ML12011A157

Amendment 61 to Final Safety Analysis Report, Appendix 3A - Plant Design Assessment Report for Safety/Relief Valves and Loss-of-Coolant Accident Loads

ML12011A157
Person / Time
Site:	Columbia
Issue date:	12/14/2011
From:	Energy Northwest
To:	Office of Nuclear Reactor Regulation
References
GO2-11-201
Download: ML12011A157 (359)

Text

C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 Appendix 3A PLANT DESIGN ASSESSMENT REPORT FOR SAFETY/RELIEF VALVES AND LOSS-OF-COOLANT ACCIDENT LOADS TABLE OF CONTENTS

Section Page 3A-i 3A.1.1 CONFORMANCE TO NRC ACCEPTANCE CRITERIA .................... 3A.1.1-1 3A.1.2 ROLE OF THE DESIGN ASSESSMENT REPORT ........................... 3A.1.2-1 3A.1.3 ASSESSMENT APPROACH .......................................................

3A.1.3-1 3A.1.4

SUMMARY

OF DESIGN ASSE SSMENT REPORT CONTENT ........... 3A.1.4-1 3A.2

SUMMARY

AND CONCLUSIONS ................................................. 3A.2.1-1 3A.2.1 GENERAL DESCRIPTION OF PLANT

......................................... 3A.2.1-1 3A.2.1.1 Structures, Piping, and Components Directly Affected by Pool Dynamic Loads

...................................................................... 3A.2.1-1 3A.2.1.2 Structures, Piping, and Co mponents Indirectly Affected by Pool Dynamic Loads

...................................................................... 3A.2.1-3 3A.2.2

SUMMARY

OF CHANGE S AND CONCLUSIONS ......................... 3A.2.2-1 3A.2.2.1 Summary of Changes to Preserve Design Margins ........................... 3A.2.2-1 3A.2.2.2 Conclusions .......................................................................... 3A.2.2-1

3A.3 CONTAINMENT DYNAMIC FORCING FUNCTIONS ........................ 3A.3.1-1 3A.3.1 LOADS ASSOCIATED WITH SAFETY/RELIEF VALVE ACTUATION ..........................................................................

3A.3.1-1 3A.3.1.1 Description of the Safety/Relief System

......................................... 3A.3.1-1 3A.3.1.2 Description of the Phe nomena and Resulting Loads .......................... 3A.3.1-1 3A.3.1.2.1 Water Cl earing Loads

............................................................ 3A.3.1-2 3A.3.1.2.2 Air Clearing Loads ...............................................................

3A.3.1-2 3A.3.1.2.3 Steam Condensation Loads

...................................................... 3A.3.1-2 3A.3.1.3 Safety/Relief Valve Air Cleari ng Loads ........................................ 3A.3.1-3 3A.3.1.3.1 Boundary Loads ...................................................................

3A.3.1-3 3A.3.1.3.1.1 Spatia l Distribution of Boundary Pressures

................................ 3A.3.1-3 3A.3.1.3.1.2 Pressure Wave Forms

......................................................... 3A.3.1-4 3A.3.1.3.1.3 Design Maximum Pressure Amplitude ..................................... 3A.3.1-4 3A.3.1.3.2 Submerged St ructure Loads

..................................................... 3A.3.1-4 3A.3.1.3.2.1 Peak Safety/Relief Valve Dynamic Loads ................................. 3A

.3.1-5 3A.3.1.3.2.2 Time Dependence of Safety/Relief Valve Loads and Dynamic Load Factors

.................................................................... 3A.3.1-5 3A.3.1.3.2.3 Safety/Relief Valv e Loads on Structures ................................... 3A.3.1-5 3A.3.1.4 Refere nces ............................................................................ 3A.3.1-6 C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 Appendix 3A PLANT DESIGN ASSESSMENT REPORT FOR SAFETY/RELIEF VALVES AND LOSS-OF-COOLANT ACCIDENT LOADS TABLE OF CONTENTS (Continued)

Section Page 3A-ii 3A.3.2 LOADS ASSOCIATED WITH LOSS-OF-COOLANT ACCIDENT........3A.3.2-1 3A.3.2.1 Description of Pressure Suppression System...................................3A.3.2-1 3A.3.2.2 Description of the Phenomena and Resulting Loads..........................3A.3.2-1 3A.3.2.3 Short-Term Lo ss-of-Coolant Accident Loads..................................3A.3.2-3 3A.3.2.3.1 Analytic al Models and Supporti ng Test Data................................3A.3.2-3 3A.3.2.3.1.1 Vent Clearing Jet and Induced Flow Field Model........................3A.3.2-3 3A.3.2.3.1.2 Loss-of-Coolant Accident Bubble Charging Model......................3A.3.2-4 3A.3.2.3.1.3 Pool Swell Analytical Model.................................................3A.3.2-5 3A.3.2.3.1.4 Fallback Model.................................................................3A.3.2-6 3A.3.2.3.2 Boundary Loads...................................................................

3A.3.2-6 3A.3.2.3.3 Structure Loads...................................................................

3A.3.2-6 3A.3.2.3.3.1 Loads on Major Structures...................................................3A.3.2-9 3A.3.2.3.3.2 Loads on Fully Submerged Piping Systems Below Elevation 454.4 ft..........................................................................3A.3

.2-10 3A.3.2.3.3.3 Loads on Partially Submerged Piping Systems...........................3A.3.2-10 3A.3.2.3.3.4 Loads on Piping System s and Structural Components Between Elevations 454.4 ft and 484.4 ft.............................................3A.3

.2-11 3A.3.2.4 Long-Term Hydrodynamic Loads................................................3A.3

.2-11 3A.3.2.4.1 Analytic al Models and Supporti ng Test Data................................3A.3.2-11 3A.3.2.4.1.1 Chugging Loads................................................................3A.3

.2-11 3A.3.2.4.1.2 Condensation Oscillation Loads.............................................3A.3

.2-12 3A.3.2.4.2 Boundary Loads...................................................................

3A.3.2-12 3A.3.2.4.2.1 Chugging Loads................................................................3A.3

.2-12 3A.3.2.4.2.2 Condensation Oscillation Loads.............................................3A.3

.2-13 3A.3.2.4.3 Submerged St ructure Loads.....................................................3A.3

.2-13 3A.3.2.4.3.1 Condensation Oscillation Loads.............................................3A.3

.2-13 3A.3.2.4.3.2 Chugging Loads................................................................3A.3

.2-13 3A.3.2.4.4 Lateral Loads on Downcomer Vents..........................................3A.3

.2-15 3A.3.2.5 Pressure and Temperature Transients...........................................3A.3

.2-16 3A.3.2.5.1 Results for CGS...................................................................

3A.3.2-17 3A.3.2.5.2 Differentia l Pressure Load on the Di aphragm Floor.......................3A.3.2-18 3A.3.2.6 Building Response to Loss-of-Coolant Accident Loads......................3A.3.2-18 3A.3.2.7 References............................................................................3A.3

.2-18 C OLUMBIA G ENERATING S TATION Amendment 57 F INAL S AFETY A NALYSIS R EPORT December 2003 Appendix 3A PLANT DESIGN ASSESSMENT REPORT FOR

SAFETY/R ELIEF V A LVES AND LOSS-OF-COOLANT ACCIDENT LOADS TABLE OF CONTENTS (Continued)

Section Page LDC N-0 2-0 0 0 3A-iii 3A.3.3 LOAD

SUMMARY

...................................................................3A.3.3-1 3A.3.4 SEQUENCE OF DYNAMIC LOADS............................................3A.3.4-1 3A.3.5 LOAD COMBINATIONS A ND ACCEPTANCE CRITERIA................3A.3.5-1 3A.3.5.1 Steel C ontainment Structure.......................................................3A.3.5-1 3A.3.5.1.1 Defi nitions.........................................................................3A.3.5-1 3A.3.5.1.2 Load Combinations...............................................................

3A.3.5-2 3A.3.5.1.3 Acceptance Criteria..............................................................3A.3.5-3 3A.3.5.2 Reinforced-Concrete Structures...................................................3A.3.5-3 3A.3.5.2.1 Defi nitions.........................................................................3A.3.5-3 3A.3.5.2.2 Load Combinations...............................................................

3A.3.5-4 3A.3.5.2.3 Acceptance Criteria..............................................................3A.3.5-5 3A.3.5.3 Steel Structures......................................................................3A.3.5-5 3A.3.5.3.1 Defi nitions.........................................................................3A.3.5-5 3A.3.5.3.2 Load Combinations...............................................................

3A.3.5-5 3A.3.5.3.3 Acceptance Criteria..............................................................3A.3.5-5 3A.3.5.4 Piping Systems.......................................................................3A.3.5-6 3A.3.5.4.1 Defi nitions.........................................................................3A.3.5-6 3A.3.5.4.2 Load Combinations...............................................................

3A.3.5-7 3A.3.5.4.3 Acceptance Criteria..............................................................3A.3.5-8 3A.3.5.5 References...........................................................................3A.3.5-8

3A.4 DESIGN ASSESSMENT...............................................................3A.4.1-1 3A.4.1 SUPPRESSION POOL BOUNDARY STRUCTURES.........................3A.4.1-1 3A.4.1.1 Assessment of Steel Containment Structure....................................3A.4.1-1 3A.4.1.1.1 Loads Used for Assessment.....................................................3A.4.1-1 3A.4.1.1.1.1 Safety/Relief Valve Loads....................................................3A.4.1-1 3A.4.1.1.1.1.1 Single Valve Discharge Case..............................................3A.4.1-2 3A.4.1.1.1.1.2 Two Valves Discharge Case...............................................3A.4.1-2 3A.4.1.1.1.1.3 Automatic Depressurization System Valves Discharge Case........3A.4.1-2 3A.4.1.1.1.1.4 All Valves Discharge Case.................................................3A.4.1-2 3A.4.1.1.1.2 Loss-of-Coolant Accident Loads............................................3A.4.1-2 3A.4.1.1.1.2.1 Chugging Loads..............................................................3A.4.1-2 3A.4.1.1.1.2.2 High and Medium Mass Flux Condensa tion Oscillations.............3A.4.1-2 3A.4.1.1.1.2.3 Other Lo ss-of-Coolant Accident Loads..................................3A.4.1-3 C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 Appendix 3A PLANT DESIGN ASSESSMENT REPORT FOR SAFETY/RELIEF VALVES AND LOSS-OF-COOLANT ACCIDENT LOADS TABLE OF CONTENTS (Continued)

Section Page 3A-iv 3A.4.1.1.1.3 Other Significant Loads.......................................................3A.4.1-3 3A.4.1.1.2 Controlling Lo ad Combinations................................................3A.4

.1-4 3A.4.1.1.3 Acceptance Criteria..............................................................3A.4.1-4 3A.4.1.1.4 Method of Analysis...............................................................

3A.4.1-5 3A.4.1.1.4.1 Formulation of the Problem..................................................3A.4.1-5 3A.4.1.1.4.2 Mathematical Model...........................................................3A.4.1-5 3A.4.1.1.4.3 Coupled Equations of Motion................................................3A.4.1-5 3A.4.1.1.4.4 Numerical Solution............................................................3A.4.1-6 3A.4.1.1.4.5 Computer Program.............................................................3A.4.1-6 3A.4.1.1.5 Results a nd Design Margin.....................................................3A.4.1-6 3A.4.1.1.5.1 Results of Analysis.............................................................3A.4.1-6 3A.4.1.1.5.2 Assessment Results............................................................3A.4.1-8 3A.4.1.2 Basemat...............................................................................3A.4.1-8 3A.4.1.2.1 Loads Used for Assessment.....................................................3A.4.1-9 3A.4.1.2.1.1 Safety/Relief Valve Loads....................................................3A.4.1-9 3A.4.1.2.1.2 Loss-of-Coolant Accident Loads............................................3A.4.1-9 3A.4.1.2.1.3 Other Significant Loads.......................................................3A.4.1-9 3A.4.1.2.2 Appli cable Load Combinations and Acceptance Criteria..................3A.4.1-9 3A.4.1.2.3 Method of Analysis...............................................................

3A.4.1-9 3A.4.1.2.3.1 Effects of E O , E SS , D, L.......................................................3A.4

.1-10 3A.4.1.2.3.2 Effect of P SR , P B................................................................3A.4

.1-10 3A.4.1.2.3.3 Critic al Load Combination...................................................3A.4

.1-10 3A.4.1.2.3.4 Capacity..........................................................................3A.4

.1-10 3A.4.1.2.4 Results and Design Margins....................................................3A.4

.1-10 3A.4.1.3 Pedestal...............................................................................3A.4

.1-11 3A.4.1.3.1 Loads Used for Assessment.....................................................3A.4

.1-11 3A.4.1.3.1.1 Safety/Relief Valve Loads....................................................3A.4

.1-11 3A.4.1.3.1.2 Loss-of-Coolant Accident Loads............................................3A.4

.1-11 3A.4.1.3.1.3 Other Significant Loads.......................................................3A.4

.1-11 3A.4.1.3.2 Appli cable Load Combinations and Acceptance Criteria..................3A.4.1-11 3A.4.1.3.3 Method of Analysis...............................................................

3A.4.1-12 3A.4.1.3.3.1 Asymmetric Action............................................................3A.4

.1-12 3A.4.1.3.3.2 Symmetric Action..............................................................3A.4

.1-12 3A.4.1.3.4 Results and Design Margins....................................................3A.4

.1-13 C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 Appendix 3A PLANT DESIGN ASSESSMENT REPORT FOR SAFETY/RELIEF VALVES AND LOSS-OF-COOLANT ACCIDENT LOADS TABLE OF CONTENTS (Continued)

Section Page 3A-v 3A.4.1.4 Diaphragm Floor....................................................................3A.4

.1-14 3A.4.1.4.1 Loads Used for Assessment.....................................................3A.4

.1-14 3A.4.1.4.1.1 Safety/Relief Valve Actuation Loads.......................................3A.4.1-14 3A.4.1.4.1.2 Loss-of-Coolant Accident Loads............................................3A.4

.1-14 3A.4.1.4.1.3 Other Significant Loads.......................................................3A.4

.1-15 3A.4.1.4.2 Controlling Lo ad Combinations................................................3A.4

.1-15 3A.4.1.4.3 Acceptance Criteria..............................................................

3A.4.1-15 3A.4.1.4.4 Method of Analysis...............................................................

3A.4.1-15 3A.4.1.4.5 Results and Design Margins....................................................3A.4

.1-16 3A.4.1.5 Diaphragm Floor Seal..............................................................3A.4

.1-17 3A.4.1.5.1 Loads Used for Assessment.....................................................3A.4

.1-17 3A.4.1.5.2 Controlling Lo ad Combination.................................................3A.4

.1-18 3A.4.1.5.3 Acceptance Criteria..............................................................

3A.4.1-18 3A.4.1.5.4 Method of Analysis...............................................................

3A.4.1-19 3A.4.1.5.5 Results and Design Margins....................................................3A.4

.1-19 3A.4.1.6 References............................................................................3A.4

.1-19 3A.4.2 SUPPRESSION POOL MAJOR STRUCTURES AND COMPONENTS..3A.4.2-1 3A.4.2.1 Down comer Bracing System......................................................3A.4.2-1 3A.4.2.1.1 Descriptio n of System............................................................3A.4.2-1 3A.4.2.1.2 Loads Used for Assessment.....................................................3A.4.2-2 3A.4.2.1.2.1 Safety/Relief Valve Actuation Loads.......................................3A.4.2-2 3A.4.2.1.2.2 Loss-of-Coolant Accident Loads............................................3A.4.2-2 3A.4.2.1.2.3 Other Significant Loads.......................................................3A.4.2-4 3A.4.2.1.3 Controlling Lo ad Combinations and Acceptance Criteria.................3A.4.2-4 3A.4.2.1.4 Method of Analysis...............................................................

3A.4.2-4 3A.4.2.1.4.1 Analysis for Horizontal Loads...............................................3A.4.2-5 3A.4.2.1.4.2 Analys is for Vertical Loads..................................................3A.4.2-5 3A.4.2.1.4.3 Design Load Conditions......................................................3A.4.2-5 3A.4.2.1.5 Results a nd Design Margin.....................................................3A.4.2-8 3A.4.2.1.5.1 Principal Results................................................................3A.4.2-8 3A.4.2.1.5.2 Design Margins.................................................................3A.4.2-9 3A.4.2.2 Columns..............................................................................3A.4.2-9 3A.4.2.2.1 Loads Used for Assessment.....................................................3A.4.2-9 3A.4.2.2.1.1 Safety/Relief Valve Actuation Loads.......................................3A.4.2-9 C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 Appendix 3A PLANT DESIGN ASSESSMENT REPORT FOR SAFETY/RELIEF VALVES AND LOSS-OF-COOLANT ACCIDENT LOADS TABLE OF CONTENTS (Continued)

Section Page 3A-vi 3A.4.2.2.1.2 Loss-of-Coolant Accident Loads............................................3A.4

.2-10 3A.4.2.2.1.3 Other Significant Loads.......................................................3A.4

.2-11 3A.4.2.2.2 Appli cable Load Combinations and Acceptance Criteria..................3A.4.2-11 3A.4.2.2.3 Method of Analysis...............................................................

3A.4.2-11 3A.4.2.2.4 Results and Design Margins....................................................3A.4

.2-14 3A.4.2.3 Downcomers.........................................................................3A.4

.2-14 3A.4.2.3.1 Loads Used for Assessment.....................................................3A.4

.2-15 3A.4.2.3.2 Load Combination and Acceptance Criteria.................................

3A.4.2-16 3A.4.2.3.3 Method of Analysis...............................................................

3A.4.2-17 3A.4.2.3.3.1 Static Analysis..................................................................3A.4

.2-17 3A.4.2.3.3.2 Response Spectrum Analysis.................................................3A.4

.2-17 3A.4.2.3.4 Results a nd Design Margin.....................................................3A.4

.2-17 3A.4.2.3.5 Fatigue Evaluations...............................................................

3A.4.2-18 3A.4.2.4 Safety/Rel ief Valve Piping Systems..............................................3A.4

.2-18 3A.4.2.4.1 Loads Used for Assessment.....................................................3A.4

.2-19 3A.4.2.4.2 Load Combination and Acceptance Criteria.................................

3A.4.2-19 3A.4.2.4.3 Allowable Stress Limits (Equation 9 of NC-3652 a nd NC-3611, Reference 3A.4.2-1).............................................................3A

.4.2-20 3A.4.2.4.4 Method of Analysis...............................................................

3A.4.2-20 3A.4.2.4.4.1 Static Analysis..................................................................3A.4

.2-20 3A.4.2.4.4.2 Response Spectrum Analysis.................................................3A.4

.2-20 3A.4.2.4.4.3 Time History Analysis........................................................3A.4

.2-20 3A.4.2.4.5 Results a nd Design Margin.....................................................3A.4

.2-20 3A.4.2.4.6 Fatigue Evaluations...............................................................

3A.4.2-21 3A.4.2.5 Quencher..............................................................................3A.4

.2-21 3A.4.2.5.1 Loads Used for Assessment.....................................................3A.4

.2-22 3A.4.2.5.2 Load Combination Acceptance Criteria.......................................

3A.4.2-22 3A.4.2.5.3 Evaluation..........................................................................3A

.4.2-22 3A.4.2.6 Platforms and Ladders.............................................................3A.4

.2-23 3A.4.2.6.1 Loads Used for Assessment.....................................................3A.4

.2-23 3A.4.2.6.1.1 Safety/Relief Valve Operation Loads.......................................3A.4.2-23 3A.4.2.6.1.2 Loss-of-Coolant Accident Loads............................................3A.4

.2-23 3A.4.2.6.1.3 Other Significant Loads.......................................................3A.4

.2-23 3A.4.2.6.2 Controlling Lo ad Combinations................................................3A.4

.2-24 C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 Appendix 3A PLANT DESIGN ASSESSMENT REPORT FOR SAFETY/RELIEF VALVES AND LOSS-OF-COOLANT ACCIDENT LOADS TABLE OF CONTENTS (Continued)

Section Page 3A-vii 3A.4.2.6.3 Acceptance Criteria..............................................................

3A.4.2-24 3A.4.2.6.4 Method of Analysis...............................................................

3A.4.2-24 3A.4.2.6.5 Results..............................................................................3A

.4.2-24 3A.4.2.7 References............................................................................3A.4

.2-25 3A.4.3 MISCELLANEOUS SUPPRESSION POOL PIPING SYSTEMS............3A.4.3-1 3A.4.3.1 Loads Used for Assessment.......................................................3A.4.3-1 3A.4.3.2 Load Combination and Acceptance Criteria....................................3A.4.3-1 3A.4.3.3 Method of Analysis.................................................................3A.4.3-1 3A.4.3.4 Results and Design Margins.......................................................3A.4.3-2

3A.5 EFFECTS DUE TO BUILDING RESPONSES TO SAFETY/RELIEF VALVE DISCHARGE AND LOSS-OF-COOLANT ACCIDENT LOADS....................................................................3A.5.1-1 3A.5.1 BUILDING RESPONSE S TO SAFETY/RELIEF VALVE DISCHARGE LOADS................................................................3A.5.1-1 3A.5.1.1 Analytical Model....................................................................3A.5.1-1 3A.5.1.1.1 Overall Bu ilding Model..........................................................3A.5.1-1 3A.5.1.1.2 Steel Containm ent Shell Model.................................................3A.5.1-1 3A.5.1.2 Method of Analysis.................................................................3A.5.1-1 3A.5.1.3 Safety/Relief Valve Discharge Load Cases.....................................3A.5.1-2 3A.5.1.3.1 Response to A ll Valve Discharge..............................................3A.5.1-2 3A.5.1.3.2 Automatic De pressurization System Valves Discharge Case.............3A.5.1-3 3A.5.1.3.3 Two Valves Discharge Case....................................................3A.5.1-4 3A.5.1.3.4 Single Valv e Discharge..........................................................3A.5.1-4 3A.5.2 BUILDING RESP ONSES TO LOSS-OF-COOLANT ACCIDENT LOADS.................................................................3A.5.2-1 3A.5.2.1 Analytical Model....................................................................3A.5.2-1 3A.5.2.2 Method Of Anal ysis and Building Response...................................3A.5.2-1 3A.5.2.2.1 Reactor Building Res ponse, Nearly Symmetric Loading -

Acceleration Response Spectra.................................................3A

.5.2-2 C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 Appendix 3A PLANT DESIGN ASSESSMENT REPORT FOR SAFETY/RELIEF VALVES AND LOSS-OF-COOLANT ACCIDENT LOADS TABLE OF CONTENTS (Continued)

Section Page 3A-viii ATTACHMENTS 3A.A Not used

3A.B Three-Dimensional Source Flows in Exact Containment Geometry........3A.B-1

3A.C Concept of Drag Forces Due to Hydrodynamic Flow Fields................3A.C-1

3A.D Calculation Models for Shor t-Term Loss-of-Coolant Accident

Phenomena............................................................................3A.D-1

3A.E Suppression Pool Temperat ure Monitoring System...........................3A.E-1

3A.F Computer Programs.................................................................3A.F-1

3A.G Not Used

3A.H Conformance of CGS Design to NRC Acceptance Criteria..................3A.H-1

3A.I Safety/Relief Valve and Lo ss-of-Coolant Accident Loads on Submerged Structures...............................................................3A.I-1

C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 Appendix 3A PLANT DESIGN ASSESSMENT REPORT FOR SAFETY/RELIEF VALVES AND LOSS-OF-COOLANT ACCIDENT LOADS LIST OF TABLES

Section Title Page 3A-ix 3A.2.2-1 Suppression Pool A ssessment Summary....................................3A.2.2-3 3A.3.1-1 Summary of Safety/Relief System Characteristics........................3A.3.1-9

3A.3.2-1 Summary of Loss-of-Coolant A ccident Affected Structures.............3A.3.2-23

3A.3.2-2 CGS Data for Loss-of-Coolant Accident Water Jet Analysis...........3A.3.2-24

3A.3.2-3 CGS Data for Vent Clearing and Pool Swell Analysis...................3A.3.2-25

3A.3.2-4 Results from Loss-of-C oolant Accident Bubble Charging Analysis for CGS...............................................................3A.3

.2-26 3A.3.2-5 CGS Drywell Pressure as a Function of Time for Loss-of-Coolant Accident (Effe cts of Pipe Inventory and Subcooling Included)...........................................................3A.3

.2-27 3A.3.2-6 Results of Pool Swe ll Analysis for CGS....................................3A

.3.2-28 3A.3.2-7 CGS Plant Parameters for Loss-of-Coolant Accident Transient Analysis..............................................................3A.3

.2-29 3A.3.2-8 Short-Term Loss-of-Coolan t Accident Loads on Structures Below El. 454.4 ft..............................................................3A.3

.2-30 3A.3.2-9 Short-Term Loss-of-Coolan t Accident Loads on Structures between El. 454 ft 4.75 in. a nd 484 ft 4.75 in.............................3A.3

.2-32 3A.3.3-1 Summary of Hydrodynamic Loads on Wetwell Structures..............3A.3.3-3

3A.3.5-1 Equivalent Static Loads for Pressure Transients and Loss-of-Coolant Accident Effects............................................

3A.3.5-9 C OLUMBIA G ENERATING S TATION Amendment 54 F INAL S AFETY A NALYSIS R EPORT April 2000 Appendix 3A PLANT DESIGN ASSESSMENT REPORT FOR SAFETY/RELIEF VALVES AND LOSS-OF-COOLANT ACCIDENT LOADS LIST OF TABLES (Continued)

Section Title Page LDC N-9 9-0 0 0 3A-x 3A.3.5-2 Acceptance Criteria for Containment Vessel Allowable Stress Limits.....................................................................3A.3

.5-10 3A.3.5-3 Load Combinations - Reinforced-Concret e Structures...................3A.3.5-11

3A.3.5-4 Load Combinations - Steel Structures.......................................3A.3.5-12

3A.3.5-5 Load Combinations and Acceptance Criteria for ASME Code Class 1, 2, and 3 Balance-of-Plant Piping and Equipment...............3A.3.5-13

3A.4.1-1 Basemat - Stress Resultants at Critical Sections...........................3A

.4.1-21 3A.4.1-2 Pedestal - Stress Re sultants at Base..........................................3A.4

.1-22 3A.4.1-3 Equivalent Stress Cycles for Fatigue Evaluation..........................3A

.4.1-23 3A.4.1-4 Summary of Stress Intensities for Diaphragm Floor Seal................3A.4.1-24

3A.4.1-5 Cummulative Usage Factor Calculation for Diaphragm Floor Seal....3A.4.1-25

3A.4.2-1 Downcomer Bracing System C ontrolling Design Margins..............3A.4.2-27

3A.4.2-2 Controlling Stress Resultants in Column....................................3A.4

.2-28 3A.4.2-3 Results Summary - Safety/Relief Valve Quencher........................3A.4.2-29 3A.4.2-4 Cummulative Usage Factor Calculation at 24 in.

Downcomer Anchor............................................................3A.4

.2-30 3A.4.2-5 Cummulative Usage Factor Calculation at 28 in.

Downcomer Anchor............................................................3A.4

.2-31 3A.4.2-6 Maximum Usage Fa ctors Table..............................................3A.4

.2-32 C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 Appendix 3A PLANT DESIGN ASSESSMENT REPORT FOR SAFETY/RELIEF VALVES AND LOSS-OF-COOLANT ACCIDENT LOADS LIST OF TABLES (Continued)

Section Title Page 3A-xi 3A.4.3-1 Miscellaneous Wetwell Piping................................................3A.4.3-3 3A.4.3-2 Piping Zone Versus Loads....................................................3A.4.3-5

3A.4.3-3 Summary of Results and Design Margins for Miscellaneous Wetwell Piping..................................................................3A.4.3-6

C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 Appendix 3A

PLANT DESIGN ASSESSMENT REPORT FOR SAFETY/RELIEF VALVES AND LOSS-OF-COOLANT ACCIDENT LOADS

LIST OF FIGURES

Section Title 3A-xii 3A.2.1-1 Primary and Seconda ry Containment Structure 3A.2.1-2 Suppression Pool Compos ite Plan at El. 435 ft 3 in.

3A.2.1-3 Suppression Pool Compos ite Plan at El. 455 ft 4 in.

3A.2.1-4 Suppression Pool Compos ite Plan at El. 486 ft 8 in.

3A.2.1-5 Suppression Pool Composite Plan at El. 494 ft 5-1/4 in.

3A.2.1-6 Suppression Pool Composite Sections "1-1" and "2-2"

3A.2.1-7 Suppression Pool Composite Sections "3-3" and "4-4"

3A.2.1-8 Suppression Pool Composite Sections "5-5" and "6-6"

3A.2.1-9 Containment Vess el Developed Elevation

3A.3.1-1 Normalized Design Circumfere ntial Distribution of Pool Boundary Pressures at Containment

3A.3.1-2 Normalized Design Vertical Distribution of Pool Boundary Pressures at Containment

3A.3.1-3 MFP Design Wave Form (Normalized) Time History 3A.3.1-4 MFP Design Wave Form (Norma lized) Amplitude of Frequency Spectrum 3A.3.1-5 SFP Design Wave Form (Normalized) Time History

3A.3.1-6 SFP Design Wave Form (Norma lized) Amplitude of Frequency Spectrum

3A.3.1-7 SRV Air Clearing Load Distribution on a Downcomer

C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 Appendix 3A PLANT DESIGN ASSESSMENT REPORT FOR SAFETY/RELIEF VALVES AND LOSS-OF-COOLANT ACCIDENT LOADS LIST OF FIGURES (Continued)

Section Title 3A-xiii 3A.3.1-8 SRV Air Clearing Load Distribution on a Concrete Column 3A.3.1-9 SRV Air Clearing Load Distribution on an SRV Discharge Line and Quencher Support 3A.3.1-10 Dynamic Load Factor Versus Fre quency to be Used fo r Defining SRV Load on Submerged Structures

3A.3.1-11a SRV Air Clearing Load Distribu tion on Piping, Supports, and Bracing Truss

3A.3.1-11b SRV Air Clearing Load Distribu tion on Piping, Supports, and Bracing Truss

3A.3.2-1 Short Term Hydrodynamic Pr ocesses Associated with a LOCA

3A.3.2-2 Downcomer Vent Water Clearing Velocity Versus Time

3A.3.2-3 Downcomer Vent Water Cl earing Acceleration Versus Time 3A.3.2-4 LOCA Bubble Charging Radial Component of in Radial Plane Containing Downcomers

3A.3.2-5 LOCA Bubble Charging Tangential Component of in Vertical Cylindrical Surface Through Middle Downcomers

3A.3.2-6 LOCA Bubble Chargi ng Vertical Component of in Radial Plane Containing Downcomers 3A.3.2-7 LOCA Bubble Radius and Source Strength Time Histor ies by PSAM Method

3A.3.2-8 Pool Swell Water Slug Velocity Versus Time

3A.3.2-9 Pool Swell Water Sl ug Acceleration Versus Time

3A.3.2-10 Pool Swell Water Slug Elevation (Top Surface) Versus Time C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 Appendix 3A PLANT DESIGN ASSESSMENT REPORT FOR SAFETY/RELIEF VALVES AND LOSS-OF-COOLANT ACCIDENT LOADS LIST OF FIGURES (Continued)

Section Title 3A-xiv 3A.3.2-11 Pool Swell Air B ubble Pressure Versus Time

3A.3.2-12 Pool Swell Wetwell Air Pressure Versus Time

3A.3.2-13 Pool Swell Water Slug Velocity Versus Elevation of Slug Top Surface

3A.3.2-14 Fallback Water Slug Velocity Versus Elevation of Water Slug Top Surface

3A.3.2-15 LOCA Boundary Load Duration

3A.3.2-16 LOCA Boundary Load Di stribution During Vent Clearing

3A.3.2-17 LOCA Boundary Load Di stribution During Pool Swell

3A.3.2-18 Distribution of Short Term LOCA Loads on Structures Below El. 454.4 ft

3A.3.2-19 Pressure Gradients Across Submerged Structures Due to Chugging

3A.3.2-20 Large Recirculation Line Brea k - Pressure Response - Minimum ECCS

3A.3.2-21 Containment Pressure Response for Large Recirculation Line Break - Cases A, B, and C

3A.3.2-22 Large Recirculation Li ne Break - Temperature Response 3A.3.2-23 Drywell Temperat ure Response for Large Reci rculation Line Break - Cases A, B, and C 3A.3.2-24 Suppression Pool Temperature Response for Large Recirculation Line Break -

Long Term Response

3A.3.2-25 Pressure Response Main Steam Line Break

C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 Appendix 3A PLANT DESIGN ASSESSMENT REPORT FOR SAFETY/RELIEF VALVES AND LOSS-OF-COOLANT ACCIDENT LOADS LIST OF FIGURES (Continued)

Section Title 3A-xv 3A.3.2-26 Temperature Re sponse - Main Steam Line Break - Minimum ECCS 3A.3.2-27 Temperature Response - Recirculati on Line Break (0.1 ft

2) 3A.3.2-28 Pressure Response - Re circulation Line Break (0.1 ft

2) 3A.4.1-1 Stiffened Containment in Wetwell Region

3A.4.1-2 Displacement Profile SRV Load - All Valves

3A.4.1-3 Displacement Profile Nearly Symmetric Chugging

3A.4.1-4 Stiffener Configuration

3A.4.1-5 Basemat Pl an and Sections

3A.4.1-6 Reactor Building Cross Section

3A.4.1-7 Reactor Pedestal, Diaphragm Floor, and Columns

3A.4.1-8 Pedestal-Interaction Diag ram Axial Load Versus Moment

3A.4.1-9 Drywell Diaphragm Floor Seal

3A.4.1.10 Finite Element M odel Diaphragm Floor Seal 3A.4.2-1 Downcomer Bracing System - Plan and Details 3A.4.2-2 Downcomer Bracing System - Model for Structural Analysis

3A.4.2-3 Diaphragm Floor Colu mns and Adjoining Structures

3A.4.2-4 Diaphragm Floor Column Model for Dynamic Analysis

C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 Appendix 3A PLANT DESIGN ASSESSMENT REPORT FOR SAFETY/RELIEF VALVES AND LOSS-OF-COOLANT ACCIDENT LOADS LIST OF FIGURES (Continued)

Section Title 3A-xvi 3A.4.2-5 Structural Model of Di aphragm Floor Beam and Column 3A.4.2-6 SRV Piping System Inner Ring Quencher Support

3A.4.2-7 Quencher Assembly

3A.4.2-8 SRV Quencher Assembly

3A.4.3-1 Hydrodynamic Loading Zones

3A.5.1-1a Axisymmetric Model of th e Reactor Building and Soil Foundation

3A.5.1-1b Reactor Building Model

3A.5.1-2 Containment Shell Model Cross-Section Details

3A.5.1-3a Top of RPV Pedestal, El. 520 ft Mass No. 44 (Radial)

3A.5.1-3b Top of RPV Pedestal, El. 520 ft Mass No. 44 (Radial)

3A.5.1-4a Top of RPV Pedestal, El. 520 ft Mass No. 44 (Vertical)

3A.5.1-4b Top of RPV Pedestal, El. 520 ft Mass No. 44 (Vertical)

3A.5.1-5a Basemat at RPV Pedest al, El. 435 ft Mass No. 141 Radial) 3A.5.1-5b Basemat at RPV Pedestal , El. 435 ft Mass No. 141 (Radial) 3A.5.1-6a Basemat at RPV Pedestal , El. 435 ft Mass No. 141 (Vertical)

3A.5.1-6b Basemat at RPV Pedestal , El. 435 ft Mass No. 141 (Vertical)

3A.5.1-7a Top of Sacrificial Shield Wall, El. 567 ft Mass No. 14 (Radial)

C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 Appendix 3A PLANT DESIGN ASSESSMENT REPORT FOR SAFETY/RELIEF VALVES AND LOSS-OF-COOLANT ACCIDENT LOADS LIST OF FIGURES (Continued)

Section Title 3A-xvii 3A.5.1-7b Top of Sacrificial Shield Wall, El. 567 ft Mass No. 14 (Radial) 3A.5.1-8a Top of Sacrificial Shield Wall, El. 567 ft Mass No. 14 (Vertical)

3A.5.1-8b Top of Sacrificial Shield Wall, El. 567 ft Mass No. 14 (Vertical)

3A.5.1-9a RPV, El. 545 ft Mass No. 27 (Radial)

3A.5.1-9b RPV, El. 545 ft Mass No. 27 (Radial)

3A.5.1-10a RPV, El. 545 ft Mass No. 27 (Vertical)

3A.5.1-10b RPV, El. 545 ft Mass No. 27 (Vertical)

3A.5.1-11a Containment Vessel, El. 547 ft Mass No. 60600 (Radial)

3A.5.1-11b Containment Vessel, El. 547 ft Mass No. 60600 (Radial)

3A.5.1-12a Containment Vessel, El. 547 ft Mass No. 60600 (Vertical)

3A.5.1-12b Containment Vessel, El. 547 ft Mass No. 60600 (Vertical)

3A.5.1-13a Containment Vessel, El. 448 ft Mass No. 50100 (Radial)

3A.5.1-13b Containment Vessel, El. 448 ft Mass No. 50100 (Radial) 3A.5.1-14a Containment Vessel, El. 448 ft Mass No. 50100 (Vertical) 3A.5.1-14b Containment Vessel, El. 448 ft Mass No. 50100 (Vertical)

3A.5.1-15a Top of RPV Pedestal , El. 520 ft Mass No. 44 (Radial)

3A.5.1-15b Top of RPV Pedestal , El. 520 ft Mass No. 44 (Radial)

C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 Appendix 3A PLANT DESIGN ASSESSMENT REPORT FOR SAFETY/RELIEF VALVES AND LOSS-OF-COOLANT ACCIDENT LOADS LIST OF FIGURES (Continued)

Section Title 3A-xviii 3A.5.1-16a Top of RPV Pedestal , El. 520 ft Mass No. 44 (Vertical) 3A.5.1-16b Top of RPV Pedestal , El. 520 ft Mass No. 44 (Vertical)

3A.5.1-17a Basemat at RPV Pedestal, El. 435 ft Mass No. 141 (Radial)

3A.5.1-17b Basemat at RPV Pedestal, El. 435 ft Mass No. 141 (Radial)

3A.5.1-18a Basemat at RPV Pedestal, El. 435 ft Mass No. 141 (Vertical)

3A.5.1-18b Basemat at RPV Pedestal, El. 435 ft Mass No. 141 (Vertical)

3A.5.1-19a Top of Sacrificial Shield Wall, El. 567 ft Mass No. 14 (Radial)

3A.5.1-19b Top of Sacrificial Shield Wall, El. 567 ft Mass No. 14 (Radial)

3A.5.1-20a Top of Sacrificial Shield Wall, El. 567 ft Mass No. 14 (Vertical)

3A.5.1-20b Top of Sacrificial Shield Wall, El. 567 ft Mass No. 14 (Vertical)

3A.5.1-21a RPV, El. 545 ft Mass No. 27 (Radial)

3A.5.1-21b RPV, El. 545 ft Mass No. 27 (Radial)

3A.5.1-22a RPV, El. 545 ft Mass No. 27 (Vertical)

3A.5.1-22b RPV, El. 545 ft Mass No. 27 (Vertical)

3A.5.1-23a Containment Vessel, El. 547 ft Mass No. 60600 (Radial)

3A.5.1-23b Containment Vessel, El. 547 ft Mass No. 60600 (Radial)

3A.5.1-24a Containment Vessel, El. 547 ft Mass No. 60600 (Vertical)

C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 Appendix 3A PLANT DESIGN ASSESSMENT REPORT FOR SAFETY/RELIEF VALVES AND LOSS-OF-COOLANT ACCIDENT LOADS LIST OF FIGURES (Continued)

Section Title 3A-xix 3A.5.1-24b Containment Vessel, El. 547 ft Mass No. 60600 (Vertical) 3A.5.1-25a Containment Vessel, El. 448 ft Mass No. 50100 (Radial)

3A.5.1-25b Containment Vessel, El. 448 ft Mass No. 50100 (Radial)

3A.5.1-26a Containment Vessel, El. 448 ft Mass No. 50100 (Vertical)

3A.5.1-26b Containment Vessel, El. 448 ft Mass No. 50100 (Vertical)

3A.5.2-1 Reactor Building Model

3A.5.2-2 Containment Vessel, El. 448 ft Mass No. 152 (Radial)

3A.5.2-3 Containment Vessel, El. 448 ft Mass No. 152 (Vertical)

3A.5.2-4 Containment Vessel, El. 459 ft Mass No. 123 (Radial)

3A.5.2-5 Containment Vessel, El. 459 ft Mass No. 123 (Vertical)

3A.5.2-6 RPV Support on Pedestal , El. 519 ft Mass No. 57 (Radial)

3A.5.2-7 RPV Support on Pedestal , El. 519 ft Mass No. 57 (Vertical)

3A.5.2-8 Containment Vessel, El. 583 ft Mass No. 12 (Radial) 3A.5.2-9 Building Wall, El.

521 ft Mass No. 55 (Radial) 3A.5.2-10 Building Wall, El.

521 ft Mass No. 55 (Vertical)

C OLUMBIA G ENERATING S TATION Amendment 57 F INAL S AFETY A NALYSIS R EPORT December 2003 LDCN-02-000 3A.1.1-1 Appendix 3A

PLANT DESIGN ASSESSMENT REPORT FOR SAFETY/RELIEF VALVES AND LOSS-OF-COOLANT ACCIDENT LOADS

This third revision of the "Plant Design Assessment Report," (DAR) together with the safety/relief valve (SRV), condensation oscillation and chugging reports, finalizes the Columbia Generating Station load definition, load application, load combination, and design margins for hydrodynamic loading conditions.

In July 1993 Energy Northwest requested an amendment to the operating license to allow an increase in the power level of the plant. The effects of power uprate on the containment system response are described in NEDC-32141-P. Specifically, Section 4.1 of NEDC-32141-P states that for short-term containment pressure response, the peak pressure values are below design values and remain virtually unaffect ed by power uprate and extended load line limit. The loss-of-coolant accident (LOCA) c ontainment dynamic loads are not affected by power uprate, and SRV containment loads will remain below their design

allowables. Appendix 3A has not been updated to reflect the mi nor changes to the LOCA and SRV load analyses described in NEDC-32141-P.

3A.1.1 CONFORMANCE TO NRC ACCEPTANCE CRITERIA

The DAR specifies the Columbia Generating Station positi on for each of the pool dynamic loads. The table further provides a detailed description of each load, the NRC evaluation, and the Columbia Generating Stati on position on the acceptance for each load. The Columbia Generating Stati on positions (Attachment 3A.H) were discussed, reviewed, and approved by NRC at various meetings with Energy Northwest. The NRC acceptance wa s formalized in the Columbia Generating Station Sa fety Evaluation Report (SER).

C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.1.2-1 3A.1.2 ROLE OF THE DESIGN ASSESSMENT REPORT

The Columbia Generating Station DAR serves th e primary purpose of assessing the adequacy of structures and equipment affected during SRV actuation or a postulated LOCA. This report utilizes the load definiti on data from the SRV repor t, chugging report, DFFR (Reference 3A.3.2-2), and applicable NRC guidelines as outlined in A.H.

Specifically Revision 3 of the DA R serves the following purposes:

a. Summarize the loads and effects agreed upon with NRC which are most important to the design of the plant,

b. Identify the design modifi cations implemented to w ithstand these loads, and

c. Identify the design margins for hydrodynamic loading conditions.

C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.1.3-1 3A.1.3 ASSESSMENT APPROACH

The information developed in the SR V and Chugging Reports (References 3A.3.1-1 and 3A.3.2-15 respectively) together with other information availabl e as outlined in A.H , was used to assess all major structures, systems, and components in the wetwell region. The effects of hydrodynamic loads outside the wetwell region are discussed in the appropriate sections of the FSAR.

C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.1.4-1 3A.1.4

SUMMARY

OF DESIGN ASSESSMENT REPORT CONTENT

The DAR, Revision 3 is summarized as follows:

Chapter 1 a. Introduction to CGS Loads,

b. Review of the purpose of the plant specific loads, and c. Discussion of the assessment of cont ainment components since DAR Revision 2.

Chapter 2

a. General description of the CGS containment, and b. Summary and conclusions.

