104.4 Containment System Pressure (psia) 15.7 Relative humidity (%) 25 Inside temperature (°F) 120 Outside temperature (°F) N/A

  • Service Water Inlet temperature (°F) 80 Refueling Water Storage Tank (RWST) water temperature (°F) 100 Safety Injection Tank (SIT) water temperature (°F) 120 o heat transfer credited from containment structure to outside environment.

14.8-13 Rev. 35

TABLE 14.8.2-3 MINIMUM CONTAINMENT HEAT SINK DATA Heat Sink Exposed Surface Area (sq ft) Containment Cylinder and Dome Cylinder 52,800 Dome 19,070 71,870 Unlined Concrete Steam Generator Compartment Walls 26,114 Miscellaneous Slabs 4,488 Elevator Foundation 643 Pressurizer Wall and Roof 2,159 Refueling Canal (Outside) 10,043 Steam Generator Pedestals 2,860 Steam Generator Buttresses 3,840 Fuel Canal Buttresses 3,273 53,420 Reactor Support ncrete 3,486 inches, exposed on one side to the containment osphere and on the other to a 150°F source to account the higher reactor cavity operating temperature) Galvanized Steel 116,497 Painted Steel Less than 0.12 in. Thick 5,605 Painted Steel 0.12 to 0.16 in. Thick 16,863 Painted Steel 0.16 to 0.24 in. Thick 36,713 Painted Steel 0.24 to 0.3 in. Thick 10,289 Painted Steel 0.3 to 0.4 in. Thick 19,366 Painted Steel 0.4 to 0.5 in. Thick 4,525 14.8-14 Rev. 35

Heat Sink Exposed Surface Area (sq ft) Painted Steel 0.5 to 0.625 in. Thick 5,338 Painted Steel 0.625 to 0.75 in. Thick 2,243 Painted Steel 0.75 to 1.0 in. Thick 2,862 Painted Steel 1.0 to 1.5 in. Thick 4,322 Painted Steel Greater than 1.5 in. Thick 1,031 Unpainted Stainless Steel 18,464 Containment Floor 8,102 Safety Injection Tanks 2,541 14.8-15 Rev. 35

ABLE 14.8.2-4 SEQUENCE OF EVENTS, MP2-MSLB: LOSS OF OFFSITE POWER AND THE FAILURE OF VITAL BUS VA-10 OR VA-20 FROM 102% POWER IME (seconds) EVENT SETPOINT/VALUE 0.0 MSLB occurs from 102% power, break size is 3.51 ft2. 0.0 Loss of offsite power. 1.29 Low RCS flow reactor trip condition reached. 89.7% of initial RCS flow 1.94 Low RCS flow reactor trip signal generated. 0.65 second delay 2.50 Containment High Pressure Signal (CHPS) 5.83 psig with condition is reached. (Due to the assumed loss of uncertainty offsite power and failure of VA-10 or VA-20, the affected steam generator FRV and main feedwater isolation valves do not close.) 3.40 High Containment Pressure MSIS generated. 0.9 second delay 6.15 Containment High-High Pressure Signal 11.08 psig with (CHHPS) condition reached. uncertainty 28.5 Containment cooling fans energize. Time based on CHPS + 26 second delay. 37.1 Peak Containment Pressure reached. 326.5°F 74.76 Containment spray flow commences. See Note 1 Time based on CHHPS + 68.6 seconds for pump start, valve stroke time, and header fill time. 180 Maximum AFW Flow to the Affected Steam 550 gpm, 100°F Generator. 552.1 Peak Containment Pressure reached. 53.8 psig 1000 Simulation ended. The minimum containment spray flows used in this analysis are provided in Table 14.8.2-5. 14.8-16 Rev. 35