Chapter 3 A discussion of the manner in which the plant specific loads for CGS are determined, based on information provided in the SRV and chugging reports, DFFR , and other associated and referenced documents.

Chapter 4 A review of the design assessment for the CGS containment system. Assessment and conclusions are included for suppression pool boundary structures (s teel containment, vertical and horizontal tees, basemat, pedestal, diaphragm floor, and diaphragm floor seal) and for suppression pool major structures (downcomer bracing, columns, downcomers, SRV piping systems, quenchers, and platform s), and for suppression pool mi scellaneous piping systems.

Chapter 5 Provides the building response due to SRV discharge, LOCA.

Attachments The attachments to this report provide

a. Attachment not utilized,

b. Theoretical formulation for the calculation of three dimensional source flows in exact containment geometry, C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.1.4-2 c. A method to calculate drag forces on submerged stru ctures caused by hydrodynamic flow fields, d. Calculation models for short term LOCA phenomena,

e. Description of the suppression pool temperature mon itoring system,

f. Description of computer programs u tilized for CGS load definition and plant assessment,

g. Attachment not utilized,

h. Table of conformance of CGS design to NRC acceptance criteria, and

i. Summary of the methodologies used for defining SRV and LOCA loads on submerged structures.

C OLUMBIA G ENERATING S TATION Amendment 54 F INAL S AFETY A NALYSIS R EPORT April 2000 LDC N-9 9-0 0 0 3A.2.1-1 3A.2

SUMMARY

AND CONCLUSIONS

3A.2.1 GENERAL DESCRIPTION OF PLANT

The Energy Northwest Nuclear Project No. 2 (CGS) is a nuclear fueled electrical generating station which utilizes a General Electric Company BWR-5 (1969 product line) nuclear reactor.

The primary containment utilizes a Mark II over/under pressure-s uppression configuration (see Figure 3A.2.1-1

). The primary c o ntainment c o nsists of a s t eel pre s sure vessel enc l osed by a concrete shield wall bot h supported by a concrete basemat. The pr imary containment is enclosed by the reactor building, a reinforced-concrete struct ure functioning as a secondary containment.

The drywell is connected to the suppression chamber by 99 downc omer pipes. Originally 102 downcomer pipes were provided but three were capped, as discussed in Sections 3A.3.2.1 and 3A.3.2.2. Steam that could be released in the drywell during a postulated loss-of-coolant accident (LOCA) is channeled through these downcomer pipes into the suppression pool where it is condensed thus effe cting pressure-suppression.

Eighteen safety/relief valves (SRVs) are mounted on the four main steam lines. When SRVs are actuated, steam from the reactor pressure vessel (RPV) fl ows through the SRV discharge lines into the suppression pool where the steam is condensed.

The discharge lines from all 18 SRVs are routed inside selected downcomers into the suppression chamber (Figures 3A.2.1-6 through 3A.2.1-8). Each discharge line t e r m inates with a quencher device having four arms. Seven of the 18 SRVs are part of the auto matic depressurization system (ADS) (Table 3A.3.1-1) which is designed to function, under certain conditions, following a postulated intermediate or small size line break.

3A.2.1.1 Structures, Piping, and Components Directly Affect ed by Pool Dynamic Loads

The structures in the suppression chamber are shown in Fi gures 3A.2.1-2 t h rough 3A.2.1-8. The structures, piping and compone nts directly affected by the hydrodynamic events associated with the LOCA pressure suppression and the SRV discharge are iden tified below. The applicable hydrodynamic load s are identified in Section 3A.4. a. Boundary elements

The suppression chamber boundary elemen ts are: the st eel containment including the vertical a n d horizontal t ee st i ffe n e rs (Figure 3A.4.1-4

), t h e concrete basemat, the concrete pedestal, the diaphragm floor and the diaphragm floor seal;

C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.2.1-2 b. Major structures and components The major vertical struc t ures are sh o w n in Figure 3A.2.1-2. They are t h e 102 downcomers, the 18 SRV lines including quenchers and support towers, and the 17 concrete columns supporting the diaphragm floor. The major horizontal s t ructures are the steel t r uss shown in Figure 3A.2.1-3 which provides lateral support to the downcomer s and the SRV lines, and the platform at el. 472 ft 4 in. shown in Figures 3 A.2.1-3 and 3A.2.1-8. The downcomer

bracing truss is submerged and the platfo rm is located in the pool swell zone; and

c. Miscellaneous piping systems

A developed elevation of the CGS containment showing the location of the

containment penetrations is shown i n Figure 3 A.2.1-9. The piping systems of the suppression pool are classi fied as described below.

1. Fully submerged piping systems

Eleven piping systems ar e fully submerged in the suppression pool (see Figure 3A.2.1-2

). Seven systems e n ter the pool through containment penetrations at el. 452 ft 0 in. One pipe [4 in.-fuel pool cooling and cleanup (FPC)] enters the pool through th e pedestal at el. 451 ft 8.25 in.

Two short lengths of pipe (instrum entation stubs) enter the pool at

el. 462 ft 0 in. A third (instrumentati on line) enters at el. 455 ft 0 in.

Pipes below the downcomer vent exit at el. 454 ft 4.75 in.

2. Partially submerged piping systems

Thirteen partially submerged pipi ng systems enter the suppression chamber through containment penetr ations at el. 467 ft 9 in. (see Figure 3A.2.1-3) and enter the pool ve r tically within 3 ft 0 in. distance

from the containment as shown in Figures 3A.2.1-6 through 3A.2.1-8. 3. Piping systems in pool swell zone

The pool swell zone is identified in Section 3A.3.2.3 to be between the elevations of the initial pool surface (466 ft 4.75 in.) and the maximum pool rise during a LOCA (design ba sis accident) (484 ft 4.75 in.).

C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.2.1-3 Piping systems in the pool swell zone include short projections into the chamber from the containment at one access hatch and 10 miscellaneous piping systems as shown in Figure 3A.2.1-3 and Figures 3A.2.1-6 through 3A.2.1-8. 4. Piping systems above the pool swell zone

Piping systems above the pool swell zone include short lengths of pipe entering at el. 491 ft 0 in. and two penetrations for the wetwell spray header also at el. 491 ft 0 in. as shown in Figu r e 3A.2.1-4 and Figures 3A.2.1-6 through 3A.2.1-8.

3A.2.1.2 Structures, Piping, and Components Indirectly Affected by Pool Dynamic Loads

In the drywell region the containment structures , piping, and components are also affected by pool dynamic loads. This is a result of loading applied to th e suppression chamber boundary (basemat, pedestal, and containment shell) wh ich would result in vibratory motion transmitted through the reactor pedestal and the primary containment. This is commonly referred to as "building response."

Primary and Secondary Containment Structure 950021.60 3A.2.1-1 Figure Amendment 53 November 1998Form No. 960690.veR.oN .warD El. 581-10 1/2" El. 548 El. 521' - 0 1/4" El. 499' - 6" El. 471 El. 435' - 3" Steel Primary Containment Vessel Dryer & Separator Storage PoolBiological Shield WallDrywell Floor Seal Sand Filled Pockets Sacrificial Shield El. 522' - 0" Suppression ChamberFoundation Mat Outer Skirt Inner Skirt El. 466' - 4 3/4" (Max)

Suppression Pool Water LevelBridge Girder 125 Ton Bridge CraneHigh Point Top of Steel El. 668'-9 1/4" Roof Truss Top Of Parapet Steel El. 671' - 2" Reactor BuildingStorage Area 77' - 0" 143' - 0"Section Looking South Columns Down Comer VentsDrywell FloorDrywell AreaStabilizer Truss Reactor Well Pool Cavity El. 606' - 10 1/2"Refueling Platform El. 598' - 8"Crane Runway Girder Reactor Pedestal 85' - 9" I.D.20' - 3" I.D.22'-8 7/8" O.D.

R.P.V.31' - 8" I.D.

El. 428' - 3 1/4" El. 435' - 3" El. 446' - 0" 47' - 9 3/4" El. 572' 0" El. 422' - 3" Bldg. Ins. Dia.127'-0" x 143'-0" El. 492'-8 1/4"18' - 9" I.D.

El. 501'El. 519'-2 1/4" 39'-6 1/2" Columns 26'-3 1/8" 7'-10 3/4" 7'-8" Fuel Storage Pool El. 583' - 1/4" 3'-6 3/4" Columbia Generating Station Final Safety Analysis Report

Draw. No.

Rev.Figure

Draw. No.

Rev.Figure

Draw. No.

Rev.Figure 3A.2.1-7 Amendment 58 December 2005Suppression Pool Composite Sections"3-3" and "4-4"SM251 Columbia Generating Station Final Safety Analysis ReportDraw. No.Rev.Figure Form No. 960690FH 0

Figure Not Available For Public Viewing C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.2.2-1 3A.2.2

SUMMARY

OF CHANGES AND CONCLUSIONS

3A.2.2.1 Summary of Changes to Preserve Design Margins

Structures, piping, and compone nts which were affected by pool dynamic loads were divided into two general categories, i.e., those directly affected by pool dynamic loads (those in and bounded by the suppression chamber) and those affected only indi rectly by pool dynamic loads (outside the suppression cham ber). The Design Assessment Report (DAR) addressed the structures, piping, and components in and bounding the suppression chamber. For these structures several changes in design were implemente d as a result of c onsideration of SRV discharge and LOCA hydrodynamic loads.

Table 3A.2.2-1 provides a list of the structures and components that were impacted by the DAR and includes the desi gn margins, controlling load combination, and the design changes that have been made. The steel containment structure has been reinforced by the addition of seven horizontal rows of tee stiffeners as shown in F i gure 3A.4.1-1. The downcomer b r acing system has been redesigned from a system of radial beams to a pipe truss system. This bracing syst em also is designed to provide lateral restraint for the SRV discharge pipes. Quenchers have been provided as exit devices for the SRV discharge pipes. Additions and mo difications of pipe s upports for miscellaneous piping systems have been provided. Other miscellaneous changes are noted in Table 3A.2.2-1.

3A.2.2.2 Conclusions

The DAR, Revision 3 concluded that the modified design of the wetwell for CGS is capable of withstanding the effects of the hydrodynamic loads resulting from SRV actuation and postulated LOCA events in conjunction with other applicable loads.

The effects due to hydrodynamic loads outside the wetwell region are discussed in the FSAR.

C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 Table 3A.2.2-1 Suppression Pool A ssessment Summary Structure Controlling Margi n a Controlling Load Combinatio n b Changes to Structures Due to SRV and LOCA Loa d s 3A.2.2-3 Steel containment 1.26 3 Added horizontal tee stiffeners, revised platform

location, and connection to

containment.

Basemat Bending - 1.14 Shear - 1.27 7 None Pedestal diaphragm

floor 1.11 Downward - 1.62

Uplift - 1.27 4

4a None None Diaphragm floor seal See Section

3A.4.1.5.5 See Section

3A.4.1.5 None Downcomer 1.68 5 Redesigned bracing system as a pipe truss sy stem Columns 1.19 1 None Downcomers 1.08 See Section 3A.4.2.3 Added stainless-steel spool piece, provided local

reinforcement where SRV

pipe penetrates downcomer, raised the vacuum breaker

valves out of the pool swell zone. Capped three downcomers.

SRV piping systems 1.05 See Section 3A.4.2.4 Provided lateral restraint at

downcomer bracing system, rerouted SRV lines. Quenchers 1.25 See Table 3A.4.2-3 Added quencher device and support to the end of SRV

lines C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 Table 3A.2.2-1 Suppression Pool Assessme nt Summary (Continued)

Structure Controlling Margin a Controlling Load Combination a Changes to Structures Due to SRV and LOCA Loads 3A.2.2-4 Platforms and ladders 1.18 5a Removed floor plate, replaced with grating, revised locations and

connections to containment, added platform supports to service vacuum breaker valves; strengthened grating

connections and supports. Miscellaneous

wetwell piping system See Section 4.3 See Section 3A.4.3 Added and modified pipe supports. Stiffened penetrations. Relocated two piping systems.

a See Section 3A.4 for a complete discussion.

b See Section 3A.3.5 for a complete discussion.

C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.3.1-1 3A.3 CONTAINMENT DYNAMIC FORCING FUNCTIONS 3A.3.1 LOADS ASSOCIATED WITH SAFETY/RELIEF VALVE ACTUATION

3A.3.1.1 Description of the Safety/Relief System

The safety/relief system is comprised of 18 safety/relief valves (SRVs) connected to the main steam lines in the drywell chamber. From each of the valves, a discharge line with two vacuum breakers is routed from the drywe ll into the wetwell where it terminates in a quencher in the suppress i on pool, as shown in F i gures 3A.2.1

-3 to 3A.2.1-8. To pass through the drywell floor, the discharge lines are r outed through selected downcomers to about el. 490 ft. At this elevation, they exit the downcomers and ar e routed horizontally, as shown in Figure 3A.2.1-4 , to points di r ectly above their respective quen c hers. They then are routed

vertically down into the suppr e ssion pool, as shown in Figu r e s 3A.2.1-6 to 3A.2.1-8. Under normal operating conditions, each quencher is filled with water and its discharge line is filled to the same level as the surf ace of the suppression pool.

The remaining volume of the discharge line up to the SR V is filled with air.

Table 3A.3.1-1 summarizes the characteristics of the safety/relief system.

3A.3.1.2 Description of the Phenomena and Resulting Loads

During plant operation, if one or more of the SRVs is actuated , three transient events occur consecutively for each:

a. The water in each quencher and discharge line is expelled into the suppression pool through the holes in the quencher arms,

b. The air which fills each discharge line is expelled into the suppression pool, and

c. The steam from the discharge line being vented is expelled into the suppression pool and condensed.

Each of these events creates di sturbances in the suppression pool.

The first creates water jets, the second creates air discharge related pressure oscillations and the third creates pressure fluctuations as the steam is condensed. These disturbances, in turn, produce hydrodynamic loads both on the structures which are submerged in the suppression pool and on the pool

boundaries. Sections 3A.3.1.2.1 , 3A.3.1.2.2 , and 3A.3.1.2.3 briefly describe the characteristics of the load producing phenomena, while Section 3A.3.1.3 discusses the loads produced.

C OLUMBIA G ENERATING S TATION Amendment 54 F INAL S AFETY A NALYSIS R EPORT April 2000 LDC N-9 9-0 0 0 3A.3.1-2 3A.3.1.2.1 Water Clearing Loads

When an SRV opens and permits steam to pass , the steam flow compre sses the air above the water standing in the discharge line increasing the line pressure. The increased air pressure forces the water into the suppression pool through the holes in each side of the quencher arms.

As the water flows from the adjacent side s of the quencher arms , it coalesces into four turbulent jets. These jets flow away from the quencher pr oducing acceleration and standard drag loads on structures in their paths.

Due to the turbulent na ture of the jets, their momentum diffuses rapidly and their effective velocity decreases to zero in a distance comparable to a quencher arm length.

Based on the results of scal ed experiments and Caorso test results (Reference 3A.3.1-3), the region throughout which the jets produce a significant load is small, at most existing to a distance comparable to a quencher arm length. There are no structures located in this region except a small intrusion of the concrete column which is designed for significantly higher air clearing loads. Detailed defi nition of water clearing loads therefore is unnecessary for the purposes of this report. It is noted that a clear demarcation between the water clearing loads and air clearing loads is not possi ble from the recorded Caorso test data. For the purpose of this report it is assumed that significant pressure peaks represent air clearing loads and that the slowly rising pressure before the first signifi cant peak represents water clearing loads.

3A.3.1.2.2 Air Clearing Loads

Following the expulsion of water from the SRV di scharge line, the air trapped in the line is forced through the holes in the sides of the quencher arms into the suppression pool. As a result of this disturbance of the pool, oscilla tions are produced in th e pool which induce time varying pressure and velocity fields. These fields crea te acceleration drag loads on the submerged structures and time varying loads on the pool boundaries. The definition of these air clearing loads is provided in Section 3A.3.1.3.

3A.3.1.2.3 Steam Condensation Loads

After the water and air have been expelled fr om the SRV discharge line, high pressure, high temperature, high mass flux stea m is discharged into the s uppression pool. As the steam condenses and collapses, vibrations or small pressure fluctuations are produced in the water.

Suppression pool temperature tran sient analyses were performe d for CGS, for a stuck open relief valve, isolation scram, and small break accident (SBA), in accordance with the requirements of NUREG-0783. The peak pool temperature for each of these cases is maintained within the limits of NUREG-0783. As a result unstable steam conde nsation due to extended SRV discharge to the s uppression pool will not occur. Details of this analysis are provided in Reference 3A.3.1-7. The suppression pool temperature monitoring system is discussed in FSAR Appendix 3A Attachment 3A.E.

C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.3.1-3 3A.3.1.3 Safety/Relief Valve Air Clearing Loads Testing and analytical efforts have been performed by the Mark II Owners' Group to define the loads resulting from discharg e through a quencher device upon actuation of the SRV. The SRV testing carried out at the Caorso plant in Italy represents the most extensive test program to date with geometry and pl ant conditions similar to CGS. An analytical effort was undertaken by Burns and Roe to ev aluate the data taken during th e Caorso Phase I and II tests (References 3A.3.1-3 and 3A.3.1-4) which resulted in an improved SRV load definition and application methodology for Mark II containmen ts (Reference 3A.3.1-1). These results and a detailed description of the an alysis have been submitted, re viewed, and approved by the NRC (Reference 3A.3.1-6) as part of the SRV reports (Reference 3A.3.1-1).

Hydrodynamic loads due to an SRV discharge affect both pool bounda ries and submerged structures. A summary of the improved load definition for the specific applications is provided in Sections 3A.3.1.3.1 and 3A.3.1.3.2.

3A.3.1.3.1 Boundary Loads

Expulsion of water and air in a discharge line during an SRV disc harge creates di sturbances in the suppression pool which induce dynamic pres sure loads on the pool boundary. Resulting dynamic effects depend upon the definition of a ri gid wall pressure incide nt on this boundary. Analytical interpretations and subsequent definitions of the spatial pressure distribution, pressure wave forms, and maximum pressure amplitudes recorded during the Caorso tests are the basis for defining the design boundary pressure load. Convers ion of these Caorso results for application to CGS requires a correlation be tween the test conditions at Caorso and the design condition at CGS. Reference 3A.3.1-1 details this correlation along with the derivation of the design boundary pressure with regard to all the possible SRV discharge events that may occur during the life of the plant (see 3A.4.1.1.1.1

). Sections 3A.3.1.3.1.1 , 3A.3.1.3.1.2 , and 3A.3.1.3.1.3 summarize the more important aspects of the derivation.

3A.3.1.3.1.1 Spatial Distribu tion of Boundary Pressures.

The spatial distribution of boundary pressures during an SRV discharge contributes to a comp lete definition of the design boundary pressure load.

Based on analytical studies of data recorded at Caorso, it was concluded that the spatial distri bution of pressure is independent of the time variable. As stated in Reference 3A.3.1-1 , this implies that the maximu m pressure amplitude can be representatively used when st udying the spatial distributions.

The circumferential pressure distribution (Figu r e 3A.3.1-1) as well as t h e vertical pressure

distribution (Figure 3A.3.1-2) adopted for t h e SRV load specification is obtained th r ough comparisons of various availa ble distributions (Reference 3A.3.1-1). It should be noted that both distributions have been normalized for 1 psi peak pressure.

C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.3.1-4 3A.3.1.3.1.2 Pressure Wave Forms.

The SRV boundary pre ssure load specification depends on the evaluation of the pressure wave forms measured during the Caorso tests.

Based on the experimental data recorded, two distinct characteristic wave forms prevail, the multiple frequency pressure (MFP) wave form and the single frequency pressure (SFP) wave form.

The design MFP wave form, shown in Figure 3A.3.

1-3 , re f lec t s the characte r istics of all s u ch MFP wave forms measured at Caorso. Initially , there are several pressure spikes as the pressure wave reaches the pool boundary. They exhibit a frequency c ontent in the range of 15.0 Hz to 40 Hz. Following the pressure spikes are damped oscillations exhibiting primarily a single frequency in the range of 6.0 Hz to 10.0 Hz. Maximum pressure amplitude occurs in the initial period and decays rap i dly thereaf t e r. Figure 3A.3.

1-4 illustra t e s the frequency spectrum of the pressure amplitude trace indicating rich frequency content in the range of 15.0 Hz to 40.0 Hz and a distinct peak in the ra nge of 6.0 Hz to 10.0 Hz. There is negligible frequency content beyond 40.0 Hz.

The design SFP wave form, shown in Figure 3A.3.

1-5 , re f lec t s the characte r istics of all s u ch SFP wave forms measured at Caorso.

Unlike the MFP wave form, a single characteristic frequency of oscillation predominates for the entire time history. As shown in the corr e sponding fr e quency spectrum (Figu r e 3A.3.1-6

), the dominant frequency is in the range of 6.0 Hz to 10.0 Hz. Again, there is negligible frequency content beyond 40.0 Hz.

Due to the randomness and variability in the characteristic/dominant frequencies of the pressure wave forms recorded , the time histories of the de sign pressure wave forms are compacted and expanded to obtain a characteristi c frequency covering the range of 4.0 Hz to 12.0 Hz at intervals of 1.0 Hz.

As a result, each type of wave form (MFP and SFP) is depicted by nine separate pressure time hist ories (see Reference 3A.3.1-1 for details).

3A.3.1.3.1.3 Design Maximum Pressure Amplitude.

Conversion of the Caorso maximum pressure amplitude computed for application to CGS yields a design maximum pressure amplitude of 15.0 psi (References 3A.3.1-1 and 3A.3.1-5). This is the rigid wall pressure incident on the suppression pool boundary resulting from an SR V actuation. Application of this design boundary pressure for all SRV discharge cases that may possibly occur during the life of the CGS plant, as specified in the DFFR (Reference 3A.3.1-2), is discussed in Section 3A.4.1.1.1.1. As discussed in Reference 3A.3.1-2 , the design pressure reflects a 90% confidence level and 90%

probability of nonexceedence.

3A.3.1.3.2 Submerged Structure Loads

The methodology for calculating the loads on submer ged structures during SRV discharge uses the predicted pressure time histories directly rather than the velocity and acceleration C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.3.1-5 transients. The pressure pr edictions are based on the pressures measured on submerged structures in Caorso tests and their correlation with the boundary pressure loads.

3A.3.1.3.2.1 Peak Safety/Relief Valve Dynamic Loads. The methodology us ed to define peak SRV dynamic loads on submer ged structures is outlined in A.I (see also Reference 3A.3.1-8).

3A.3.1.3.2.2 Time Dependence of Safety/Relief Valve Loads and Dynamic Load Factors.

The pressure time histories re corded on submerged structures at Caorso show waveform characteristics sim ilar to those recorded at pool boundary. As indicated in Reference 3A.3.1-1 , boundary pressure time histories cons ist of SFP time histories and MFP time histories.

Dynamic load factor (DLF) ver s us frequency curves are pre s ented in Figure 3A.3.

1-10.

A typical curve, such as the curve labeled SF P, 1% damping, was calculated as follows:

a. Response spectra that co rrespond to all the SFP time histories described in Section 3A.3.1.3.1.2 are computed using 1% damping,

b. The response spectra obtain e d in (a) are enveloped, and

c. The DLF curve is obtained from the response spectrum envelope by dividing the responses at various frequencies by the zero period response.

3A.3.1.3.2.3 Safety/Relief Valve Loads on Structures. Loads on submerged structures are shown for the submerged structures listed below.

Unless otherwise noted, calculation of an equivalent static load is completed via a DLF as obta i ned from Figure 3A.3.1-10. a. Downcomers Figure 3A.3.1-7 shows dynamic load on a downcomer;

b. Columns Figure 3A.3.1-8 shows dynamic load on a conc r e te column. Note, for the subsequent actuation load case the maximum direct pressure load is shown (see Reference 3A.3.1-8). Reference 3A.3.1-9 computes the maximum dynamic reaction of the column, and these time-history analyses results are applied in Reference 3A.3.1-10 for final column structural (i.e., code) qualification.

Table 3A.4.2-2 tabulates maximum column reaction load results;

C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.3.1-6 c. Safety/relief valve lin e and quencher supports Figure 3A.3.1-9 shows dynamic load on S R V line and the quencher support.

Horizontal flow past an SRV line due to actuation of its quen c her is negligible;

and

d. Piping, supports, and bracing truss

Figure 3A.3.1-11 shows equ i valent static loads on p i ping, supports, and bracing truss. 3A.3.1.4 References

3A.3.1-1 "SRV Loads - Improved Definition and Application Met hodology for Mark II Containments," Technical Re port, Burns and Roe, Inc

., July 1980. Transmitted to NRC by letter GO2-80-172, of August 8, 1980.

3A.3.1-2 "Mark II Containment Dynamic Forcing Functions Information Report (DFFR)," General Electric Company, NEDO-21061 , Revision 4, November 1981.

3A.3.1-3 "Mark II Containment Supporting Program Caorso SRV Discharge Tests, Phase I Test Report," General El ectric Company, NEDE-25100-P , May 1979.

3A.3.1-4 "Mark II Containment Supporting Program Caorso SRV Discharge Tests, Phase II ATR Report," General El ectric Compa ny, NEDE-25118.

3A.3.1-5 Letter, GO2-82-35, 1/13/82, "Responses to CSB Open Items 44 through 48,"

G. D. Bouchey (WPPSS) to A. Schwencer (NRC).

3A.3.1-6 "Safety Evaluation Report Related to the Operation of WPPSS Nuclear Project No. 2," NUREG-0892, Supplement No. 1, USNRC, August 1982.

3A.3.1-7 "Suppression Temperature Analysis," Stone and Webster Report 14057-4 (D-1). Transmitted to NRC by letter GO2-81-524 of December 15, 1981.

3A.3.1-8 Supply System Calculation No. ME-02-93-22, "SRV Air Clearing Loads on Diaphragm Slab Columns."

3A.3.1-9 Supply System Calculation No. CE 93-12, "Diaphragm Slab Column Evaluation for SRV Air Clearing Loads."

C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.3.1-7 3A.3.1-10 Burns & Roe Calculation No. 6.19.36, Book SV-830, "Calculation for Reactor Building Diaphragm Floor Column and Plant DAR for Hydr odynamic Loads, Rev. 3."

C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 Table 3A.3.1-1 Summary of Safety/Rel i e f Sys t em C h aracter i s t i cs 3A.3.1-9 1. Number of SRVs 18 2. SRV manufacturer Crosby P.S.P 3. Designations of SRVs and pressure setpoints (Refer to Figure 3A.2.1-2) Valves (psi) 1 B a , 1 C a 1165 1 A a , 2 B a , 2C a , 1 D a 1175 2A, 3B, 3C, 2D 1185 3A, 4B, 4C, 3D 1195 4A, 5B, 5C, 4D 1205 4. Number of automatic depressurization valves

[automatic depressurization system (ADS)]

7 5. Designations of aut o matic depressurization valves MS-RV-3D, 4A, 4B, 4C, 4D, 5B, 5C

6. Nominal range of valve opening ti m es (ms)20-150 7. Nominal range of nameplate steam rates (lbm/sec) 236.33-251

.81 8. Number of vacuum br e a kers per discharge line 2 (in parallel)

9. Size of each vacuum br e a ker (in.)

10 10. Each discharge line pipe siz e s (in./schedule) 10/80 expanded to 12/80 at approximate

el. 493 ft

11. Discharge line rage of lengths to normal water level (ft) 107.99-161

.99 12. Discharge line range of air volumes to normal water level (ft

3) 57.2-88.1 13. Depth of suppression p ool at RPV pedestal to normal water level (ft) 31.0 14. Submergence of quenche rs to high water level (ft) 17.4 C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 Table 3A.3.1-1 Summary of Safety/Rel i e f Sys t em C h aracter i s t i cs (Continued) 3A.3.1-10

15. Elevation of inner/out er ring quenchers above basemat (ft) 13.6-8.2 16. Quencher area defined by circumscribed circle (ft 2) 74.6 a These val v es a r e the l o w setpoint valves whi c h are prone to subsequent actuation. However, subsequent actuation may occur (though unlikely) in the higher set pressure SRV g r oups (see also discussion contained in Reference 3A.3.1-8).

Normalized Design Circumferential Distribution ofPool Boundary Pressures at Containment 950021.64 3A.3.1-1 Figure Amendment 53November 1998 Form No. 960690Draw. No.Rev.Quencher 1.0 R = 39.5'Columbia Generating StationFinal Safety Analysis Report Normalized Design Vertical Distribution of PoolBoundary Pressures at Containment 950021.65 3A.3.1-2 Figure Amendment 53November 1998 Form No. 960690Draw. No.Rev.1.0 0.0 20 15 10 5 0 Sq= Quencher Submergence (NWL)Pressure (PSI)

Depth (Ft.)

Columbia Generating StationFinal Safety Analysis Report MFP Design Wave Form (Normalized)Time History 950021.66 3A.3.1-3 Figure Amendment 53November 1998 Form No. 960690Draw. No.Rev.Pressure (psi)Time (Sec)0.000.200.400.600.801.00 1.20-1.0-0.5 0.0 0.5 1.0 1.5

2.0 Columbia

Generating StationFinal Safety Analysis Report MFP Design Wave Form (Normalized) Amplitudeof Frequency Spectrum 950021.67 3A.3.1-4 Figure Amendment 53November 1998 Form No. 960690Draw. No.Rev.Pressure (10

-3 PSI)Frequency (Hz.)020406080100120 0 8 16 24 32 40 48 140 160 Columbia Generating StationFinal Safety Analysis Report SFP Design Wave Form (Normalized)Time History 950021.68 3A.3.1-5 Figure Amendment 53November 1998 Form No. 960690Draw. No.Rev.Pressure (psi)Time (Sec)0.000.400.801.201.602.00

-0.80-0.40 0.00 0.40 0.80 1.20 2.40 2.80-1.20 Columbia Generating StationFinal Safety Analysis Report SFP Design Wave Form (Normalized)Amplitude of Frequency Spectrum 950021.69 3A.3.1-6 Figure Amendment 53November 1998 Form No. 960690Draw. No.Rev.Frequency (Hz)0204060 100 40 80 120 160 200 240 120 140 0 80 160 Columbia Generating StationFinal Safety Analysis Report Pressure (10

-3 psi)

SRV Air Clearing Load Distribution on a Downcomer 950021.70 3A.3.1-7 Figure Amendment 53November 1998 Form No. 960690Draw. No.Rev.0.0 2.0 4.0 6.0 8.0 10.0 12.0012345 0.75 psi 3.0 psi Submergence (ft)

El. 466' - 4 3/4" El. 454' - 4 3/4"Initial Actuation Loading CasesSubsequent Actuation

Loading Cases El. 461 - 0" Horizontal Dynamic Load (psi) 24" or 28" Downcomer Notes: 1. Dynamic load factors are obtained from Figure 3A.3.1-10

.2. Horizontal loads on downcomers are applied in any horizontal direction producing maximum load effects.

L C Columbia Generating StationFinal Safety Analysis Report Amendment 55 May 2001SRV Air Clearing LoadDistribution on a Concrete Column 950021.71 3A.3.1-8 Figure Form No. 960690Draw. No.Rev.EL. 466' - 4 3/4" EL. 442' - 4 3/4" C L Submergence (ft) 0.0 2.0 4.0 6.0 8.0 10.0 12.00246810 1.33 PSI 1.67 psiSubsequent Actuation Loading Case -Effective max Column Frontal Pressure Based

on Nodal Forces of Ref 3A.3.1-8 on 3 ft Column Increments.

EL. 461 - 0" Horizontal Dynamic Load (psi) 14.0 16.0 18.0 20.0 22.0 24.0 EL. 449' - 0"1214Initial Actuation Loading Cases 12" 8.61 psi 13.05 psi 5.67 psi 8.33 psi 42" Concrete

Column Notes 1. Dynamic load factors for initial actuation load cases are obtained from

Figure 3A.3.1-10 , dynamic load response reactions for the subsequent actuation

case are tabulated in Table 3A.4.2-2 (see also References 3A.3.1-9 and 3A.3.1-10).2. Horizontal loads on concrete column are applied in any horizontal direction producing maximum load effects.

Columbia Generating StationFinal Safety Analysis Report SRV Air Clearing Load Distribution on anSRV Discharge Line and Quencher Support 950021.72 3A.3.1-9 Figure Amendment 53November 1998 Form No. 960690Draw. No.Rev.Notes 1. Dynamic load factors are obtained from Figure 3A.3.1-10

.2. Horizontal loads on SRV lines are applied in any horizontal direction producing maximum load effects.

EL. 449' - 0" 1.2 psi 3.0 psi 1.5 psi Initial Actuation Loading Cases Subsequent actuation

Loading Cases EL. 461' - 0" 0.8 psi 0 5 10 15 20 25 30 Submergence (ft) 0.3 psi01.02.03.04.0 Horizontal Dynamic Load (psi)

EL. 466' - 4 3/4" EL. 435' - 3" C L12" SRV Discharge Line 12" 0.8 psi Columbia Generating StationFinal Safety Analysis Report Dynamic Load Facto r Versus Frequencyto be Used for Defining SRV Load onSubmerged Structures 950021.73 3A.3.1-10 Figure Amendment 53November 1998 Form No. 960690Draw. No.Rev.Note:1. The curve labeled as MFP is used with initial actuation cases and the curves labeled as SFP

are used with the subsequent actuation cases.

0 20 40 60 80 100 Frequency (Hz)

Dynamic Load Factor 5 4 3 2 1 5.1 4.2 3.25 2.6 2.24 1.76 2.52 2.34 2.2 2.1 2.05 1.92 1.52 1.4 1.3 1.08SFP, 1% Damping for Pipes 12" DiameterSFP, 2% Damping for Pipes > 12" DiameterSFP, 4% Damping for Concrete ColumnMFP, 1% Damping for all Structures and Piping Columbia Generating StationFinal Safety Analysis Report SRV Air Clearing Load Distribution onPiping, Supports and Bracing Truss 950021.743A.3.1-11a Figure Amendment 53November 1998 Form No. 960690Draw. No.Rev.ELE VA TION PLAN PedestalContainment WallVertical Pipe Segment ofInterest (Typ.)Vertical Pipe Segment of Interest (Typ.)

Pedestal Pipe Support(Typ.)Containment Wall El. 466' - 4 3/4"Bracing Truss El. 455' - 4" El. 449' - 0" El. 446' - 0" El. 435' - 3" Columbia Generating StationFinal Safety Analysis Report Amendment 55 May 2001SRV Air Clearing Load Distribution onPiping, Supports and Bracing Truss 950021.753A.3.1-11b Figure Form No. 960690Draw. No.Rev.Notes:1. The load is applied along the line joining the center of the nearest actuating quencher and the geometric center of the pipe segment as shown in Fig. 3A.3.1-11a

.2. The equivalent static load (p 1 lb) on piping, supports and bracing truss is: P = 0.32D 2Lfor initial actuation. P = 0.25KD 2L for subsequent actuation and for pipes near containment or pool bottom. P = 0.64KD 2Lfor subsequent actuation and for pipes near pedestal.

where:

D = diameter of pipe or the cylinder circumscribing a non-circular cross-section, in units of inches. L = length of segment, in units of inches

K = see note 5

3. For load on piping, supports and bracing truss, the load component parallel to the pipe is neglected.

4. Since the load direction on piping, supports and bracing truss generally varies from one point to another, segmentation of the piping

or structural component may be needed.

5. If the fundamental natural frequency of the piping, support or bracing truss member is greater than or equal to 17HZ, K = 1. This is applicable If the fundamental natural frequency is less than 17HZ, then K = , where DLF is the dynamic load facor and is determined from Fig. 3A.3.1-10

.If K, as calculated above, is less than 1, K = 1 is used.

in most cases.

2.6 DLF Columbia Generating StationFinal Safety Analysis Report C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.3.2-1 3A.3.2 LOADS ASSOCIATED WITH LOSS-OF-COOLANT ACCIDENT

A loss-of-coolant accident (LOCA) occurs when the integrity of the reactor coolant pressure boundary is breached and coolant is released. In order to contain the coolant which flashes to steam, CGS utilizes a GE Mark II pressure suppression system. The LOCA loading phenomena are discussed in Section 3A.3.2.2. The short-term and long-term LOCA loads are discussed in deta il in Sections 3A.3.2.3 and 3A.3.2.4, respectively. Section 3A.3.2.5 describes the LOCA pressure and temperature transients and Section 3A.3.2.6 describes the CGS building response to the LOCA loads.

3A.3.2.1 Description of Pressure Suppression System

The CGS primary containment utilizes a GE Mark II over/under pressure-suppression configuration (see Figure 3A.2.1-1

). The dry w ell and supp r ession cha m ber (or wetwell) are

large sealed volumes designed to contain and c ondense escaping reactor coolant. Both contain structures and piping systems with the suppr ession chamber approximate ly half filled with water (suppression pool) for steam quenching. The drywell is connected to the suppression pool by 99 downcomer pipes th at channel steam released dur ing a LOCA for quenching and pressure suppression. Details of the downcomers, other piping systems and structures in the suppression chamber are shown in F i gures 3A.2.1-2 t h rough 3A.2.1-8.

Originally, 102 downcomer pipes we re provided. During the asse ssment of the wetwell piping for hydrodynamic loads, it was found that the LOCA water jet loads on three containment

vessel penetrations and their supports were excessive. To el iminate these load s, three of the 102 downcomers are capped in the drywell region, leaving 99 for venting steam to the wetwell.

The piping systems involved are at penetrati on X-31, X-34, and X-36. The three capped downcomers are on the outer circle of downcomers closest to the containment vessel, at azimuths 95°, 63°, and 42° (Figu r es 3A.2.1-2 and 3A.2.1-4).

3A.3.2.2 Description of the Phenomena and Resulting Loads

The sequence of LOCA-generated hydrodynamic events descri bed below cause dynamic loads on the containment and on structur es and components located in th e wetwell. Th ese transient dynamic forces (see Table 3A.3.2-1) are termed dynamic forcing functions and are discussed in detail in Reference 3A.3.2-2 and summarized below. Section 3A.3.4 discusses the sequence of LOCA generated loads.

Following a postulated LOCA, released coolant causes the drywell pr essure to rise rapidly and to accelerate the column of wa ter in each downcomer downward due to the pressure rise. As the water exits the downcomers and enters the suppression pool , it forms a jet-pool interface which rolls into a mushroom shaped vortex ri ng. Expulsion of water out of each of the C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.3.2-2 downcomers results in a water jet which pr oduces loads on submer ged structures and suppression pool boundary pressure loads. Because bulk pool ve locities are small during vent clearing, the corresponding impact and induced fl ow field drag loads are generally small.

However, locally, significan t drag loads may result.

Immediately after vent clearing, air

• in the downcomer vents from the drywell beings to flow into the suppression pool. The LOCA air

• bubbles are formed at the exits of the vents which charge and expand under the entire pool surface causing thr ee dimensional drag loads on submerged structures. On contact with each other, the i ndividual bubbles coalesce and accelerate the pool water above the downcomer vent exit plane vertically with no significant horizontal water motion.

Pool swell is the upward movement of suppressi on pool water above the exit plane of the vents due to injection of drywell ai r* below the pool surface. The velocity and acceleration of the water slug associated with th is phenomenon produces impact , drag, and lift forces on structures within the swell zone.

The containment boundaries also experience loads due to drywell pre ssurization, air bubble pressure and wetwell free airspa ce compression. The rising pool surface motion is slowed due to the compression of air in the suppression cham ber airspace. At about this time the rising bubbles break up the remaining pool slug which fa lls back to its original position terminating pool swell.

During pool swell, the bottom of the rising wate r slug continually falls back to the suppression pool due to instabilities at the bubble/water slug interface. This phenomenon and the large scale falling back of the rema ining water slug at pool swell termination is known as fallback and causes drag and lift loads on structures in the pool swell zone, but a negligible containment boundary load.

Pool swell and the subsequent fallback of the remaining water slug are followed by an air/steam mixture flow through the downcomers until the drywell air is completely purged and the mass flux becomes pure steam. The loading phenomena associ ated with a high or medium steam mass flux is te rmed steam condensati on oscillation (CO).

The air content and steam mass flow rate along with the pool temperature determines the behavior of the steam/suppression pool water interface. At high steam flow the interface location is essentially constant. As the flow rate decreases, due to reactor depressurization and associated drywell pressure decrease, the interface takes on an oscillatory character. The rate

• During a LOCA an air (nitrogen)/steam mixture would be blown down the downcomer vents; however, the analytical models of LOCA phenomena conservativel y assume that only noncondensables are injected into the suppression pool.

C OLUMBIA G ENERATING S TATION Amendment 54 F INAL S AFETY A NALYSIS R EPORT April 2000 LDCN-99-000 3A.3.2-3 of change of the displacement of the interface is reflected in s ubmerged structure and suppression pool boundary loads.

When the steam mass flux decreases below a critical le vel a hydrodyna mic phenomenon termed chugging occurs. Chugging is associ ated with low steam mass flow and high suppression pool boundary pressu re spikes relative to CO. The phenomenon appears to be random in time and is caused by the complex in teraction of water/steam condensation surface instabilities with the physical properties of the downcom ers, the suppression pool, and the suppression pool boundaries.

Chugging causes loads on the s uppression pool boundary and submerged structures.

To determine the effects of downcomer capping on the hydrodynamic load definition, displacement time histories due to chugging are obtained at selected nodes on the containment vessel. Two sets of time histories are obtained. The first in cludes the effects of downcomer capping and the second excludes th ese effects. A comparison of the maximum displacements indicates that capping results in a slight reduction in the maximum displacements. Since lower displacements are associated with lower loads and stresses, it is conservative to assume uncapped downcomers for esta blishing the chugging load de finition. The Containment Functional Design Analysis described in Section 6.2 was based on the venting capacity of 99 downcomers.

3A.3.2.3 Short-Term Loss-of-Coolant Accident Loads

Short-term LOCA loads are asso ciated with hydrodynamic relate d phenomena that occur within a few seconds after LOCA initiation. The short-term loading phenomena include downcomer vent clearing, LOCA bubble charging, pool swell, and fallback.

Figure 3A.3.2-1 illustrates the short-term loading phenomena. The flow fields during downco mer vent clearing and LOCA bubble charging are three-di mensional. Flow fields during pool sw ell and fallback are vertical and exist only above the vent exit elev ation. Loads on submerged structures due to downcomer vent clearing and LO CA bubble charging are compared and the larger loads are employed for subse quent evaluations.

Section 3A.3.2.3.1 and Table 3A.H-1 provide a detailed summary of the short-term LOCA loading phenomenon and load calculation procedures used to assess CGS structures.

3A.3.2.3.1 Analytical Models and Supporti ng Test Data 3A.3.2.3.1.1 Vent Clearing Jet and Induced Flow Field Model. To calculate the CGS vent clearing jet and induced flow field, a LOCA Water Jet Analy tical Model was developed for CGS. The model development and s upporting test data are documented in Reference 3A.3.2-9. The calculation is performed for a unit cell with a downcomer at the center. Input data included the downc omer vent water clearing time history.

C OLUMBIA G ENERATING S TATION Amendment 59 F INAL S AFETY A NALYSIS R EPORT December 2007 LDCN-06-000 3A.3.2-4 In order to calculate the dow ncomer vent water clearing tim e history and to provide a continuous pool surface displacement time history, a VENT computer code was developed. The model development and supporting test da ta for VENT are disc ussed in detail in A.D.2.

CGS input data for the LOCA water jet model and the VENT computer code are shown in Tables 3A.3.2-2 and 3A.3.2-3, respectively. As appropriate, maximums or minimums of CGS parameters are used to make the input data conservative in order to maximize vent water clearing velocities.

The vent exit water velocity and acceleration calculated by VENT are increased by 10% as requested by the NRC in Reference 3A.3.2-1. The velocity and acceleration time histories (including the 10% increase) are shown in Figures 3A.3.2-2 and 3A.3.2-3, respectively. As indicated in Reference 3A.3.2-9 , tests have shown that LOCA jet continues to propagate downward for a short duration beyond the vent cl earing instant due to rapidly charging air.

Although the vent clearing time for CGS is 0.654 sec, the jet tip reache s a maximum velocity of about 15.8 ft/sec at t = 0.704 sec.

Submerged boundary loads during downcomer vent water clearing is specified to be a static addition of an overpressure of 24 psi to the local hydrostatic pressure below the downcomer vent exit (walls and basemat) with a linear attenuati on to zero at the pool surface.

3A.3.2.3.1.2 Loss-of-Coolant Accident Bubble Charging Model. In order to calculate the flow field associated with the LOCA bubble charging phenomenon, a LOCA bubble charging model was developed. The model development and supporting test data are discussed in detail in Attachment 3A.D.4.

As discussed in A.D.4 , the LOCA bubble charging flow field is calculated using a numerical technique for potential flows in the exact CGS suppression pool geometry. The uniformly charging LOCA bubbles are modeled as equal strength point sources located one downcomer radius below the CGS ve nt exit at el. 453 ft-4.75 in.

Figure 3A.B-2 shows the modeled geometry for the CGS LOCA bubble charging phenomenon and Figures 3A.3.2-4 through 3A.3.2-6 show contour plots of the maximum radial, tangential, and vertical components of the gradient of the velocity potential.

The growth rate of the CG S LOCA bubble and the corres ponding source strengths of the point sources are determined by continuity from the rate of displacem ent of the pool surface after vent clearing. The rate of displacement of the pool surface is determined from the pool swell analytical model (PSAM) (see 3A.3.2.3.1.3

). Figure 3A.3.2-7 shows the velocity source strength, Q(t), the acceleration source st rength, Q(t), and the radius, R(t), of the CGS LOCA bubbles chargi ng process. Section 3A.B.4.2 discusses how to determine the CGS LOCA bubble charging transient flow field using the data in Figures 3A.3.2-4 through C OLUMBIA G ENERATING S TATION Amendment 54 F INAL S AFETY A NALYSIS R EPORT April 2000 LDC N-9 9-0 0 0 3A.3.2-5 3A.3.2-7. Tables 3A.3.2-4 and 3A.D-5 summarize results from the LOCA bubble charging analysis for CGS.

Since LOCA bubble charging represents the ear ly portion of pool swell, the associated suppression pool boundary load s are discussed in Section 3A.3.2.3.1.3.

3A.3.2.3.1.3 Pool Sw ell Analytical Model.

The pool swell transient is conservativ ely defined for CGS by the computer code SWELL base d on the model presented in Reference 3A.3.2-10. The model development and supporting te st data are discussed in Section 3A.C.3.

The CGS input data for SWELL are given in Tables 3A.3.2-3 and 3A.3.2-5. As appropriate, maximums and minimums of CGS parameters are used to make the input data conservative in order to maximize pool swell velocities. The CGS pool swell velo city, acceleration, displacement, bubble p r essure, a nd wetwell air pressure versus time are given in Figures 3A.3.2-8 through 3A.3.2-12 , respectively.

F i gure 3A.3.2-13 shows the pool lug velocity vs.

displacement and Table 3A.3.2-6 shows results from the pool swell analysis for CGS. Velocities and accelerations calculated by the computer code SWELL are multiplied by a factor of 1.1, as requested by the NRC in Reference 3A.3.2-1. The 10% increase in velocities and acce l eratio n s is included in Figures 3A.3.2-8 , 3A.3.2-9 , and 3A.3.2-13.

Test results indicate that no si gnificant froth will occur when the water slug breaks up during pool swell in Mark II containments. Hence, the load due to froth is negligible (Reference 3A.3.2-11) and no froth impingement is considered in the as sessment of the design of structures.

Loads on submerged boundaries dur ing pool swell are calculated by specifying that basemat and wall loads be determined from static addition of the maxi mum bubble pressure predicted by the PSAM to the local hydrostatic pressure below the downcomer vent exit plane and with a linear attenuation to the maximum wetwell air space pressure at maximum pool swell

elevation.

Wetwell wall loads due to air compression during pool swell are specified by the direct application of the PSAM calculated wetwell air compression pressure to the wetwell walls above the pool surface.

The short-term drywell pressure history during pool swell is spec ified as the drywell pressure transient presented in Table 3A.3.2-5. Test data have shown that a small short duration upward pressu re differential on the diaphragm floor may occur due to rapid pressurization of the wetwell air space during the pool swell transient. The 4T Test results, disc ussed in Section 4.

4.6.6 of Reference 3A.3.2-11 , show a net upward load on the diaphragm floor of less than 2.2 psi. These low value 4T test results C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.3.2-6 were confirmed by small scale pool swell tests which showed an average scaled up net upward pressure differential of 1.

83 psi. However, as discussed in S ection 4.4.6.6 of Reference 3A.3.2-11 , the Bodega Bay test data, for a wide variety of blowdown conditions, shows that the wetwell air space pressure never exceeds the drywell pressure.

Conservatively, therefore, the CG S diaphragm floor is assessed for an uplift pressure, P, of 5.5 psi. This is a maximum dynamic load and is in agreemen t with the loads specified in References 3A.3.2-2 and 3A.3.2-8. 3A.3.2.3.1.4 Fallback Model. The model presented in Reference 3A.3.2-11 is used to calculate the CGS fallback phenomenon. The model development and supp orting test data are discussed in Reference 3A.3.2-11. The fallback model conserva tively assumes the pool swell slug remains intact during pool swell and fa lls back from its maximum height 1.5 x H o , where H o is the pre-LOCA downcomer submergence, at full water density and under the constant acceleration due to gravity.

Figure 3A.3.2-14 show s a plot of the fallback water slug velocity vs. elevation of water slug top surface.

Reference 3A.3.2-10 indicates that fallback loads on submerged boundaries are negligible and are therefore not specified. This is ba sed on review of existing fallback data.

3A.3.2.3.2 Boundary Loads

The analytical models used to describe containment boundary loads during short-term LOCA loading ph e nomena are discussed in Section 3A.3.2.3.1. Figures 3A.3.2-15 th r ough 3A.3.2-17 show the du r a tion and distribution of the short-term containment boundary loads.

3A.3.2.3.3 Structure Loads

The CGS structures affected by short-term LOCA loading phenomena are id entified in Section 3A.2.1.1 and shown in Figures 3A.2.1-2 throu g h 3A.2.1-8. Short-term LOCA loads on submerged structures are given in Tables 3A.3.2-8 , 3A.3.2-9 , and Figure 3A.3.2-18

.

Piping and structural component s below el. 454.4 ft are subject ed to loads caused by water clearing/air charging. Piping and structural components betw een elevations 454.4 ft and 434.4 ft are subjected to drag and lift loads ca used by pool swell/fall back. Also, piping and structural components between elevations 466.4 ft and 484.4 ft ar e subjected to impact loads caused by pool swell. Direct hydrodynamic loads do not exist for piping and structures above el. 484.4 ft.

The following is a description of how the shor t-term LOCA definition models, discussed in Section 3A.3.2.3.1, are applied to CGS structures.

C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.3.2-7 a. Vent clearing jet load The loads on submerged structures lo cated inside the jet boundaries are calculated using the vent exit velocity and acceleration at t = 0.654 sec. The loads on submerged structures located out side the jet boundaries are calculated using velocity and acceleration fields at t = 0.704 sec.

For elbows of 24-in. diameter pipes, the impact load for full momentum transfer of the intercepted jet (K=2) (Reference 3A.3.2-10) is calculated to be smaller than the drag load. Hence, drag load is used in the assessment.

Submerged structures located below the vent exit elevation are subjected to drag loads caused by the vent clearing jet. Th e types of drag loads and the formulas for calculating them are presented in item c below and in Attachment 3A.C. For the case of the vent clearing jet, the standard drag coefficient, C D , is equal to 1.2; the accelerati on drag coefficient, C M, is equal to 2.0; and the lift coefficient, C L, is equal to 0. In order to acc ount for the dynamic nature of the loading phenomenon, a multiplier of 2.0 is us ed to calculate the equivalent static pressure.

b. Loss-of-coolant ac cident bubble load

Submerged structures located below the vent exit elevation are subjected to drag loads caused by LOCA bubble charging.

The types of drag loads and the formulas for calculating them are presented in Item c below and in A.C. The velocity and acceleration at any point in the pool are calculated by applying the methods outlined in 3A.3.2.3.1.2.

For the case of LOCA bubble charging, the standard drag coefficient, C D, is equal to 1.2; the acceleration drag coefficient, C M, is equal to 2.0; and the lift coefficient, C L, is equal to 0. These numerical values are consistent with References 3A.3.2-4 and 3A.3.2-6. In order to account for the dynamic nature of the LOCA bubble charging load phe nomenon, a DLF of 2.0 is used to calculate the equivalent static pressure.

c. Pool swell load

The structures above the initial pool su rface and below the maximum pool swell height are subject to impact loads due to the rising pool surface. The maximum dynamic pressure due to pool swell im pact is calculated by using the methodology outlined in Reference 3A.3.2-2 and in Section III.B.3.c.1 of Reference 3A.3.2-1. In order to account for th e dynamic nature of the pool swell loading phenomenon, the equivalent static pressure is obtained by C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.3.2-8 multiplying the maximum dynamic impact pressure by a DLF. The DLF is given in Figure 5 of Reference 3A.3.2-3. No impact loads on grating surfaces are specified since the CGS grating ba rs are narrow (typically 3/16-in. wide).

Structures located above the vent ex it elevation and below the maximum pool swell height are subjected to drag forces during the pool swell transient. Drag loads have three components: standa rd drag, acceleration drag, and lift. Formulas for calculating these three types of load are presented in A.C. Standard drag load is velocity square dependent and acts in the flow direction. Acceleration dr ag load is proportional to the flow acceleration and acts in the flow direction. Lift load is velocity square dependent and is normal to the flow direction.

In order to calculate the three components of drag lo ad, three coef ficients are determined. These are: standard drag coefficient, C D; acceleration drag coefficient, C M; and lift coefficient, C L. For circular cross-sections, coefficients C D, C M, and C L are calculated using Reference 3A.3.2-6. For non-circular cross-sections, coefficients C D and C M are calculated using Reference 3A.3.2-7 , and C L is assumed equal to 1.6 as indicated in Reference 3A.3.2-5. Interference eff ects from adjacent structures are taken into account by using the methodology presented in Reference 3A.3.2-4. In order to account for the dynamic nature of the pool swell loading phenomenon, appropriate DLFs are

used to calculate the e quivalent static loads.

The gratings are subject to drag loads during pool swell due to resistance to the flow through them. The dynamic drag load is determined by the product of the

differential pressure across the grating and the total plan area of the grating.

The open area fraction for the grating is greater than 60%. The peak dynamic differential pressure across the grating is obtained from Figure 4-1 of Reference 3A.3.2-2. This figure is based on a pool swell approach velocity of 40 fps. Since the maximum pool swell velocity for CGS is smaller than 40 fps, the differential pressure values in Figure 4-1 of Reference 3A.3.2-2 are multiplied by a factor equal to the squa re of the ratio given by the maximum velocity divided by 40 fps. The maximum pool swell velocity is given in Table 3A.3.2-6 (plus a velocity multiplier of 1.

1). In order to account for the dynamic nature of the pool swell load ing phenomenon, a multiplier of 2.0 is used to calculate the equi valent static pressure.

Vertical drag pressure and horizontal lif t pressure are applie d simultaneously to the projected area of a cylinder circum scribing the structural member under consideration. The projec tion is always made on a plane normal to the direction of pressure. Vertical impact load is a pplied separately (not in combination with drag and lift loads) to structures lo cated between eleva tions 466.4 ft and C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.3.2-9 484.4 ft on the projected area of the stru cture on the horizontal plane. Impact load is not used to check local bending effects of flanges or webs or ovaling effects of a pipe. The full drag, lift, or impact load calculated as outlined above is applied normal to a pipe or a st ructural component. There are no loads parallel to the longitudinal axis of the structure.

d. Fallback load Structures located above the vent ex it elevation and below the maximum pool swell height are subjected to drag an d lift forces during fallback. The same methodology and the same drag and lift coefficients used in calculating pool swell drag and lift loads reused to define fallback drag and lift loads. To account for the dynamic nature of the fallback loading phenomenon, a multiplier

of 2.0 is used to calculate the equivalent static load. No impact loads are specified during fallback (Reference 3A.3.2-11).

Fallback drag loads on gratings are specified to be equal in magnitude to the pool swell drag loads but opposite in direction.

The method of applying drag and lift pr essure on a structural member is the same as that outlined at the end of Item c above.

3A.3.2.3.3.1 Loads on Major Structures.

The major vertical st ructures are shown in Figure 3A.2.1-2. They are: the 102 do w nc o m ers (three were capped), the 18 SRV lines including quenchers and support towers, a nd the 17 concrete columns supporting the diaphragm floor. The major horizontal structures are the steel truss shown in Figure 3A.2.1-3 , which prov i des lateral support to the downcomers and the SRV lines, and the

platform at el. 472 ft 4 in. shown in Figures 3 A.2.1-3 and 3A.2.1-8. The truss is submerged and the platform is located in the pool swell zone.

Hydrodynamic loads on the vertically oriented columns may occur only due to horizontal flow across the columns.

Since the columns are vertical, pool swell a nd fallback loads are negligible. The column s are located along circumferential and radial lines of symmetry between downcomers; however, some asymmetry is created because some of the downcomers have a diameter of 24 in. while others have a diameter of 28 in. This small asymmetry in the downcomer arrangement causes sm all loads on the columns during water clearing/air charging.

The magnitude and distribution of loads on the columns are shown in Table 3A.3.2-8 and Figure 3A.3.2-18 respectively.

As with the columns, hydrod ynamic loads on the vertically oriented downcomers may occur only due to horizontal flow across the downcomers since the downcomers are located along circumferential and radial lines of symmetry between other downcomers and the suppression pool boundaries, the LOCA water jet, LOCA bubble charging, pool swell, and fallback C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.3.2-10 horizontal flows are small. Because the horizont al flows are small the resulting loadings are negligible.

Table 3A.3.2-8 and Figure 3A.3.2-18 show the magn itude and distribution of loads on a vertical SRV line as well as quencher arms dur ing the water clearing/ai r charging phases of LOCA. Since quencher supports are below el. 448 ft-0 in., they ar e not subjected to short-term LOCA (Fi g ure 3A.3.2-18

). Pool s w ell and fal l back exert negligible loads on the SRV lines due to the vertical orientation of the SRV lines above th e downcomer vent exit plane. Because of their location below the downcomer vent exit plan e, no pool swell or fallback loads are experienced by the quencher supports and quencher arms.

The downcomer bracing truss is oriented in a ho rizontal plane one foot above the downcomer vent exit plane. Since the truss is located above the downcomer vent exit the induced flow field and, therefore, drag load due to LOCA water jet during vent clearing is small.

Horizontal LOCA bubble load on the bracing truss is small. Because th e truss is above the downcomer vent exit plane, pool swe ll and fallback load s are experienced.

The only major horizontal structure in the pool swell zone is a plat form at el. 472 ft-4 in. The platform consists of a gra ting with an open area fracti on of 0.776 and is supported by 0.5-in. thick vertical members. The dynamic drag load is cal culated to be 4.2 psi for pool swell and fallback induced loads and is multiplied by a DLF of 2.0. The equivalent static load is then applied to the total plan area of the platform. Because the platform is above the initial pool surface it does not experience vent clearing or LOCA bubble loads.

3A.3.2.3.3.2 Loads on Fully Submerged Piping Systems Below Elevation 454.4 ft.

The 24-in. high-pressure core spray (HPCS) (X-31), 24-in. residual heat removal (RHR)-B (X-32),

8-in. reactor core isolation cooling (RCIC) (X-33), 24-in. low-pressu re core spray (LPCS) (X-34), 24-in. RHR-A (X-35), 24-in. RHR-C (X

-36), 6-in. suppressi on pool cleanup (X-100) and the 4-in. fuel pool cooling and cleanup (FPC) are the eight piping systems fully submerged in the CGS suppression pool. Seven systems ente r the pool through containment penetrations

at el. 452 ft 0 in. as shown in Figu r e 3A.2.1-2. The 4-in. FPC enters the pool thr o ugh the pedestal at el. 451 ft 6 in. and is also shown in Figure 3A.2.1-

2. Each of these pipes runs

along the CGS suppression pool bo undaries with a maximum dist ance between pipe centerline and boundaries of 46 in. Since these pipes ar e below the downcomer vent exit el. 454 ft 4.75 in. they are not subjected to pool swell or fallback loads during a LOCA. The magnitude and distribution of short-term LOCA loads on submerged pipes are given in Table 3A.3.2-8 and Figure 3A.3.2-18 r e spectively.

3A.3.2.3.3.3 Loads On Part ially Submerged Piping Systems.

The 13 partially submerged piping systems enter the suppression chamber thr ough containment penetr ations at el. 467 ft 9 in. as shown in Figu r e 3A.2.1-3 and enter th e pool vertically within a 39 in. distance from the containment as shown in Fig u res 3A.2.1-6 t h rough 3A.2.1-8.

C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.3.2-11 The pipes' horizontal portions are subjected to vertical drag and horizontal lift loads caused by pool swell/fallback. Also, the pipes' horizontal portions experience ver tical impact pressure due to pool swell. Inclined braces above the penetrations are not subjected to pool swell impact since they are shielded by the pipes.

Horizontal supports be low penetrations do not experience pool swell impact since these supports are below el. 466.4 ft.

3A.3.2.3.3.4 Loads on Piping Systems and Structural Components Between Elevations 454.4 ft and 484.4 ft.

The portion of the wetw ell between elevations 454.4 ft and 484.4 ft is affected by pool swell and fallback only. Piping sy stems located in this zone ar e subjected to drag, and lift loads. In addition piping be tween the normal pool surface, el. 466.4 ft, and el. 484.4 ft is subjected to impact loads unless shielded by pipe supports below. These piping systems include penetrati on sleeves, pipe stubs, pi pes, and pipe supports.

3A.3.2.4 Long-Term Hydrodynamic Loads

Long-term hydrodynamic loads refer to the LOCA related loads exerted on the pool boundaries and on submerged wetwell structures following the fallback transient. These loads are associated with steam condensation at the downc omer vent exits as st eam from the drywell flows into the suppression cham ber via the vents. Depending on the steam mass flux through the vents, two types of loading conditions are anticipated. During hi gh or medium mass flux so-called COs occur. During low mass flux, chugging loads occur. Chugging loads and CO loads are discussed in the following sections.

3A.3.2.4.1 Analytical Models and Supporti ng Test Data

3A.3.2.4.1.1 Chugging Loads.

The design specification for chugging loads herein is based on the major test programs on the effects of CO and chugging, which were conducted for the Mark II Containment Program during the period of 1975 to 1981 (References 3A.3.2-16 and 3A.3.2-17). It takes account of the generic chugging load definition devel oped for the Mark II type plants (References 3A.3.2-18 and 3A.3.2-19), but departs from the generic definition in order to provide for important structural differences between the CGS plant and the generic plant. The design specification (Reference 3A.3.2-15) is called the Revised Definition as it represents a revision of an earlier de finition called the Improved Definition (Reference 3A.3.2-20) which was based only on the first stage of the test results, the 4T Program (Reference 3A.3.2-16). The current design specif ication, i.e., the Revised Definition, was submitted to and approved by the NRC (Reference 3A.3.1-6). The scope of the design specification covers bot h single vent loads and multiple vent loads app licable to the CGS plant.

The single vent load definition is based on two series of full scale single vent tests, namely, the 4T tests and the 4TCO tests (References 3A.3.2-16 and 3A.3.2-17). The load is defined in terms of a series of source loads located at the vent exit with significant load features selected on the basis of the pressure readings at the test tank and the characteristics disclosed by the C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.3.2-12 vent-pool-tank system. Both test programs indicate the impuls ive and random strength nature of chugging, and it is noted that the strength, although random, is related to system conditions.

The greater chugs of the 4TCO program are controlling relative to pressure amplitude, frequency content, and spatial distribution of pressure. Thus, the Revised Definition is a group of seven source loads sele cted on the basis of the seven greatest chugs from the 4TCO data. A design load envelope is obtained by applying the source load s in turn at the vent exit; each load consists of an impulsive pressure gradient with appropriate system parameter values. To calculate results at the wetwell tank, the theory of acoustic fluids is used with a fully coupled model consisting of th e vent steam, the pool water, and the wetwell tank. The calculated results, due to application of the defining source loads, envelope both the 4TCO data and the 4T data, relative to all pertinent characteristics.

The multiple vent load definition utilizes the preceding single vent definition in conjunction with the 102 vents in the CGS plan

t. It is described in Section 3A.3.2.4.2.1.

3A.3.2.4.1.2 Condensation Oscillation Loads.

The generic definiti on of CO loads, as developed for the Mark II Containmen t Program, is given in Reference 3A.3.2-21. The definition is based on direct appl ication of the bounding pressure measurements from full-scale single vent 4TCO tests to the Mark II containment boundaries.

A comparison has been made between this load definition for CO loads, the JAERI CO results, and the preceding definition for chugging loads. The comparison, as reported in Reference 3A.3.2-12 shows that the CO loads and the c hugging loads are similar with respect to pertinent characteristics such as randomly varying amplitude, frequency content, and desynchronization of vent loads. However, it is also shown (References 3A.3.2-12 and 3A.3.2-23), that in a Mark II multivent containment, the controlling boundary pressures due to the CO load are less than those due to the chugging load. It is therefore concluded that the CO load does not represent a governi ng load and that it need not be considered in the assessment of the design adequacy of the CGS structures.

3A.3.2.4.2 Boundary Loads

3A.3.2.4.2.1 Chugging Loads.

The current design specificati on, i.e., the Revised Definition (Reference 3A.3.2-15), includes definition of the chugging boundary loads due to multivent chugging together with the asso ciated application methodology.

The chugging pressures on the suppression pool boundary are de fined as resulting from the a pplication of the single vent design chugging loads at all 102 vent exits in the CGS pool-c ontainment structure. To determine the boundary pressure s, the analysis is based on a fully coupled model which accounts for all important plant parameters: vent length, three-dimensional multivent pool geometry, pool with sloped bottom, and flexibil ity of pool structural boundaries. Detailed analytical methods are described in Reference 3A.3.2-15.

C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.3.2-13 The methodology for application of the source load s to the 102 vents is generally similar to that of the generic load definition (Reference 3A.3.2-19), and reflects the characteristics of multivent behavior disclosed by the JAERI and CREARE test programs. Two deterministic spatial distributions of chug stre ngths, similar to but not the same as the generic distributions, are specified to maximize axisy mmetric and nonaxisy mmetric responses.

Random variation of chug initiation tines from vent to vent is recognized with desynchronization of the start times as in the generic definition. Th e seven basic single vent loads of the design specification are applied in turn at the 102 vents with variation of strength and initiation time between vents as previously noted; an envelope of containment pressures is calculated. The calculated results exceed the test results in the 4TC0 and the JAERI tests with respect to maximum containment pressures and the Fourier amplitude spectra for containment pressures.

3A.3.2.4.2.2 Condensation Oscillation Loads.

The discussion in Section 3A.3.2.4.1.2 on the relative magnitudes of chugging loads and CO loads is applicable. As noted therein, the controlling boundary pressu res due to chugging exceed those due to CO. Hence, the CO load is not a governing load and it is not considered in the assessment of the design adequacy of the CGS structure.

3A.3.2.4.3 Submerged Structure Loads

Loads on submerged structures caused by CO and chugging are presented in the following

sections.

3A.3.2.4.3.1 Condensation Oscillation Loads.

There is no need to assess the CGS plant for a CO load definition since CO loading is le ss critical and is bounded by the chugging load.

3A.3.2.4.3.2 Chugging Loads.

The LOCA chugging loads on submerged structures are defined consistently with the lo ad definition for the pool boundary. Pressure field in the fluid is obtained using chuggi ng design sources develope d for pool boundary loads (Reference 3A.3.2-15). From the pressure fi eld, pressure gradient s and loads on submerged structures are calculated.

In the method described above, it is assumed that the flow in th e vicinity of the vent during chugging is unaffected by the presence of pool boundaries or other sources in the pool.

Therefore, fluid pressures in a single cell/single vent pool are re presentative of pressures near the vent in the CGS pool.

C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.3.2-14 For structures located beyond a distance of 4R (R being the downcomer ra dius) from the vent exit center, the chugging loads are negligible and need not be considered in the design assessment of the structures.

Equivalent static chugging loads for structures located within a distance of 4R from the vent exit cent er are calculated by using the formula:

pCDDLF m 4 p n where

p = equivalent static pressure on structure (psi)

C m = hydrodynamic mass coefficient = 2.0

DLF = dynamic load factor = 1.5

D = diameter in inches of the pipe or of the cylinder circumscribing the cross-section of a support member

p n = pressure gradient across a submerg e d structure (p si/in). N u merical values for this term a r e g i ven in Figure 3A.3.2-19.

Chugging loads calculated as outlin ed above are listed as follows:

a. Bracing Truss Members: A load of 10 psi is applied vertically upward or downward. This load is applied simulta neously to all members connecting to a downcomer. Chugging event under each vent occurs with random phasing from events at other vents. Therefore, this load is not applied simultaneously to all

members of the truss;

b. Inner Row SRV Line: A load of 12 ps i is applied radially outward or inward from the vent axis of the adjacent inne r row downcomer. The load distribution is assumed uniform between el. 454.4 ft and el. 452.4 ft and then linearly decreasing to zero at P o int A (Figure 3A.3.2-19

); and

c. Pipes and Supports: A pr essure is applied vertically upward or downward. The magnitude and distribution of the vertical pressure are calculated using Equation 3A.3.2-1 and Figure 3A.3.2-19. The maximum vertical pressure is

2.4D psi. Simultaneously, a radial pressure is applied inward or outward from the vent axis of the adjacent outer row downcomer.

The magnitude and distribution of the radial pressure are calculated using E quation 3A.3.2-1 and Figure 3A.3.2-19. The radial load is assumed zero above the vent exit.

C OLUMBIA G ENERATING S TATION Amendment 54 F INAL S AFETY A NALYSIS R EPORT April 2000 LDC N-9 9-0 0 0 3A.3.2-15 The pressure loads listed above are multiplied by the projected area of the structure segment normal to the direction of the loading to obtain the total force on each segment. The total load on each segment of the structure is applied at the geom etric center of the segment

3A.3.2.4.4 Lateral Loads on Downcomer Vents

This section provides the definition of the late ral loads which occur n ear the downcomer exits during chugging. The de finition conforms with the requireme nts for such loads as prescribed in NUREG-0808 (Reference 3A.3.2-8). These lateral loads are de fined herein in relation to the downcomer bracing system which is described and asse ssed in Section 3A.4.2.1. The principal features included in the definition are single vent loads, loads on multiple vents, overall loading of the bracing system, and loads and downcomer size.

a. Single Vent Load - The maximum latera l exit load on one 24-in. downcomer is defined as a dynamic single pul se load having a half sine wave shape with load amplitude of 65 klbf (kips) and duration of 3 msec. For the assessment of the bracing system, a dynamic analysis of th e system acted on by this single vent load is made.

b. Loads on Multiple Vents -

With multiple vent loading, the force on each loaded vent is also defined as a single pulse half sine wave dynamic load. However, in line with NUREG-0808, the pulse durati on is taken to vary over a range of values and the force amplitude is taken to depend on the pulse duration, T, and on the number of loaded vents. The ve nt force, F(t), is evaluated by the expression:

F(t) = M A(T) sin (t/T) 0 t T (Eq. 3A.3.2-2 )

where A(T) = (50 - 20 T/3) klbf 3 T < 6 ms (Eq. 3A.3.2-3)

The factor M is a load reduction factor which depends on the number of loaded vents and the required level of exceedence probability. For a given number of

loaded downcomers, it is evaluated at the exceeden ce probability level of 10

-4 using the diagram specified in NUREG-0808.

To determine the controlling value of the amplitude factor A, the bracing system was analyzed over the range of the duration, T.

Thus, values of T in the range of 3 to 6 ms and the associated values of A were utilized. It was determined that the controlling value of T was always 3 ms.

Consequently, the associated value of A in the above equa tion for F(t) is ev aluated as 30 kips.

C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.3.2-16 For the assessment under multiple vent load ing, the bracing system is analyzed dynamically with lateral loads on a given set of vents. In the analysis, the force on each loaded vent is defined by E quation 3A.3.2-2 where M is obtained as previously described, A equa ls 30 klbf, and T equals 3 ms.

c. Overall Loading of the Bracing System - The design assessment of the downcomer bracing system under mu ltiple lateral exit loads in 3A.4.2.1 describes the method used to determine the controlling number of loaded vents and the controlling direction of the loads.

The methodology is generally similar to that of Section 2.3.2.2 of NUREG-0808.

d. Loads and Downcomer Size - The downcomer system consists of 102 downcomers of which 84 are 24-in. diameter and 18 are 28-in. diameter.

The lateral loads on the 24-in. diameter vents have been defined above in paragraphs a and b for the cases of single vent loading and multiple vent loading. In line with NUREG-0808, th e lateral loads on th e 28-in. vents are defined as 1.34 times as great as those on the 24-in. vents.

3A.3.2.5 Pressure and Temperature Transients

A LOCA causes a pressure and temperature transient in the drywell and wetwell due to mass and energy released from the line break. The drywell and wetw ell pressure and temperature histories are employed to establish the structural loading conditions in the containment. The response must be determined for a range of parameters such as break size, reactor pressure, and initial (preincident) containm ent conditions. The analytical models used to evaluate the pressure and temperature re sponse of the containment ha ve been developed by GE (References 3A.3.2-13 and 3A.3.2-14).

The assumptions made for analyzing the LOCA transients were ba sed on conservatively predicting the blowdown mass and energy rates into the drywell and suppression pool. The following assumptions were made:

a. Initial drywell pressure of 0.75 psig,

b. Downcomer submergen ce at high water level, c. Minimum drywell free air volume,

d. Minimum wetwell free air volume,

e. Minimum water volume in wetwell, and

C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.3.2-17

f. Suppress ion pool initial temperature and service water temperature at the maximum Technical Spec ifications limit.

(Assumptions b, d, and e are inconsistent w ith each other but provide conservative results.)

For the intermediate size breaks, it was assumed that all the mass and energy releases from the broken pipe discharge via the drywell into the suppression pool. Normally, a portion of the mass and energy release will be dispersed ove r the drywell volume causing the drywell pressure and temperature to rise.

3A.3.2.5.1 Results for CGS

The drywell pressure transients have been calculated with in ventory effects included in the analysis. The results for the recirculation and main steam line breaks are presented in Figures 3A.3.2-20 th r o ugh 3A.3.2-28.

The plant parameters used are given in Table 3A.3.2-7. The spectrum of accident conditions covered are:

a. Large double-ended break of a recirculation line,

b. Large double-ended break of one main steamline, and c. Intermediate recirculation line break (0.1 ft 2 break area).

Table 3A.3.2-5 provides the drywell pressure transient during the 2 sec period immediately following a recirculation line break. This table differs fr om the values plotted in Figure 3A.3.2-20 , since it include s t h e influence of reactor sub c ooling on the initial blowdown flow rate from the break. This is a short-term effect that occurs during the first few seconds of the accident and does not influence the maximum drywell pressure.

Since this LOCA leads to the most severe short-term pressure conditions, the data in Table 3A.3.2-5 are used to calculate the pool swell velociti es and associat ed effects.

Figures 3A.3.2-20 th r o ugh 3A.3.2-24 give the pr e ssure/temperature trans i ents resulting from a large reci r c ulation line break. As shown in Figure 3A.3.2-20

, the drywell pre s sure increases to a maximum value of 35 psig in about 20 sec.

Also, the pressure of the wetwell air, while increasing with time, is less than the drywell pre ssure until the ECCS flow starts to spill from the break. The drywell temperature increases to about 2 8 0°F (Figure 3A.3.2-23

), while the

temperature of the suppression pool increases to about 220°F.

Similar response curves for a main steam line break are also shown in Figures 3A.3.2-25 and 3A.3.2-26 , and tho s e for an intermedia t e break accident (IBA) are shown in Figures 3A.3.2-27 and 3A.3.2-28.

C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.3.2-18 3A.3.2.5.2 Differential Pressure Load on the Diaphragm Floor As illust r a ted in Figures 3A.3.2-20

, 3A.3.2-21 , 3A.3.2-25 , and 3A.3.2-28 , the net pressure on the diaphragm floor is downward throughout a LOCA transient (refer to FSAR

Section 6.2.1). However, the diaphragm floor is conservatively designed for a net uplift pressure of 5.5 ps i as discussed in 3A.3.2.3.1.3. The maximum net downward pressure on the diaphragm occurs during the short-term pa rt of the large break LOCA transients and reaches a maximum value of 20 psi.

During the initial phase of a LOCA transient, drywell air is blown down through the downcomer vents into the wetwell by the steam from the break. The steam rapidly replaces the air in the drywell.

Initially, a steam-air fixture fl ows through the vents into the suppression pool and forms bubbles that rise to the pool's surface. Condensati on of the steam in the bubbles allows only the air component to reach the wetwell airspace. As the air collects, the air space becomes pressurized and heated.

As the reactor vessel blowdown ends, the emergenc y core cooling system floods the core with water which starts to spill out of the break into the drywell. This resu lts in rapid condensation of the steam, which has replaced the air, and co nsequent rapid drywell depressurization below that in the wetwell. Before the net upward pressure beco mes large, however, nine 24-in. vacuum breakers, which are attach ed to nine selected downcomers, open to equalize the pressure difference by returning the air (nitrogen) co llected in the wetwell to the drywell. These vacuum breakers are adjusted to open when the differential pressure across them is in the range of 0.15 to 0.35 psi.

3A.3.2.6 Building Response to Loss-of-Coolant Accident Loads

The analysis and response of the reactor building under the action of long-term LOCA loads is discussed in Section 3A.5.2.

3A.3.2.7 References

3A.3.2-1 "Mark II Containment Lead Plant Program Load Evaluation and Acceptance Criteria," NUREG-0487, NRC, October 1978.

3A.3.2-2 Mark II Containment Dynamic Forc ing Functions Information Report (DFFR), General Electric Company, NEDO-21061, Revi sion 4, November 1981.

3A.3.2-3 "Impact Loads on Structures Above Mark II Containment Pools," G. Maise, Brookhaven National Laboratory, February 28, 1978.

C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.3.2-19 3A.3.2-4 "Submerged Structure Methodology," Appendix G, Zimmer Power Station, Unit 1, DAR Amendment No. 13, October 1980.

3A.3.2-5 "Mark II Containment Lead Plant Program Load Evaluation and Acceptance Criteria," NUREG-0487, Supplement No.1, NRC, September 1980.

3A.3.2-6 "In-Line and Transv erse Forces on Cylinders in Oscillatory Flow at High Reynolds Number," Journal of Ship Rese arch, Volume 21, No. 4, pp. 200-216, December 1977.

3A.3.2-7 "Forces on Cylinders and Plates in an Oscillating Fluid," G. H. Keulegan and L. H. Carpenter, Journal of Research of the National Bureau of Standards, Volume 60, No. 5, May 1958.

3A.3.2-8 "Mark II Containment Program Load Evaluation and Acceptance Criteria,"

NUREG-0808, USNRC, August 1981.

3A.3.2-9 "An Analytical Model for LOCA Water Jet in Mark II C ontainments (The Ring Vortex Model)," Burns and Roe, Inc., September 1980.

3A.3.2-10 Mark II Containment Dynamic Forc ing Functions Information Report (DFFR), General Electric Company, NEDO-21061, Revi sion 3, June 1978.

3A.3.2-11 Mark II Containment Dynamic Forc ing Functions Information Report (DFFR), General Electric Company, NEDO-21061, Revi sion 2, September 1976.

3A.3.2-12 "Comparison of C ondensation Oscillation and C hugging Loads for Assessment of WPPSS Nuclear Project No. 2," Su mmary Report, Proprietary, Burns and Roe, Inc., December 1981 transmitted to NRC by letter GO2-81-552 dated December 24, 1981.

3A.3.2-13 The General Electric Pressure Suppression Containment Analytical Model, General Electric Company, NEDO-10320, April 1971.

3A.3.2-14 Safe System Analysis for Standby Core Cooling Equipment, General Electric Company, NEDE-10169, Pr oprietary, September 1977.

3A.3.2-15 "Chugging Loads - Revised Definition and Applicati on Methodology for Mark II Containments (Based on 4TCO Test Results)," Technical Report, Burns

and Roe, Inc., July 1981. Transm itted to NRC by letter GO2-81-189 of July 22, 1981.

C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.3.2-20 3A.3.2-16 "Mark II Pressure Suppression Te st Program - Phase I, II, and III of the 4T Tests - Application Memo," NED E-23678-P, Rev. 0, General Electric Company, January 1977.

3A.3.2-17 "4T Condensation Oscillation Test Program Final Test Report,"

NEDE-24811-P, General Electric Company, May 1980.

3A.3.2-18 "Mark II Improved Chugging Methodol ogy," NEDE-24822-P, General Electric Company, May 1980.

3A.3.2-19 "Generic Chugging Load Definition Report," NEDE-24302-P, General Electric Company, April 1981.

3A.3.2-20 "Chugging Loads - Improved Defi nition and Applicat ion Methodology to Mark II Containments," Te chnical Report, Proprietary, Burns and Roe, Inc., June 1979.

3A.3.2-21 "Generic Condensation Oscillat ion Load Definition Report," NEDO-24288, General Electric Company, February 1981.

3A.3.2-22 "Mass Energy Report," Gene ral Electric Report, GEWP-2-77-533, March 15, 1977.

3A.3.2-23 Letter report on C.O.

loads GO2-82-351 of April 1, 1982.

3A.3.2-24 "WPPSS Nuclear Project No. 2, Fi nal Safety Analysis Report," Washington Public Power Supply System, Chapter 6.2. 3A.3.2-25 AEC-TR-6630, "Handbook of Hydraulic Resistance - Coefficients of Local Resistance And Friction,"

I. E. Idel'chik, 1960.

3A.3.2-26 "Flow of Fluids Through Valv es, Fittings, and Pipe

," Technical Paper No. 410, Crane Company, 1980.

3A.3.2-27 "CONTEMPT-LT--A Computer Program For Predicting Containment Pressure-Temperature Response to A Loss-of-Coolant Accident," Aerojet Nuclear Company, June 1975.

3A.3.2-28 "Mark II Pressure S uppression Containment Systems:

An Analytical Model of the Pool Swell Phenomenon," NEDE-21544-P, General Elec tric Company, December 1976.

C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.3.2-21 3A.3.2-29 Response to NRC Question 020.071, transmitted via letter MFN-275-78 to Mr. J. F. Stolz, Chief, Light Water Reactor Branch No. 1, USNRC, from Mr. L. J. Sobon, Manager BWR Contai nment Licensing, General Electric Company on "Responses to NRC Reque st for Additional Information (Round 3 Questions)," da ted June 30, 1978.

C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT N o v e mb e r 1998 3A.3.2-23 Table 3A.3.2-1 Summary of Loss-of-Coolant A ccident Affected Structures Type of Load i ng Condition Short-Term LOCA Long-Term LOCA Structur e s Ex p e riencing LO CA Loads Water Jet LOCA Bubble Pool Swell Fallback Condensation Oscillations Chugging Fully submer g e d piping systems 3A.3.2.3.1.1/3A.3.2.3.3.2 3A.3.2.3.1.2/3A.3.2.3.3.2 3A.3.2.4.1.2/ 3A.3.2.4.3.1 3A.3.2.4.1.1/

3A.3.2.4.3.2 Partially sub m erged piping systems 3A.3.2.3.1.1/3A.3.2.3.3.3 3A.3.2.3.1.2/

3A.3.2.3.3.3 3A.3.2.3.1.3/

3A.3.2.3.3.3 3A.3.2.3.1.4/

3A.3.2.3.3.3 3A.3.2.4.1.2/

3A.3.2.4.3.1 3A.3.2.4.1.1/

3A.3.2.4.3.2 Piping systems fully above initial pool surface 3A.3.2.3.1.3/ 3A.3.2.3.3.4 3A.3.2.3.1.4/

3A.3.2.3.3.4 Grating 3A.3.2.3.1.3/

3A.3.2.3.3.1 3A.3.2.3.1.4/

3A.3.2.3.3.1 Drywell floor 3A.3.2.3.2 Containment wall 3A.3.2.3.1.1 3A.3.2.3.2 3A.3.2.3.2 3A.3.2.4.2.2 3A.3.2.4.2.1 Pedestal 3A.3.2.3.1.1 3A.3.2.3.2 3A.3.2.3.2 3A.3.2.4.2.2 3A.3.2.4.2.1 Ba s e mat 3A.3.2.3.1.1 3A.3.2.3.2 3A.3.2.3.2 3 A.3.2.3.2 3A.3.2.4.

2.2 3A.3.2.4.2.1 Columns 3A.3.2.3.1.1/3A.3.2.3.3.1 3A.3.2.3.1.2/

3A.3.2.3.3.1 Downcomers 3A.3.2.3.1.1

/3A.3.2.3.3.1 3A.3.2.3.1.2

/3A.3.2.3.3.1 3A.3.2.4.1.2

/ 3A.3.2.4.3.1 3A.3.2.4.1.1

/3A.3.2.4.3.2 Downcomers bracing system 3A.3.2.3.1.3

/ 3A.3.2.3.3.1 3A.3.2.3.1.4

/3A.3.2.3.3.1 3A.3.2.4.1.2

/ 3A.3.2.4.3.1 3A.3.2.4.1.1

/3A.3.2.4.3.2 SRV system 3A.3.2.3.1.1

/3A.3.2.3.3.1 3A.3.2.3.1.2

/3A.3.2.3.3.1 3A.3.2.4.1.2

/ 3A.3.2.4.3.1 3A.3.2.4.1.1

/3A.3.2.4.3.2

Note: Numbers refer to section of Appendix 3A.

C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.3.2-24 Table 3A.3.2-2 CGS Data for Loss-of-Coolant Accident Water Jet A n alysis Unit cell diameter 40.87 in.

Unit cell depth 354.0 in.

Downcomer inner radius 13.625 in.

Downcomer submergence 144.0 in.

Downcomer water c l ear i ng velocity time history Figure 3A.3.2-2

C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.3.2-25 Table 3A.3.2-3 CGS Data for Vent Clearing And P ool Swell Analysis Drywell 1. Temperature (init i al) 135 F 2. Drywell pressure transient Table 3A.3.2-5

3. Relative humidity 0%

Suppression Chamber

1. Free air volume (maximum) 147,290 f t 3 2. Net suppression pool surface area 4520 f t 2 3. Pressure (initial) 0.7 psig

4. Air specific heat ratio 1.4 a/1.2 b Downcomer vent system

1. Submergence (minimum/maximum) 11 ft-8 in./12 ft a 2. Nominal diameter 2 ft

3. Number of vents 102

4. Vent exit area 321 ft 2 5. Vent loss coefficient 1.9 a Value used to determine ma ximum pool swell elevation.

C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.3.2-26 Table 3A.3.2-4 Results from Loss-o f-Coolant Accident Bubble Charging A n alysis for CGS

Downcome r s Inner Radius Middle Radius Outer Radius Bubble coalesc e nce time (sec

)a 0.09 0.16 0.24 Bubble radius at coalescence (ft) 2.12 2.77 3.41 a Times rep r esent time aft e r vents have cleared.

C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.3.2-27 Table 3A.3.2-5 CGS Dryw e ll Pressure as a Function of Time for

Loss-of-Coolant Acciden t a (Effects of Pipe Invento r y and S u bcooling Included)

Time After Loss-of-Coolant Accident (s ec) Drywell Pressure (ps i a) 0.0 15.45 0.00159 15.32 0.00171 15.30 0.00549 14.72 0.0641 17.61 0.127 20.18 0.252 24.83 0.502 33.27 0.720 35.69 0.740 35.42 1.099 35.11 1.537 35.84 1.568 35.92 2.037 36.00 a See Re f e rence 3A.3.2-2 2.

C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.3.2-28 Table 3A.3.2-6 Results of Pool Sw e ll Analysis for CGS Vent clearing tim e a - t c 0.65 sec Pool water s u rface vel o cit y b at time t c 5.3 ft/sec Time of maximum pool swell velocity 1.12 sec Maximum pool swell velocit y b 28.7 ft/sec Time of maximum pool height a 1.48 sec Maximum pool swell height (H ma x) 18.0 ft Ratio of H max to H o (downcomer submergence)

1.5 Maximum

air bubble p r essure 35.75 psia Maximum wetwell airspace pressure 47.55 psia Maximum pool swell elevation 484 ft 4.75 i n. msl Pool swell terminatio n a 1.48 sec a Times rep r esent time after L O CA initiation.

b Does not include velocity multiplier of 1.1.