ABLE 14.8.2-5 ENGINEERED SAFETY FEATURES PERFORMANCE FOR MSLB CONTAINMENT ANALYSIS Safety Features Value Notes Containment spray ater temperature 100°F SAS setpoint 11.08 psig TS value plus uncertainty inimum flow rate 1361 gpm at 54 psig Values based on a minimum 1375 gpm at 51 psig RWST level of 30 feet above 1394 gpm at 47 psig tank bottom. 1428 gpm at 40 psig elay time with normal AC 49 seconds Includes 33 seconds for header wer available fill time and 16 seconds for signal generation, pump start and valve stroke. elay time with the loss of 68.6 seconds Includes 33 seconds for header rmal AC power fill time and 35.6 seconds for signal generation, pump start and valve stroke. Containment Air circulation (CAR) Cooling ns umber of fans 4 2 for certain single failure cases ctivation setpoint 5.83 psig TS value plus uncertainty elay time with normal AC 15 seconds wer available elay time with loss of normal 26 seconds power eat removal capability of one 80 million BTU/hr based on A GOTHIC Fan Cooler Model R Fan air inlet temperature of is used. This model is 289°F and a fan flow rate of benchmarked to the cited 34,800 cfm, along with a specification data. cooling water inlet temperature of 130°F and flow rate of 2000 gpm. Safety Injection None Assumed. 14.8-17 Rev. 35

MPS-2 FSAR FIGURE 14.8.2-1 MAIN STEAM LINE BREAK ANALYSIS - 102% POWER WITH LOSS OF OFFSITE POWER AND FAILURE OF VITAL BUS CABINET VA-10 OR VA CONTAINMENT PRESSURE VS. TIME 60 50 ~ ------ --- 1:1.040

    'iii Q.

V ai

I VI VI
      <11                                                                               I
     ~
     .... 30 C
      <11 E

c -

    'iii II c

o u 20 10 o o 100 200 300 400 500 600 700 800 900 1000 Time, seconds Rev. 31.3

MPS-2 FSAR FIGURE 14.8.2-2 MAIN STEAM LINE BREAK ANALYSIS - 102 % POWER WITH LOSS OF OFFSITE POWER AND FAILURE OF VITAL BUS CABINET VA-10 OR VA CONTAINMENT TEMPERATURE VS. TIME 350 300 u.. I"--,---t-+--+--+-+--+----J-_ o QI j

 'l...U 250 QI Q.

E {:!.

 ...r::

QI E

 .5     200 III
 ...r::

o U 150 100 o 100 200 300 400 500 600 700 800 900 1000 Time, seconds Rev. 31.3

MPS-2 FSAR FIGURE 14.8.2-3 MAIN STEAM LINE BREAK ANALYSIS - 102 % POWER WITH LOSS OF OFFSITE POWER AND FAILURE OF VITAL BUS CABINET VA-10 OR VA MASS FLOW RATE VS. TIME 7000 6000 5000 u OJ III

~4000 OJ IV a::

o

~       3000 III IV
~

2000 \ \ : o w from Brea k 1000 FW Flashing Flow

                    -         I o

o 100 200 300 400 500 600 Time, seconds Rev. 31.3

MPS-2 FSAR FIGURE 14.8.2-4 MAIN STEAM LINE BREAK ANALYSIS - 102 % POWER WITH LOSS OF OFFSITE POWER AND FAILURE OF VITAL BUS CABINET VA-10 OR VA ENERGY RELEASE RATE VS. TIME 8.0E+06 7.0E+06 6.0E+06 5.0E+06 4.0E+06 3.0E+06

            \

2.0E+06

             \

Energy from Break 1.0E+06

                       --~

FW Flashing Energy I O.OE+OO o

               "        100            200           300     400          500         600 Rev. 31.3

MPS-2 FSAR FIGURE 14.8.2-5 MAIN STEAM LINE BREAK ANALYSIS - 102 % POWER WITH LOSS OF OFFSITE POWER AND FAILURE OF VITAL BUS CABINET VA-10 OR VA INTEGRATED MASS FLOW VS. TIME 350 300

                                                                       ~

250 ~ ~

 .c                                      ~           Mass from Break
   ~
 &200 VI VI
                                 /

III

  ~                         /
 "0 QI "iU 150            /

aD QI

  ....l:

100 I 50 I FW Flashing Mass I o I o 100 200 300 400 500 600 Time, seconds Rev. 31.3

MPS-2 FSAR FIGURE 14.8.2-6 MAIN STEAM LINE BREAK ANALYSIS - 102 % POWER WITH LOSS OF OFFSITE POWER AND FAILURE OF VITAL BUS CABINET VA-10 OR VA INTEGRATED ENERGY RELEASE VS. TIME 400 350

J 300 -"'"
                                                        ~

t; t: ~ergy from Break

 .2
~ 250 /

1Il~ III III III

 ~ 200
  ~

V III t: w

 ] 150 OJ
                    /
  ~

tl.O III OJ t:

 - 100            I 50 FW Flashing Energy
               /I o

o 100 200 300 400 500 600 Time, seconds Rev. 31.3

MPS-2 FSAR FIGURE 14.8.2-7 MAIN STEAM LINE BREAK ANALYSIS - 102 % POWER WITH LOSS OF OFFSITE POWER AND FAILURE OF VITAL BUS CABINET VA-10 OR VA AFFECTED STEAM GENERATOR PRESSURE VS. TIME 1000 900 800 700

     ~
    *iii c.

oJ 600

     ~

III III CI) Q. sao c:J II) CI) u CI) 400 \

    ~

300 \ 200 \ 100 ............... a a 100 200 300 400 SOD 600 Time, seconds Rev. 31.3

MPS-2 FSAR FIGURE 14.8.2-8 MAIN STEAM LINE BREAK ANALYSIS - 102 % POWER WITH LOSS OF OFFSITE POWER AND FAILURE OF VITAL BUS CABINET VA-10 OR VA UNAFFECTED STEAM GENERATOR PRESSURE VS. TIME 1100 1000 f'----.- 900 800

 'iii
  "'c.

QJ"' 700

s III III QJ 600 0.

C) V)

 "C     500 QJ u
 -QJ
 ~
  "'c: 400
J 300 200 100 a

a 100 200 300 400 500 600 Time, seconds Rev. 31.3

MPS-2 FSAR FIGURE 14.8.2-9 MAIN STEAM LINE BREAK ANALYSIS - 102 % POWER WITH LOSS OF OFFSITE POWER AND FAILURE OF VITAL BUS CABINET VA-10 OR VA AFFECTED STEAM GENERATOR LIQUID MASS VS. TIME 160000 140000 120000

c
\I'
\I' 100000 ctl 2

"'C C'

i 80000

(!) V) -~ "'C CI) oIJ ~ u 60000 CI) <C 40000 ~ 20000

                                               ~
                                      '"                            ~-

0 o 100 200 300 400 500 600 700 800 900 1000 Time, seconds Rev. 31.3

14.8-27 Rev. 35 FIGURE 14.8.3-2 DELETED BY FSARCR 04-MP2-018. 14.8-28 Rev. 35

FIGURE 14.8.3-3 DELETED BY FSARCR 04-MP2-018. 14.8-29 Rev. 35

FIGURE 14.8.3-4 DELETED BY FSARCR 04-MP2-018. 14.8-30 Rev. 35

FIGURE 14.8.3-5 DELETED BY FSARCR 04-MP2-018. 14.8-31 Rev. 35

FIGURE 14.8.3-6 DELETED BY FSARCR 04-MP2-018. 14.8-32 Rev. 35

TABLE 14.8.4-1 LOSS OF COOLANT ACCIDENT (OFF SITE ASSUMPTIONS) Assumption Core power level = 2754 MWt Core released fractions: Consistent with Table 2 of Regulatory Guide 1.183 Iodine composition: Containment Sump Particulate 95% 0% Elemental 4.85% 97% Organic 0.15% 3% Containment leak rate: 0.5%/day 24 hrs. 0.25%/day > 24 hrs. Enclosure Building Filtration System (EBFS) charcoal filter efficiencies: particulate/elemental/organic 90 / 90 / 70% (1) Time Before Enclosure Building Filtration System (EBFS) is Fully 110 seconds Functional EBFS Bypass Leakage (% by Weight of Containment Air per Day) 0 - 110 sec: 0.5% 110 sec - 24 hours: 0.007% 24 - 720 hours: 0.0035% NOTE: Prior to EBFS draw down at 110 seconds, bypass leakage is the full containment leak rate of 0.5%. After 110 seconds, bypass leakage is based on 1.4% of containment leak rate and half that at 24 hours. X/Qs: Location Time Period Elevated Ground Release EAB (0-2) hrs. 1.00 E-04 3.66 E-04 LPZ (0-4) hrs 2.69 E-05 4.80 E-05 (4-8) hrs. 1.07E-05 2.31 E-05 (8-24) hrs. 6.72E-06 1.60 E-05 (24-96) hrs. 2.46E-06 7.25 E-06 (96-720) hrs. 5.83E-07 2.32 E-06 14.8-33 Rev. 35