C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.3.2-29 Table 3A.3.2-7 CGS Plant Parameters for Loss-of-Coolant Accident Trans i ent Analysis

1. Drywell

a. Free air volume 200,540 f t 3 b. Temperatu r e (ini tia l) 135 F c. Pressure (initial) 0.75 psig d Relative humidity 50% 2. Wetwell

a. Free air volume 144,184 f t 3 b. Water volum e a 107,850 ft 3 c. Pool temperature (ini tia l) 90 F d. Pressure (initial) 0.75 psig

e. Relative humidity 100% 3. Break a r ea a. Design basis accident (DBA) -

recirculation line 3.106 ft 2 b. DBA - steam line 3.92 f t 2 c. Intermedia t e break 0.1 f t 2 4. Main vent

a. Maximum submergence 12 ft

b. Nominal diameter 2 ft c. Number of vents 10 2 b d. Vent entrance flow area 304.6 f t 2 e. Downcomer loss f actor 1.9 a Water volume b Refer to S ection 3A.3.2.2 for a discussion of the effect of capping three downcomers.

C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.3.2-30 Table 3A.3.2-8 Short-Term Loss-of-Coolant Accident Loads

on Structur e s Below Elevation 454.4 ft Radial Location r of Geometric Center of the Structure or Segment of Structure Zone I 0 r 2.3R Zone II 2.3R r 5.0R Structure p r max (psi) p v max (psi) p r max (psi) p v max (psi) 42 in. diameter vert i cal column +2 12 in. diameter vertical SRV line 20 inner row

+2 outer row Pipes and Supports Diamete r a > 12 in. (X-31,32,34,35,36) 60 +212 +6 -45 Diamete r a 12 in. (X-33,100,4 in. FPC, quencher arm) 60 +100 +6 -25 a For noncy l indrical structures, the diameter of a cylinder circumscribing the cross section of the structure is used.

Notes:

1. Vertical load is positive in the downward d i rection. Radial load is positive in the radially outward direction from the axis of symmetry of the downcome

r.

2. The vertical distribution associated with the ta b u lat e d peak values is shown in Figure 3A.3.2-18. The distribution in the r a dial direction in each zone i s uniform and

is aximsy m m etric in t h e circumferential direction.

3. Long structures are divid e d into smaller segments, L D; D being the diameter of the structure, and L being t h e segment length.

C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.3.2-31 Table 3A.3.2-8

Short-Term Loss-of-Coolant Accident Loads

on Structur e s Below Elev ation 454.4 ft (Continued)

4. Radial or vertical load is ca l cula t e d at the geome t ric cen t e r of each struc t ure or segment (see Figure 3A.3.2-18) by multiplying the pressure value at the geometric center (obtained from this table) with the leng t h and the diame t er of the structure.

For noncylindrical structures, diameter of the circumscribing c y linder is used.

5. Radial or vertical load h a s, in general, two c o mponents. One is p a rallel and the other is normal to the structure or segment. All com p onents of load that are p a ral l el to the structure or segment are neglec t ed.

6. The radial location, r, of the structure or segment is tak e n from the nearest vent. Since the flow from vents occurs in-phase du r ing water clearing/a i r charging phases of

LOCA, the flow field calculations and the above specified loads have already accounted

for the multi-vents effect. The r efore, effects of flow from other adjacent vents need not be added or subtra c t ed.

7. The loads specified a r e equivalent static and utilize a DLF equal to 2.

C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.3.2-32 Table 3A.3.2-9

Short-Term Loss-of-Coolant Accident Loads on

Structures Between Elevations 454 ft 4.75 in. and 484 ft 4.75 in.

Equivalent Static Load Pool Swell Fallback Pool Swell/ Fallback Structures Located Between El. 454 ft 4.75 in. and El. 484 ft 4.75 in.

Pool Swell/

Impact P (psi) Drag P (psi) Horizontal Lift Ph (psi) I. Downcomer bracing truss 25 5 II. Platform at el. 472 ft 4 in.: grating perimeter members 8.4 25 III. Piping at 467 ft 9 in. and supports A. Horizontal portions of pipes

a. X-48,118,117,4 7 ,26,63,49,101, X-23,24, X-4 (pipe and sleeve) 41 25 15 b. X-64,65 16 20 15 B. Inclined braces above penetra t ions 50 25 C. Horizontal supports below penetrations 60 25 IV. Penetration sleeves, p i pe stubs, and electrical box protective structures

a. X-51,66 sleeves for:

X-81,82,83,84,116 209 25 5 b. X-87A 166 25 5 c. X-86A 62 25 5

d. X-88,87B,86B,116 (piping) 25 5 e. X-81,82,83,84 (piping except X-82e) 209 20 5 f. X-107A, X-107B (electrical box protective structures) 116 25 5 Short Term Hydrodynamic Processes Associated with a LOCA 950021.76 3A.3.2-1 Figure Amendment 53 November 1998Form No. 960690.veR.oN .warD MaximumPool Swell Height 4. FALLBACK

3. POOL SWELL

2. LOCA BUBBLE CHARGING

1. VENT CLEARING Increasing Drywell PressureWetwell Air Compression Increasing Drywell PressureWetwell Air Compression Initial Pool Surface Elevation (Typ.)Air Bubble Pressure Air Bubble

Pressure Water Slug Columbia Generating Station Final Safety Analysis Report Downcome r Vent Water Clearing VelocityVersus Time 950021.77 3A.3.2-2 Figure Amendment 53November 1998 Form No. 960690Draw. No.Rev.10 20 30 40 50 60 70 80 90 100Time = 0.654 Sec.Velocity = 82.22 Ft/Sec0.10.20.30.40.50.60.70Time After LOCA (Sec)

Columbia Generating StationFinal Safety Analysis Report Vent Clearing Velocity (Ft/Sec)

Downcome r Vent Water Clearing AccelerationVersus Time 950021.78 3A.3.2-3 Figure Amendment 53November 1998 Form No. 960690Draw. No.Rev.Time = 0.654 Sec.

Acceleration = 1254.26 Ft/Sec 2 0.10.20.30.40.50.60.70Time After LOCA (Sec) 0 2 4 6 8 10 12 14 16-2 Columbia Generating StationFinal Safety Analysis Report Vent Clearing Acceleration (Ft/Sec 2)x10 2

Draw. No.Rev.Figure

Draw. No.Rev.Figure

Draw. No.Rev.Figure LOCA Bubble Radius and Source Strength Time Histories by PSAM Method 950021.79 3A.3.2-7 Figure Amendment 53November 1998 Form No. 960690Draw. No.Rev.0.02.04.06.08.10

.18.24Time After Vent Clearing (Sec) 1.6 2.0 2.4 2.8 3.2 3.6 4.0 4.4 0.4.8 1.2.12.14.16.20.22 R(t)Q(t)4.8 1.2 1.4 1.6 1.8 2.0 2.2 2.4 2.6 2.8 3.0 3.2 3.4 3.6 1.0 R(t)1.1 x 10 4 Q(t)1.1 x 10 4 ft in/s()ft in()ft Columbia Generating StationFinal Safety Analysis Report Q(t)Q(t)

Pool Swell Water Slug Velocity Versus Time 950021.80 3A.3.2-8 Figure Amendment 53November 1998 Form No. 960690Draw. No.Rev.Time After LOCA (sec)Slug Velocity (ft/sec) 0 4 8 12 16 20 24 28 32 3600.40.81.21.62.02.42.8 Columbia Generating StationFinal Safety Analysis Report Pool Swell Water Slug Acceleration Versus Time 950021.81 3A.3.2-9 Figure Amendment 53November 1998 Form No. 960690Draw. No.Rev.Time After LOCA (Sec)

-160-120-80-40 0 40 80 120 160 20000.40.81.21.62.02.42.8 Columbia Generating StationFinal Safety Analysis Report Slug Acceleration (Ft/Sec

2)

Pool Swell Water Slug Elevation(Top Surface) Versus T ime 950021.82 3A.3.2-10 Figure Amendment 53November 1998 Form No. 960690Draw. No.Rev.Time After LOCA (Sec) 466 468 470 472 474 476 478 480 482 484 00.30.60.91.21.51.82.1 Columbia Generating StationFinal Safety Analysis Report Elevatin of Slug Top Suface (Ft.)

Pool Swell Air Bubble Pressure Versus T ime 950021.833A.3.2-11 Figure Amendment 53November 1998 Form No. 960690Draw. No.Rev.Time After LOCA (sec)

Pressure (psia) 0 5 10 15 20 25 30 35 40 45.60.75.901.051.201.351.501.65 Columbia Generating StationFinal Safety Analysis Report Pool Swell Wetwell Air Pressure Versus Time 950021.84 3A.3.2-12 Figure Amendment 53November 1998 Form No. 960690Draw. No.Rev.Time After LOCA (Sec)

Pressure (psia) 0 6 12 18 24 30 36 42 48 54.600.30.60.91.21.51.82.1 Columbia Generating StationFinal Safety Analysis Report Pool Swell Water Slug Velocity Versus Elevationof Slug Top Surface 950021.85 3A.3.2-13 Figure Amendment 53November 1998 Form No. 960690Draw. No.Rev.Elevation of Slug Top Surface (ft)Slug Velocity (ft/s) 0 4 8 12 16 20 24 28 32 36 466469472475478481484487 Columbia Generating StationFinal Safety Analysis Report Fallback Water Slug Velocity Versus Elevationof Water Slug Top Surface 950021.86 3A.3.2-14 Figure Amendment 53November 1998 Form No. 960690Draw. No.Rev.Elevation of WaterSlug Top Surface (Ft) 0 5 10 15 20 25 30 35 40Fallback Water Slug Velocity (Ft/Sec)El. 484' - 4 3/4" El. 466' - 4 3/4" Columbia Generating StationFinal Safety Analysis Report LOCA Boundary Load Duration 950021.87 3A.3.2-15 Figure Amendment 53November 1998 Form No. 960690Draw. No.Rev.Event Event A B C D Loading PhenomenonSubmerged Boundary Loads During Vent Clearing Drywell Pressurization

Loads on Submerged Boundaries During Pool Swell Wetwell Air Compression 0 0.5 1.0 1.5 2.0 D C B ATime After Break (Sec)

Columbia Generating StationFinal Safety Analysis Report LOCA Boundary Load DistributionDuring Vent Clearing 950021.88 3A.3.2-16 Figure Amendment 53November 1998 Form No. 960690Draw. No.Rev.Drywell Air Pressure (see Table 3A.3.2-5

)Downcomer Vent Initial Pool Surface Elevation Suppression PoolWetwell Air Pressure (see Fig 3A.3.2-12

)Overpressure Hydrostatic Pressure Columbia Generating StationFinal Safety Analysis Report Amendment 55 May 2001LOCA Boundary Load Distribution During Pool Swell 950021.89 3A.3.2-17 Figure Form No. 960690Draw. No.Rev.Drywell Air Pressure (See Table 3A.3.2-5

)Downcomer Vent Suppression PoolWetwell Air Pressure (See Fig 3A.3.2-12

)Hydrostatic Pressure Maximum Bubble Pressure (See Table 3A.3.2-6

)Maximum Pool Swell Elevation (See Table 3A.3.2-6)Maximum Wetwell Air Pressure (See Fig. 3A.3.2-12

)P Columbia Generating StationFinal Safety Analysis Report

.50Distribution of Short Term LOCA Loads onStructures Below El. 454.4 ft 950021.90 3A.3.2-18 Figure Amendment 53November 1998 Form No. 960690Draw. No.Rev.Pv i= 12L i x .25 x (100)

Pr i= 12L i x .25 x ( 60)Pv j= 12L j x .25 x (-25)

Pr j= 12L j x .25 x (6)

Example: Segment j Zone 2 Zone 1 Segment i Pr iL iL j12"Ø Quencher Arm 1.0 1.0.75.25 Pv j Pv iVent C Lof Axisymmetry R2.3R5.0R Elevation 454. 4' 453.0'452.0'451.0'450.0'449.0'448.0'Vertical Distributionof Pv, Pr Pr Pv Zone 1 Zone 2 Pr Pv Pr j Columbia Generating StationFinal Safety Analysis Report

.6 psi/in.Pressure Gradients Across Submerged Structures Due to Chugging 950021.91 3A.3.2-19 Figure Amendment 53November 1998 Form No. 960690Draw. No.Rev.C L.6 psi/in.

DowncommerBracing TrussContainment VesselWater Level El. 466.4'12" Ø SRV Line RVent DowncomerVent Exit El. 454.4' 4R.8 psi/in.1.0 psi/in.

.4 psi/in.

Piping System 24"Ø Entering Containment at El. 452' 4R A 2.4R C L Columbia Generating StationFinal Safety Analysis Report Large Recirculation Line Break - Pressure Response - Minimum ECCS 950021.92 3A.3.2-20 Figure Amendment 53November 1998 Form No. 960690Draw. No.Rev.Time (Sec)

DrywellWetwell 0.1 1 10 10 2 10 3 40 30 20 10 0 Columbia Generating StationFinal Safety Analysis Report Drywell (DW) and Wetwell (WW) Pressure (psig)

Containment Pressure Response for LargeRecirculation Line Break -Cases A, B, and C 950021.93 3A.3.2-21 Figure Amendment 53November 1998 Form No. 960690Draw. No.Rev.Time (Sec)

Containment Pressure (psig) 10 2 10 3 10 4 10 5 10 6 40 30 20 10 0 c b a a) 3 LPCI, 1 HPCS, 1 LPCS, 2 HX, KHX = 578 b) 1 LPCI, 1 HPCS,1 HX, KHX = 289

b) 1 LPCI, 1 HPCS,1 HX, KHX = 289, No Containment Spray Columbia Generating StationFinal Safety Analysis Report Large Recirculation Line Break -Temperature Response 950021.94 3A.3.2-22 Figure Amendment 53November 1998 Form No. 960690Draw. No.Rev.Time (sec) 0.1 1 10 10 2 10 3 400 300 200 100 0 DrywellWetwell Columbia Generating StationFinal Safety Analysis Report Drywell and Wetwell Temperature (°F)

Drywell Temperature Response forLarge Recirculation Line Break -Cases A, B, and C 950021.95 3A.3.2-23 Figure Amendment 53November 1998 Form No. 960690Draw. No.Rev.Time (sec) 10 1 10 2 10 3 10 5 10 6 400 300 200 100 0 a) 3 LPCI, 1 HPCS, 1 LPCS, 2 HX, KHX = 578 b) 1 LPCI, 1 HPCS,1 HX, KHX = 289

c) 1 LPCI, 1 HPCS,1 HX, KHX = 289, No Containment Spray 10 4 c b a Columbia Generating StationFinal Safety Analysis Report Drywell Temperature (°F)

Suppression Pool Temperature Response for Large Recirculation Line Break -Long Term Response 950021.96 3A.3.2-24 Figure Amendment 53November 1998 Form No. 960690Draw. No.Rev.Time (sec)

Suppression Pool Temperature (°F) 600 10 3 10 5 10 6 400 300 200 100 0 a) 2 HX, 3 LPCI, 1 HPCS, 1 LPCS, KHX = 589 W/Spray b) 1 HX, 1 LPCI, 1 HPCS, KHX = 289, W/Spray

c) 1 HX, 1 LPCI, 1 HPCS, KHX = 289, No Containment Spray 10 4 b, c a Columbia Generating StationFinal Safety Analysis Report Pressure ResponseMain Steam Line Break 950021.97 3A.3.2-25 Figure Amendment 53November 1998 Form No. 960690Draw. No.Rev.Time (Sec)

Pressure (psig) 1 10 2 10 3 40 30 20 10 0 10Wetwell Drywell 0.1 Columbia Generating StationFinal Safety Analysis Report Temperature Response -Main Steam Line Break -

Minimum ECCS 950021.98 3A.3.2-26 Figure Amendment 53November 1998 Form No. 960690Draw. No.Rev.Time (Sec)Temperature ( F)1 10 2 10 3 400 300 200 100 0 10Wetwell Drywell 0.1 Columbia Generating StationFinal Safety Analysis Report Temperature Response -Recirculation Line Break (0.1 ft 2)950021.99 3A.3.2-27 Figure Amendment 53November 1998 Form No. 960690Draw. No.Rev.Time (Sec)Temperature ( F)1 10 2 10 3 400 300 200 100 0 10Wetwell Drywell 0.1 Columbia Generating StationFinal Safety Analysis Report Amendment 54 April 2000 Figure Form No. 960690Draw. No.Rev.Pressure Response -Recirculation Line Break (0.1 ft 2)960222.84 3A.3.2-28Time (Sec)

Pressure (psig) 0.1 1 10 10 3 40 30 20 10 0 10 2 DrywellWetwell Columbia Generating StationFinal Safety Analysis Report C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.3.3-1 3A.3.3 LOAD

SUMMARY

A load summary is given in Table 3A.3.3-1 to provide guidance in identifying loads and to prov i de references to more d e tailed discussions of them.

T h e table lis t s the loads being evaluated and the peak load magnitude directly ap plied to each structure, or a reference to the DAR sections where the information is derived. It also lists the classification of the load as primary or secondary as defined by NRC.

C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT N o v e mb e r 1998 3A.3.3-3 Table 3A.3.3-1 a Summary of Hydrodyn a mic Loads on Wetwell Structures

Piping S y stem Load Category Stl. Contain-ment Vert. and Horiz. Tees (6)

Base-mat Pedestal Dia. Floor Dia. Floor Seal Down-comer Bracing Column Down-comer SRV Piping System Quencher Plat-for m s and Ladd e r F u lly Sub. (3)

Part. Sub. In P ool Swell Zone (3) DAR Ref. Sec.

Load C l ass SRV Water Clearing sm sm sm sm N/A N/A sm sm sm sm sm N/A sm sm N/A 3.1.2.1 Sec. Air Clearing 151 151 151 151 N/A N/A 3.1.3.2 3.1.3.2 3.1.3.2 3.1.3.2 3.1.3.2 N/A 3.1.3.2. 3.1.3.2 N/A 3.1.3 Prim. Steam Cond. sm sm sm sm N/A N/A sm sm sm sm sm N/A sm sm N/A 3.1.3.2 Sec. L O CA Vent Water Clearing 24 24 24 sm N/A N/A sm 3.2.3.3 sm 3.2

.3.3 3.2.3.3 N/A 3.2

.3.3 sm N/A 3.2.3 Prim. Air Bu b b le Charging 21 b 21 b 21 b s m N/A N/A a Conformance to NRC acceptance criteria for each load in this table is presented in A.H. b Values are in psig.

Notes: 1. Peak dynamic pressure.

2. Unless identified as dynamic pressure, the numerical value given is equivalent static peak pressure (i.e., includes a DLF).

3. All fully-submerged piping systems enter the pool below vent exit.

4. sm denotes load as small, not requiring consideration.

5. Between elevations 466 ft-4.75 in. and 484 ft-4.75 in.

6. LOCA induced vent, thrust loads are neglected as secondary loads (see Reference 3A.3.2-1).

C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.3.4-1 3A.3.4 SEQUENCE OF DYNAMIC LOADS The effects of various dynamic loads on structures are analyzed separately. It is important to establish the relative time sequenc e of all dynamic events to obt ain a realistic assessment of design margins. The DFFR (Reference 3A.3.2-2) established relative sequence of dynamic loads during a single SRV discharge event and a pos tulated LOCA. It is noted that during an SRV actuation the water clearing loads, the ai r clearing loads, and the steam condensation loads occur in a sequence and th e peak dynamic effects due to each need not be combined.

Similarly, the LOCA water clearing, air clearing (bubble char ging), pool swell impact, pool swell drag, pool fallback, and the high, inte rmediate, and low mass flux steam condensation occur in a sequence and the peak dynamic effects due to each need not be combined. During pool swell or fallback, drag pressure parallel to the flow and lift pressure normal to the flow occur simultaneously, and the tw o loads are combined.

Also, for some submerged structures, pool swell impact occurs at the upper parts of th e structure while pool sw ell drag and lift loads act on the lower parts of the structure. In this case, the three lo ads are combined. The continuity of short-term wate r and air clearing phases during an SRV discharge or during a LOCA and the pool swell impact and drag loads during a DBA LOCA is recognized either by specifying a conservative DLF for use with peak lo ad value of the combined time history or by use of the combined time history in the dyna mic analysis in assessing the structures.

The DFFR provides guidance about the SRV actu ations occurring during the normal operation of the plant and during the small, intermedia te, and large break LO CA. In absence of information about relative tim e sequence of the th ree events, the fo llowing conservative assumptions for submerged structures are made in combining the effects due to these events in this assessment:

a. Short-term LOCA and SRV loads For structures submerged in the pool, the direct hydrodynamic loads due to the LOCA jet, LOCA bubble, pool swell, a nd fallback are not combined with the direct hydrodynamic loads due to the SRV water jet or air bubbles. The presence in the pool of air bubbles from the SRV line is assumed to have negligible effect on short-term pool dynamics during a LOCA. The seismic effects are combined with the short-term LOCA load effects; and

b. Long-term LOCA and SRV loads For all miscellaneous subm erged piping systems and ma jor structures (columns, downcomer bracing, downcomers and SRV line with que ncher) the worst case LOCA steam condensation loads are combin ed with seismic effects and with a single adjacent SRV actuation load or with the actuation of ADS valves.

C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.3.5-1 3A.3.5 LOAD COMBINATIONS AND ACCEPTANCE CRITERIA Load combinations and accep tance criteria for events wh ich include suppression pool hydrodynamic loads are described in this section. Four categor ies of structural components affected by these events are iden tified and the applicable load combinations and acceptance for each category are listed. These structural categories are the steel containment structure (suppression chamber portion), rein forced concrete structures, st eel structures, and piping and piping systems. Symbols representing generic lo ad types are used in the load combinations; these symbols are defined below.

3A.3.5.1 Steel Containment Structure

3A.3.5.1.1 Definitions

D Dead loads

L Live loads

E O Loads generated by operating basis earthquake

E SS Loads generated by safe shutdown earthquake

H Loads associated with pool swell phenomenon following L the clearing of the downcomer vents including fallback

P Containment pressure associated with the large break A (DBA) LOCA

P B Containment pressure associated with IBA or SBA

P Loads associated with chugging phenomena

P E Design external pressure on the containment

P o Normal operating pressure PSR Loads associated with main steam SRV actuation P V LOCA related hydrodynamic loads on suppression chamber, including H L, P c R A Pipe reactions under therma l conditions generated by the postulated accidents and including R o

R E Pipe reactions under thermal conditions during event causing external pressure C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.3.5-2 R O Pipe reactions during startup, normal operating or shutdown conditions, based on the most critical transi ent or steady-state condition

R R Reaction and jet forces associated with the pipe break

T A Thermal loads under thermal cond itions associated with LOCA

T E Thermal loads under thermal conditions during event causing external pressure

T O Thermal effects and load s during normal operation

3A.3.5.1.2 Load Combinations

The following load combinations for the CGS steel containment are in agreemen t with those specified in the NRC Standard Review Plan, 3.8.2, Revision 0, and properly include the new SRV and LOCA hydrodynamic loads.

(1) D+L+P O+T O+R O+P SR (2) D+L+E O+P O+T O+R O+P SR (3) D+L+E O+T A+R A+(P A or P B)+P V+P SR (4) D+L+E O+T E+R E+P E+P SR (5) D+L+E ss+P O+T O+R O+P SR (6) D+L+E ss+T A+R A+(P A or P B)+P V+P SR (7) D+L+E ss+T E+R E+P E+P SR Notes:

1. In all combinations the hydrostatic pressure due to th e presence of water in the pressure-suppression chamber pool is considered with dead and live loads.

2. Restraint due to the presence of filler ma terial between the containment vessel and the biological shield wall (equivalent to an extern al pressure of 2 psi) is considered where critical.

C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.3.5-3 3. For independent, short duration, vibratory loads such as seismic, SRV discharge, chugging loads, the peak dyna mic responses due to the individual loads are combined by the square-root-of-sum-of-t he-squares (SRSS) method.

4. For time relationship between concurrently applied loads in combinations (3) and (6) see Section 3A.3.4.

5. In the case of P SR and P C both the axisymmetric a nd nonaxisymmetric loading conditions are investigated.

6. Maximum equivalent static pressures for P A , P B , and P V are given in Table 3A.3.5-1.

The design assessment of the c ontainment vessel is made on th e basis of the SRSS method as stated above. However, subse quent investigation i ndicates that if combination of peak dynamic responses due to seismi c, SRV discharge loads, and chugging loads is done by the absolute sum method, the resulta nt design margins for the contai nment vessel are greater than 1.0.

3A.3.5.1.3 Acceptance Criteria

The design rules for the steel containment are in accordance with ASME Code Section III, 1971 Edition through the 1972 Summer addenda, Subsection NE, Class MC Components.

The acceptance criteria for each load combina tion are summarized in Table 3A.3.5-2.

3A.3.5.2 Reinforced-Concrete Structures

Structures to which the criteria below apply include the basemat, the RPV pedestal, the columns supporting the diaphragm floor , and the diaphragm floor slab.

3A.3.5.2.1 Definitions

D Dead loads

E O Loads generated by operating basis earthquake E SS Loads generated by safe shutdown earthquake H L Hydrodynamic forces associated with pool swell phenomenon following the clearing of the downcomer vents including fallback forces

L Live loads

C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.3.5-4 P A All loads associated with the large break (DBA) LOCA including drywell and suppression chamber transient pressure loads and H L, P c, and P co as defined herein P B All loads associated with IBA or SBA type of LOCA including transient pressure loads and P c P c Loads associated with chugging phenomena during LOCA

P O Normal operating pressure

P SR Loads associated with main steam SRV actuation

R A Pipe reactions under therma l conditions generated by the postulated accidents and including R O

R O Pipe react ions during startup, normal operating or shut down conditions, based on the most critical transient or steady-state condition

R R Reaction and jet forces associated with the pipe break

T A Thermal loads resulting from thermal cond itions generated by postu lated accidents and including T O

T O Thermal effects and loads during normal operation.

3A.3.5.2.2 Load Combinations

The load combinations for the base mat and for the reinforced concrete structures internal to the containment are listed in Table 3A.3.5-3. The following notes are applicable to Table 3A.3.5-3. a. In combinations 4, 4a, 5, 5a, 7, and 7a, the maximum values of P A , P B , T A , R A , and PSR including a DLF are used unless a dynamic analysis is performed.

b. Thermal loads may be neglected when it can be shown that they are secondary and self-limiting in nature.

c. All the loads listed are not necessarily applicable to all the structures.

d. For independent short duration vibratory loads such as seismic, SRV discharge loads and chugging loads, the peak dyna mic responses are combined by the SRSS method. Also, peak responses due to SRV direct pressure loads and due to SRV building motion response spectra are combined by SRSS.

C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.3.5-5 e. The design assessment of the structures of Section 3A.3.5.2 is made on the basis of the preceding notes. Ho wever, subsequent investigation indicates that if combination of peak dynamic responses due to seismic, SRV discharge loads and chugging loads is done by the absolu te sum method, the resultant design margins for the structures of Section 3A.3.5.2 are greater than 1.0.

3A.3.5.2.3 Acceptance Criteria

For all load combinations in Table 3A.3.5-3 , the allowable limit on section strength is the section strength required to re sist design loads based on the st rength design methods described in ACI 318-77 (Reference 3A.3.5-1). 3A.3.5.3 Steel Structures

Structures to which the criteria below apply include the downcomer bracing system, the diaphragm floor beams, and platforms and ladders attached to the containment shell.

3A.3.5.3.1 Definitions

Definitions of load symbols in Table 3A.3.5-4 are the same as those in 3A.3.5.2.1.

3A.3.5.3.2 Load Combinations

The load combinations for steel structures internal to the containment are listed in Table 3A.3.5-4. Notes a, b, c, and d listed for Table 3A.3.5-3 in 3A.3.5.2.2 are also applicable to Table 3A.3.5-4.

The design assessment of the structures of 3A.3.5.3 is made on the basis of the preceding notes. However, subsequent investigation indicates that if combination of peak dynamic responses due to seismic, SRV discharge loads and chugging loads is done by the absolute sum method, the resultant design ma rgins for the structures of 3A.3.5.3 are greater than 1.0.

3A.3.5.3.3 Acceptance Criteria

The allowable limits for structural acceptance for the load combinations of Table 3A.3.5-4 using the elastic working stress method are defined as follows.

C OLUMBIA G ENERATING S TATION Amendment 54 F INAL S AFETY A NALYSIS R EPORT April 2000 LDC N-9 9-0 0 0 3A.3.5-6 Combination Limit 1 S 2,3 1.5S 4,4a,5,5a,6 1.6S

7,7a 1.7S

In the above, S is the required section strengt h based on the elastic design methods and the allowable stresses defined in Part 1 of the AISC Specification (Reference 3A.3.5-2). The 33% increase in allowable stresses for concrete and steel due to seis mic loadings is not permitted.

The allowable limits for structural acceptance for the load combinations of Table 3A.3.5-4 using the plastic design met hod are defined as follows:

Combination Limit 1,2,3 Y 4,4a,5,5a,6,7,7a 0.9Y

In the above, Y is the section strength required to resist design loads based on the plastic design methods described in Part 2 of the AISC Specifi cation (Reference 3A.3.5-2).

3A.3.5.4 Piping Systems

3A.3.5.4.1 Definitions

The loads for the piping component s are: dead weight, seismic, and loads associated with SRV actuation and LOCA effects. The SRV and LO CA loads have been described in detail in previous sections of this report. A description of the symbols as they appear in the piping and component load combination table (Table 3A.3.5-5) follows:

Load Symbol Load Description

P Operating pressure D.W. Dead weight

OBE Loads due to opera tional basis earthquake

SSE Loads due to safe shutdown earthquake

C OLUMBIA G ENERATING S TATION Amendment 54 F INAL S AFETY A NALYSIS R EPORT April 2000 LDC N-9 9-0 0 0 3A.3.5-7 SRV Loads due to sequential pre ssure setpoint actuation of all (18) SRVs

a. SRV bubble loads on submerged piping and components in the suppression pool

b. Building motion induced loads.

SBA/IBA Loads associated with SBA/IBA:

a. Chugging/CO loads on submerged structures. (Chugging bounds CO loads.)

b. Building motion due to chugging loads. (Chugging bounds CO loads.)

DBA Loads associated with DBA:

a. Water jet

b. LOCA bubble

c. Pool swell

d. Fallback

e. Chugging loads on submerged structures. (Chugging bounds CO loads.)

f. Building motion induced loads due to CO and chugging. (Chugging bounds CO.)

3A.3.5.4.2 Load Combinations

Load combinations for the loads listed in Section 3A.3.5.4.1 are given in Table 3A.3.5-5. These load combinations are based on Table 6.1 of the DFFR (Reference 3A.3.2-11) and modified conservatively. For independent short duration vibratory loads such as seismic, SRV discharge loads, and chugging loads, the peak dynamic respons es are combined by the SRSS method as described in the DFFR. The time relationship for the loads are described in Section 3A.3.4.

C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.3.5-8 3A.3.5.4.3 Acceptance Criteria

Piping and components are designed for norma l, upset, emergency, and faulted plant conditions, as delineated in the Load Combination Table using the stress values for the respective normal, upset, emergency, and faul ted limits as defined in the appropriate subsection of the ASME Boiler and Pressure Vessel Code (Reference 3A.3.5-3).

3A.3.5.5 References 3A.3.5-1 "Building Code Requirements fo r Reinforced Concrete," ACI 318-71/77.

3A.3.5-2 "Specification for the Design, Fabri cation, and Erection of Structural Steel for Buildings," American Institute of St eel Construction, February 12, 1969/

November 11, 1978.

3A.3.5-3 ASME Boiler and Pressure Vessel Code,Section III, Divisi on 1, Subsection NC "Class 2 Components," American Soci ety of Mechanical Engineers, 1971 through Winter 1973 Addenda.

• Faulted conditions appeared for the first time in Winter 1976 addendum.

C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.3.5-9 Table 3A.3.5-1 Equivalent Static Loads fo r Pressure Trans i ents and Loss-of-Coolant Accident Effects

P A (Max) psig P B (Max) psig H L (Max) psig Loading Combinations (See Section 3A.3.5.1.2) Drywell Wetwell Drywell Wetwell Vent Clearing Pressure (1), (2), (4), (5), and (7)

- - - - - (3) and (6)

+34 +28 +30 +25 +24 C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT N o v e mb e r 1998 3A.3.5-10 Table 3A.3.5-2 Accep t a nce Criter i a for Containment Ves s el Allowable Stress Limits Primary Stresses Loading Combinations a General Membrane (P m)Local Membrane (P L) Bending+ Local Membrane (P B+P L) Secondary Stresses Peak Stresses Buckling (1) and (2)

S m 1.5 S m 1.5 S m 3S m Consider for fatigue analysis Allowable given by ASME III Section NE-3133 (3) and (4)

S m 1.5 S m 1.5 S m N/A N/A Allowable given by ASME III Section NE-3133 (5), (6), and (7) For elements not integral and

continuous S m 1.5 S m 1.5 S m N/A N/A Allowable given by ASME III Section NE-3133 For elements integral and continuous The greater of

1.2S m or S y The greater of

1.8S m or 1.5S y The greater of

1.8S m or 1.5S y N/A N/A 120% of Allowable given by ASME III Section NE-31311 a For definition of loading combinations, see Section 3A.3.5.1.

Notes:

1. Thermal stresses need not be considered in computing P m , P L , and P B.

2. Thermal effects are considered in:

a. Specifying stress intensity limits as a function of temperature. b. Analyzing effects of cyclic operation ASME III Section NE-3222.4 C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT N o v e mb e r 1998 3A.3.5-11 Table 3A.3.5-3 Load Combinations - Reinf o rced-C o n crete Structures Number Load Condition D L P O T O R O E O E SS P B P A T A R A P SR R R Service Lo a d Conditions

1 Normal w/o t emperat u r e 1.4 1.7 1.0 1.5 2 Normal w/ t e mperat u r e 1.0 1.3 1.0 1.0 1.0 1.3 3 Normal sev ere envir o nme n t 1.0 1.0 1.0 1.0 1.0 1.25 1.25 Factored L o ad Conditions

4 Abnormal (I BA/SBA) 1.0 1.0 1.25 1.0 1.0 1.25 4a Abnormal (D BA) 1.0 1.0 1.25 1.0 1.0 1.

0 a 5 Abnormal (I BA/SBA) severe envir o nment 1.0 1.0 1.1 1.1 1.0 1.0 1.1 5a Abnormal (D BA) severe envir o nment 1.0 1.0 1.1 1.1 1.0 1.0 1.

0 a 6 Normal ex t r e me envir o n m ent 1.0 1.0 1.0 1.0 7 Abnormal (I BA/SBA) extreme envir o nment 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 7a Abnormal (D BA) extreme envir o nment 1.0 1.0 1.0 1.0 1.0 1.0 1.

0 a 1.0 a Single valve actuation.

Note: See Section 3A.3.5.2.2.

C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT N o v e mb e r 1998 3A.3.5-12 Table 3A.3.5-4 Load Combinations - Steel Structures Number Load Conditi o n D L P O T O R O E O E SS P B P A T A R A P SR R R Using Elast i c Working Stress Design Met h od - Part 1 of AISC Specs, 1969 Service Load Conditions

1 Normal w/o temperat u r e 1.0 1.0 1.0 1.0 2 Normal w/ temperat u r e 1.0 1.0 1.0 1.0 1.0 1.0 3 Normal sev.

envir. 1.0 1.0 1.0 1.0 1.0 1.0 1.0 Fact o r ed L o ad Conditions

4 Abnormal (IBA) 1.0 1.0 1.0 1.0 1.0 1.0 4a Abnormal (DBA) 1.0 1.0 1.0 1.0 1.0 1.0 a 5 Abnormal (IBA) sev. env.

1.0 1.0 1.0 1.0 1.0 1.0 1.0 5a Abnormal (DBA) sev. env.

1.0 1.0 1.0 1.0 1.0 1.0 1.0 a 6 Normal ext.

env. 1.0 1.0 1.0 1.0 1.0 1.0 7 Abnormal (IBA) ext. env.

1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 7a Abnormal (DBA) ext. env.

1.0 1.0 1.0 1.0 1.0 1.0 1.0 a 1.0 Using Plastic Design Meth o d - Part 2 of A I SC Specs, 1969 Service Load Conditions

1 Normal w/o temperat u r e 1.7 1.7 1.7 1.5 2 Normal w/te m perat u r e 1.0 1.3 1.0 1.0 1.0 1.3 3 Normal sev.

env. 1.0 1.0 1.0 1.0 1.0 1.25 1.25 Fact o r ed L o ad Conditions

4 Abnormal (IBA) 1.0 1.0 1.25 1.0 1.0 1.25 4a Abnormal (DBA) 1.0 1.0 1.25 1.0 1.0 1.25 a 5 Abnormal (IBA) sev. env.

1.0 1.0 1.1 1.1 1.0 1.0 1.1 5a Abnormal (DBA) sev. env.

1.0 1.0 1.1 1.1 1.0 1.0 1.1 a 6 Normal ext.

env. 1.0 1.0 1.0 1.0 1.0 1.0 1.0 7 Abnormal (IBA) ext. env.

1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 7a Abnormal (DBA) ext. env.

1.0 1.0 1.0 1.0 1.0 1.0 1.0 a 1.0 a Single valve actuation.

Note: See Section 3A.3.5.2.2

.

C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 Table 3A.3.5-5 Load Combinations and Acce p t a nce Criter i a for ASME Code

Class 1, 2, and 3 Balance

-of-Plant Piping and Equipmen t a 3A.3.5-13 Load Ca s e s Load Combinat i ons (1, 2, and 3)

Design A ssessment Acceptance Cr iteria 1 P+D.W. Normal (A) 2 N+OBE+SR V ONE Upset (B) 3 N+OBE+SR V TWO Upset (B) 4 N+OBE+SR V ALL Upset (B) 5 N+OBE+SR V ALL+SBA Emergenc y b (C) 6 N+OBE+SR V TW O+SBA Emergenc y b (C) 7 N+SSE+SR V AD S+SBA/IBA Faulte d b (D) 8 N+SSE+SR V TW O+SBA/IBA Faulte d b (D) 9 N+SSE+SR V ONE Faulte d b (D) 10 N+SSE+SR V TWO Faulte d b (D) 11 N+SSE+SR V ALL Faulte d b (D) 12 N+SSE+DBA Faulte d b (D) a Equipment includes pumps, valves, supports, a nd vessels. For bolting used in connection with the support of ASME Code Class 1, 2, and 3 components, vendor load capacity data sheets are used, or where design is by the architect engineer, stre ss levels are maintained less than specified minimum yield at temperature.

b All ASME Code Class 1, 2, a nd 3 piping systems which are re quired to function for safety shutdown under the postulated events shall meet the requirements of NRC's memorandum, "Evaluation of Topical Report - Piping Functional Capability Cr iteria," date July 17, 1980.

C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 Table 3A.3.5-5 Load Combinations and Acceptance Criteria for ASME Code Class 1, 2, and 3 Balance-of-Plant Piping and Equipment a (Continued) 3A.3.5-14 Notes: 1. As required by the appropri ate subsection, i.e., NB, NC, or ND of ASME Section III, Division I. Other loads, such as thermal tran sient, thermal gradient s, and anchor point displacement portion of the OBE or SRV, may require c onsideration in addition to those primary stress-producing loads listed.

2. SBA, IBA, and DBA include all event induced loads, as applicable, such as chugging, pool swell, drag loads, a nnulus pressurization, etc.

3. Seismic and hydrodynamic loads are combined by the SRSS technique and added to the applicable static loads.

Load Definition Legend

Normal (N) Normal loads include in ternal pressure and dead weight

OBE Operational basis earthquake loads

SSE Safe shutdown earthquake loads

SRVTWO SRV discharge induced loads from two adjacent valves

SRV ALL The loads induced by act uation of all SRVs

SRV ADS The loads induced by the actuation of SRVs associated with the automatic depressurization system

SRV ONE The loads induced by the actuation of one SRV

C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.4.1-1 3A.4 DESIGN ASSESSMENT 3A.4.1 SUPPRESSION POOL BOUNDARY STRUCTURES

3A.4.1.1 Assessment of Steel Containment Structure

The primary containment structure in the suppression chamber area, as shown in Figure 3A.4.1-1 , is an orthogonally sti f fened steel shell.

S e e FSAR Section 3.8 for a description of the steel containment structure.

The thickness of the steel she ll plate in the suppression chamber region is approximately 1.5 in. and varies with height. Vertical "Tee" stiffeners, at a spacing of 40.5 in., are welded to the inside face of the she ll plate and extend about 8 ft be yond the knuckle (cylinder-cone inter-face) elevation. In th e pool region of the suppression chamber, additional horizontal "Tee" stiffener rings at an approximate spacing of 36 in., are welded to the inside face of the shell plate. The purpose of addi ng these t e es (F i gure 3A.4.1-4) is to inc r ea s e the l o ad carrying capacity of the containment shell. The vertical tees are required for resisting compressive loads due, mainly, to seismic effects. Th e horizontal tees are intended to carry the hydrodynamic loads which were not c onsidered in the original desi gn of the containment shell. There are no heavy attachments to the cont ainment in the suppression chamber region.

The drywell floor slab is radially separated from the containment and the gap is sealed by means of a radially and vertic a lly flexible sea l , as shown in Figure 3A.4.1-9. The s l ab is connected to the containment in the tange ntial direction by means of shear lugs.

3A.4.1.1.1 Loads Us ed For Assessment

The methods used for calculating hydrodynamic loads on the pool boundary, as described in Sections 3A.3.1 and 3A.3.2 , provide a conservative definition of loads for design assessment.

3A.4.1.1.1.1 Safety/Relief Valve Loads.

The suppression pool bounda ry pressure loading is determined in accordance with procedures described in Reference 3A.4.1-1 on the basis of operating conditions at CGS.

A discussion on the derivation of the safety/relief valve (SRV) load definition is provided in 3A.3.1. Several different incidents may occur which would cause one or more SRVs to actuate. For example, the valves may operate either manually, or on pressure setpoints following a turbine trip, automatically through the automatic depressurization system (ADS) system.

The critical modes of SRV actuations considered in the design are detailed in Reference 3A.4.1-1. For the purposes of design assessmen t, consideration is given to all the SRV discharge cases that are postulated to occu r during the life of the CGS plant. A summary of the various cases is explained below.

C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.4.1-2 3A.4.1.1.1.1.1 Single Valve Discharge Case.

Actuation of any si ngle SRV is postulated during a loss-of-coolant accident (LOCA) invol ving a large or intermediate break. Two possible cases of single SRV discharge are considered, the single inner quencher discharge and the single outer quencher discharge. A single inner quencher discharge is more likely to occur because of its lower pressure setpoint. Ho wever, a single outer quencher discharge is conservatively assumed for the assessment of the containment vessel.

3A.4.1.1.1.1.2 Two Valves Discharge Case.

For this event, two SRVs are considered to discharge concurrently through two adjace nt quenchers.

3A.4.1.1.1.1.3 Automatic Depressurization System Valves Discharge Case.

The seven ADS valves for CGS are assigned to outer quenchers in a configura tion that is nearly axisymmetrical. The ADS is characterized by an automatic and simultaneous actuation as discussed in the DFFR (Reference 3A.3.2-2). The ADS discharge is not considered to occur during a large pipe break LOCA. However, it is assumed that the AD S may discharge during an intermediate pipe break or a small pipe break LOCA.

3A.4.1.1.1.1.4 All Valves Discharge Case. Under certain plant c onditions, the actuation of all 18 SRVs in CGS is assumed. Two different conditions may occur during this event: the axisymmetric all valves discharge conservatively assumes that all 18 SRVs discharge simultaneously; the nearly symmetric all valv es discharge assumes that there is some imbalance during the discharge event. As discussed in the DFFR (Reference 3A.3.2-2), the all valves discharge is not considered to occur during a large pipe break LOCA. However, it is assumed that the all valves disc harge may occur during an intermediate pipe break or a small pipe break LOCA.

3A.4.1.1.1.2 Loss-of-C oolant Accident Loads.

Loss-of-coolant accidents are associated with postulated large pipe breaks [design basis accident (DBA)], intermediate pipe breaks [intermediate break accident (IBA)], or sma ll pipe breaks [small br eak accident (SBA)].

Various transient LOCA pressure loads on the pool boundary cons idered in the assessment of the containment include vent wa ter clearing jet loads, air bubble pressure loads, pool swell, fallback, drywell and wetwell pressu re transients, and chugging loads.

3A.4.1.1.1.2.1 Chugging Loads.

A general discussion of the chugging phenomenon is included in Section 3A.3.2.4. Design pool boundary lo ads discussed in Section 3A.3.2.4.2.1 are used for the structural assessment.

3A.4.1.1.1.2.2 High and Medium Mass Flux Condensati on Oscillations.

During the sequence of a LOCA event, condensation oscillations take place after pool swell and fallback.

Depending on the steam mass flux ra te, they are identified as either (a) high mass flux or (b) medium mass flux condensa tion oscillations. Howeve r, as noted in Section 3A.3.2.4.1.2 , the controlling boundary pressu re loads due to chugging exc eed those due to condensation C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.4.1-3 oscillations. Therefore, condensation oscillation loads are not c onsidered in the assessment of the containment vessel.

3A.4.1.1.1.2.3 Other Loss-of-Coolant Accident Loads. Loss-of-coolant acci dent loads, other than those due to condensation oscillations and chugging, include

a. Pressure and temperature transients

These transients represent sy mmetric loadings. An equi valent static loading is used for a LOCA pressure transient which takes due account of the time history of the pressure buildup. Thermal effects (T A) on stress intensities and cyclic operation have been considered. Pressure and temperature transients considered in the a s se ss m ent are s h own in Figu r es 3A.3.2-20 th r ough 3A.3.2-28; b. Reaction from downcomer vent horizontal exit load

The horizontal loads acting at the downcomer exits are transmitted via the downcomer bracing system and result in tangential reactions at the steel containment structure;

c. Pool swell bubble pressure

The air slug pressure in the suppression pool during pool swell acts symmetrically around the inside of the c ontainment structure. Its time history, in relation to the appropriate period of the containment structure, is used in determining the dynamic load factor (DLF).

In addition to the preceding case of sy mmetric loading, the case of an asymmetric bubble pressu re acting on the submerged boundary in accordance with Reference 3A.3.2-5 is also included; and

d. Reactions from components supported by the steel containment structure

Drag and impact loads occur on structural components supported by the steel containment structure as a result of pool swell and fallback. Reactions from these structural components are carri ed by the containment structure.

3A.4.1.1.1.3 Othe r Significant Loads.

a. Seismic loads

Loads due to the operating basic ear thquake (OBE) and th e safe shutdown earthquake (SSE), developed in the proj ect design, are applicable. Seismic C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.4.1-4 loads include the effect of the water in the suppression pool. The effects due to water sloshing (pressure load) have been accounted for in the containment pool swell assuming a concurrent seismi c event (SSE) is insignificant;

b. Dead load, live load, and hydrostatic pressure The hydrostatic pressure due to the s uppression pool is included in the dead load; and

c. Design external pressure

An external pressure of 2 psi resulting from atmospheric conditions inside and

outside the drywell is used. When ex ternal pressure governs the design, an additional external pressure of 2 psi due to the reaction of the compressible foam between the containment and the bi ological shield wall is used. This reaction results from the thermal expansion of the containment shell.

3A.4.1.1.2 Controlling Load Combinations

The applicable load combinations for the pr essure-suppression chambe r portion of the steel containment structure ar e defined in Section 3A.3.5.1.2. Combinations presented therein are also applicable to the horizontal and vertical stiffening tees. Load combination (3), stated below, is found to control the design of the stiffened steel c ontainment structure.

Load Combination (3): D+L+E O+T A+R A+P B+P V+P SR The interpretation and contribution of each of the terms depends on the event being considered. In considering the overall steel containment struct ure, the controlling combination of events involves ADS actuation during an intermediate break LOCA with clugging.

Consequently, P B and T A refer to the IBA. In th is load combination, T A and R A are relatively insignificant. The term P SR refers to ADS pool boundary pressure lo ading. Thus, the effective controlling load combination involves:

D+L+E O+P B+P SR+P C 3A.4.1.1.3 Acceptance Criteria

The acceptance criteria for design of steel containment st ructure, including horizontal and vertical tees, is in compliance with the 1971 ASME Code, Edition through the 1972 Summer Addenda,Section III, Division 1, S ubsection NE is given in Section 3A.3.5.1.3.

C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.4.1-5 3A.4.1.1.4 Method of Analysis

3A.4.1.1.4.1 Formulation of the Problem. In accordance with th e methods presented in Sections 3A.3.1 and 3A.3.2 for defining hydrodynamic loads, it is assumed that the incident pressures on the pool boundary, re sulting from bubble oscillations and steam conde nsation, are acting on the boundary as extern ally applied loads. In computing the responses of the structure, the fluid-shell interaction effects are accounted for by solving the coupled partial differential equations gove rning the fluid and the shell struct ure using finite elements. The stresses due to chugging used for assessing the containment structure are obtained by using the building model and the method of analysis presented in Section 3A.5.2. For SRV loads, a refined containment mode l described in Section 3A.5.1 is used.

For pool boundary hydrodynamic loads, since the vertical an d horizontal tees are integrated in the model of the suppression chamber portion of the containment, the fluid-shell interaction analysis also gives the dyna mic stresses in the tees.

3A.4.1.1.4.2 Mathematical Model. The mathematical model used in the containment analysis for SRV loads is discussed in Secti o n 3A.5.1.1.2 and sho w n in Figure 3A.5.1-2. Prominent features of this axisymmetric model are discussed in Reference 3A.4.1-1. The mathematical model used in the containment analysis fo r chugging loads is discussed in Section 3A.5.2.1 and shown in F i gure 3A.5.2-1. A detailed descr i ption of the chugging model can be found in

Reference 3A.4.1-4.

3A.4.1.1.4.3 Coupled Equations of Motion.

a. Shell equations

The partial differential e quations governing the motion of the containment shell are based on equations given in Reference 3A.4.1-3; b. Fluid equations

The partial differential e quations governing the dynamics of compressible fluid are the continuity equation and the equation of motion establishing the relationship between the pressure and the velocity of a particle in the fluid. For reasons cited in Reference 3A.4.1-4, the water in the suppression pool is considered to be compre ssible for the chugging model.

However, for reasons cited in Reference 3A.4.1-1 , the water is considered to be incompressible for the SRV model;

C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.4.1-6 c. Fluid boundary conditions The fluid boundaries can be seen in Figure 3A.4.1-8. The pressure at the fluid surface is specified to be zero. The continuity requi rement at the fluid-structure interface is satisfied by sp ecifying the radial component of fluid motion at the interface to be equal to that of the shell;

d. Shell boundary conditions As discussed in Section 3A.5.1.1.2 , a refined containment model is used for the analysis of the containment structure unde r SRV loading. This refined model is connected to the overall building mode l at the following locations (see Reference 3A.4.1-1): basemat in the radial, ve rtical, and circumferential directions; diaphragm floor in the circum ferential direction; and stabilizer truss and refueling bellows in the radial and circumferential directions.

For the analysis of the containment structure due to chugging loads, the overall building model discussed in Section 3A.5.2.1 , is used;

e. Geometric symmetry

The axisymmetric geometry of the containment shell-fluid system is utilized in the solution of the equations in cylindrical coordinates. The azimuth coordinate

is eliminated from the governing equations by representing azimuthal dependence of each variable by a Fourier series. The equations are thus solved for each Fourier term or harmonic of the series and the final solution is obtained by summation of solutions for each term.

3A.4.1.1.4.4 Numerical Solution. Numerical solutions to equations described in Section 3A.4.1.1.4.3 are obtained by using finite elemen ts. An integration time step of 0.001 sec is used in the analyses for both SRV and chugging loads.

3A.4.1.1.4.5 Computer Program.

A Burns and Roe computer program "HYDI-2" (Attachment 3A.F) was developed for Mark II containmen t configurations, and subsequently used in the analysis of the containment stru cture due to chugging loads. The numerical solutions were verified by the commercially available finite elem ent program "NASTRAN" (Attachment 3A.F). The analysis of the refined containment model for SRV loads was made with the "NASTRAN" program.

3A.4.1.1.5 Results and Design Margin

3A.4.1.1.5.1 Results of Analysis.

The hydrodynamic pressure loads, as described in Section 3A.4.1.1.1 , are applied to the containment wall of the fluid-shell interaction models C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.4.1-7 (Figures 3A.5.1-2 and 3A.5.2-1). Utilizing the Burns and R o e, Inc. computer program HYDI-2 for chugging loads, and the NASTRAN pr ogram for SRV loads, the responses of the containment are computed. The maximum stresses (in time) are evaluated in the applicable load combinations for determining the cont rolling load combination and corresponding design margin for the containment structure. Maximum time-wise profiles of radial displacements are presented in Figures 3 A.4.1-2 and 3A.4.1-3 for the contai n m ent struct u r e.

Responses to various load s discussed in Section 3A.4.1.1.1 are summarized below:

a. ADS discharge case

The ADS actuation combined with IBA or SBA is the most critical axisymmetric pressure load on the containment. For the purposes of assessment, the ADS pressure loading is conservatively assumed to be the same as the larger of each of the two all valves discharge pressure loadings, i.e., the design boundary pressure. Thus, the re sponse of the containment to the ADS actuation is shown in F i gure 3A.4.1-2. The re s u lting stres s es a r e used i n the controlling load combination for cal culating the design margin of the containment structure;

b. All valve discharge case

In considering the two different all valves discharge events, the nearly symmetric pressure loading is slightly greater than the axisymmetric pressure loading (see References 3A.4.1-1 and 3A.4.1-2); c. Single valve discharge case Actuation of a single SRV combined w ith LOCA (DBA) loads results in the most critical non-axisymmetric pressure loading on the contai nment. However, responses of the containment to the resu lting load combination is less severe than case (a), above;

d. Two valves discharge case Responses of the containmen t to this load are conser vatively assumed to be the same as case (c) (see Reference 3A.4.1-1); e. Chugging Reference 3A.4.1-4 , described in Section 3A.3.2.4.2.2 , presents the responses of the containment shell to the chugging pressure load. The resulting stresses are used in the controlling load combina tion for calculating the design margin of C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.4.1-8 the containment structure. As discussed in Section 3A.5.2.2, the nearly symmetric chugging load is used for asse ssment purposes.

The response of the containment to these design chugging l o ads is s h own in F i gure 3A.4.1-3

and f. High and medium mass flux condensation oscillations

Condensation oscillation loads are not considered in the assessment of the containment structure since they are bounded by chugging loads.

3A.4.1.1.5.2 Assessment Results. The containment assessment performed in accordance with Sections 3A.3.5.1.2 and 3A.3.5.1.3 shows that in load combina tion (3) the general membrane stress intensity controls the containment design.

Based on this calculation, the design margin for the containment shell is 1.29.

The buckling strength of the CGS containment in resisting the external pressure and axial compression acting on the suppression pool region increases substa ntially as a result of adding the horizontal stiffening rings. The most critical load combination under events causing net external pressure and axial compression is load combination (7). Under this load combination the ratio of the allowable buckling pressure load of the containment to the applied external pressure load is 3.1. The ratio of the allowable buckling axial load of the containment to the axial compressive load is 1.37.

In this analysis, the intera ction effect is accounted for by assuming that the horizontal stiffening tees resist only external pressure while the vertical stiffening tees resist only axial compression.

For the containment tees, the stress intensiti es for load combination (3) are governed by primary bending plus local memb rane stresses. These stresses occur at the webs of the horizontal tees and at the root of the flange for the vertical tees. Based on these stress values, the design margins for the vertical and horizontal tees are 2.23 and 1.26, respectively.

The combined stresses in both the containment and tees are calculated for the controlling load combination by adding the stress resulting from static loads alge braically and stresses due to dynamic oscillating loads by the square-root-of-sum-of-the-squares (SRSS) method. The

resulting design margin for the c ontainment structure (including the horizontal and vertical tees) is 1.26.

3A.4.1.2 Basemat The assessment of the capacity of the basemat relative to load combinations involving suppression pool hydrodynamic load s is made in this section.

The basemat and adjoining structures are shown in Figu r es 3A.4.1-5 and 3A.4.1-6.

C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.4.1-9 3A.4.1.2.1 Loads Used for Assessment A complete description of all th e hydrodynamic loads used in the assessment of the basemat is provided in Section 3A.3. Symbols, equations, and load combinations referred to in this section are detailed in Section 3A.3.5.2.

3A.4.1.2.1.1 Safety/Relief Valve Loads.

Loads on the suppressi on pool boundary due to SRV actuations are detailed in Section 3A.3.1. Specific SRV loading cas es considered in the basemat assessment include symmetric loads due to the actuation of all 18 valves and asymmetric loads due to the actuation of a single SRV. Fo r both cases, dynamic stresses in the basemat (bending and shear) are developed on the basis of a time-histor y application of the loads. The analytical m odel u s ed for the as s e ss m ent is sh o w n in Figu r e 3A.5.1-lb. Prominent features of the model, including the use of axisymmetric shell elements is discussed in Section 3A.5.1.

3A.4.1.2.1.2 Loss-of-C oolant Accident Loads.

Loads on the suppression pool boundary due to a LOCA are detailed in Section 3A.3.2. Of all the LOCA loads, chugging pressures are the most significant with respect to the basemat. Other LOCA loads, including jet loads and bubble pressures, are not significant with respect to the basemat assessment. Dynamic stresses in the basemat (bending and shear) are developed on the basis of a time-hi story application of the loads. The analyt i c al m odel used for the assessment is shown in F i gure 3A.5.

2-1. Prominent features of the model, including the use of axisymmetric shell elements, are discussed in Section 3A.5.2.

3A.4.1.2.1.3 Othe r Significant Loads. Seismic loads constitute a principal loading in the basemat assessment. The seismic loads on the basemat from the superstructure (exterior walls, biological shield wall, and pedestal) are the same as those used in the original building design.

In that original design, a dynami c analysis was made using a disc rete mathematical idealization of the entire reactor building. The stress resultants at the base of the superstructure (overturning moment, axial forc e) due to the OBE and the SSE as developed in the original design are used.

Dead and live loads as developed in the original structural design are also used.

3A.4.1.2.2 Applicable Load Combinations and Acceptance Criteria

The load combination and acceptance criteria described in Section 3A.3.5.2 are applicable to the basemat.

3A.4.1.2.3 Method of Analysis

The structural capacity of the ba semat is investigated for the applicable load combinations with loads as listed above. The gene ral approach in the basemat as sessment is to determine the C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.4.1-10 values of the controlling stress resultants in the basemat (be nding and shear) on the basis of elastic analysis under applied desi gn loads and to calculate the cap acity of the basemat in terms of these stress resultants by the strength method of the ACI 318-71 Code (Reference 3A.4.1-5). Critical sections for bend ing and shear are located with respect to the face of the biological shield wall in complia nce with code requirements.

3A.4.1.2.3.1 Effects of E O , E SS , D, L. The investigation of the basemat for the combined effect of dead, live, and seismic loads is ba sed on the analysis performed in the original building design. Depicted as a plate on an el astic foundation, the base mat is modeled as a series of plate elements while the supporting soil is modeled as a group of elastic springs situated at designated nodes. The effects of the seismic overturning moment and the vertical acceleration of the dead and live loads are converted to nodal lo ads. Resulting stresses from the model and loads described above are calculated with the use of the computer program NASTRAN. Values of the c ontrolling bending moment, beam shear, and punching shear due to combined dead, live, and SSE loads are tabulated in Table 3A.4.1-1 for the critical section of the basemat.

3A.4.1.2.3.2 Effect of P SR , P B. The values of bending mome nts and shears at the critical section in the basemat due to SRV actuation and chugging are available fr om the analysis of the reactor building models described in Section 3A.5. The symmetric mode of SRV actuation and the nearly symmetric mode of chugging result in the comparatively la rger values of stress resultants. The controlling stress resulta nts for these loads are tabulated in Table 3A.4.1-1 for the critical section.

3A.4.1.2.3.3 Critical Load Combination. Review of the stress resultant values in connection with the applicable load combinations shows that the critical load comb ination for all stress resultants is load combination (7); this combination is noted below with only the significant terms included.

(7) D + L + E SS + P B + P SR 3A.4.1.2.3.4 Capacity.

The capacity of the basemat with respect to bending, beam shear, and punching shear is determined in accordan ce with the strength met hod of the ACI 318-71 Code (Reference 3A.4.1-5). These stress resultant capacities are listed in Table 3A.4.1-1. 3A.4.1.2.4 Results and Design Margins

Comparison of the design values for the stress resultants in Table 3A.4.1-1 with the capacity values in the table shows that the basemat provides adequate capacity. The ratio of bending capacity to design bending moment is 1.14. The ratio of beam shear capacity to design beam shear is 1.48. The ratio of punching shear capacity to design punching shear is 1.27.

C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.4.1-11 3A.4.1.3 Pedestal The assessment of the capacity of the reactor pressure vessel (RPV) pedestal re lative to load combinations involving suppression pool hydrodynamic loads is ma de in this section. The pedestal and adjoining structures in the s uppression chamber are sho w n in Figure 3 A.4.1-7.

3A.4.1.3.1 Loads Used for Assessment

A complete description of a ll the hydrodynamic loads used in the assessment of the RPV pedestal is provided in Section 3A.3. Symbols, equations, and load combinations referred to in this section are detailed in Section 3A.3.5.2.

3A.4.1.3.1.1 Safety/Relief Valve Loads.

Loads on the suppressi on pool boundary due to SRV actuations are detailed in Section 3A.3.1. Specific SRV loading cas es considered in the pedestal assessment include symmetric loads due to the actuation of all 18 valves and asymmetric loads due to the ac tuation of a single SRV. Fo r the asymmetric case, dynamic stresses in the pedestal are developed on the basi s of time history applic ation of the load. For the symmetric case, dynamic stress es are developed on the basis of applied pressures increased by an appropriate DLF which is determined from the time histor y analysis. The analytical model used for the assessment is shown in Figu r e 3A.5.1-lb. Prominent features of the model, including the use of axisymmetric shell elements, are discussed in Section 3A.5.1.

3A.4.1.3.1.2 Loss-of-C oolant Accident Loads.

Loads on the suppression pool boundary due to LOCA are detailed in Section 3A.3.2. Of all the LOCA loads, chugging pressures are the most significant with resp ect to the pedestal. Other LOCA loads, including pool swell, jet loads, and bubble pressures are not significant wi th respect to the pe destal assessment.

Dynamic stresses in the pedestal are developed on the basis of the model and load application described in Section 3A.4.1.2.1.2.

3A.4.1.3.1.3 Othe r Significant Loads.

Seismic loads (E O , E SS) constitute a principal loading in the pedestal assessment. The seismic loadings and associated analysis in this assessment are the same as those used for the original design. A dynamic analysis was made using a discrete mathematical idealization of the entire reactor building including the pedestal. The stress resultants in the pedestal (overall bending moment, horizontal shea r force, and axial force) due to the OBE (E O) and the SSE (E SS) as developed in the original design are used. Dead loads as developed in the original bu ilding design are also utilized.

3A.4.1.3.2 Applicable Load Combinations and Acceptance Criteria

The load combinations and accep tance criteria for internal reinforced concrete structures described in Section 3A.3.5.2 are applicable to the pedestal.

C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.4.1-12 3A.4.1.3.3 Method of Analysis

The structural capacity of the pedestal is investigated for the load combinations with two types of loading, namely, asymmetric and symmetric. The general approach in the pedestal assessment is to determine the va lues of the controlling stress re sultants on the basis of elastic analysis under design loadings and to calculate the capacity of the pedestal in terms of these stress resultants in accord ance with the strength method of the ACI 318-71 Code (Reference 3A.4.1-5). 3A.4.1.3.3.1 Asymmetric Action. The loads which contribute to asymmetric action of the pedestal are seismic loads (E O and E SS), loads due to single SRV actuation (P SR), and loads due to chugging phenomena (P B). The significant stress resultants associated with these loads are overturning moment and total shear.

For seismic loading, the values of these stress resultants in the original design are used. For P SR and P B , the stress resultants are obtained by integrating over the entire pedestal section the stresses obtained from the elastic analysis of the reactor building structural model. Controlling values of the stress resultants which occur at the base of the pedestal are tabulated in Table 3A.4.1-2. Review of the stress resultant values in connection w ith the load combina tions shows that the critical load combination for both moment and sh ear is load combinati on (7), stated below, with only the significant load terms included.

(7) D + ESS + P B + P SR The capacity of the pedestal relative to overturning moment and concurrent axial load is expressed by the interaction curve shown in Fig u re 3A.4.1-10. Points along the interaction curve representing different capacity combinations of axial load (P u) and bending moment (M u) have been calculated in line with the AC I 318-71 Code (Reference 3A.4.1-5). The minimum and controlling value of axial load occurs with upward seismic ac tion; this axial load value (12,380 kips) and the controlling overturning moment from Table 3A.4.1-2 (212,380 ft kips) are also plotted in Figure 3 A.4.1-10. As noted in the figure, the bending moment capacity coincident with the controlling axial load is 375,000 ft kips.

The capacity of the pedestal relative to overall horizontal (tangential) shear, calculated in accordance with the ACI 318-71 Code (Reference 3A.4.1-5), is 14,500 kips. From Table 3A.4.1-2 , the controlling design shear is 2760 kips.

3A.4.1.3.3.2 Symmetric Action. Loads which contribute to symmetric action of the pedestal are due to actuation of all 18 SRVs (P SR) and chugging phenomena (P B). The hydrostatic pressure (D) also causes symmetric action. For P SR and P B, an appropriate DLF is included as noted in Section 3A.4.1.3.1.1. Symmetric action is investigat ed with respect to radial and circumferential normal stresses a nd with respect to the effect of the end fixity at the base.

C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.4.1-13 To obtain the radial and circumferential normal stresses, the pede stal is analyzed as a thick walled cylinder. Maximum compressive stre ss (circumferential) occurs for load combination (1):

(1) 1.4D + 1.7L + 1.OP O + 1.5P SR In this load combin ation, terms L and P O do not contribute to the stre sses being considered and are omitted. For calculation purposes, the maximum values of D, and P SR (positive value) are used. Maximum tensile stress (circumferential) occurs for load combination (4):

(4) 1.OD + 1.OL + 1.OT A + 1.OR A + 1.25 (P B + P SR)

In this load co mbination, L, T A, and R A are omitted as they do not affect the stress resultant under consideration.

Radial shear and moment occur unde r the symmetric loads due to th e fixity of the pedestal at its junction with the basemat. Th e analysis is based on a general theory of the elastic behavior of cylindrical shells. Maximum values of radial shear and moment occur at the pedestal base with load combination (4).

In this load co mbination, L, T A, and R A are omitted as they do not affect the stress resultant under consideration.

3A.4.1.3.4 Results and Design Margins

Results for asymmetric loading are summarized below:

a. The controlling value of the pedestal overturning moment un der design loadings is less than the pedestal moment capacity. The ratio of the moment capacity to the controlling overturni ng moment is 1.77; and

b. The controlling value of the pedestal base shear under design loadings is less than the pedestal shear capacity. The ratio of shear capacity to controlling applied shear is 5.25.

Results for symmetric load ing are summarized below:

a. The calculated normal stresses occurring during sy mmetric action are found to be less than the allowable strength values. The ratio of pedestal capacity to stress under controlling load ing is 5.53 for tensile ci rcumferential stress and 15.83 for circumferential compressive stress; and

C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.4.1-14

b. The calculated stresses due to radial shear and moment due to pedestal fixity at its base during symmetric ac tion are found to be less th an the allowable strength values. The ratio of radial shear capacity to maximum shear due to load is 1.11 and the corresponding ratio for radial moment is 5.06.

Review of the preceding results shows that the overall controlling de sign margin is 1.11 applicable to radial shear under symmetric loading.

3A.4.1.4 Diaphragm Floor

Assessment of the capa c ity of the diaphragm floor (see Fig u res 3A.4.1-6 and 3A.4.1-7) relative to load combinations involving suppression pool hydrodynam ic loads is made in this section.

3A.4.1.4.1 Loads Used for Assessment

A complete description of all hydrodynamic loads is given in Section 3A.3. This subsection discusses the loads used for the assessment of the diaphragm floor.

3A.4.1.4.1.1 Safety/Relief Valve Actuation Loads.

Safety/relief valve discharge does not result in pressure loads directly on the diaphragm floor, but causes dynamic horizontal pressure differentials across the downcomers, the SRV piping, and the columns, all of which are supported at the diaphragm floor, and dynami c vertical pressure differential across the downcomer bracing which is transferred to the diaphragm floor by the downcomers.

In addition, building response spectra from SRV discharge result in acceleration of the diaphragm floor which induces dynamic stre sses in the components of the floor.

3A.4.1.4.1.2 Loss-of-C oolant Accident Loads.

The maximum net downward pressure on the diaphragm floor during a DBA LOCA is 20 psi (see Section 3A.3.2.5.2). Since the time required to develop the maximum net downward differential pressure resulting from a recirculation line break is approximately 0.7 sec, dynamic effects ar e not significant and temperature transients at time of peak downward pressure differential do not contribute to the floor loading.

The maximum net upward pressure on the diaphragm floor is of short duration, and is due to wetwell atmosphere compre ssion resulting from pool swell during a DBA LOCA (see Section 3A.3.2.5.2). A value of 5.5 psi maximum net upward pressure is considered in the assessment of the dia phragm floor (see Section 3A.3.2.3.1.3

).

Pool swell and fallback drag loads on the downcomer bracing (see Section 3A.3.2.3.3.1) are transferred to the diaphragm floor by the downcomers.

C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.4.1-15 Other significant LOCA loads in clude pipe break jet impinge ment and steam condensation accelerations obtained from the response spectra at the diaphragm floor support locations.

3A.4.1.4.1.3 Othe r Significant Loads.

Other loads which result in significant stresses in the diaphragm floor are dead loads, live load s, and vertical seismic accelerations.

Dead leads include the weight of the diaphragm floor reinforced concre te slab and supporting steel beams, downcomers, horizontal run of SRV piping, includi ng vertical supports, downcomer bracing supported vertically by the downcomers and, hence, the diaphragm floor.

Live loads include personnel and equipment weights on the diaphragm floor. Seismic accelerations are obtained from the seismic re sponse spectra at the support locations for the diaphragm floor.

3A.4.1.4.2 Controlling Load Combinations

The load combination cr iteria for structures in ternal to the pressure-suppression chamber (see Sections 3A.3.5.2 and 3A.3.5.3) are applicable to the diaphragm floor. In particular, the combinations for steel structures, using the elastic working stress design method with both service load conditions and factored load conditions, are investig ated in the analysis for the structural steel components of the floor (see Section 3A.3.5.3). The combinations for reinforced concrete structures, using the ultimate strength design method with both service load conditions and factored load conditions, are investigated for the reinfo rced concrete slab component of the fl oor (see Section 3A.3.5.2). The controlling load combinations are specified in Section 3A.4.1.4.5.

3A.4.1.4.3 Acceptance Criteria

The acceptable stress levels for the steel components of the dia phragm floor are specified in Section 3A.3.5.3.3.

The acceptable allowable limit for the concrete components is the ultimate strength as determined by the ultimate strength design method of the ACI 318-71 Building Code (Reference 3A.4.1-5).

3A.4.1.4.4 Method of Analysis The diaphragm floor components, c onsisting of the reinforced concre te slab, the st ructural steel circumferential and radial beams, and connect ions, were investigated individually for the effects of both the upward and downward loads. To determine design loads for each of the components, the diaphragm floor was analyzed as a slab (one-way);

beam (circumferential beams), and girder (radial beam s) structural system with th e radial beams supported at the pedestal and on the columns, but not at the containment vessel shell.

C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.4.1-16 3A.4.1.4.5 Results and Design Margins

The diaphragm floor reinforced c oncrete slab and the steel circ umferential and radial beams, including connections, were found to have sufficient capacity to withstand the governing load combinations. The critical component under downward load, as defined by the governing load combination (7a) for steel structures, is the radial beam. The design margin (i.e., ratio of the allowable stress to the maximum absolute calculated stress) for the radial beam is 1.62, and is based on the following input into the loadi ng combination (7a) for steel structures:

(7a) 1.7S 1.0D+1.0L+1.0E SS+1.0P A+1.0T A+1.0R A+1.0P SR+1.0R R D = 4.28 psi

L = 6.94 psi E SS = 1.52 psi

P A = 20 psi

T A = 0 R A = 0 P SR = 1.65 psi

R R = 534,000 lb

where D, L, E SS , P A , T A , R A , P SR , and R R are as defined in Section 3A.3.5.2.

The critical components under upward load are the anchor bolts at the radial b eam to column connection. The design margin (i.e., ratio of the allowable st ress to the maximum absolute calculated stress) far the anchor bolts is 1.27, based on the following input into the governing loading combination (4a) for reinforced concrete structures:

(4a) U 1.0D+1.0L+1.25P A+1.0T A+1.0R A+1.0P SR D = 4.28 psi

L = 0

P A = 5.57 psi

T A = 0 C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.4.1-17 R A = 0 P SR = 1.65 psi

where D, L, P A , T A , R A , and P SR, are as defined in Section 3A.3.5.2.

3A.4.1.5 Diaphragm Floor Seal The diaphragm floor seal is located at the in side surface of the primary containment vessel periphery at el. 493 ft 5 in. It provides a flex ible, pressure tight seal between the primary containment vessel and the diaphragm floor a nd is capable of accommodating differential thermal expansion between them.

The diaphragm floor seal is a 270 omega-shaped configuration of stainless steel a nd is drained to the floor drain sy stem with four drain pipes, as shown in F i gure 3A.4.1-9.

3A.4.1.5.1 Loads Used for Assessment

a. Normal plant condition

This condition is defined as reactor startup, operation at power, and normal reactor cold shutdown.

These loads are due to th ermal expansion of the component and thermal disp lacement between the conc rete diaphragm floor and primary containment during normal plant operation;

b. SRV loads

These loads are not directly applied to the diaphrag m floor seal, but do cause displacement of the primary containment shell relative to the diaphragm floor resulting in stress in the seal;

c. LOCA loads

The LOCA combination governing the design of the seal includes the effects of relative thermal displacement, differential pressure, and hydrodynamic effects.

Other LOCA effects, given in Section 3A.3.2, include the temperature and pressure transients, pool swell air compression load, and primary containment displacement due to build ing response. A discussi on of direct load, i.e., temperature and pressure transients a nd pool swell phenomenon is presented in Section 3A.4.1.5.2

and C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.4.1-18

d. Other loads The effect of dead load, OBE, and SSE seismic loads, as applicable, are included in the analysis. Loads from the drain pipes are also included.

3A.4.1.5.2 Controlling Load Combination

The controlling load combination for the diaphragm floor seal is that which includes loads due to a DBA. The load due to pool swe ll air compression is given in Section 3A.3.2.3. The LOCA pressure and temperature tran sients are described in Section 3A.3.2.5.

The following individual loads were utilized for load comb inations, per ASME Code Section III, Subsection NE in the design of the diaphragm floor seal. The load combinations used are defined in Table 3A.3.5-5. a. Normal Plant Conditions,

b. OBE,

c. SSE,

d. LOCA,

e. SRV loads with all four valve actuations,

f. Piping loads due to all dynamic loads listed above,

g. The maximum differe ntial pressure, and

h. Relative displacement due to the movement of the primary containment vessel and diaphragm floor.

The fatigue evaluation wa s performed conservativel y using the cycles in Table 3A.4.1-3.

3A.4.1.5.3 Acceptance Criteria The acceptance criteria for the analysis of the diaphr agm floor seal is as follows:

a. Achieving a positive margin of safety on critical elastic buckling of the seal when considering the maximum conv ex pressure on the seal due to hydrodynamic loads;

C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.4.1-19

b. Stress based on elastic analysis of the seal is not to exceed the following:

1. Average membrane stress intensity is not to exceed the allowable values defined in ASME Code Section III 1971 through Summer 1972 Addenda Paragraph NE 3320;

c. The cumulative usage factor as defined in ASME Code Section III, 1971 through Summer 1972 Addenda, Paragraph NB 3222.4(e) is not to exceed unity. 3A.4.1.5.4 Method of Analysis

The ANSYS (see A.F) finite element model of the diaphragm floor seal consists of a 5.2° segment of the omega shaped configuration and the weldolet welding fitting for the drain pipe. Refer to Figure 3A.4.1-10. Unit differential pressure, u n it displacements in the

vertical and radially horizontal directions to represent the differential displacement of the primary containment vessel and diaphragm floor, unit piping loads at the weldolet, and a linearized thermal gradient are specified load steps in the analys is. Stresses ar e calculated by applying scaling factors to the unit load analyses and superimposing the results. Note that circumferential differential displacement of the primary containment vessel and the diaphragm floor in the horizontal plane is prevented by sh ear lugs furnished along the outer periphery of the diaphragm floor.

3A.4.1.5.5 Results and Design Margins

The differential displacements, differential pressures, and th e piping loads for the critical loading combinations, as tabulated in Section 3A.4.1.5.2 , were used to perform the stress analysis. The design margin on the elastic buc kling of the omega se al as described in Section 3A.4.1.5.3.a is 34.7. The calculated stress inte nsity and fatigue values are presented in Tables 3A.4.1-4 and 3A.4.1-5 respectively.

3A.4.1.6 References

3A.4.1-1 "SRV Loads - Improved Definition and Application Met hodology for Mark II Containments," technical report, Burns and Roe, Inc. Transmitted to NRC by letter GO2-80-172 date d August 8, 1980.

3A.4.1-2 Letter GO2-82-35, "Responses to CSB Open Items 44 through 48,"

G. D. Bouchey (WPPSS) to A. Sc hwencer (NRC), January 13, 1982.

3A.4.1-3 Gosh, S. and Wilson, E., "Dynamic St ress Analysis of Axisymmetric Structures Under Arbitrary Loading," University of California at Berkeley, EEEC, 69-September 10, 1969.

C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.4.1-20 3A.4.1-4 "Chugging Loads - Revised Definition and Appl ication Methodology for Mark II Containments (Based on 4TCO Test Results)," technical report, Burns and Roe, Inc. Transmitted to NRC by letter GO2-81-189 dated July 22, 1981.

3A.4.1-5 ACI Standard 318-71/

77, Building Code Requirement s for Reinforced Concrete, American Concrete Institute.

C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.4.1-21 Table 3A.4.1-1 Basemat -

S tress R e sul t a n ts at Crit i cal S ec t ions Bending Moment (kips per ft)

Beam Shear (kips per ft)

Punching Shear (kips per ft)

D+L+E SS 3132 125 315 P SR 511 32 32 P B 76 1 20 Comb. 7 3649 157 353 Capacity 4230 232 465 C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.4.1-22 Table 3A.4.1-2 Pedestal - S t ress Resultants at B a se Load Overturning Moment - M (ft - kips)

Base Shear - H (kips)

E O 123,600 1,530 E SS 212,300 2,665 P SR 2,825 719 P B 5,088 29 (Comb. 7) 212,380 2,760 Capacity 375,000 14,500 C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.4.1-23 Table 3A.4.1-3 Equivalent Stress Cycles for Fatigue Evaluation Load Number of Events Number of Equivalent Stress Cycles per Event Total Number of Stress Cyc l es Operating basis earthquake 5 10 50 Safe shutdown ear thquake 1 10 10 SRV a 4,478 3 13,434 Chugging 1 1,000 1,000 a This incl u d es the cyc l es due to building motion, direct press u re, and fluid transients during SRV actuations.

C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.4.1-24 Table 3A.4.1-4 Summary of Stress Inte n s ities for Diaphragm Floor Seal Primary Membrane Stress Intensity Primary Membrane Plus Secondary Stress Intensity Loading Condition Table 3A.3.5-5 Calculated Stress Int.

Pm (ksi) ASME Allowable Stress Limit (k si)

Design Margin Calculated Stress Int.

Pm + Q Range (ksi)

ASME Allowable Stress

Limit (ks i) Design Margin/ Remar k s Normal 4.46 Sm 16.56 3.71 26.83 3 Sm 49.68 1.85 Upset 11.84 Sy 25.0 2.11 44.36 3 Sm 49.68 1.12 Emer g e ncy 12.89 1.2 Sm 19.9 1.54 Evaluation not required Faulted 13.82 1.2 Sy 30.0 2.17 Evaluation not required D e sign M a r gin = Al l owab l e S t r e ss C a l c ul a t e d S t r e ss C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.4.1-25 Table 3A.4.1-5 Cumulative Usage Factor Calculati o n for Diaphr a g m Floor S e al Load Combination Set a Expected Number of Cycles (ni)

Alt. Stress Salt (psi)

Allowable Number of Cycles N Ca l c ula t ed Usage Factor Ui =ni Ni 1 1 387.0 33 0.03 2 9 204.0 150 0.06 3 50 178.5 250 0.2 4 60 153.4 350 0.172 5 880 74.0 3700 0.238 6 12,434 18.0

0.0 Cumulative

Usage Factor U = 0.7 < 1 a Load combination set definition 1 - NPC + LOCA + SSE + SRV + CHUGGING

2 - NPC + SSE + SRV + CHUG G I NG 3 - NPC + OBE + SRV + CHU G GING 4 - NPC + SRV+ CHUGGING

5 - SRV + CHUGGING

6 - SRV

Figure Amendment 53 November 1998 Form No. 960690 Draw. No.Rev.Stiffened Containment in Wetwell Region 960222.85 3A.4.1-1 El. 446'3' - 2" 3' - 2" 2' - 10" 3' - 0" 3' - 0" 2' - 10" 1' - 8" Containment Shell Typical Detail Typical Detail Containment

Shell 20" 1 1/2" Horizontal Tee

Stiffener Ring Maximum Water Level. El. 466' - 4 3/4" WT 8x18 Vertical Tee

Stiffener WT 8x18 18" 1 1/2" Columbia Generating Station Final Safety Analysis Report Figure Amendment 53November 1998 Form No. 960690Draw. No.Rev.Displacement ProfileSRV Load - All Valves 960222.86 3A.4.1-20.0.02.04.06.08.10 80 120 160 200 240 280 320 360 400 440 480 520 560 600 640 680 0 Distance from Base (Inches)

Radial Displacement (Inches) 40 El. 446.0' Columbia Generating StationFinal Safety Analysis Report Figure Amendment 53November 1998 Form No. 960690Draw. No.Rev.Displacement Profile Nearly Symmetric Chugging 960222.87 3A.4.1-30.0.02.04.06.08.10 80 120 160 200 240 280 320 360 400 440 480 520 560 600 640 680 0 Distance from Base (Inches)

Radial Displacement (Inches) 40 El. 446.0' Columbia Generating StationFinal Safety Analysis Report Amendment 53November 1998Stiffener Configuration 970187.22 3A.4.1-4 Figure Form No. 960690Draw. No.Rev.Vertical Tee Stiffener Containment ShellHorizontal TeeStiffener Ring Columbia Generating StationFinal Safety Analysis Report Figure Not Available For Public Viewing Figure Not Available For Public Viewing

Draw. No.Rev.Figure Figure Amendment 53 November 1998 Form No. 960690 Draw. No.Rev.Pedestal-Interaction Diagram Axial Load Versus Moment 960222.90 3A.4.1-8 36 33 30 27 24 21 18 15 12 9 6 3 0051015202530354045505560 Mu: Bending Moment in Foot-KIPS x 10 4 Pu = 30,000 k Mu = 513,000 Pu: Axial load in KIPS x 10 3 ft-k P'u = 12380 k M'u = 212,380 ft-k Pu = 12380 k Mu = 375,000 ft-k Ao 0.1Po = 19400 k.9 Ø .7 P'u = Minimum

= M/P = Constant

Ø = .7 Columbia Generating Station Final Safety Analysis Report Figure Amendment 53November 1998 Form No. 960690Draw. No.Rev.Drywell Diaphragm Floor Seal 960222.88 3A.4.1-9 Continuous P L 3' - 0 3/4" 2' - 0" Female Shear Lug Male Shear Lug ContainmentVessel Shell El. 493' - 5" R = 41' - 6" 2" Drain Diaphragm Floor Seal W 36 Radial Beam El. 499' - 6" R = 40' - 5 9/16" Cover P Jet Deflector P Curb P L L L Columbia Generating StationFinal Safety Analysis Report Amendment 55 May 2001Finite Element Model Diaphragm Floor Seal 990578.01 3A.4.1-10 Figure Form No. 960690Draw. No.Rev.Weldolet Welding Fitting Diaphragm Floor Seal Columbia Generating StationFinal Safety Analysis Report C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.4.2-1 3A.4.2 SUPPRESSION POOL MAJO R STRUCTURES AND COMPONENTS Assessment of the capac ities of the major structures and components of the suppression pool chamber relative to load combinations involving suppression pool hydrodynamic loads is made in this section. The structures consid ered are downcomer bracing systems columns, downcomers, SRV piping system , quenchers, and platforms.

3A.4.2.1 Downcomer Bracing System An assessment was made of the capacity of the original design of the downcomer bracing system relative to load combinations involving suppression pool hydrodyn amic loads. It was determined that this original bracing, consisting of a system of radial beams, had inadequate capacity. Consequently, a replacement pipe truss bracing system was designed and installed. The assessment of the capacity of this pipe truss system relati ve to the load combinations involving suppression pool hydrodynamic loads is made in this section.

3A.4.2.1.1 Description of System

The pipe truss system of downcomer bracing is s hown in Figure 3A.4.2-

1. Like the original system, the function served by th e pipe truss system is to pr ovide horizontal support for the 102 downcomers, (three of which are capped as discusse d in Sections 3A.3.2.1 and 3A.3.2.2) and the 18 SRV discharge pipe s at a level near the lowe r end of the downcomers.

The pipe truss system consists of a horizontal planar truss loca ted with center line at the same elevation as the original system. The model of the truss used in the structural analysis is shown in F i gure 3A.4.2-2. In the t r uss sys t em, the downcomers and the SRV lines are located at the truss nodes. Structural rings are pr ovided around each downcomer and each SRV pipe for connections by the truss members. The truss members are 4-in. and 6-in. double extra strong steel pipes. Connections of the truss to the RPV pedestal and to the containment vessel are at the same connection points as the original radial beams.

As described in Section 3A.4.2.1.2, the pipe truss system is subjected to both horizontal and vertical loads. Horizontal re actions from the down comers and from the SRV pipes are applied to the encircling structural rings which form the truss nodes. Horiz ontal forces applied directly to the truss members are also carried by the members to the truss node

s. By truss actions these horizontal loads are transmitted to the supports at the RPV pedestal and at the containment vessel. The pede stal connection can sustain bot h radial and circumferential reaction components due to horizon tal loading; however, the vessel reaction is circumferential because the connection is free to move radially.

Vertical loadings, due to the various causes listed in Section 3A.4.2.1.2, act directly on the pipe truss system. To carry these vertical loadings, supports are provided against upward and downward motion at each of the downcomers; also the connections to th e pedestal and vessel C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.4.2-2 are restrained vertically. Vertical forces acting on each truss member are carried to its ends at the structural rings, pedestal, or vessel connections. The structural rings around the downcomers are independent of the downcomers but stops are welded to the downcomers to prevent differential vertical motion. The structural rings around the SRV lines are independent of these lines and no restraint against differential vertical motion is provided. Vertical loads from the rings on the downcomers are transmitted by the downcomers to the drywell floor.

3A.4.2.1.2 Loads Used for Assessment A complete description of all hydrodynamic loads is given in Section 3A.3. In this section only the loads used for the assessment of the bracing system are discussed. Symbols used in this section are defined in Section 3A.3.5.3 in connection with load combinations.

3A.4.2.1.2.1 Safety/Relief Valve Actuation Loads. Safety/relief valv e actuation causes horizontal and vertical loading on the bracing system as unbalanced pressures and induced accelerations of supported components occur.

The pressures and accelerations acting on the downcomers, the SRV pipe lines , and the bracing members cause the horizontal loading; the vertical loading is due to these actions on the bracing system alone. The spatial distribution of these loads is discussed under Methods of Analysis (Section 3A.4.2.1.4

).