Assumption Dose Conversion Factors Federal Guidance Reports 11 and 12 ) Containment Free Air Volume = 1.899 x 106 ft3 ) Breathing Rates (0 -8) hr. = 3.5 x 10-4 m3/sec (8-24) hr. = 1.8 x 10-4 m3/sec (24-720) hr. = 2.3 x 10-4 m3/sec ) Release Points:Filtered - Millstone Stack Bypass - MP-2 Containment ) Containment Sprayed Volume: 35.4% ) Containment Spray Removal Coefficients: elemental = 20 per hour particulate = 6.42 per hour ) Containment Spray Effectiveness Time: elemental: 75 seconds - 3.03 hours particulate: 75 seconds - 3.23 hours ) ESF Leakage: 24 gallons per hour ) ESF Leakage begins at 27.5 minutes post LOCA ) Sump Volume: 3.773E+04 ft3 ) RWST Back leakage begins at 6.45 hours ) RWST Back leakage amount: 0.05 - 0.70 gpm ) Iodine DF: 100 ) Sump pH 7.0 70% is a conservative analysis assumption for the organic iodine. Technical Specifications can support assumptions for efficiencies of 90% for all iodine species. 14.8-34 Rev. 35

BLE 14.8.4-2

SUMMARY

OF DOSES FOR LOSS OF COOLANT ACCIDENT DOSE (rems) TEDE EAB 2.9E+00 LPZ 1.7E+00 14.8-35 Rev. 35

Control Room Volume = 3.565 E+04 ft3. Control Room Unfiltered Inleakage in Recirculation Mode 200 cfm Control Room Normal Makeup Air Flowrate 800 cfm Time from MP-2 LOCA Initiation to Time when Control Room Intake Dampers Close 20 seconds Time when Control Room Emergency Ventilation (Filtration) System Operating at Full Capacity 1 hour Control Room Emergency Ventilation (Filtration) System Flowrate 2,250 cfm CREV Filter Efficiency 70% (1) Control Room Shielding: North Wall: 2' concrete. West Wall: 1.5 feet concrete except 8 foot section which is 2 feet concrete. South Wall: 24.5 feet of 1 foot concrete except glass wall 86.75 feet long. East Wall: 2 feet concrete. Roof: 2 feet concrete. Floor: 2 foot concrete. 70% is a conservative analysis assumption for particulate, elemental and organic iodine. Technical Specifications can support assumptions for control room filter efficiencies of 90% for all iodine species. 14.8-36 Rev. 35

ABLE 14.8.4-4 ATMOSPHERIC DISPERSION DATA FOR MILLSTONE UNIT 2 CONTROL ROOM Release Point Ground X/Q, sec/m3 RWST X/Q, sec/m3 Stack X/Q, sec/m3 2 hr 3.00 E-03 9.54 E-04 2.51 E-04 4 hr 1.87 E-03 7.56 E-04 2.51 E-04 8 hr 1.87 E-03 7.56 E-04 1.96 E-05 24 hr 6.64 E-04 2.72 E-04 5.46 E-06

- 96 hr           5.83 E-04           2.17 E-04       3.43 E-07
- 720 hr          4.97 E-04           1.51E-04        6.44 E-09 14.8-37                         Rev. 35

ABLE 14.8.4-5 DOSE TO MILLSTONE UNIT 2 CONTROL ROOM OPERATORS Release TEDE (1) Millstone 2 (LOCA) 3.0E+00 es: Dose through wall and ceiling from external sources included. 14.8-38 Rev. 35}}