Loads due to the SRV pressures and induced accelerations are a pplied to the bracing system as equivalent static loads. Th e magnitudes of the loadings fr om the downcomers and SRV lines are based on analyses of each of these components as described in the assessments of the components in Sections 3A.4.2.3 and 3A.4.2.4 respectively. The pr essure loadings on the bracing members proper are equivalent stat ic pressures as defined in Section 3A.3. Reactions due to the pressures on the downcomers, SRV line s, and truss members are applied at the truss nodes.

The forces due to the induced accelerations of the downcomers, SRV lines, and bracing members are obtained by analysis of these struct ures using the response spectra developed for SRV actuation. These forces from downcomers, SRV lines, and bracing members are also applied as reactions at the truss nodes.

3A.4.2.1.2.2 Loss-of-C oolant Accident Loads.

Loss-of-coolant accidents are characterized by several phenomena causing nonconcurrent loadings on the bracing system. The principal nonconcurrent loadings are shor t-term pool swell pressure a nd fallback pressure loads and long-term chugging loads. Ot her significant loads are short-term jet and bubble loads and long-term condensation os cillation loads. Pool swell a nd fallback are basi cally vertical motions and generally applied vertical pressure s are associated with each. The chugging loads result from lateral forces at the downcomer exits, from horizontal and vertical accelerations induced by building motion, and from chuggi ng pressures on the wetwell components supported by the bracing system. The horizontal accelerations of the downcomers, SRV lines, C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.4.2-3 and bracing members contribute to the horizontal loading of the bracing, whereas the vertical acceleration of only the bracing members cau ses vertical loading of the bracing.

a. Pressure due to pool swell The pool swell pressures acting on the br acing system are determined to be 25 psi upward and 5 psi horizontal including the DLF, applied concurrently;

b. Pressure due to fallback

Fallback pressures acting on the bracing system are determined to be 25 psi downward and 5 psi horizontal including the DLF, applied concurrently;

c. Lateral forces at downcomer exits

The characteristics of these late ral exit loads are described in Section 3A.3.2.4.4. As noted therein, the late ral exit loads are dynamic loads whose amplitude depends on the size of the downcomer and the number of concurrently loaded downcomers. Fo r an assumed number of concurrently

loaded downcomers, the axial forces in the members of the bracing system and the reactions at the supports are determined by dynamic analysis of the system.

Determination of the critical number of downcomer exits which are subjected simultaneously to lateral force is desc ribed subsequently under Methods of Analysis (Section 3A.4.2.1.4

); d. Induced accelerations due to chugging

The forces due to the induced accelerations of the downcomers, SRV lines, and bracing members are obtained by analyses of these structures using the response spectra developed for chuggi ng action. Forces from these components are applied on the bracing system as reactions at truss nodes;

e. Chugging pressures Chugging pressures are applied on the me mbers of the bracing system and on the inner row SRV lines as described in Section 3A.3.2.4.3.2. Equivalent static loads constituting the end reactions due to the pressures on th ese components are applied at the truss nodes; and

f. Other LOCA loads

The other LOCA loads mentioned above ar e now considered. It is noted that the short-term loads due to LOCA jet and bubble are smaller in magnitude than C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.4.2-4 the previously considered SRV pressure loads which cannot occur simultaneously with them. Hence, these short-term lo ads are not controlling.

Similarly, it is determined that the l ong-term condensation os cillation loads are smaller than loads due to chugging whic h occur subsequently. Therefore, the condensation oscillation loads also are not controlling.

3A.4.2.1.2.3 Othe r Significant Loads. Seismic forces represent a significant loading on the bracing system. The forces due to the seismi c accelerations of the downcomers, SRV lines, and bracing members are obtained by analysis of these structur es using the response spectra developed for OBE and SSE. The forces from downcomers, SRV lines, and bracing members are applied as reactions at the truss nodes.

Dead load of the bracing system and thermal loads are also included in the assessment. Thermal loads result from temperature change of the bracing system and from reactions on the bracing system from supported piping.

3A.4.2.1.3 Controlling Load Combinations and Acceptance Criteria

The load combinations and acceptance criteria per tinent to the downcomer bracing are those listed in Section 3A.3.5.3. Based on the results of the analys is, it is determined that, for the loading conditions involved, the controlling combin ations with associated allowable stresses are those listed below. Only the significant load terms are included.

Service load conditions:

(1) S D + P SR Factored load conditions:

(5) 1.6 S D+T A+E O+P B+P SR (7) 1.7 S D+T A+E SS+P B+P SR Further description as to the loads included in the design loading conditions is presented under Method of Analysis.

3A.4.2.1.4 Method of Analysis

Structurally, the pipe truss downcomer bracing system is treated as a plane truss with respect to horizontal loads and as an assembly of beams with respect to vertical loads. The horizontal planar pipe truss is supported externally at 17 equally spaced points around the pedestal and at

an equal number of points at the containment vessel. Prior to the truss analysis, the reactions due to the distributed loads along the downcomers and along the SRV lines and bracing C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.4.2-5 members are calculated and these reactions are applied di rectly at the tr uss nodes. The analysis methods, design loading conditions, a nd principal results are described below.

3A.4.2.1.4.1 Analysis for Horizontal Loads.

a. The analysis uses the proprietary computer program "Mc Auto STRUDL".

b. The structu r al model u s ed in the ana l ysis is shown in F i gure 3A.4.2-2. The model represents the actual configurati on of the north half of the symmetrical structure except for adjustments along the omitted half structure.

c. The truss and its connec tions are treated as pin c onnected so that the truss members carry on ly axial load.

d. All loads are taken to act at the nodes in the truss analysis. The effect of truss member bending under distri buted normal loads is considered in member design where combined axial load and member bending is provided for.

e. The STRUDL analysis furnishes truss member axial loads and reaction components as well as displacements of nodes.

3A.4.2.1.4.2 Analysis for Vertical Loads.

a. Vertical loads on the bracing system, i.e., on bracing members and on rings, are transmitted through bending and shear to supports on the downcomers and at the pedestal and vessel. Combined axial load and member bending is provided for in the design.

b. Downcomers and diaphragm floor are i nvestigated for the resultant vertical loads. 3A.4.2.1.4.3 Design Load Conditions. The spatial distribution, direction, and magnitudes of possibly coincident loads as described below ar e used in the controlling load combinations; two critical loading conditions are adopted as a conservative basis of design. By these loading conditions it is intended to maximize stresses and reactions in specific portions of the half-structure. These critical stresses and reactions are then utilized to design all similar members and supports around the entire structure.

The two conditions utili zed in the design are noted below.

a. Condition 1: Maximum Horizontal Loading

This loading is intended to result in greatest stresses in the members and supports in the vicinity of the north-south and east-west axes.

The directions of C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.4.2-6 building motions and applied pressures and loads are selected so as to maximize overall loading in the south to north and west to east directions. (Refer to truss layout in Figure 3A.4.2-2

.) Associated vert ical loads are also included.

1. SRV actuation

a) Building motions due to all valve actuation and due to single valve actuation are considered. The building motions cause reactions at all downcomers and SRV lines and accelerations of all truss members. Radial forces and accelerations are directed outward; circumferential forces and accelerations are clockwise. Reactions of downcomers, SRV lines, and members act on the

truss nodes.

b) Unbalanced pressures due to a ll valve and single valve actuation are included. Varying pressures depending on distance from the active nodes are applied to a ll downcomers, SRV lines, and bracing members. The resulting r eactions from these components act at the truss nodes. The direction of reactions from downcomers and SRV lines is generally along the line from the

active node to the structure.

For the bracing member, the pressures and reactions are normal to the bracing member.

c) Vertical loads on the pipe tru ss are due to dead load, building motion induced acceleration, and unbalanced pressure.

2. Long-term LOCA loads (chugging)

a) In the analysis, the number of downcomers subjected to lateral exit loads is varied from one to all 51 downcomers in the model structure. Two directions of load application are considered:

south to north and west to east. It is determined that controlling effects are always due to loading of a single vent. In this regard it is noted that the lateral load definition in Section 3A.3.2.4.4 provides a high intensity load for single vent loading as compared to multiple vent loading.

b) Building motion due to chugging causes reactions at all downcomers and SRV lines and accelerations of all truss members. Radial forces and accelerations are directed outward; circumferential forces and accelerations are clockwise. Reactions of the downcomers, SRV lines, and truss members act on the truss nodes.

C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.4.2-7 c) Chugging pressures on the interior SRV lines result in radially outward reactions applied at the truss nodes.

d) Vertical loads on bracing members are due to building motion induced accelerations and chuggi ng pressure. In addition, inclination of bracing members during LOCA caused by

downcomer temperature growth resu lts in vertical component of member axial force as an additional reaction on the downcomer.

3. Seismic

a) Eastward, northward, and vertical seismic actions are included. Both OBE and SSE are considered.

b) In west-east seismic action, all downcomers and SRV lines react against the bracing towards the ea st. All bracing members have accelerations eastward. Corresponding nort hward effects result from south-north seismic action.

c) Vertical loads on bracing members are due to vertical accelerations;

b. Condition 2: Maximum Vertical Loading

The vertical loads asso ciated with Condition 1 which involve controlling horizontal loads have been previously stated. To determine the maximum vertical loading, short-term LOCA events are investigated. Loads included under Condition 2 are described below.

1. SRV actuation

The acceleration and pressure loads due to single valve actuation are included.

2. Pool swell or fallback

The vertical statically equivalent pressures due to pool swell or fallback are 25 psi on pipe truss members and on truss rings at downcomers and

SRV lines.

C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.4.2-8 3. Seismic Horizontal and vertical loadings are the same as de scribed above for Condition l.

4. Additional DBA effects

Inclination of bracing members during DBA caused by downcomer temperature growth results in a ve rtical component of member axial force as an additional reaction on the downcomers. Also, building motion results in vertical acceleration of the bracing members.

However, this acceleration ha s relatively small magnitude.

5. Coincident me mber axial loads

Such axial loads are caused by single SRV actuation and by seismic loading.

3A.4.2.1.5 Results and Design Margin

The principal results of the an alysis and the resu ltant design margins are stated below.

3A.4.2.1.5.1 Pr incipal Results.

a. Bracing members These members are designed for combin ed stress involving axial force and biaxial bending. Members connecting to the pedestal are 6-in. double extra strong steel pipe; all othe r members are 4-in. double extra strong steel pipe;

b. Node rings

These rings are designed for a combination of radial loads which is conservative based on the design loading conditions.

As a result of this analysis, built-up H-sections 5 in. wide by 7 in. high at the downcomers and 4 in. wide by 6 in.

high at the SRV lines are used. Materi al is high strength structural steel (ASTM A 572, Fy = 60 k.s.i.);

C OLUMBIA G ENERATING S TATION Amendment 59 F INAL S AFETY A NALYSIS R EPORT December 2007 LDCN-06-000 3A.4.2-9 c. Pedestal connections

The existing embedded plates for the origin al radial beam system are utilized in the pipe truss system.

Additional strengthening wa s installed based on the design loading conditions. Six concrete anchors were added at each pedestal

connection; and

d. Containment vessel connection The pipe truss end bearing at the containm ent vessel fits into the existing socket provided for the original radial beam

s. The containment vessel redesign includes the reactions due to the down comer bracing design loading conditions.

3A.4.2.1.5.2 Design Margins. The design margins provided by the pipe truss bracing system are discussed in this section. The design margin for a structural component or system refers to the controlling design parameter su ch as bending stress for a fle xural member and sum of the amplified stress ratios for the cas e of combined axial and bending stress. The design margin is then defined as the ratio of the permissible value of the design parameter to the value of the design parameter under design loading.

Table 3A.4.2-1 lists the controlling desi gn margins for each of the six principal structural components of the pipe truss bracing system, namely, 6-in. pipe me mbers, 4-in. pipe members, pedestal connection, vessel connection, downcomer ring, a nd SRV line ring. The calculated design margins are listed in the table for the cases of maximum horizontal or vertical loading (Conditions 1 or 2), under both service load and factored load conditions.

3A.4.2.2 Columns

The assessment of the capacity of the columns relative to load combinations involving suppression pool hydrodynamic loads is presented in this section.

The columns and adjoining structures are shown in Figure 3A.4.2-3.

3A.4.2.2.1 Loads Used for Assessment

A complete description of all hydrodynamic loads is given in Section 3A.3. Seismic loads are described in Section 3A.3.5. In this section, only the loads used for the assessment of the columns are discussed. Symbols used in this section are defined in Section 3A.3.5.2 in connection with load combinations.

3A.4.2.2.1.1 Safety/Relief Valve Actuation Loads.

Actuation of the SRVs results in four different load effects on the columns. These are unbalanced bubble pressure on the

columns, column accelerations associated with resultant build ing motions, quencher discharge water jet loads, and secondary effects from pressure loading of the downcomer bracing truss.

C OLUMBIA G ENERATING S TATION Amendment 54 F INAL S AFETY A NALYSIS R EPORT April 2000 LDCN-99-000 3A.4.2-10

Maximum bubble pressures applie d laterally on the column are defined for two cases, namely, initial actuation and subsequent actuation. The spatial distribu tions of the maximum pressure loading on the column corresponding to these two cases are shown in Figure 3A.3.1-8. The DLFs associated with these maximum pressure loadings are functions of the column modal frequencies as shown in Figure 3A.3.1-10. The maximum dynamic co lumn reactor loads for the subsequent actuation load case are documented by Reference 3A.3.1-9 and are tabulated in Table 3A.4.2-2. The structural model used for th e dynamic analysis of the column under lateral load is described in Section 3A.4.2.2.3 and shown in Figure 3A.4.2-4.

The horizontal and vertical accelerations due to building mo tion are based on the response spectra developed for SRV actuation in Section 3A.5.1. The horizontal (lateral) accelerations are used in the dynamic analysis of the co lumn by the response spectrum method in conjunction with the structural model described in Section 3A.4.2.2.3 and shown in Figure 3A.4.2-4. The axial (vertical) forces in the column are obtained from the vertical response spectra by a dynamic analysis by the re sponse spectrum method; a structural model, which includes both the diaphragm floor beam and the column, is used and is shown in Figure 3A.4.2-5.

The quencher jet drag load on the column is sm all in comparison with the bubble load. Since the two loadings do not occur simu ltaneously, the jet loading is not a controlling loading. The direct vertical pressure loading on the downcomer bracing members is transmitted into the diaphragm floor by the downcomers. The diaphr agm floor in turn loads the column. The net loading on the column during the phase of maxi mum bubble pressure on the column is small and is included in the assessment.

3A.4.2.2.1.2 Loss-of-C oolant Accident Loads. Both short-term and long-term LOCA loads are significant in the assessment of the columns.

Several axial loads on the columns due to shor t-term LOCA events ar e considered in the analysis. The overall pressure transient in the drywell and wetwell results in a net downward pressure (20.55 psi equivalent static) on the diaphragm floor a nd hence an axial compressive column load. The design net upward pressure (5.50 psi) on the diaphragm floor during the

pool swell transient causes an upw ard load on the columns. Duri ng LOCA, vertical pressures act on the downcomer bracing system due to bubble charging and due to pool swell and fallback. The resultant vertical forces are transmitted via th e downcomers and the diaphragm floor into the columns. Howe ver, these forces are not cont rolling in comparison with the aforementioned net downward and upward pressu res on the diaphragm floor. In addition to the pressure loads, the load on the column, due to pipe break/jet impingement on the diaphragm floor, is incl uded in the assessment.

The significant lateral loads associated w ith LOCA, which are included in the column assessment, are those due to c hugging. Lateral loads due to water clearing and due to air C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.4.2-11 clearing are negligible. The load s due to condensation oscillati on are less critic al than those due to chugging, and co nsequently, are omitted from the assessment. The effects of chugging include both direct horizontal pressures on the columns and induced horizontal and vertical accelerations associated w ith building motions, but only the effects due to building accelerations are significant.

The horizontal and vertical ac celerations due to chugging ar e based on the response spectra described in Section 3A.5.2. Column loading is then developed by dynamic analysis of the column by the response spectrum method in th e manner described a bove for SRV induced building motion. The analysis for horizontal loading uses the structural model of Figure 3A.4.2-4 , and the analy s is f o r vertical loading uses the structural model of Figure 3A.4.2-5.

3A.4.2.2.1.3 Othe r Significant Loads. Dead load, live load, seis mic loads, and loads due to annulus pressurization constitute additional significant loads incl uded in the column analysis.

The dead load of the diaphragm floor and columns is carried as an axial column load. The live load on the diaphragm floor is transmitted to the columns as an axial load. Horizontal and vertical column loadings are obt ained from the seismic response spectra in the same manner as described above for the SRV and LOCA response spectra.

Pressurization of the annulus between the RPV and the sacrificial sh ield wall, due to a postulated break in either the circulation line or the feedwater line, as described in 6.2.1.2 of the FSAR, results in building mo tions with associated respons e spectra. Horizontal and vertical column loadings are obtained from the response spectra in the same manner as described above for the SRV and LOCA response spectra.

3A.4.2.2.2 Applicable Load Combinations and Acceptance Criteria

The load combinations and accep tance criteria for internal reinforced-concrete structures described in Section 3A.3.5.2 are applicable to the columns.

3A.4.2.2.3 Method of Analysis

The structural capacity of the column is investig ated for the applicable load combinations with loads as listed above. The general approach in the column assessment is to determine the values of the controlling stress resultants in the co lumn (bending moment, axial force, and shear) on the basis of elastic analysis under desi gn loads and to calculate the capacity of the column in terms of these stress resultants by the strength method of the ACI 318-71 Code (Reference 3A.4.1-5). Three loading cases are investigated. Because of the shape of the column interaction curve for bending and axia l load, maximum bending in the column is checked first with minimum coincident axial load and then with maximum coincident axial load. The third loading cas e involves maximum shears.

C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.4.2-12 The moments, shears, and axial forces are obtained for each of the significant loads. The column is analyzed dynamically via a modal time-history technique (Reference 3A.3.1-9) for the directly applied SRV pressure s and for the loads due to the building motion accelerations as defined by the SRV chugging, seismic, and annulus pressurization response spectra; dynamic analysis is by the response spec trum method. In the analysis, th e column is treated as fixed at its base and simply supported at its top in accordance with actual construction. For the dynamic analysis under lateral (horizontal) loading, the actual struct ure is modeled as a 17-node beam as shown in Figure 3A.4.2-4. Ma sses are l u mped at t h e nodes with additional mass provided below pool level to account for the effect of th e suppression pool. The axial (vertical) column forces are determined by the dynamic analysis using a structural model which represents t h e diaphragm floor beam - c o l u mn system. As shown in Figure 3A.4.2-5 ,

the diaphragm floor mass is included in 26 n odes along the beam and the column mass is 14 nodes along the column. Dynami c analysis for lateral loads a nd for vertical loads are made using the commercial computer pr ograms STRUDL and IMAGES-3D.

The critical locations along the column for bending moment and shear are determined from the analysis results. Maximum be nding moments and shear occur at the column base. The magnitude of the horizontal shear at the top of the column is also noted to verify the adequacy of the top connection. The values of the controlling bending moments, shears, and axial force are listed for each significant loading in Table 3A.4.2-2. a. Maximum bending moment with minimum axial load

It is determined that the controlling values of maximum bending moment with

coincident minimum axial load occur for load combination (1). This combination is stated below for minimu m axial load with only the significant load terms listed.

(1) 1.0 D + 1.5 P SR The maximum bending moment is caused by the direct lateral pressure on the column, due to SRV subsequent actuation, and the associated horizontal building motion acceleration. The coincident minimum compression axial load occurs with dead load (D) and upward load due to the vertical building motion acceleration caused by the SRV actuation (P SR). The controlling values of bending moment and axial lo ad are 17.67 ft kips and 173.3 kips (compression), respectively.

The column capacity for co mbined loading and axial load is determined in accordance with the strength desi gn method of the ACI 318-71 Code (Reference 3A.4.1-5). From the applicable in teraction curves, the bending moment capacity of the co lumn, with the above axia l load of 173.3 kips, is found to be 2182.0 ft kips; C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.4.2-13

b. Maximum bending moment with maximum axial load The controlling coincident values of bending moment and axial load occur for load combination (1). This combination is stated below for maximum axial load with only the significant load terms listed.

(1) 1.4 D + 1.7 L + 1.5 P SR The maximum bending moment is caused by SRV subs equent actuation as in (a) above. The coincident maximum compre ssive axial load is due to dead load (D), live load (L), and downwa rd load due to the SRV actuated building motion (P SR). The controlling value of the axial load is 745.2 kips. As a result of moment magnification due to the axial lo ad, the controlling va lue of the bending moment is increased from 1700 ft kips to 1869 ft kips. Utilizing the column interaction curves, the bending moment cap acity is found to be 2230.5 ft kips;

c. Maximum shear As noted previously, maximum column shear occurs at th e column base. Column shear capacity is affected by the axial force at the section. The controlling load combination for column sh ear is load combination (1) which is stated below for minimum axial load with only the significant loads listed.

(1) 1.0 D + 1.5 P SR The maximum shear is caused by the same event as in (a) above, namely, SRV subsequent actuation (P SR). The coincident minimum compressive axial load occurs with dead load (D) and SRV actuation (P SR). The controlling values of shear and axial load are 142.

4 kips and 173.3 kips (compression), respectively.

Utilizing the strength design met hod of the ACI 318-71 Code (Reference 3A.4.1-5), the column shear capacity is calculated to be 376.9 kips; and

d. Column top shear connection

The controlling top horizontal reaction in relation to the capacity of the top connection in shear occurs with load combination (1) with minimum axial (compressive) load. This combination is stated below.

(1) 1.0 D + 1.5 P SR C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.4.2-14 The top horizontal reaction is caused in this case by SRV initial actuation (P SR); its magnitude is 62.1 kips.

The shear capacity of the top connection is due to shear friction associated with the connecting anchor bolts and the supe rimposed axial load. Conservatively, the superimposed load is due to dead load (D) reduced by the upward load due to SRV actuation (P SR). Utilizing the strength de sign method of the ACI 318-71 Code (Reference 3A.4.1-5), the shear capacity of the top connection is 75.5 kips.

3A.4.2.2.4 Results and Design Margins

It is determined that the column s have adequate capacity with regard to the applicable load combinations involving suppression pool hydrodynamic loads.

The design margins with respect to the significant stress resultants are noted below:

a. Maximum bending moment with minimum axial load - The smallest design margin representing the ratio of colu mn bending capacity to applied bending moment, with minimum axial load, is 1.54;

b. Maximum bending moment with maximum axial load - The smallest design margin representing the ratio of colu mn bending capacity to applied bending moment, with maximum axial load, is 1.49;

c. Maximum column shear - The smallest design margin representing the ratio of the column shear capacity to the applied shear is 3.23; and

d. Column top shear connec tion - The smallest design ma rgin representing the ratio of the capacity of the column top connecti on in shear to the applied top shear is 1.18. 3A.4.2.3 Downcomers

The primary function of the downcomer vent system is to channel the steam accumulating in the drywell chamber during a LOCA into the wetwell chamber to accomplish pressure suppression (see Section 3A.3.2.1).

The downcomer vent system consists of ei ghty-three 24-in. OD and sixteen 28-in. OD standard schedule carbon steel pipes running vertically downward from the diaphragm floor (except that the ends of th e downcomers are stainless steel as described in FSAR Section 3.8.3.4). Originally 102 downcomers were provided, but three (one 24-in. and two 28-in.) were capped, as discussed in Sections 3A.3.2.1 and 3A.3.2.2. The downcomers are embedded in the diaphragm floor and extend down to el. 454 ft 4.75 in. All downcomers C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.4.2-15 are restrained laterally at el. 455 ft 4 in. by the downcomer bracing system , which is vertically restrained by the downcomers. Vertical lo ads are imposed by the bracing system onto the downcomers and transmitted to the diaphragm f l oor. (See Figures 3 A.2.1-5 and 3A.2.1-6.)

Nine of the 24-in. OD downcomers have an ex tra strong welding tee at el. 491 ft 11 in. to accommodate 24-in. dual inline vacuum breaker valves. In addition, to provide extra strength, the eighteen 28-in. downcomers have been stiffe ned by the insertion of a 4-ft 8-in. long by 2-in. thick spool piece. This piece accommodates the penetration for the main steam SRV (MSRV) piping which is welded to these downcomers at el. 493 ft 0 in.

Figure 3A.2.1-4 shows locations where the v a cuum breaker valve a s se m b lies and the MSRV piping penetrate the downcomers just below the diaphragm floor.

3A.4.2.3.1 Loads Used for Assessment

The downcomer piping is subjected to stat ic, dynamic, and hydrodynamic loads under the various plant operating c onditions identified as normal, upset, emergency, and faulted. Each of these loads in vari ous combinations is identified in Section 3A.3.5.4.

The individual loads acting on the downcomers are identified below:

a. Deadweight (W)

b. Thermal expansion and thermal transient

c. Pressure (P)

The pressure differentia l between the drywell and suppression chamber atmospheres produces loads on the downcomer walls since it acts as a pressure retaining boundary during a LOCA.

d. OBE

e. SSE

f. SRV discharge dynamic loads The spatial distribution of the maximum direct bubble pressure loading used for downcomer assessment was obtained by multiplying a dynamic pressure load (Figure 3A.3.1-7) by the maximum DLF.

A maximum DLF of 4.2 was conserva tively obtained from the response spectrum of DLFs shown in Figure 3A.3.1-10.

C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.4.2-16 The inertia loading effects due to the acceleration of the structure are described

in Section 3A.5.1. The response spectra used for downcomer assessment are the enveloped spectra whic h were developed by enveloping the spectra due to four SRV actuation cases at the appropriate locations for the downcomers.

g. LOCA loads

The loads on the downcomer associated with LOCA are chugging pressure, condensation oscillation pr essure, and the building response loading during LOCA event.

The spatial distribution of condensat ion oscillation and chugging pressure loadings on downcomers are considered as equivalent static pressure loads.

The pressure distribution for the condensation oscillation on downcomers is bounded by chugging load and ther efore is not a controlling load.

The LOCA jet, LOCA bubble, pool swell, and fallback loads are identified to be negligible on downcomers as described in Section 3A.3.2, hence, these loads are not considered in downcomer analysis.

The input response spectra for chugging and condensation osc illation are based on the spectra described in Section 3A.5.2 and are enveloped in the same manner as described in the pr evious section for SRV load.

3A.4.2.3.2 Load Combination and Acceptance Criteria

The resultant stresses experienced by the downcom ers are considered accep table if they satisfy the ASME Boiler and Pressure Code,Section III, Subsection NC (Reference 3A.4.2-1).

The allowable stress "S h" for both the 24-in. and 28-in. OD downcomers is 15,000 psi. This value was obtained from the ta bulated values in Section III, Attachment 3A.I for "S h" at a design temperature of 340°F, for carbon st eel SA 155 KCF 70 a nd SA 106, GR.C.

Allowable Stress Limits (Equation 9 of NC-3652 a nd NC-3611, Reference 3A.4.2-1) The stress includes the primary membrane plus the primary bending stresses. The limits of these stresses depend on the lo ading conditions as follows :

a. The limit of stress under the upset condition is 1.2S h = 18,000 psi, b. The limit of stress under th e emergency condition is 1.8S h = 27,000 psi, and c. The limit of stress under the faulted condition is 2.4S h = 35,000 psi.

C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.4.2-17 3A.4.2.3.3 Method of Analysis

Downcomers were analyzed fo r the appropriate loading combination using the computer program ADLPIPE (Attachment 3A.F

).

The mathematical model for the dow ncomer is a vertical pipe an chored at the underside of the drywell floor and guided at the downcomer bracing system. The inertia effect of water surrounding the submerged portion of the downcomer was obtained by the addition of a virtual mass of water distributed along the submerged portion. The mass of water inside the submerged portion of the downcom ers was conservatively consid ered in the model for all dynamic loadings. The SRV discharge lines were incorporated in the model of the 28-in. downcomer.

3A.4.2.3.3.1 Static Analysis.

Static analysis techniques were used to determine the stresses due to dead weight, internal pressure, thermal and hydrodynami c loads using an equivalent sta t ic pre s sure load, or an appropria t e DLF, as shown in Figure 3A.3.1-10.

3A.4.2.3.3.2 Response Spectrum Analysis.

The response spectrum method of analysis was performed, for seismic and hydrodynamic loads, using the ADLPIPE program. Modal responses were combined in accordance with Regulatory Guid e 1.92 while damping values were selected per Regulatory Guide 1.61.

Spatial components were combined by the SRSS method, with the exception of seismic which used the higher of the absolute sum of (a) north-s outh and vertical or (b) east-west and vertical (see Section 3.7 of the FSAR).

3A.4.2.3.4 Results and Design Margin

The downcomers were analyzed for all lo ad combinations described in Section 3A.3.5.4. The stresses in the 24-in. OD and 23-in. OD downc omers pipe show that they are structurally adequate for all plant operati ng conditions. The design margin s for the 24-in. and 28-in. downcomers in each criteria category are summarized below.

The lowest design margin is shown to be 1.08.

24-in. Downcomer Acceptance

Criteria From Allowable Calculated Design Table 3A.3.5-5 Stress Stress Margin Upset 18,000 15,513 1.16 Emergency 27,000 15,709 1.72 Faulted 36,000 18,810 1.91

C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.4.2-18 28-in. Do w n comer Acceptance

Criteria From Allowable Calculated Design Table 3A.3.5-5 Stress Stress Margin Upset 18,000 16,654 1.08 Emergency 27,000 16,744 1.61 Faults 36,000 18,371 1.96 Design margin is defined as follows:

DM = Allowable S t r e ss C a l c u l ated S t r e ss 3A.4.2.3.5 Fatigue Evaluations

The fatigue evaluation presented below was an NRC request and is not an ASME requirement.

The fatigue evaluation of 24-in. and 28-in. downcomer lines in the wetwell air volume was performed using ASME Section III, Class 1 rules (NB-3600). A governing loading scenario, based on the DFFR (Reference 3A.3.2-1), was developed. The load ings which were evaluated are:

a. Internal pressure,

b. Thermal expansion and transients,

c. Seismic,

d. Pressure differential effects be tween drywell and suppression chamber, e. SRV pool load and building response, and f. Chugging pool load and building response.

Equivalent numbers of fatigue cycles were determined for dynamic lo ads. The 24-in. and 28-in. downcomers were analyzed for the appropriate load combinations and their associated number of cycles as presented in Table 3A.4.1-3. The combined stre sses and corresponding equivalent stress cycles were computed to obtain the fa tigue usage factor. The maximum fatigue usage factor for both downcomers are presented in Tables 3A.4.2-4 and 3A.4.2-5.

3A.4.2.4 Safety/Relief Valve Piping Systems

The MSRV piping in the suppression chamber consists of 18 independent piping systems, each comprised of 10-in. and 19-in. OD Schedule 80 carbon steel pipe. The wetwell portion of each SRV piping system in the wetwell origin ates from a 28-in. downcomer (anchor point),

C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.4.2-19 which penetrates at el.

493 ft 0 in., and then runs horizontally for a sufficient length to provide enough thermal flexibility. The horizontal run also allows the quenchers to be distributed evenly about the suppression pool. The piping then drops vertically downward to the quencher body which is bolted to the que ncher support at el. 447 ft 0 in. The quencher support is modeled as an integral part of the SRV piping system and as such has flexibility taken into account. A schematic showing a typical SRV piping layout is shown in Figure 3A.4.2-6. Lateral guides are provided at the downcomer bracing.

3A.4.2.4.1 Loads Used for Assessment

The SRV piping systems are subj ected to static, dynamic, and hydrodynamic loads due to normal, upset, emergency, and faulted plant operating conditi ons. The loading cases and combinations are described in Section 3A.3.5.4.

The hydrodynamic loads resulting in significant effects on the SRV piping are listed below. For a description of these loads see Section 3A.4.2.3.1. a. Deadweight (w), b. Thermal,

c. Pressure (P),

d. OBE,

e. SSE,

f. SRV pressure (Figure 3A.3.1-9

),

g. SRV building response,

h. SRV blowdown,

i. LOCA jet (Figure 3A.3.2-18 and Table 3A.3.2-8),

j. LOCA bubble (Figure 3A.3.2-18 and Table 3A.3.2-8), k. Chugging d r ag (Figure 3A.3.2-19

), and l. Chugging building response.

The SRV pressure used in the analysis is applied in the same ma nner as described in Section 3A.4.2.3.1 for the downcomers. The building response loads are based on the spectra described in Section 3A.5.

3A.4.2.4.2 Load Combination and Acceptance Criteria

The stresses within the SRV piping are acceptable if they satisfy the ASME Boiler and Pressure Vessel Code, Subsection NC. The allowable stress "S h" used for primary stress evaluation is 15,000 psi. This value was obtained from the tabulated values in Section III, Attachment 3A.I for "S h" at a design temperatur e of 475°F for carbon steel.

C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.4.2-20 3A.4.2.4.3 Allowable Stress Limits (Equation 9 of NC-3652 and NC-3611, Reference 3A.4.2-1) The stress for the SRV piping in cludes the primary membrane plus the primary bending stresses. The limits of thes e stresses depend on the lo ading conditions as follows:

a. The limit of stress under the upset condition is 1.2S h = 18,000 psi, b. The limit of stress under th e emergency condition is 1.8S h = 27,000 psi, and c. The limit of stress under the faulted condition is 2.4S h = 36,000 psi.

3A.4.2.4.4 Method of Analysis

The SRV piping was analyzed for the appropriate loading combinations (Table 3A.3.5-5) using the computer programs ADLPIPE and ANSYS (Attachment 3A.F

). Analysis was performed for all 18 SRV lines. The quenche r towers were included in the piping models to account for quencher tower flexibility. The in ertial effects of water were acc ounted for in the same matter as described in Section 3A.4.2.3.3. Analysis results are summarized in Section 3A.4.2.4.5.

3A.4.2.4.4.1 Static Analysis.

Static analysis techniques were used to determine the stresses due to dead weight, internal pressure, thermal and hydrodynami c loads using an equivalent sta t ic pre s sure load or an appropria t e DLF, as shown in Figure 3A.3.1-10.

3A.4.2.4.4.2 Response Spectrum Analysis.

The ADLPIPE program was utilized to perform dynamic response spectra analyses. Modal re sponses were combined in accordance with Regulatory Guide 1.92 while damping values were selected per Regulatory Guide 1.61.

3A.4.2.4.4.3 Time History Analysis. The ANSYS program was utilized as required for critical SRV lines in order to obtain more r ealistic piping response for SRV building response loads. For SRV blowdown transient loads, ADLPIPE was us ed to perform a force time

history analysis.

3A.4.2.4.5 Results and Design Margin

The calculated stresses for th e design configuration of all SRV piping systems show that the piping is structurally adequate for all plant operating conditions. The maximum calculated stresses and the resulting minimum design margins for the SRV piping systems for the controlling load combinations from Section 3A.3.5.4 are shown below.

C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.4.2-21 Acceptance Criteria from Allowable Calculated Design Table 3A.3.5-5 Stress (psi) Stress (psi)

Margin Upset 18,000 17,206 1.05

Emergency 27,000 17,207 1.57 Faulted 36,000 17,280 2.08

Design margin is defined as follows:

DM = Allowable StressCalculated Stress 3A.4.2.4.6 Fatigue Evaluations

The fatigue evaluation presented below was an NRC request and is not an ASME requirement.

The fatigue evaluation on all 18 SRV lines in the wetwell air volume was performed using

ASME Section III, Class 1 rules (NB-3600). A governing loading scenario, based on the DFFR (Reference 3A.3.2-11), was developed. The loadings which were evaluated are:

a. Internal pressure,

b. Thermal expansion and transients,

c. Seismic,

d. SRV blowdown,

e. SRV pool load and building response, and f. Chugging pool load and building response.

Equivalent numbers of fatigue cycles were determin ed for dynamic loads. All 18 SRV discharge lines in the wetwell region were analyzed for the appr opriate load combinations and their associated number of cycles as presented on Table 3A.4.1-3. The combined stresses and corresponding equivalent stre ss cycles were computed to obta in the fatigue usage factor. The maximum fatigue usage factor for all 18 SRV discharge lines in the wetwell air volume was found to be below ASME allowable limits. Th e results of the maxi mum usage factors is presented on Table 3A.4.2-6.

3A.4.2.5 Quencher

Quenchers have been installed on the discharge end of the SRV lines to reduce air clearing loads and to promote effectiv e heat transfer be tween the suppression pool water and the discharging steam-air mixture durin g SRV actuation. The quenchers are an integral part of the SRV piping system and are bolted to the quencher support at th e base plate. The quencher support assessment is included in the SRV piping system assessment, Section 3A.4.2.4.

C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.4.2-22 3A.4.2.5.1 Loads U s ed for Assess m e nt

The quenchers, in common with the other piping components, are subjected to static, dynamic, and hydrodynamic loads due to normal, upset, emergency, and faulted plant conditions. The loading cases arc described in Section 3A.3.5.4.

The mechanical loads are from Table 3-16 of the DFFR (Reference 3A.4.2-2) and modified to account for CGS plant specific conditions. The load from DFFR Table 3-16, which are not plant specific, are the quencher arm and body loads arising from the SRV water and air clearing transients. These generic boundi ng loads are described in the DFFR (Reference 3A.4.2-2). The other loads on Table 3A.3-16 of the DFFR are modified to account for (a) quencher arm loads caused by the various submerged hydrodynamic loading described in Section 3A.3 , and (b) the static and dynamic loads resulting from the SRV piping system analysis de scribed in Section 3A.4.2.4.

3A.4.2.5.2 Load Combination Acceptance Criteria

The assessment of the quenchers for the plant loads is performed in accordance with ASME Section III, Subsection NC (Reference 3A.4.2-1). The code stamp and hydrotest requirements are not applicable since the quencher is not a pressure retaining component. The code jurisdiction ends at the weld between the SR V discharge piping and th e quencher inlet nozzle (12 in. x 24 in. reducer).

3A.4.2.5.3 Evaluation

The quencher body and the quencher arms are ex amined to determine their adequacy for conditions of loadi ng described above.

The quencher body together with the quencher ar ms were modeled thr ough finite element program ANSYS. The model us es a quadrilateral shell elem ent which has both bending and membrane capabilities. The element has six degrees of freedom at each node. This element also has an option for variable thickness. The modeled structu r e is shown in Figures 3A.4.2-7

and 3A.4.2-8. Element loading capabilities include surface temperatures and pressures. Also, concentrated loads can be applied at each node point.

The significant loads affe cting quencher body and the quencher arms are

a. The loads arising from the SRV water and air clearing transients,

b. The SRV loads caused by pool velocity a nd acceleration fields from an adjacent firing quencher, and C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.4.2-23

c. SRV induced building r e sponse loads.

The quencher assembly was analyzed statically with the load vectors relate d to type of loads as discussed above and fo r load combinations presented in Section 3A.4.2.5.2. Since the quencher has linear properties the superposition of load combinations was used for evaluation of results.

The calculated stress intensities for the various loading conditions are presented in Table 3A.4.2-3. The tabulated design margins for the governing ups et condition indicate the quencher assembly is st ructurally adequate.

3A.4.2.6 Platforms and Ladders

The assessment of the capacities of the platforms and ladder relative to load combinations involving suppression pool hydrodynamic loads is made in this section.

3A.4.2.6.1 Loads Used for Assessment

Complete description of all hydrodynam ic loads has been given in Section 3A.3. The loads used for the assessment of the plat forms and ladder are discussed below.

3A.4.2.6.1.1 Safety/Relief Valve Operation Loads. No direct loads on the platforms at el. 472 ft 4 in. and el. 486 ft 8 in. or the ladder between el. 472 ft 4 in. and 490 ft 1 in. result from operation of the SRV system. However, building response to dynamic pressures at the pool boundary during SRV discha rge result in dynamic stresses in the platform and ladder components which are supported from the steel containment structure.

3A.4.2.6.1.2 Loss-of-C oolant Accident Loads.

As discussed in Section 3A.3.2.3, a LOCA results in pool swell impact and drag loads and fallback drag loads on the platform at el. 472 ft 4 in. and the ladder between el. 472 ft 4 in. a nd el. 484 ft 4.75 in. Also, the ladder and the knee braces below the platform at el. 472 ft 4 in. are subjected to a horizontal lift load caused by both pool swell and fallback.

Additional LOCA loads include building respon ses to condensation os cillation and chugging which are obtained from response spectra at tile points of att achment of the platforms and ladder to the containment vessel. Since the condensation oscillation load is bounded by the chugging load, no separate plat form assessment has been pe rformed for the condensation oscillation loading condition.

3A.4.2.6.1.3 Othe r Significant Loads.

Other loads which result in significant stresses in the platform and ladder components are dead load s, live loads, and se ismic accelerations.

C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.4.2-24 Dead loads include the weights of the platform s and ladder. Live lo ads include personnel on the platforms and ladder, equipment weights on the platforms, and the monorail loads on the platform at el. 486 ft 8 in. Se ismic accelerations are obtained fr om seismic response spectra at the points of attachment of the platforms and ladder to the containment vessel.

3A.4.2.6.2 Controlling Load Combinations

The load combination criteria for structures internal to the suppression chamber (see Section 3A.3.5.3) are applicable to the pl atforms and ladder. In particular, the combinations for steel structures using the elastic working stress method are investigated for both service load conditions and factored load conditions.

3A.4.2.6.3 Acceptance Criteria

The acceptable stress level for the platform and ladder compone nts using the elastic working stress method is as defined for th e loading combinations in Section 3A.3.5.3.

3A.4.2.6.4 Method of Analysis

Both platforms and the ladder were investigated for the effects of the loading combinations in Section 3A.3.5.3. The components of the platform at el. 472 ft 4 in. and the ladder were analyzed individually, wher eas the platform at el. 486 ft 8 in.

is above the effects of pool shell and therefore was not reanaly zed. As noted in Section 3A.3.2.3 , the grating does not sustain impact loads. Therefore, the only significant lo ad on the grating is the drag load component. The supporting beams and bracing members were anal yzed for drag and lif t pool swell/fallback loads. The portion of the ladder between el. 472 ft 4 in. and 404 ft 4.75 in. was analyzed for impact loading on the highest rung in the pool swell region and drag and lift on all rungs below the rung with impact.

3A.4.2.6.5 Results

The governing load combination (5a) for steel structures is given below with corresponding input for the platform at el. 472 ft 4 in. and the portion of the ladder between el. 472 ft 4 in.

and 484 ft 4.75 in.:

(5a) 1.6S 1.0D+1.0L+1.0E o+1.0P A+1.0T A+1.R A+1.0P SR D = 0.10 psi

L = 0

E O = 0.11 psi

C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.4.2-25 P A = 8.4 psi (drag on gross area of grating)

= 25.0 psi (d r a g on ladder rungs and bracing members)

= 44.0 psi (impact on ladder rungs)

= small (horizontal lift pressure on supporting bars and handra i l members)

= 11.0 psi (horizontal lift pressure on bracing members)

= 5.0 psi (horizontal lift pressure on ladder rungs)

= small (impact and drag on su p porting bars and handrail members)

T A = 0 R A = 0 P SR = 0.05 psi where D, L, E o , P A , T A , R A , and P SR are as defi n e d in Secti o n 3A.3.5.2.

The portion of the ladder between el. 472 ft 4 in. and 484 ft 4.75 in. was found to have sufficient capacity to withstand the governing load comb ination with a design margin (i.e., ratio of the allowable st ress to the maximum absolute calculated stress) of 2.15. However, the existing platform at el. 472 ft 4 in. was f ound to be deficient (under the above load combination. The critical com ponent is the grating under upward load. To make the grating sufficient, every other bearing bar (instead of every fourth as is in the original design) has been welded to all supporting members including the platform member supporting the ladder.

With this reinforcement a design margin fo r the grating of 1.80 is attained. Also, two platform members have been reinforced, and the rectangul ar bracing members have been replaced with circular members. With this reinforcement the critical supporting member has a design margin of 1.18.

3A.4.2.7 References

3A.4.2-1 ASME Boiler and Pressure Vessel Code,Section III, Division 1, Subsection NC, "Class 2 Components," American Society of Mechanical

Engineers, 1971 through Winter 1973 Addenda.

• Faulted conditions appeared first in Winter 1976 Addenda.

C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.4.2-26 3A.4.2-2 Mark II Containment Dynamic For c ing Functions Information Report (DFFR), NEDO-21061, Revision 2, September 1976.

C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.4.2-27 Table 3A.4.2-1 Downcomer Bracing System Controlling Design Margins Component Service Load Conditions Maximum Horizontal Loading Factored L o ad Maximum Horizontal Loading Conditions Maximum Vertical Loading 6 in. D.E.S. pipe 3.45 2.96 4 in. D.E.S. pipe 2.5 1.68 Pedestal connection 2.00 2.02 Vessel connection 5.26 2.75 Downcomer ring 1.92 1.95 SRV line ring 3.39 4.12 Notes:

1. For service load conditions, load combination (1) control

s. Only maximum horizontal loading is applicable.

2. For factored load condition s , load combination (5) contr o ls. Only the design margin for the controlling loading is listed.

C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.4.2-28 Table 3A.4.2-2 Controlling Stress R e sultants in Column

Shear (kips)

Axial Force a (kips) Base Moment (ft kips)

Base Top Base Top Safety/Relief Valve Actuation All valves initial 771.1 70.2 41.4 102.1 86.4 1 valve subseque ntial 1133.0 94.9 48.9 37.1 37.1 Loss-of-Coolant Accident Chugging 13.5 2.5 1.4 11.7 10.2 Annulus pressurization 136.4 11.1 6.5 6.4 5.3 Pipe break jet 745.0 745.0 Pool swell

-172.0 -172.0 Transient p r essure di f f e r ence (drywell-wetwell) +642.0 +642.0 Seismic Operating basis earthquake 127.1 9.9 5.7 31.2 27.1 Safe shutdown earthquake 164.2 12.8 7.4 46.1 39.6 Dead +229.0 +134.0 Live +217.0 +217.0 a Positive a x ial force is com p ressive; negative is tensile.

C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.4.2-29 Table 3A.4.2-3 Results Su mmary - Safet y/Relief Va l v e Quencher Principal S t ress (p si) Loading Condition Stress Category 1 2 3 Critical Stress Intensity (psi)

Allowable Stress (psi)

D.M. (1) Normal P m P L P L + Q 2525 11,092 13,678 1390 1204 1052-275-275 0 2800 11,367 13,678 15,978 23,967 47,934 5.71 2.11 3.50 Upset P L + Q P m P L 28,399 4102 13,480 6985 535 2540 0-275-275 38,399 4377 13,755 47,934 17,576 26,364 1.25 4.01 1.92 Emerge P m P L 3989 16,964 1392 1711-275

-275 4264 17,239 23,967 28,761 5.62 1.67 Fault (1) P m P L 3994 16,982 1395 1712-275

-275 4267 17,257 31,956 38,347 7.48 1.99 Fault (2) P m P L 3334 14,738 1223 1633-275

-275 3609 15,013 31,956 33,347 8.85 2.55 a Generally the maximum stresses occurs between the quencher a r ms, where the a r ms meet the body.

Notes:

1. Design margin is defined as follows:

DM = All o w a ble S t r e s s C a l c u l ated S t r e ss 2. The definition of the above terms is proved i n ASME Code Section III, Paragraph NC-3217 of W i nter 1976 Addenda.

C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT N o v e mb e r 1998 3A.4.2-30 Table 3A.4.2-4 Cumulative Usage Factor Calcul a tion at 24 in. Downcomer Anchor Load Combinatio n a Set Expected Number of Cycles (ni)

Bending Moment Mi lb/ft Peak Stress Sp (psi) Alt. Stress Salt (psi) Allowable Number of Cycles N Ca l c ula t ed Usage Factor Ui ni Ni 1 1 244,008 60,116 30

, 0 5 8 20,000 0.0001 2 9 244,008 32,575 16

, 2 8 8 200,000 0.0001 3 50 202,180 26,991 13

, 4 9 6 300,000 0.0002 4 940 216,065 28,845 14

, 4 2 2 300,000 0.0031 5 12,434 183,244 24,463 12,2 3 2 400,000 0.0311 Cumulative Usage Factor U= 0.0346 < 1.0 a Load combination set definition 1 - NPC + LOCA + SSE + SRV + CHUGGING

2 - SSE + SRV + CHUGGING

3 - OBE + SRV

4 - SRV + CHUGGING

5 - SRV C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT N o v e mb e r 1998 3A.4.2-31 Table 3A.4.2-5 Cumulative Usage Factor Calcul a tion at 28 in. Downcomer Anchor Load Combinatio n a Set Expected Number of Cycles (ni)

Bending Moment Mi lb/ft Peak Stress Sp (psi) Alt. Stress Salt (psi) Allowable Number of Cycles N Ca l c ula t ed Usage Factor Ui ni Ni 1 1 346,771 61,562 30

, 7 8 1 18,000 0.0001 2 9 346,771 33,775 16

, 8 8 8 150,000 0.0001 3 50 307,359 29,937 14

, 9 6 8 200,000 0.0002 4 940 557,714 54,321 27

, 1 6 1 30,000 0.0313 5 12,434 275,671 26,850 13,4 2 5 400,000 0.0311 Cumulative Usage Factor U= 0.0629 < 1.0 a Load combination set definition 1 - NPC + LOCA + SSE + SRV + CHUGGING

2 - SSE + SRV + CHUGGING

3 - OBE + SRV

4 - SRV + CHUGGING

5 - SRV C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.4.2-32 Table 3A.4.2-6 Maximum Usage Factors Table

First Actuation Second Actuation Subsequent Actuations Total Cumulative Usage Factor Low set SRV lines (MSRV-1C)

(MSRV-1B) 0.626 0.033 0.237 0.896 Non-low-set SRV

lines 0.527 0.202 N/A 0.729

Draw. No.

Rev.Figure Form No. 960690 Amendment 53November 1998 FigureDraw. No.Rev.Downcomer Bracing System - Model For Structural Analysis 950021.61 3A.4.2-2 75 76 78 80 68 71 61 64 69 72 73 70 65 62 57 54 49 46 66 58 6 3 Y X 90 East North 0 41 7 15 27 35 43 51 59 67 RPV Pedestal 418 21 410 202 203115Typical Joint At ContainmentTypical Member NumberTypical Node Number 42 34 26 1911Typical SRV LineTypical DowncomerTypical Column 61 22 18 14 17 25 30 24 10 9 13 8 12 16 20 23 28 423 29 32 33 220122 31 36 37 433 38 40 441 39 44 444 48 45 47 52 53 56 55 60 63 152 148 463 134 130 121116 103 85 80 67 62 49 44 31 26 13 98 222 223 17 27 35 45 53 63 71 81 89 99 107117 125 135 143 8 153 161 1 14 18 23 28 32 36 41 46 50 54 59 64 68 72 77 82 86 90 95 100 104 108 122 126 131 136 140 144 149 154 162 6 164 165 166 167 159 155 156 150 218 145 146 217 465 147 151 142 141 137 138 127 132 216 457 129 133 124 214 128 123119 120113118114110 11 1 50 106 105 109 101 102 96211 91 92 213 212 93 97 88 87 83 78 84 209 73 74 69 208 75 70 79 66 65 60 55 56 51 57 52 48 47 42 206 37 38 33 205 207 39 43 34 30 29 24 204 20 19 1511 12 16 21 25 158 225 224 210 219 Columbia Generating StationFinal Safety Analysis Report 5

Diaphragm Floor Column Modelfor Dynamic Analysis 950021.62 3A.4.2-4 Figure Amendment 53November 1998 Form No. 960690Draw. No.Rev.Model for Horizontal Loads Fixed at Base El. 439' - 4" 1 2 3 4 5 6 7 8 9 1011 12 13 14 15 16 17 1 2 3 4 5 6 7 8 9 1011 12 13 14 15 16 12 at 3' = 36" 4 at 4' = 16' 52' - 0"Typical Member No.Typical Node No.

El. 466' - 4"Water Surface El. 491' - 4" Pinned atTop Diaphragm Floor Columbia Generating StationFinal Safety Analysis Report Structural Model of DiaphragmFloor Beam and Column 950021.63 3A.4.2-5 Figure Amendment 53November 1998 Form No. 960690Draw. No.Rev.Model for Vertical Loads 1' - 4" 40 39 38 37 36 35 34 33 32 31 30 29 28 13 at 4' = 52' 27 39 38 37 36 35 34 33 32 31 30 29 28 27 26 26 22 18 14 10 6 2 124201612841 17 at 1' = 17' 18' - 4" 7 at 1' = 7' Member: Node: 1' - 6" Columbia Generating StationFinal Safety Analysis Report Figure Amendment 53November 1998 Form No. 960690Draw. No.Rev.SRV Piping SystemInner Ring Quencher Support 960222.91 3A.4.2-6 El. 493' - 0" El. 493' - 0" El. 489' - 5" Anchor at 28" Ø Downcomer Rigid Restraint (Vertical Only) 12x10 Reducer Upper GuideWater Level Downcomer Bracing El. 466' - 3 1/4" El. 455 '- 4" El. 435' - 3" Columbia Generating StationFinal Safety Analysis Report Figure Amendment 53November 1998 Form No. 960690Draw. No.Rev.Quencher Assembly 960222.92 3A.4.2-7 10 12.75" 1/2" 2.343" 18.5" 1" 4.4375'6" 12" Sch.80 PipeT = .687"Temp. Design = 474 F Press. Design = 550 psigReducer, 1 1/4" Thk. Rolled SA155 KCF 70 24" Sch.160 PipeSA376, Type 304 12" Sch. 80 Pipe SA312, Type 304, or SA358, Type 304, Cl. 124" - 600# Blind Flg., F.F., F&D

SA182, GR.F 304 17" 1.1927" 2.6565" 5.5989'Columbia Generating StationFinal Safety Analysis Report Figure Amendment 53November 1998 Form No. 960690Draw. No.Rev.SRV Quencher Assembly 960222.93 3A.4.2-8 Columbia Generating StationFinal Safety Analysis Report 24" Sch. 160 Pipe (Quencher Body) 12" Sch. 80 Pipe (Quencher Arm) 24"-600# Blind Flange C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.4.3-1 3A.4.3 MISCELLANEOUS SUPPRESSION POOL PIPING SYSTEMS

A tabulation of the miscellane ous systems located in the su ppression chamber is given in Table 3A.4.3-1. The results of the stress analysis for suppression pool piping other than SRV piping and downcomers is presented in Table 3A.4.3-3.

Depending upon the location in the wetwell, the suppression pool piping will be subjected to different loading associated with SRV discharge and LOCA. Fo r identification purposes, the miscellaneous wetwell piping has been broken into four zones, i.e., fully submerged piping, partially submerged piping, piping in the pool swell zone, and piping above the pool swell zone (Figure 3A.4.3-1

). Pip i ng in each zone is noted in Table 3A.4.3-

1.

3A.4.3.1 Loads Used for Assessment

The wetwell piping systems are subjected to static, dynamic, and hydrodynamic loads due to normal, upset, emergency, and fa ulted plant operating conditions. Each zone is characterized by certain applicable loads shown in Table 3A.4.3-2. A description of each of these loads is provided in Sections 3A.3 and 3A.5.

3A.4.3.2 Load Combination and Acceptance Criteria

The design limits, as set forth in the ASME Boiler and Pressure Ve ssel Code (ASME Code)Section III, Subsection NC (Reference 3A.4.2-1) were utilized for th e assessment of the suppression pool piping systems.

The various piping systems are c onsidered acceptable if they satisfy the equations of Paragraph NC-3652 of Secti on III of the ASME Code.

The piping, as listed on Table 3A.4.3-1 , is fabricated of low all oy carbon steel pipe having an allowable stress "S h" primary evaluation of 15,000 psi up to a temperature of 275°F.

Allowable Stress Limits (Equation 9 of NC-3652 a nd NC-3611, Reference 3A.4.2-1)

The stress for the miscellaneous wetwell pipi ng includes the primary membrane plus bending stresses. The limits of these stresses, depending on loading conditions, are the same as those listed in Section 3A.4.2.4.3. 3A.4.3.3 Method of Analysis The miscellaneous suppre ssion pool piping systems were analy zed for the appropriate loading combinations using ADLPIPE and ANSYS computer programs (Attachment 3A.F

). Hand calculation methods were used to analyze short cantilevered pi pes attached to the primary containment. Static and response spectrum analyses were handled in the same manner as described in Section 3A.4.2.3.3. However, displacement time history analyses were performed, as required, to obt ain more realistic piping respon ses due to SRV building response C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.4.3-2 loads. For residual heat removal (RHR) blowdown transient loads, ADLPIPE was used to perform force time history analysis.

3A.4.3.4 Results and Design Margins

Table 3A.4.3-3 presents pipe stress re sults and design margins for piping systems in the wetwell. The pipe supports for the wetwell were also assessed and determined to be within acceptable limits.

Table 3A.4.3-1 Miscellaneous Wetwell Piping Ref. Penetration Number X-Line/Description Penetration Elevations Penetration Azimuths Remarks C OLUMBIA G ENERATING S TATION Amendment53 F INAL S AFETY A NALYSIS R EPORTNovember 1998 3A.4.3-3 Zone I - Fully Submerged Piping 33 8 in. RCIC Pump Suction 452 ft-0 in.

336° 100 6 in. Fuel pool cleanup 452 ft-0 in.

295° 35 24 in. RHR pump "A" suction 452 ft-0 in.

276° - 49 ft Similar in config. to X-32 32 24 in. RHR pump "B" suction 452 ft-0 in.

263° - 11 ft 31 24 in. HPCS pump suction 452 ft-0 in.

97° - 36 ft 34 24 in. LPCS pump suction 452 ft-0 in.

66° - 05 ft Similar in config. to X-36 36 24 in. RHR pump C suction 452 ft-0 in.

45° - 39 ft 88 3 in. instrumentation 455 ft-0 in.

180° 86B 2 in. instrumentation 462 ft-0 in.

45° Short length of pipe 87B 2 in. instrumentation 462 ft-0 in.

245° Short length of pipe N/A 4 in. FPC - SYPHON N/A N/A Zone II - Partially Submerged Piping 64 1.5 in. RCIC vacuum pump discharge 467 ft-9 in.

345° Similar in config. to X-65 4 10 in. RCIC line (24 in. header) 467 ft-9 in.

318° 65 2 in. RCIC pump min. flow 467 ft-9 in.

312° 101 6 in. RPC test 467 ft-9 in.

0° 47 18 in. RHR pump A test 467 ft-9 in.

288.7° Similar in config. to X-48 117 18 in. RHR pressure relief 467 ft-9 in.

279° Similar in config. to X-11B 118 18 in. RHR pressure relief 467 ft-9 in.

261° 48 18 in. RHR pump B test 467 ft-9 in.

251.29° 23 3 in. EDR equipment drain 467 ft-9 in.

132° 24 3 in. FDR floor drain 467 ft-9 in.

111° 49 12 in. HPCS pump test 467 ft-9 in.

103° - 31 ft 63 12 in. LPCS pump test 467 ft-9 in.

60° Similar in config. to X-49 26 18 in. RHR pump "C" test 467 ft-9 in.

20° Table 3A.4.3-1 Miscellaneous Wetwel l Piping (Continued)

Ref. Penetration Number X-Line/Description Penetration Elevations Penetration Azimuths Remarks C OLUMBIA G ENERATING S TATION Amendment59 F INAL S AFETY A NALYSIS R EPORTDecember 2007LDCN-06-039 3A.4.3-4Zone III - Pool Swell Zone 87A 2 in. instrumentation 471 ft-6 in.

245° Short length of pipe 86A 2 in. instrumentation 471 ft-6 in.

45° Short length of pipe 51 42 in. HATCH 473 ft-6 in.

155° Short length of pipe 107B 12 in. electrical line 475 ft-0 in.

240° Short length of pipe 107A 12 in. electrical line 475 ft-0 in.

52° - 30 ft Short length of pipe 66 24 in. CSP 475 ft-0 in.

222° - 30 ft Short length of pipe 116 12 in. RCIC 477 ft-6 in. Short length of pipe 81 14 in. instrumentation 479 ft-4 in.

235° Short length of pipe 82 10 in. CAS 479 ft-4 in.

230° 83 10 in. instrumentation 479 ft-4 in.

240° 84 10 in. instrumentation 479 ft-4 in.

40° Similar in config. to X-83 Zone IV - Above Pool Swell Zone 67 24 in. CEP 491 ft-0 in.

0° Short length of pipe 102 6 in. Plugged (4 in. CAC pipe) 240° Short length of pipe 103 6 in. Plugged (4 in. CAC pipe) 180° Short length of pipe 104 6 in. Plugged (4 in. CAC pipe) 323° Short length of pipe 105 6 in. Plugged (4 in. CAC pipe) 140° Short length of pipe 57 12 in. SPARE 47° Short length of pipe 58 12 in. SPARE 132° Short length of pipe 59 12 in. SPARE 258° Short length of pipe 60 12 in. SPARE 280° Short length of pipe 25A 6 in. RHR 70° 25B 6 in. RHR 235° - 30 ft 119 24 in. vacuum breaker 151° Short length of pipe C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.4.3-5 Table 3A.4.3-2 Piping Zone Versus Loads Zone Loading Fully Submerged Partially Submerged Pool Swell Zone Above Pool Swell Zone Deadweight X X X X Thermal X X X X Pressure X X X X Operating basis earthquake X X X X Safe shutdown earthquake X X X X SRV pressure X X SRV response spectra X X X X LOCA jet X X LOCA bubble X X Pool swell and fallback X X X Chugging p r essure X X Chugging response spectra X X X X C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1 998 Table 3A.4.3-3 Summary of Results a nd Design Margins for Miscellane o u s Wetwell Piping Penetration Number Controlling Load Case Allowable Stress (psi) Calculated Stress (psi)

Design Margin 3A.4.3-6 X-33 Upset 18,0 0 0 12,264 1.46 X-100 Upset 18,000 5,231 3.44 X-32 Upset 18,000 9,923 1.81 X-31 Upset 18,0 0 0 14,096 1.28 X-36 Upset 18,0 0 0 11,687 1.54 X-4 Emergency 27

, 000 14,453 1.86 X-101 Faulted 36

, 000 19,826 1.81 X-47 and X-117 Emergen c y 27,000 16,952 1.59 X-23 Upset 18,0 0 0 13,963 1.28 X-49 Emergency 27

, 000 13,001 2.07 X-26 Emergency 27

, 000 19,580 1.37 X-24 Emergency 27

, 000 19,344 1.39 X-35 Upset 18,000 9,923 1.81 X-34 Upset 18,0 0 0 17,939 1.003 X-48 and X-118 Emergency 27,000 24,400 1.03 a For Most Severe Condition X-104 Emergency 27

, 000 17,687 1.53 X-82 Upset 18,0 0 0 14,388 1.25 X-25A and X-25B Upset 18,000 17,500 1.03 b X-83 Faulted 44

, 640 33,084 1.35 C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1 998 Table 3A.4.3-3 Summary of Results a nd Design Margins for Miscellane o u s Wetwe l l Piping (Continued)

Penetration Number Controlling Load Case Allowable Stress (psi) Calculated Stress (psi)

Design Margin 3A.4.3-7 X-84 Faulted 44

, 640 33,084 1.35 X-88 Upset 22,7 0 0 16,508 1.37 X-63 Faulted 36

, 000 6,220 5.78 X-64 Upset 18,0 0 0 15,193 1.19 X-65 Upset 18,0 0 0 14,271 1.26 4 in. FPC Upset 18,000 2,774 6.49 a The design margin for this line in reality is larger because of the conservative approach taken, that the flexibility of the containment was not considered for the governing load in this case - RHR transient.

b The design margin for this line is also larger due to conservatism in the envelope spectrum analysis for SRV and chugging as compared to multi-input or time-history analysis that could have been utilized.

Notes:

1. The effects of short pipes cantilevered from the containment to hydrodynamic loads is considered minimal. These short stubs are listed in Table 3A.4.3-1.

2. The information provided within this table are the maximum stresses between the nozzle and the termination of th e pipe within the wetwell.

3. Results of the stresses at the shell will be provided later.

Figure Amendment 53 November 1998Form No. 960690.veR.oN .warD Hydrodynamic Loading Zones1-3.4.A349.222069 4 3 2 1 Zone Zone Zone ZoneDowncomer VentDiaphragm FloorMaximum Pool Swell Height (El. 484'-4 3/4")Minimum Water Level (El. 466'-0 3/4")

Downcomer Vent Exit Plane (El. 454'-4 3/4")

Columbia Generating Station Final Safety Analysis Report C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.5.1-1 3A.5 EFFECTS DUE TO BUILDING RE SPONSES TO SAFETY/RELIEF VALVE DISCHARGE AND LOSS-OF-COOLANT ACCIDENT LOADS Response spectra generated at each floor lo cation define the input motions used for qualification and assessment of all the safety-related piping and equipment. In load combinations which include sa fety/relief valve (SRV) discharge and seismic loads, or loss-of-coolant accident (LOCA) steam condensation and seis mic loads, seismic response spectra based upon a finite element soil-stru cture model were used in design and plant assessments for structures, piping, and equipm e n t. The use of t h e fini t e ele m ent so il-structure model generally results in lower structural responses than the soil spring and dashpot model

used in the original seismic design.

3A.5.1 BUILDING RESPONSES TO SAFETY/RELIEF VALVE DISCHARGE LOADS

This section presents the dyna mic responses of the containm ent and internal structures subjected to loads resulting from SR V discharge as defined in Section 3A.3.1.3. The analytical model and method of analysis for determining the building structural response to SRV discharge loads are described in the following sections.

3A.5.1.1 Analytical Model

3A.5.1.1.1 Overall Building Model

Figure 3A.5.1-1a prese n ts the so il-s t r ucture model. Figure 3A.5.1-1b p r esents the axisymmetric overall building model of the CGS r eactor building. It should be noted that the thick reactor pressure vessel (R PV) pedestal and building mat we re accounted for in the model by utilizing multiple layer axis ymmetric solid elements.

These models were utilized to determine the response to the loading conditions stated in Section 3A.3, and provided responses at the RPV pedestal, foundation mat, biological shield wall, and in the reactor building walls and floors outside of primary containment.

3A.5.1.1.2 Steel Cont ainment Shell Model

Figure 3A.5.1-2 shows the more refined th r ee-dimensional finite element model of the containment shell. This model is interconnected to the rest of the building at the basemat (el. 446 ft), diaphragm floor (el.

503 ft), stabilizer truss (el. 565 ft), and the refueling bellows (el. 583 ft).

3A.5.1.2 Method of Analysis

The containment shell model wa s analyzed for SRV rigid wall pressure loads acting on the wetwell boundary and specified displacements at al l the points where the sh ell is connected to C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.5.1-2 the rest of the building. The displacements were obtained from the solutions of the overall building model. The refined containment shell model, as di scussed in Section 3A.5.1.1.2 , was used to determine the res ponse of the steel shell.

A more detailed discussion of the two models, analytical approach, and results are found in "SRV Loads - Improved Definition and Application Methodology" (Reference 3A.3.1-1).

3A.5.1.3 Safety/Relief Valve Discharge Load Cases Several SRV discharge cases were considered for CGS design ev aluation as discussed below.

3A.5.1.3.1 Response to All Valve Discharge

All 18 SRVs are conservatively assumed to discharge during certain plant conditions.

Two design conditions are associated with all the valves discharge case - the "axisymmetric" condition and the "nearly symmetric" condition. Each of these loading c onditions is applied in load combinations involving the all valves discharge case. The axisymmetric loading condition assumes all valves will discha rge simultaneously in the pool, thus maximizing the response of the axisymmetric features of the containment and reactor pedestal. The nearly symmetric loading condition assumes some imbalance may occur during actuation of all SRVs. The imbalance may occur from sequen tial discharging at different se t points, variability in valve opening time, differences in SRV discharge line geometry, etc.

Listed below is a sample of the response spectra used in pl ant assessments involving the axisymmetric loading condition. For the "near ly symmetric" loading condition, the response spectra used are constructed by adding 0.5 times the axisym metric response spectra to

0.6 times

the single valve response spectra.

Location Direction Figure Top of RPV Pedestal, el. 520 ft Radial 3A.5.1-3a ,

Mass No. 44 3A.5.1-3b Top of RPV Pedestal, el. 520 ft Vertical 3A.5.1-4a , Mass No. 44 3A.5.1-4b Basemat at RPV Pedestal, el. 435 ft Radial 3A.5.1-5a , Mass No. 141 3A.5.1-5b

Basemat at RPV Pedestal, el.

435 ft Vertical 3A.5.1-6a ,

Mass No. 141 3A.5.1-6b

C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.5.1-3 Top of Sac. Shield Wall, e

l. 567 ft Radial 3A.5.1-7a , Mass No. 14 3A.5.1-7b Top of Sac. Shield Wall, e

l. 567 ft Vertical 3A.5.1-8a , Mass No. 14 3A.5.1-8b RPV, el. 545 ft Radial 3A.5.1-9a ,

Mass No. 27 3A.5.1-9b RPV, el. 545 ft Vertical 3A.5.1-10a , Mass No. 27 3A.5.1-10b

Containment Vessel, el. 547 ft Radial 3A.5.1-11a

,

Mass No. 60600 3A.5.1-11b

Containment Vessel, el. 547 ft Vertical 3A.5.1-12a

,

Mass No. 60600 3A.5.1-12b

Containment Vessel, el. 448 ft Radial 3A.5.1-13a

,

Mass No. 50100 3A.5.1-13b

Containment Vessel, el. 448 ft Vertical 3A.5.1-14a

,

Mass No. 50100 3A.5.1-14b

Notes: 1. Figures denoted "a" refer to the conventional [single frequency pressure (SFP)] load case while the figures denoted "b" ref e r to the multiple frequency pressure (M FP) load ca s e (s ee R e fe r e nce 3A.3.1-1 for detai l s).

2. The tangential loads we r e also u tilized in the assess m ent of CGS. The tangential v a lues are not included in this submit t al as they a r e much smaller in

magnitude then the presented valves.

3A.5.1.3.2 Automatic Depressurizat i on System Valves Discharge C a se

This case corresponds to the d i scharge of the SRVs of the a u tomatic depressurization system (ADS) which are automatica l ly act u a ted. For assess m ent purposes, the more conservative all valves resp o n se spect r a values were utilized.

C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.5.1-4 3A.5.1.3.3 Two Valves Discharge Case In this case, two SRVs are considered to discharge concurrently through two adjacently located quenchers. For assessment purposes, the more conservative single outer valve discharge response spectra were utilized.

3A.5.1.3.4 Single Valve Discharge

The single outer quencher discharg e, which is less likely to ha ppen, is considered in the assessment because it is found to be more conservative, both for containment and pedestal response.

Location Direction Figure Top of RPV Pedestal, el.

520 ft Radial 3A.5.1-15a , Mass No. 44 3A.5.1-15b

Top of RPV Pedestal, el. 520 ft Vertical 3A.5.1-16a

,

Mass No. 44 3A.5.1-16b

Basemat at RPV Pedestal, el. 435 ft Radial 3A.5.1-17a , Mass No. 141 3A.5.1-17b

Basemat at RPV Pedestal, el.

435 ft Vertical 3A.5.1-18a , Mass No. 141 3A.5.1-18b

Top of Sac. Shield Wall, e

l. 567 ft Radial 3A.5.1-19a , Mass No. 14 3A.5.1-19b

Top of Sac. Shield Wall, e

l. 567 ft Vertical 3A.5.1-20a , Mass No. 14 3A.5.1-20b

RPV, el. 545 ft Radial 3A.5.1-21a ,

Mass No. 27 3A.5.1-21b

RPV, el. 545 ft Vertical 3A.5.1-22a , Mass No. 27 3A.5.1-22b

Containment Vessel, el. 547 ft Radial 3A.5.1-23a

,

Mass No. 60600 3A.5.1.23b

Containment Vessel, el. 547 ft Vertical 3A.5.1-24a

,

Mass No. 60600 3A.5.1-24b

C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.5.1-5 Containment Vessel, el. 448 ft Radial 3A.5.1-25a

,

Mass No. 50100 3A.5.1-25b

Containment Vessel, el. 448 ft Vertical 3A.5.1-26a

,

Mass No. 50100 3A.5.1.26b

Notes: 1. Figures denoted "a" refer to the conventional (SFP) l o ad case while the figures denoted "b" refer to the MFP load case (see Re f e re n ce 3A.3.1-1 for details).

2. The tangential loads we r e also u tilized in the assess m ent of CGS. The tangential v a lues are not included in this submit t al as they a r e much smaller in

magnitude then the presented valves.

Figure Amendment 53 November 1998Form No. 960690.veR.oN .warD Axisymmetric Model of the Reactor Building and Soil Foundationa1-1.5.A359.222069See Enlarged DetailFig. 3A.5.1-1b El. 427.75' El. 374.25' El. 331.25' El. 281.25' El. 206.75' El. 106.25' C L 3 at 48.0' = 144.0' 60.0'96.0'Columbia Generating Station Final Safety Analysis Report Figure Not Available For Public Viewing Figure Not Available For Public Viewing Figure Not Available For Public Viewing Figure Not Available For Public Viewing Figure Not Available For Public Viewing Figure Not Available For Public Viewing Figure Not Available For Public Viewing Figure Not Available For Public Viewing Figure Not Available For Public Viewing Figure Not Available For Public Viewing Figure Not Available For Public Viewing Figure Not Available For Public Viewing Figure Not Available For Public Viewing Figure Not Available For Public Viewing Figure Not Available For Public Viewing Figure Not Available For Public Viewing Figure Not Available For Public Viewing Figure Not Available For Public Viewing Figure Not Available For Public Viewing Figure Not Available For Public Viewing Figure Not Available For Public Viewing Figure Not Available For Public Viewing Figure Not Available For Public Viewing Figure Not Available For Public Viewing Figure Not Available For Public Viewing Figure Not Available For Public Viewing Figure Not Available For Public Viewing Figure Not Available For Public Viewing Figure Not Available For Public Viewing Figure Not Available For Public Viewing Figure Not Available For Public Viewing Figure Not Available For Public Viewing Figure Not Available For Public Viewing Figure Not Available For Public Viewing Figure Not Available For Public Viewing Figure Not Available For Public Viewing Figure Not Available For Public Viewing Figure Not Available For Public Viewing Figure Not Available For Public Viewing Figure Not Available For Public Viewing Figure Not Available For Public Viewing Figure Not Available For Public Viewing Figure Not Available For Public Viewing Figure Not Available For Public Viewing Figure Not Available For Public Viewing Figure Not Available For Public Viewing Figure Not Available For Public Viewing Figure Not Available For Public Viewing Figure Not Available For Public Viewing Figure Not Available For Public Viewing C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.5.2-1 3A.5.2 BUILDING RESPONSES TO LO SS-OF-COOLANT ACCIDENT LOADS The analysis of the containment under the action of long-term LOCA loads and the resultant

responses are described in this section. The LOCA loads cons idered are those described in Section 3A.3.2.4 , namely, chugging and condensation oscillation. However, the condensation oscillation load is not a governing load as comp ared to the chugging load, therefore it was not considered in the design assessme nt. The discussion in this sec tion applies to the effects of chugging only. A complete de finition of the chugging loads a nd the methodology of their application to the reactor build ing is contained in Reference 3A.3.2-15.

3A.5.2.1 Analytical Model

The mathematical model used for the analysis of the structure subjected to chugging loads includes the reactor building and the supporting soil. The model of the reactor building is shown in Figure 3A.5.2-1. The model of the supporting soil is the same as that shown in Figure 3A.5.1-1a in connection with the analysis for SRV discharge load

s. The soil is represented by solid axisymmetric elements with asymmetric load capability. Spatial variation of soil shear modulus and unit weight and so il-structure interacti on are accounted for.

As shown in Figure 3A.5.2-1 , two types of finite elements ha ve been used in modeling of the building, namely, axisymmetric conical shell elements and axisymmetric solid elements; both types of elements have asy mmetric loading capab ility. The wetwell co lumns, stabilizer trusses, bellows, and shear lugs between the diaphragm floor and containment are modeled using springs.

In the suppression pool region, where the hydrodynamic loads are applied and where a more accurate representation is require d, the node locations are closel y spaced. The horizontal rings attached to the containment vessel are treated as discrete rings and the additional stiffness due to the vertical stiffe ners is included with the vessel properties.

The RPV and internals are represented by axisy mmetric shell elements. The dynamic coupling effect of the fluid in the RP V is accounted for by adding hy drodynamic masses to the nodal points of the mathematical model.

3A.5.2.2 Method of Analys is and Building Response

The structural response of the reactor building, when subjected to the chugging phenomenon, is determined from the application of the seven distributions of the chugging pressures on the wetwell pool boundary in the analytical model.

These seven distribu tions of pool boundary pressures result from the seven design chugging sources developed for the multivent chugging definition described in Section 3A.3.2.4.2.1. Two loading cases are considered for each of the seven design sources, namely, nearly symmetric and asymmetric conditions. Detailed C OLUMBIA G ENERATING S TATION Amendment 57 F INAL S AFETY A NALYSIS R EPORT December 2003 LDC N-0 2-0 0 0 3A.5.2-2 analytic methods for determination of the structural response are given in Reference 3A.3.2-15.

As described in Reference 3A.3.2-15 , the nearly symmetric a nd asymmetric results are comparable. For the purpose of this assessment, the nearly symmetric loading condition is used.

The response of the building is obtained in terms of accelerati on response spectra. These were calculated for significant locations in the reactor building for the nearly symmetric and asymmetric loading conditions. The envelope spectrum curves we re plotted with peaks spread by +/-15% for damping valu es of 0.5%, 1.0%, 2.0%, and 4.0% of critical damping. Nearly symmetric response spectra plots for different locations in the building are illustrated in the figures listed below.

It should be noted that the design values presented in Figures 3A.5.2-2 through 3A.5.2-10 were increased by a factor of 1.10 to account fo r the differences in vent size (28 in. for Columbia Generating Station as compared to 24 in. for the 4T and 4TCO test) and an additional factor of 1.16 over the values presented in the chugging report.

3A.5.2.2.1 Reactor Building Response, Nearly Symmetric Loading - Acceleration Response Spectra

Location Direction Figure Containment Vessel, el. 448 ft Mass No. 152 Radial 3A.5.2-2 Containment Vessel, el. 448 ft Mass No. 152 Vertical 3A.5.2-3 Containment Vessel, el. 459 ft Mass No. 123 Radial 3A.5.2-4 Containment Vessel, el. 459 ft Mass No. 123 Vertical 3A.5.2-5 RPV Support on Pedestal, el. 519 ft Mass No. 57 Radial 3A.5.2-6 RPV Support on Pedestal, el. 519 ft Mass No. 57 Vertical 3A.5.2-7 C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.5.2-3 Containment Vessel, el. 583 ft Mass No. 12 Radial 3A.5.2-8

Building Wall, el. 521 ft

Mass No. 55 Radial 3A.5.2-9

Building Wall, el. 521 ft

Mass No. 55 Vertical 3A.5.2-10 Note that the tangential loads were also utilized in the assessment of CGS. The tangential values are not included in this submittal as they are much smaller in magnitude then the presented values.

Figure Not Available For Public Viewing Figure Not Available For Public Viewing Figure Not Available For Public Viewing Figure Not Available For Public Viewing Figure Not Available For Public Viewing Figure Not Available For Public Viewing Figure Not Available For Public Viewing Figure Not Available For Public Viewing Figure Not Available For Public Viewing Figure Not Available For Public Viewing C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.B-1 Attachment 3A.B THREE-DIMENSIONAL SOURCE FLOWS IN EXACT

CONTAINMENT GEOMETRY 3A.B.1 INTRODUCTION AND

SUMMARY

The method specified in Reference 3A.B-1 to calculate the flow fiel d due to bubbles [caused by a loss-of-coolant accident (LOCA) or safety/relief valve (S RV) actuation] in a Mark II suppression pool is the method of images (MOI).

The MOI is a potential flow technique and uses point source(s) or sink(s) to represent the bubble(s) and a number of image sources and/or sinks to simulate flow at the pool boundaries to satisfy the kinematical boundary conditions. The MOI has the following limitations with regard to its application to CGS:

a. The annular suppression pool geometry n eeds to be idealized into a rectangular pool using an "equivale nt radius" concept,

b. The sloping suppression pool bottom needs to be idealized in to a flat bottom, and

c. Computer flow field calculation costs are high.

In view of these limitations, the MOI is not used to calculate the poten tial flow field due to stationary source(s) in the CG S suppression pool. Instead, a numerical method is used to determine three-dimensional poten tial flows induced by source(s) in the exact CGS containment geometry.

The principle of the method can be stated rather simply. The three-dimensional potential flow induced by sources in any arbitrary suppression pool geometry is provided by solution of Laplace's equation for th e velocity potential, . This is done by splitting the function into two components: s which is due to all the sources and sinks (represe nting expanding or contracting bubbles) and is calculated analytically, and , which is a smooth function that is calculated numerically. The function calculated from the original boundary conditions for with the boundary values of s subracted off. Single s analytically known elsewhere and can be determined from its boundary conditions by iteration, the total value of the velocity potential, , can be easily determined.

This attachment describes the formulation of the problem, the method of solution, the

numerical procedures utilized, and presents and analyzes some results. Section 3A.B.2 discusses the problem formulation and the solution method, Section 3A.B.3 discusses the numerical procedure, Section 3A.B.4 discusses how the flow field is calculated, C OLUMBIA G ENERATING S TATION Amendment 54 F INAL S AFETY A NALYSIS R EPORT April 2000 LDC N-9 9-0 0 0 3A.B-2 Section 3A.B.5 discusses the initial estimate to start the iteration solution for . Section 3A.B.6 discusses the convergence and accu racy of the method, and Section 3A.B.7 presents and analyzes some result s for single and multiple bubble cases.

3A.B.2 PROBLEM FORMULATIO N AND SOLUTION METHOD

For a potential flow, the fluid velocity component s may be expressed in terms of a velocity potential function, as shown below:

Vxyz x , V , V yz The negative sign is purely a convention and mean s that the fluid flows in the direction of potential drop (Reference 3A.B-2).

The formulation of the boundary va lue problem for the potential is:

2 0 everywhere in the fluid excep t appropriate delta functions at sources and sinks, 0 0 n on the rigid boundari es (containment wall, basemat, and pedestal), and 0 on the free surface (follows from pressure = 0).

The velocity potential, is split into two parts:

= s + The function s represents contributions from the si ngularities (sources and/

or sinks) and its analytical expression at any point within the fluid is given below:

SS S Q4 where Q represents the source strength of the th source or sink and represents the distance from the th source or sink to the point in question (Reference 3A.B-1).

C OLUMBIA G ENERATING S TATION Amendment 55 F INAL S AFETY A NALYSIS R EPORT May 2001 LDC N-0 1-0 0 0 3A.B-3 The function is a smooth function and satisf ies the boundary value problem:

2 0 everywhere, nn s = on the rigid boundaries and any planes of symmetry, and

= s on the free surface.

The boundary values for were derived by subtracting the boundary values of s from the original boundary value conditions far the total velocity potential, . Because is extremely smooth, it can be accurately ca lculated by finite differences.

3A.B.3 NUMERICAL PROCEDURE

The geometry of the Columbia Generating St ation suppression pool and the location of quenchers and downcomers are shown in Figures 3A.2.1-1 t h rough 3A.2.1-8 , inclusive. A cylindrical grid of points is overl ayed such that all boundaries with /n0 (i.e., walls and symmetry planes) are all cente red between planes of grid points, while the water surface z = z max is a plane of grid points.

Figures 3 A.B-1 and 3A.B-2 show how the suppression pool geometry has been modeled for a single SRV actuation case and a LOCA bubble case to take advantage of symmetry.

The lengths between grid points in the r, , and z direction are denoted by r, , and z and are indexed in the three directi ons by i, j, and k, respectively.

r i = rmin = (i - 1.5) r j = (j - 1.5) z k = (z - 1.5) z Any function, f, when regarded as a function defined on the grid points will be denoted as f (r i , j ,z k) = f i ,j,k. The coordinates of the source(s) are indicated by the subscript s. As mentioned before, the potential is split into s + . s is the source potential (or the sum of each source's contribution in the multi-bubble cases):

s v Q4 C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.B-4 where

= ((r cos

-r )2 + (r si n)2 = (z-z)) 1/2 s is eval u a ted exactly on all grid points.

The numerical procedure of calculating is by the standard cover-relaxat i on method, which is summ a rized succinctly here.

satisfied the Laplace equation which, in cylind r ical coordinates, is (R e f erence 3A.B-2): 2 2 2 2 2 2 0 1 0 10 0 rr r r rz (3A.B-1) The finite difference approximation of equation 3A.B-1 is chosen as (ffinite aprroxof 22.2): fiiijkijkiiijkijk i rrrr r 21111 2 2,,,, ,, ,, r + ,,,,,,ijkijkijk i r11 2 2 2 + ,, ,, ,,ijkijkijk z11 2 2 = 0 (3A.B-2) where:

i=2,3...i max-1, j=2,3...j max-1, k=2,3...k max-1 Equation 3A.B-2 constitutes a linear system of (i max-2) x (j max-2) x (k max-2) equations. The number of unknowns is larger, since they are unknowns at the boundaries, which are to be related by the boun dary conditions.

As stated before, the boundary conditions for are: s on the water surface and nn s on walls, bottom, and planes of symmetry.

For planes of symmetry a simplification from //nn s to /n0 occurs.

C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.B-5 The finite difference equations have corresponding boundary conditions for flat bottom containments:

,, maxijks i,j,kmax (on the free surface), ,,,,ijijs i,s i, 12j,1j,2 (on basement floor), ,,,,12jkjks 1,s 2,j,kj,k (on the pedestal wall), ijkijkijkijkmaxmaxmaxmax ,, ,, ,, ,,11 (on the containment wall), ,,,,ikik 12 (on the surface 0 o), and ,,,,,,,,maxmaxmaxmaxijkijks ijks ijk11 (on the surface max). These equations, coupled with equation 3A.B-2, now provide a complete system of i max , kmax equations for the same number of unknowns.

There are many techniques available to solve the finite-difference equati ons generated from the Laplace equation. Some, using fast Fourier transform techniques or direct elimination techniques, are indeed very fast. The successive over-relaxation procedure (SOR) was selected because it is adequate in speed

in addition, it can handle general boundaries, whereas, use of other (possibly faster) methods require rectangular domains, periodic boundary conditions, or other restrictions.

Equation 3A.B-2 is solved for the center point in terms of the values at the neighboring points by iterating: (Superscript (n) indicates the n th iteration, (n+1) the next iteration, etc.)

,, (),,,,,,,, ,, ,,ijk nijkijk iijkijk ii iijk ii iijk i r rr rr rr rr1 11 2 2 11 2 1 2 1 1 2 1222 2 2 2 2 22 z z r or ()() n o n V1 where:

V o denotes an "averaging" operator.

C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.B-6 Whenever a point is updated, the new value is to be used in calculating its neighboring points ()()()nnn VV1 12 1. This is straightforward itera tion or relaxation. In SOR, which is much faster, the ch ange is anticipated by using an acceleration parameter, (Reference 3A.B-3): ()()()() nn V nn1 0 (3A.B-3)

Numerical analysis theory shows that convergence occurs for 0< <2, but the optimal depends on the geometry. Va rious test runs for the si ngle SRV bubble, and for the three bubble LOCA geometries were run with = 1.98 chosen for rapid convergence (Section 3A.B.6 gives more details on convergence).

(Note: for = 1, equation 3A.B-3 becomes ()() nn V1 0 which is the straightforward iteration case).

3A.B.4 CALCULATION OF FLOW FIELD

3A.B.4.1 Steady State Flow Field Calculation

The calculation of the velocity due to a poten tial function is performe d by taking the negative gradient of the total potential, . As discussed earlier, the total potential is determined by summing s, the potential due to the singularities plus , which is a smooth function. To ensure accuracy near singularities the velocity field is also calculated as the sum of two components:

VVV ss The velocity field due to the smooth function is determined numerica lly at points one-half a grid distance away from the velocity potenti a l grid system as shown in Figure 3A.B-3a. Essentially this scheme averages values of in the two neighboring grid planes that are normal to the direction of the desired veloc ity component. Once th ese two averages are established, the velocity component is determined by subtrac ting the two values and dividing by the grid width (r, r , z) in the appropriate direction (a representative example is given in Figure 3A.B-3

). The r, , and z velocity components due to are shown below:

Vr i,j,k ,,,,,,,, ,,,, ,,,,

ijkijkijkijkijkijkijkijk111 111111111 / 4 r C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.B-7 V0 i,j,k /,,,,,, ,,,,,,,, ,,

ijkijkijkijkijkijkijkijk r111 111111111 4 0 Vz i,j,k /,,,,,,,,,,,,,,,,ijkijkijkijkijkijkijkijk111 111111111 4 z The velocity field due to a source is determined analytically at the same grid points used for the velocity field due to the sm ooth function. The magnitude of the total velocity at a point due to a source is defined as:

V Q s4 2 The x, y, z velocity co mponents are defined by:

VxV xx ss s VyV yy ss s VzV zz ss s and the cylindrical components are defined by:

VrVxVyssscossin VVyVsssscossin VzVz ss For multiple sources, the veloc ity component are determined by summing the contribution of each of the sources. The total velocity magnitude is determined by the square root of the sum of the squared of the three components.

The usual way of calculating the velocity field is by differentiating the total potential function by the use of finite differences in exactly the same manner as was describe for the smooth function's velocity calcul ation. The defect here is that the resultant velocity field becomes inaccurate as one approaches the source(s). This is due to the singularity in and the resultant inability of a finite difference scheme to approximat e a derivative. Also, finite differences taken across the sour ce(s) will be totally inaccurate and meaningle ss. The above method of calculating the velocity field avoids this problem becau se (1) the velocities due to C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.B-8 s are an exact solution and, (2) is a smooth function which allows for a good approximation of a derivative by finite differences.

3A.B.4.2 Transient Flow Field Calculation

The flow field due to a time varying source can be calculated by assuming that the source strength varies with time. Since the velocity field is now a function of time and space, an acceleration field can also be calculated:

VrztQtrz,,,,, VrztQtrz,,,,, where:

Q(t) is the equivalent time varying point sour ce strength representing the hydrodynamic source.

()Qt is the time rate of change of Q(t).

4,, z gradient of the velocity potential at a given point due to a normalized point source strength.

V(r, ,z,t) = total velocity at a given point.

(,,,)Vrzt total acceleration at a given point.

3A.B.5 INITIAL ESTIMATE FOR ITERATION

Since n connects two grid planes which make up a rigid boundary, it is a "soft" boundary condition and "clamps" down the solution less than does a boundary condition. With known at only one out-of-six possi ble boundaries, it is necessary to start the iterations with as good an estimate as possi ble for the initial function (0). For single source cases, the initial function (0) is easy to construct: simply take - s on the surface, and = 0 on the basemat, and interpolate linearly in z along each vertical line. Subsequent convergence for is not too far from zero on all the boundaries.

For the multi-bubble LOCA cases, th e situation is quite different.

Above the sources the flow is essentially a vertical slug with uniform velocity up to the surface, while below the sources the flow is very small. Thus rises from zero on the surface to some large value near the C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.B-9 bubbles. From this knowledge of the phenomenon the following construction for the initial function was developed:

(1) Calculate the average surface velocity:

VQarea of susurfrface (2) The total behaves like:

Vzzsurfsurf far above the sources sources near the source Vzzsurfsurfs far below the source (3) From (2) above use:

Vzzsurfsurf s above the source Vzzsurfsurfs s , zz s12" Vzzsurfsurfs , zz s12" The prescription is designed to obtain the correct away from the sources quickly, and not to introduce any singular parts into near the sources. Naturally, such estimates are arbitrary, but absolute accuracy of the initial estimates are not required anyway.

3A.B.6 CONVERGENCE AND ACCURACY

This section is concerned with the accuracy of the finite differ ence solution. A good numerical solution is one that accurately approximates the exact solution.

The approach of the numerical solution to the exact solution as the grid is refined is called convergence of the numerical solution. If an iterative scheme is used, as in our case, converg ence of the iterations is defined by the difference between any two iterations ap proaching zero as the number of iterations is increased.

For this scheme, theory assures us th at both kinds of convergence take place (Reference 3A.B-3) provided 0<<2 (see Section 3A.B.3) and provided sm all enough grids and large enough number of itera tions are used. The real ques tions, of course, are whether small enough grids and enough iterations have been used to guarantee an accurate enough approximate solution.

C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.B-10 To check convergence of the num erical solution LOCA bubble charging, SRV, etc. were recalculated with several different grid sizes until the change in the solution was insignificant.

Figure 3A.B-4 shows the results of several different grids.

F i gure 3A.B-1 shows the geometry for a single SRV case.

Convergence of iterations is ea sier to check by printing the total percent change between two successive iterations, and ascertaining that this change is not more than 0.001% (for single

bubble, 300 iterations, for LOCA, 2000 iterations). Note that this suggests accuracy, but does not guarantee it, for even such small changes can be accumulated over a large number of steps and can cause divergence of the solution.

(For example: 1 n diverges as M approaches infin ity but the percen t change between successive terms n=1 approaches zero.)

There are additional ch ecks made, namely, overall conserva tion. To do this, the flow nds over all the surface areas due to the singular part of the potential, s and the smooth part of the potential, are calculated. On the solid walls, they are equal and opposite and add up to zero (this is exactly so on vertical walls and on the flat botto m, but only closely so on the slant bottom). On the free surface, the sum of the tw o should equal exactly the total of all the sources. For the num ber of iterations performed the solutions are always within 0.1%. In addition, the integral VdV 2 is calculated. For the smoot h part, this should equal zero, while for the singular part, this shoul d equal the total flow of N sources, N4. The first is accurate always to within 0.1 in.

3/sec, the latter agrees with the sum of the free surface flows to within 0.1%. This however, is not coincidental, since the finite difference approximation of the Laplacian ope rator has been chosen to be "conservative", i.e., the Gauss theorem holds for the difference approximation.

Finally, the integral, VdV 2 3, is also calculated and va nished to within 0.1 in.

3/sec for the smooth part. This same integral is calculated for the singular part and should also equal the total source flows which in fact it does not. This is unimportant, sin ce the singular part is never calculated by finite differ ence except at the boundary, and th e finite difference Laplacian is not expected to be accura te for the singular functions.

3A.B.7 RESULTS

In this section, some computed results for two representative ca ses are presented and analyzed: (a) slanted bottom pool, LOCA bubble charging event and, (b) flat bottom pool, LOCA bubble charging event.

C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.B-11 In all the velocity results presented, the source strengths have been normalized to:

Q410000,sec in.3 i.e., the flow velocity is 100 in

./sec at a radius of 10 in. from the source center, or the total flow is 410 4 in.3sec. However, to keep the potentials in a more manageable range, the printed values are 1/10 of the potentials corre sponding to the above nor malized source strength (i.e., the reader should multiply all printed potentials by 10 to get the correct value).

3A.B.7.1 Loss-of-Coolant Accident Bubble Charging

Loss-of-coolant accident bubble charging calculations were made under the assumption that all bubbles are of th e same strength (Q410000,). Then by symmetry, it suffices to perform the calculation for a wedge of angle 234106.o radius (since there are three rows of 34 downcomers). By further symme try, it suffices to consider just a half-wedge, or 5.29 o , since the plane exactly halfway between two downcomer planes is a symmetry plane. Figure 3A.B-2 gives the geometry of the LOCA bubble chargi ng calculations. The fl at bottom calculations used 16 points in r, 26 points in z, and 10 points in ; corresponding to grid sizes of r=23.75 in., =0.01155 rad , and z=15.26 in.

Figure 3A.B-5 shows the potential distribution at a plane near the sources and Figure 3A.B-6 shows the same at the plane midway between two source planes. Figures 3A.B-7 and 3A.B-8 show the velocity in the planes 0 o and 529.o , respectively.

There are some physically interestin g features revealed in these calculations. (a) In contrast to one or a few single bubbles, when LOCA bubbles grow in phase, flow below the source is essentially negligible. (b) There is very little variation in the flow pattern in the azimuthal () direction, except of course, at the exact elev ation of the sources. (c) Although the sources have the same strength, the innermost source ha s a stronger influence on the flow than does the outermost one, simply because of the increased density of the bubbles as one approaches the pedestal. (d) From calculations for three-dimensional LOCA flow for the flat bottom case (Figure 3A.B-9) the smallness of the velociti es (or essentia lly constant values) below the source levels, indicates that the effect of the slant bottom is negligible.

3A.B.8 REFERENCES

3A.B-1 Analytical Model for Estimating Drag Forces on Rigid Submerged Structures Caused by LOCA and SRV Ramshead Air Discharges, General Electric Company, NEDE-21471 (Propr ietary), September 1977.

3A.B-2 Theoretical Hydrodynamics , by L. M. Milne-Thompson, the Macmillan Company, N.Y., 1950.

C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.B-12 3A.B-3 Analysis of Numerical Methods , by E. Isaacson and H. B. Keller, John Wiley & Sons, Inc., N.Y., 1966.

2) Indicates Source/Bubble Location Figure Amendment 53November 1998 Form No. 960690Draw. No.Rev.Geometry for Single SRV Case 900547.82 3A.B-1 = 0 =180 C L Z C LContainment WallPedestal WallA) Plan View of Suppression Pool Showing Single SRV Symmetry About = 0 to 180 B) Conceptualization Showing

Geometry that was Modeled =180 j Max i Max i,j,k = 1 r 395.125" 165.0" k Max = 0 Notes: 1) Indicates Downcomer Location

3) C Indicates Suppression Pool Centerline4) Indicates Velocity Potential Grid Points L Columbia Generating StationFinal Safety Analysis Report A) Plan View of Suppression Pool Showing Downcomer Symmetry About = 0 Figure Amendment 53November 1998 Form No. 960690Draw. No.Rev.Geometry for LOCA Bubble Charging Case 900547.81 3A.B-2 C L Z/2 C L = 0 32.5 C L = 0Pedestal WallContainment WallContainment Wall B) Conceptualization Showing Geometry that

was Modeled iMax, jMax, kMax i,j = 1 k = kMax 217.75" r 1 = 276" r 2 = 360" r 3 = 444" Notes: 1) Indicates Downcomer Location

2) C Indicates Suppression Pool Center Line

3) Indicates Source/Bubble Location

4) Indicates Velocity Potential Grid Points LPedestal Wall Columbia Generating StationFinal Safety Analysis Report Figure Amendment 53November 1998 Form No. 960690Draw. No.Rev.Velocity Grid 900547.83 3A.B-3 r ri, j, k +1 k j i k j ir i, j+1, k+1 i+1, j+1, k+1 i+1, j, k i,j,kA) Relationship of Velocity Potential Grid Versus Velocity Grid

øøøøøz v i, j, k i +1, j, k +1 v i,j,k B) Conceptualization of How v Is Calculated.ø's Averaged on This Wall

Equal ø1ø Averaged on This Wall

Equal ø2ø2=(øi, j +1, k +øi, j +1, k +1

+ ø i +1, j +1, k +1

+ø i +1, j +1, k

)/4ø1-ø2 r v =øø1=(øi, j, k +øi +1, j, k

+ ø i +1, j, k +1

+ ø i, j, k +1

)/4ø i, j +1, k Columbia Generating StationFinal Safety Analysis Report Figure Amendment 53November 1998 Form No. 960690Draw. No.Rev.Grid Convergence Test 900547.80 3A.B-4 Z Grid Size - imax, jmax, kmax Fine Grid - 31, 31, 31Medium Grid - 11, 31, 11

Coarse Grid - 9, 15, 9 0.5 1.0 Elevation BasematNormalized Total Velocity at 0 = 0 and R = 182 (1/1)Columbia Generating StationFinal Safety Analysis Report

Draw. No.Rev.Figure

Draw. No.Rev.Figure

Draw. No.Rev.Figure

Draw. No.Rev.Figure

Draw. No.Rev.Figure C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.C-1 Attachment 3A.C

CONCEPT OF DRAG FORCES DUE TO HYDRODYNAMIC FLOW FIELDS

3A.C.1 CONCEPT

The concept of drag forces is described in Reference 3A.C-1 as a means to estimate loads on submerged structures due to flow fields crea ted in a Mark II containment suppression pool by the hydrodynamic events described in Section 3A.3. Loads resulting from the actual distorted flow around a structure may be estimated by postulating an equi valent locally uniform flow field due to the safe forcing f unction in the pool without any stru ctures. This uniform flow is characterized by the velocity and acceleration fields present at the geometric center of the structure or structural segment. The loads on submerged structures are characterized by drag forces due to locally uniform velocity and acceleration fields. The velocity field causes a standard drag force and a lift force, and the acceleration field causes an acceleration drag force. The total load on the structure or structural co mponent is obtained by the vectorial summation of these forces.

Information essential for calculating the drag loads is identified below.

3A.C.2 FORMULAS FOR DRAG LOADS

In the following three sections, formulas are presented to calculate velocity drag load, acceleration drag load, and lif t load. The methodol ogy described belo w is in general agreement with Reference 3A.3.2-4.

Long structures are divided into segments for mo re precise evaluation.

This is done to account for the variations of the velocity and acceleration al ong the structure.

3A.C.2.1 Velocity Drag Load

The velocity drag load is calculated using the following formula:

P sD1 2 C Vmax 2 where: P s = velocity drag pressure amplitude (psi). This pressure acts in the flow direction.

= mass density of water (lb sec 2/in.4).

C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.C-2 C D = standard drag coefficient. Numerical values for C D are given in the applicable sections of Section 3A.3. V max = maximum velocity in the direction of flow (in./sec).

3A.C.2.2 Acceleration Drag Load

The acceleration drag load is calc ulated using the following formula:

PD V A CMmax 2 4 where:

A = acceleration drag pressure amplitude (psi). This pressure acts in the flow direction.

C M = acceleration drag coefficient. Numerical values for C M are given in the applicable sections of Section 3A.3. D = diameter of cylindrical structure (i n.). If the structure is not cylindrical, D is the diameter of a cylinde r circumscribing the structure.

V max = maximum acceleration in the direction of flow (in./sec 2).

3A.C.2.3 Lift Load

The lift load is calculated using the following formula:

P L1 2 2 C VLmax where:

P L = lift pressure amplitude (psi). Th is pressure is normal to the flow direction.

C L = lift coefficient. Numerical values for C L are given in the applicable sections of Section 3A.3.

C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.C-3 3A.C.3 REFERENCES

3A.C-1 "Analytical Model for Estimating Dr ag Forces on Rigid Submerged Structures caused by LOCA and Safety/Relief Valve Ramshead Air Disc harges," General Electric Company, NEDO -21471, September 1977.

C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.D-1 Attachment 3A.D CALCUL A TION MODELS FOR SHORT-TE R M LOSS-OF-COOLANT ACCIDE N T PHENOMENA

3A.D.1 INTRODUCTION

This attachment provides additional informati on concerning the numerical techniques used to model short term hydrodynamic phe nomena. The vent clearing, pool swell, and loss-of-coolant accident (LOCA) bubble numerical models are discussed in Sections 3A.D.2 , 3A.D.3 , and 3A.D.4 respectively. In each section, the model assumptions, equations, numerical techniques, and verifica tion are either discussed in detail or referenced to the appropriate General Electric document.

3A.D.2 DOWNCOMER VENT CLEARING MODEL

The pool swell analytical model (PSAM), (Reference 3A.D-1), models the pool swell event subsequent to downcomer vent water clearance. Initial cond itions required to start the PSAM include the time of vent clearing and the pool surface displa cement, velocity, acceleration, and wetwell pressure at the time of vent clearing. In order to provide a time history of the suppression pool surface during th e downcomer vent clearing pro cess and a conservative input to the PSAM, the computer code VENT was deve loped. VENT is a s ubroutine for the PSAM computer code SWELL (see Section 3A.D.3). By continuity the downcomer vent exit water velocity and acceleration time histories are also calculated. These transients are used as conservative input to the LOCA water jet code (see Section 3A.3.2.3.1.1

).

Section 3A.D.2.1 discusses the downcomer vent clearing model development and Section 3A.D.2.2 discusses the experimental verificati on and the conservatism of the model.

3A.D.2.1 Model Development

Assumptions used in developing the vent clearing model are as follows:

a. The frictional losses of the pool system are conservatively neglected, b. The wetwell free air volume is isen tropically compressed by the upward moving water slug,

c. Heat losses are neglected,

d. The air velocity within the downcomers is small, therefore the air pressure in the vents is conservativel y assumed to equal the cu rrent drywell pressure, C OLUMBIA G ENERATING S TATION Amendment 55 F INAL S AFETY A NALYSIS R EPORT May 2001 LDC N-0 1-0 0 0 3A.D-2 e. Downcomer vent losses are conservatively neglected, and

f. Viscous effects are neglected.

Figure 3A.D-1 shows a schematic of the vent c l earing model. The mat h ematical derivation of

this model is similar to the model in Reference 3A.D-2 with the excep t i on that the vent clearing model described here couples the equati on of motion for the vent system with the equation of motion for the pool system.

3A.D.2.1.1 Drywell Pressure

The time varying drywell pressure, P D , is the driving function for the vent clearing analysis.

P D is not calculated by VENT but is input as data.

3A.D.2.1.2 Water in the Downcomer Vents

The mass of water within the downcomer vents, m', that is being accelerated downward by the increasing drywell pressure transient is given by:

mHhAwoV (3A.D-1) where w = the density of water A V = the total downcomer vent exit area

H o = the initial submergence of the downcomers

h = the displacement of the in ternal downcomer water surface.

3A.D.2.1.3 Water Slug in the Suppression Pool The mass of the water slug in the suppression pool, m, which is being accelerated upward by the increasing drywell pr essure is given by:

mHzAwop (3A.D-2)

C OLUMBIA G ENERATING S TATION Amendment 59 F INAL S AFETY A NALYSIS R EPORT December 2007 LDCN-06-000 3A.D-3 where

A p = the net suppression pool water surface area

z = the displacement of the pool surface.

3A.D.2.1.4 Suppression Chamber Air Space

From assumption 2, the transient pressure in the suppression chamber air space, P s, is calculated from:

()PPVVssosos k= where

V s = V so - A v h V so = initial wetwell free air space volume P so = initial wetwell air pressure k = specific heat ratio of air.

Combining and solving yields:

()()PPVVAhssososov k= 3A.D.2.1.5 Fluid Dynamics

Refer to Figure 3A.D-1. From Newton's second law, MA

= F, the equation of motion for the water inside of the vents is:

()()==m dh dtPPAAHhgDvwvo 2 2 where P = the pressure at the downcomer vent exit.

C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.D-4 Combining with equation 3A.C-1 and solving for PP D w yields: ()()PP Hh dn dtHhg D w oo=2 2 (3A.D-3)

Also from Newton's second law, the equation of motion for the wa ter outside of the vents is:

()()m dz dtPPAAHzgspwpo 2 2=+ Combining with equation 3A.D-2 and solving for PP s w yields: ()()PP Hz dz dtHzg s w oo=+++2 2 (3A.D-4)

Substituting h A v/A p for z (see Figure 3A.D-1) in equation 3A.D-4 and then summing equations 3A.D-3 and 3A.D-4 and solving for d 2 h/dt 2 yields: ()()()()dh dt PP g h HAAhAA Ds wovpvp 2 2 2 11=+1+AA vp (3A.D-5)

Integration of d 2 h/dt 2 yields the downcomer vent water ve locity, dh/dt, and displacement, h, transients:

dh d dh dt t 5 2 2 0= dt (3A.D-6) h dh dt dt t=0 (3A.D-7) where

t = time after LOCA initiation.

C OLUMBIA G ENERATING S TATION Amendment 54 F INAL S AFETY A NALYSIS R EPORT April 2000 LDC N-9 9-0 0 0 3A.D-5 The pool surface acceleration, velocity, and displacement transients are related by continuity to the vent wa t e r transien t s by a factor of A v/A p (see Figure 3A.D-1

). It is important to note that at h = H o, equation 3A.D-5 is the same as the equation of motion used in the PSAM (Reference 3A.D-7). This implies that equation 3A.D-5 plus the PSAM will provide a continuous and consistent time history of the suppression pool surface displacement during a LOCA.

3A.D.2.1.6 Numerical Integration

Sections 3A.D.2.1.1 and 3A.D.2.1.4 show that the drywell pressure, P D , and the wetwell air space pressure, P S are functions of time and vent water displacement, respectively. From equation 3A.D-5 this shows that the downcomer vent water acceleration, d 2 h/dt 2 , is a function of time and vent water displacem ent only. This means that equation 3A.D-5 is a second order differential equation of the functional form: d 2 h/dt 2 = f(t,h), where h = h(t) by equation 3A.D-7. This allows numerical integration of equation 3A.D-5 using a fourth order Runge-Kutta technique given by e quation 25.5.22 of R e ference 3 A.D-3. Integration of d 2 h/d t 2 gives dh/dt and h (equations 3A.D-6 and 3A.D-7, respectively).

3A.D.2.1.7 Termination of Vent Clearing Analysis Termination of the vent clea ring analysis occurs when H o - h O. This is the moment that the vent clearing is completed; or t o , vent clearance time.

3A.D.2.1.8 Input Data and Results

Input to the VENT subroutine requires data on the following plant characteristics: net pool area (A p), total downcomer vent exit area (A v), initial submergence of the downcomers (H o), initial wetwell pressure (P so), initial wetwell free air volume (V so), and the drywell pressure transient (P D). Table 3A.3.2-3 shows the CGS input data for the VENT computer code.

The vent exit water velocity and acceleration calculated by VENT are increased by 10% as indicated by the NRC in Reference 3A.3.2-1. The velocity and acceleration time histories (including the 10% inc r ease) a r e shown in Figures 3A.3.2-2 and 3A.3.2-3 , respectively.

3A.D.2.2 Experime ntal Verification

VENT has been verified against downcomer vent water displacement data from the 4T Test Series 5101, runs 21, 22, 24, and 37 (Reference 3A.D-4). These runs were chosen because they ran the full range of Mark II submergences and drywell pressurization rates. In each run, three conductivity probes were used to determine the displacement of the downcomer vent air/water interface. These probes, shown in Figure 3-3 of Reference 3A.D-5, sense the difference in conductivity between air and water. For each test, the three probes were located C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.D-6 in each downcomer at 0.5 ft, 6.5 ft, and 9.5 ft above the downcomer ve nt exits (see Figure 3-2 of Reference 3A.D-5). Tables 3A.D-1 and 3A.D-2 show the input data to VENT for the verification runs. Table 3A.D-3 shows the measured and calculated 4T Test downcomer vent probe water clearing times.

Figure 3A.D-2 summarizes the data in Table 3A.D-3 and shows that the comparison is excellent.

3A.D.3 POOL SWELL ANALYTICAL MODEL

In order to conservatively calculate the pool swell transient, the computer code SWELL was developed after the model discussed in References 3A.D-1 , 3A.D-6 , and 3A.D-7. The equations used in the SWELL computer code are documented in those references. The PSAM

is schematically shown in Figu r e 3 A.D-3 and its verification a g ainst empirical data and its conservatism is discussed in References 3A.D-1 , 3A.D-6 , 3A.D-7 , and 3A.D-8. Input to SWELL requires data on the following plant char acteristics: net pool area, total downcomer vent exit area, initial submerge nce of the downcomers, initial drywell air pressure, initial wetwell air pressure, initial dr ywell air temperature, initial wetwell free air volume, initial drywell humidity, downcomers loss coefficient, time of vent clea ring, vent clearing velocity, and drywell air pressure transient.

Figures 3A.D-4 through 3A.D-9 are plots of p o ol surface v e locity and pool surface elevation obtained with SWELL for the three benchm ark plants presented in Reference 3A.D-7. (Note:

for these verification runs, the vent cl earing subroutine described in Section 3A.D.2 is not used since the vent clearing velocity and time are given in Reference 3A.D-7.) These plots are provided for comparison with th e data included in Reference 3A.D-7 as benchmark problems for the SWELL code and show good agreement.

3A.D.4 LOSS-OF-COOLANT ACCIDENT BUBBLE CHARGING MODEL

The one-dimensional PSAM (see Section 3A.D.3) describes the bulk flow process in the suppression pool during a postulate d pool swell event. This mode l assumes a flow field in the vertical direction only. The assumption of pred ominately vertical flow has been verified by small scale multivent pool swell tests. However, obs ervation of these tests have shown that prior to LOCA bubble coalescence and the form ing of an air blanke t under the pool water slug, a significant three-dimensional flow fi eld is developed. In response to these observations, new analytical te chniques were developed in or der to model the LOCA bubble charging event. The purpose of the LOCA bubble charging model, therefor e, is to describe the three-dimensional flow fields during the early portion of the pool swell phenomenon.

In order to calculate the tran sient flow fields in the CGS suppression pool during the LOCA bubble charging portion of a postulated pool sw ell event, the computer code SOURCE was developed (see A.B

). The application of the SOURCE code to the LOCA bubble C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.D-7 charging phenomenon is schemati c ally shown in Figure 3A.B-2. This method uses point sources with the appropriate s ource strength to represent the LOCA bubble charging event in the exact CGS suppression pool geometry. In us ing this method it is assumed that all vents uniformly charge air into spherical bubbles with cente rs one downcomer radi us below the vent exits. To calculate th e transient LOCA bubble ch arging flow field, the rate of bubble growth is determined by continuity from the pool surface rise obtained from the PSAM.

A comparison of the similarity between this method and the method discussed in Reference 3A.D-9 is provided in Table 3A.D-4. 3A.D.4.1 Potential Flow Field The three-dimensional potential fl ow field calculation method that is the basis for the SOURCE computer code is described in A.B. Also described in A.B are the numerical techniques, the flow field calculation procedure, and the conservation, convergence, and accuracy checks of the method. Results for the CGS LOCA bubble charging case is discussed in Sections 3A.B.7 and 3A.3.2.

Although both methods use the same assumptions of potential flow, point sources to represent bubbles, and uniformly charging spherical bubbles, the SOURCE code is used instead of the method of images (MOI) (Reference 3A.D-9) to determine finite pool e ffects. This is because the SOURCE code models the exact CGS suppr ession pool geometry, whereas the MOI has to idealize the pool's annular boundaries and sloping floor characteristics.

3A.D.4.2 Source Strength Calculation

The rate of air charging (and, therefore, the ra te of bubble radius grow th) is determined by continuity from the pool surface rise obtained from the PSAM. In using this method, it is assumed that all vents uniformly charge air into spherical bubbles with cen ters 1 ft 0 in. below the vent exits. The PSAM method of calcul ating a transient source strength for use in determining the flow field during LOCA bubble ch arging is used for the CGS instead of the method presented in Reference 3A.D-9. Source strengths calculat ed using both methods are presented in the following secti ons where it is shown that the PSAM method is preferable to the method of Reference 3A.D-9 in that it is more conser vative and has experimental verification. General Electric has developed a method to cal culate the equivalent bubble charging velocity and acceleration source strengths for a point source in a finite pool. It is described in Reference 3A.D-9. Air bubbles at the downcomer vents during the LOCA bubble charging process are assumed to be spherical. The bubble radius growth time history is obtained by assuming the bubble dy namics are represented by th e Rayleigh equation coupled with a mass and energy balan ce for the bubble. Because the Rayleigh equation models the bubble dynamics in an infinite pool, a factor "K" must be solved for each bubble to correct for finite pool boundary effects. This factor is then multiplied by the Rayleigh bubble velocity and acceleration source strengths to solve for the finite pool velocity and acceleration source strengths at each bubble as s hown in Figure 3A.D-10. Table 3A.D-5 p r ovides some CGS C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.D-8 LOCA bubble charging pool surface velocities and accelerations obtained using the source strength method or Reference 3A.D-9 along with the SOURCE computer code. Extensive small scale multivent pool swell tests have shown that the pool surface remains relatively flat during the LOCA bubble charging process. For CG S, the pool surface transient is calculated by the PSAM (see Section 3A.D.3). It is evident in References 3A.D-1 and 3A.D-7 that the PSAM estimates of the pool surf ace transient during the early por tion of pool swell (which is LOCA bubble charging) consisten tly bounds all experimental da ta. The small scale multivent pool swell tests also indicate bubble sphericity during the early portio n of the transient.

With these observations it is possible to obtai n the bubble velocity and acceleration source strengths from the PSAM calculated pool surface transient by continuity as shown in Figure 3A.D-11. Table 3A.D-5 provides some LOCA bubble c h arging pool surface velocities and accelerations obtained fo r CGS using the Reference 3A.D-9 method and the PSAM methods for comparison. It is seen that while pool surface accelerations are similar, the pool surface velocities from the Reference 3A.D-9 method are less than 50% of the PSAM values.

Therefore, for CGS the PSAM method is conservatively accepted for LOCA bubble source strength definitions.

Figure 3A.3.2-7 shows the vel o city, Q(t) and acceleration, (), Qt source strengths which are used in CGS load calculations.

3A.D.5 REFERENCES 3A.D-1 Ernst, R. J., Ward, M. G., Ma rk II Pressure Suppression Containment Systems: An Analytical Model of the Pool Swell Phenomenon, General Electric Company, NEDE-21544-P (Proprietary), December 1976.

3A.D-2 The General Electric Pressure Suppression Contai nment Analytical Model, General Electric Company, NEDM-10320, April 1971.

3A.D-3 Handbook of Mathem atical Functions with Formulas, Graphs, and Mathematical Tables , National Bureau of Standa rds - Applied Mathematics Series .55, May 1968.

3A.D-4 Phases I, II, and III of the Te mporary Tall Tank Test (4T) Program, an Applications Memorandum, Preliminary Draft, General El ectric Company, December 1976.

3A.D-5 Mark II - Pressure Suppression Te st Program, General Electric Company, NEDE-13442-P-01, May 1976.

C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.D-9 3A.D-6 Response to NRC Question 020.68, Appendix A, WNP-2, DAR, Revision 2, August 1979.

3A.D-7 Mark II Containment Dynamic Forc ing Functions Information Report (DFFR), General Electric Company, NEDE-21061, Revision 3, June 1978.

3A.D-8 Comparison of the 1/13 Scale Mark II Containment Multivent Pool Swell Data with Analytical Methods, General Electric Company, NEDO-21667, August 1977.

3A.D-9 Moody, F. J., Analytical Mode l for Estimating Drag Forces on Rigid Submerged Structures Caused by LOCA and Safety Relief Valve Ramshead Air Discharges, General Electric Co mpany, NEDE-21471 (Proprietary), September 1977.

C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.D-11 Table 3A.D-1 Vent Clearing Analyt i cal Model Input Data

Parameter 21 22 24 37 Net pool are a a , ft 2 35.17 35.17 35.17 35.17 Downcomer vent flow area, ft 2 2.0211 2.0211 2.0211 2.0211 Downcomer submergency, ft 13.15 9.0 13.5 11.0 Integration time step, sec 0.0001 0.0001 0.0001 0.0001 Initial wetwell free air volume, ft 3 950 1108 950 1038 Drywell pressure time history, psia See Table 3 A.D-2 a Excludes area of downcomer.

C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.D-12 Table 3A.D-2 Measured 4T Test Series 5101 Drywell Dome

Pressure T i me History Run 21 Run 22 Run 24 Run 37 Time (sec) Pressure (psia) Time (sec) Pressure (psia) Time (sec) Pressure (psia) Time (sec) Pressure (psia) 0. 14.63 0. 14.57 0. 14.55

0. 14.54 0.037 14.65 0.037 14.80 0.040 14.53 0.041 14.52 0.055 15.12 0.058 15.30 0.058 14.66 0.060 14.60 0.076 15.61 0.079 15.87 0.076 15.35 0.078 14.93 0.097 15.73 0.097 16.16 0.097 16.12 0.096 16.04 0.115 15.91 0.115 16.42 0.113 16.52 0.114 17.28 0.134 16.18 0.133 16.70 0.133 17.11 0.133 17.91 0.152 16.42 0.152 17.01 0.154 17.80 0.153 18.46 0.170 16.54 0.170 17.47 0.173 18.41 0.174 19.35 0.189 16.80 0.191 17.98 0.191 19.04 0.193 19.94 0.210 17.27 0.212 18.47 0.210 19.54 0.211 20.44 0.230 17.80 0.230 10.89 0.228 20.03 0.230 20.97 0.249 13.10 0.249 19.42 0.246 20.52 0.248 21.54 0.267 18.41 0.267 19.79 0.267 21.07 0.266 22.21 0.285 13.79 0.285 20.33 0.288 21.67 0.287 23.00 0.303 19.08 0.304 20.80 0.306 22.22 0.308 23.77 0.322 19.32 0.324 21.29 0.324 22.73 0.326 24.40 0.343 19.70 0.345 21.87 0.343 23.19 0.344 24.99 0.363 20.03 0.364 22.36 0.361 23.64 0.363 25.54 0.382 20.37 0.382 22.95 0.380 24.11 0.381 26.16 0.400 20.64 0.400 23.29 0.400 24.68 0.400 26.77 0.419 20.94 0.419 23.72 0.421 25.12 0.420 27.40 0.437 21.27 0.437 24.23 0.440 25.61 0.441 27.91 0.455 21.69 0.458 24.75 0.458 26.10 0.460 28.40 0.476 22.06 0.479 25.34 0.476 26.68 0.478 29.00 0.497 22.46 0.497 25.85 0.494 27.21 0.496 29.63 0.515 22.81 0.515 26.30 0.513 27.78 0.514 30.24 0.534 23.15 0.533 26.78 0.533 28.33 0.533 30.81 0.552 23.46 0.552 27.23 0.554 28.87 0.553 31.40 0.570 23.80 0.570 27.76 0.573 29.32 0.574 31.95 0.539 24.13 0.591 28.14 0.591 29.79 0.593 32.43 0.610 24.51 0.612 28.61 0.610 30.21 0.611 32.90 0.630 24.88 0.630 28.99 0.628 30.70 0.630 33.37 0.649 25.26 0.649 29.40 0.646 31.10 0.668 33.79 0.667 25.59 0.667 29.81 0.667 31.63 0.666 34.22 0.685 25.91 0.685 30.21 0.688 32.14 0.687 34.80 0.703 26.22 0.704 30.68 0.706 32.57 0.708 35.35 0.722 26.58 0.724 31.16 0.724 33.01 0.726 35.78 C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.D-13 Table 3A.D-2 Measured 4T Test Series 5101 Drywell Dome Pressure Time History (Continued)

Run 21 Run 22 Run 24 Run 37 Time (sec) Pressure (psia) Time (sec) Pressure (psia) Time (sec) Pressure (psia) Time (sec) Pressure (psia) 0.743 27.01 0.745 31.61 0.743 33.44 0.744 36.23 0.763 27.33 0.764 31.98 0.761 33.90 0.763 36.71 0.782 27.61 0.782 32.32 0.780 34.37 0.781 37.08 0.800 27.86 0.800 32.58 0.800 34.78 0.800 37.46 0.819 29.18 0.819 32.87 0.821 35.20 0.820 37.79 0.837 28.39 0.837 33.11 0.840 35.49 0.841 38.05 0.855 28.81 0.858 33.31 0.858 35.81 0.860 38.25 0.876 29.12 0.878 33.52 0.876 36.13 0.878 38.35 0.897 29.48 0.897 33.68 0.894 36.52 0.896 38.52 0.915 29.78 0.915 33.80 0.913 36.84 0.914 38.68 0.934 30.07 0.933 33.88 0.933 37.17 0.933 38.78 0.952 30.39 0.952 33.96 0.954 37.47 0.953 33.84 0.970 30.66 0.970 34.36 0.973 37.74 0.974 38.88 0.989 30.92 0.991 34.10 0.991 37.96 0.993 38.98 1.010 31.29 1.102 34.15 1.010 38.18 1.011 39.00 1.030 31.57 1.030 34.21 1.028 38.31 1.030 39.04 1.049 31.87 1.049 34.19 1.046 38.41 1.048 39.06 1.067 32.08 1.067 34.17 1.067 38.57 1.066 39.08 1.085 32.26 1.085 34.17 1.088 38.59 1.087 39.11 1.103 32.54 1.104 34.18 1.106 38.61 1.108 39.10 1.122 32.67 1.124 34.08 1.124 38.65 1.126 39.08 1.142 32.85 1.145 34.38 1.143 38.65 1.144 39.08 1.163 33.07 1.169 34.32 1.161 38.63 1.163 39.06 1.182 33.19 1.182 33.90 1.180 38.61 1.181 39.06 1.200 33.33 1.200 33.88 1.200 38.61 1.200 39.06 1.219 33.38 1.218 33.80 1.221 38.55 1.220 39.02 1.237 33.44 1.237 33.76 1.240 38.49 1.241 38.94 1.255 33.50 1.258 33.64 1.258 38.43 1.260 38.92 1.276 33.60 1.278 33.54 1.276 38.35 1.278 38.90 1.297 33.60 1.297 33.50 1.294 38.26 1.296 38.88 1.315 33.66 1.315 33.44 1.313 38.20 1.314 38.84 1.333 33.62 1.333 33.35 1.333 38.08 1.353 38.89 1.352 33.56 1.352 33.29 1.354 37.98 1.353 38.89 1.370 33.46 1.370 33.19 1.373 37.92 1.374 38.82 1.389 33.42 1.391 33.13 1.391 37.78 1.393 38.82 C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.D-14 Table 3A.D-3 Comparison of Measured and Calculated 4T Test

Probe Water Clearing Times

Probe Elevation Above Downcomer Exit (ft)

Probe Water Clearing Time (sec) 21 a 22 a 24 a 37 a 9.5 0.667 Initially 0.573 0.381 (0.687) dry (0.582) (0.388) 6.5 0.800 0.458 0.688 0.533 (0.811) (0.467) (0.688) (0.533) 0.5 0.952 0.667 0.800 0.687 (0.961) (0.654) (0.813) (0.670) a 4T run number.

Note: Unbracketed numbers are measured data. Bracketed Numbers ar e calculated data.

C OLUMBIA G ENERATING S TATION Amendment 54 F INAL S AFETY A NALYSIS R EPORT April 2000 LDC N-9 9-0 0 0 3A.D-15 Table 3A.D-4 Comparison Between Source Method

and the Re f e rence 3A.D-9 Method for Calculation of the Los s-o f-Coolant Acc i dent Bubble Charging Event Item Source Reference 3A.D-9 Comments 1. Uses potential flow assumption Yes Yes Same 2. Uses point source to represent charging LOCA bubbles Yes Yes Same 3. All vents assumed to charge uniformly into spherical bubbles one radius below downcomers Yes Yes Same 4. Finite pool effects Uses numerical scheme discussion in Attachment

3A.B of this Report Uses MOI - 5. Models CGS annular suppression pool geometry Yes No For MOI, the CGS annular pool geometry must be idealized into a rectangular pool using an "equivalent radius" concept. 6. Models CGS sloping pool bottom Yes No For MOI, the CGS sloping pool bottom needs to be idealized into a flat pool bottom. 7. Transient source strength Determined by continuity from pool surface rise obtained from the PSAM Determined from Rayleigh bubble dynamics equation

in an infinite pool and a finite pool correction factor "K" Source method results in the same conservatism as PSAM flow field calculations. Source method yields higher velocities

and accel e r ations than the method discussed in Reference 3A.D-9. 8. Experimental verification of source strength Yes No Source method determined directly from PSAM. PSAM has extensive experimental verification as to its conservatism.

C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.D-16 Table 3A.D-5 Comparison of Results of Source and Method

of Reference 3A.D-9 Bubble Charging Source Strength Methods

GE Metho d a PSAM Method b Time After Vent Clearing (sec)

V(ft/sec) V (ft/se c 2) V(ft/sec) V (ft/se c 2) 0. 1.215 83.1 3 6 5.308 80.977 0.06 6 c 4.872 81.767 10.613 79.772 0.24 d - - 22.074 53.983

V = average pool sur f ace velocity V = average pool surface acceleration a Source strength method documen t e d in Reference 3A.D-9. Flow field data from SOURCE code. b Data obtained from F i gures 3A.3.2-4 through 3A.3.2-7. c Near bubble coa l e s cence by Reference 3A.D-9 method. d Bubble coalescence time by SOURCE.

Figure Amendment 53November 1998 Form No. 960690Draw. No.Rev.Schematic Representation of the Vent Clearing Model 900547.84 3A.D-1 th PDrywell Air Pressure, PD Pressure of

Suppression Chamber Air

Space, PS Downcomer VentTransient Pool Surface Initial Pool Surface Suppression Pool+z Ho By Continuity:z = h Av/Ap=Av/Ap=Av/Ap dz dt dh dt dt 2 d z 2 d 2 h dt 2 Columbia Generating StationFinal Safety Analysis Report Figure Amendment 53November 1998 Form No. 960690Draw. No.Rev.Comparison of Measured and Calculated 4T TestProbe Water Clearing Times 900547.85 3A.D-2 1.0.9.8.7.6.5.4.3.2.1 00.1.2.3.4.5.6.7.8.91.0 t c = t m Probe 9.5 feet above downcomer vent exit Probe 6.5 feet above downcomer vent exit

Probe 0.5 feet above downcomer vent exit CalculatedTime, t c (Sec)Measured Time, t m (Sec)t c = 1.0047t m + 0.0013 (Least Squares Linear Regression Curve Fit)

Columbia Generating StationFinal Safety Analysis Report Figure Amendment 53November 1998 Form No. 960690Draw. No.Rev.Schematic Representation of the Pool Swell Model 900547.86 3A.D-3 Vw y H oDrywell Air Pressure, PD Pressure of Suppression Chamber

Air Space, PSBubble Volume VBDowncomer VentPool Surface AtAny Time, t t = 0 InitialWater Slug Air Slug Pressure, PB Suppression Pool Pool Surface Columbia Generating StationFinal Safety Analysis Report Figure Amendment 53November 1998 Form No. 960690Draw. No.Rev.Benchmark I Plant -Slug Velocity Versus T ime 900547.87 3A.D-4Time after LOCA (Sec) 36.00 32.00 28.00 24.00 20.00 16.00 12.00 8.00 4.00 0.000.300.600.901.201.501.802.10 0.00 Columbia Generating StationFinal Safety Analysis Report Slug Velocity (Ft/S)

Figure Amendment 53November 1998 Form No. 960690Draw. No.Rev.Benchmark I Plant -Slug Velocity Versus Displacement 900547.88 3A.D-5 Slug Displacement (Ft) 36.00 32.00 28.00 24.00 20.00 16.00 12.00 8.00 4.00 0.003.006.009.0012.0015.0018.0021.00 0.00 Columbia Generating StationFinal Safety Analysis Report Slug Velocity (Ft/S)

Figure Amendment 53November 1998 Form No. 960690Draw. No.Rev.Benchmark II Plant - Slug Velocity Versus T ime 900547.89 3A.D-6Time after LOCA (Sec) 36.00 32.00 28.00 24.00 20.00 16.00 12.00 8.00 4.00 0.000.300.600.901.201.501.802.10 0.00 Columbia Generating StationFinal Safety Analysis Report Slug Velocity (Ft/S)

Figure Amendment 53November 1998 Form No. 960690Draw. No.Rev.Benchmark II Plant -Slug Velocity Versus Displacement 900547.90 3A.D-7 Slug Displacement (Ft) 36.00 32.00 28.00 24.00 20.00 16.00 12.00 8.00 4.00 0.003.006.009.0012.0015.0018.0021.00 0.00 Columbia Generating StationFinal Safety Analysis Report Slug Velocity (Ft/S)

Figure Amendment 53November 1998 Form No. 960690Draw. No.Rev.Benchmark III Plant -Slug Velocity Versus Time 900547.91 3A.D-8Time after LOCA (Sec) 36.00 32.00 28.00 24.00 20.00 16.00 12.00 8.00 4.00 0.000.300.600.901.201.501.802.10 0.00 Columbia Generating StationFinal Safety Analysis Report Slug Velocity (Ft/S)

Figure Amendment 53November 1998 Form No. 960690Draw. No.Rev.Benchmark III Plant -Slug Velocity Versus Displacement 900547.92 3A.D-9 Slug Velocity (Ft/S) 36.00 32.00 28.00 24.00 20.00 16.00 12.00 8.00 4.00 0.003.006.009.0012.0015.0018.0021.00 0.00 Columbia Generating StationFinal Safety Analysis Report Figure Amendment 53November 1998 Form No. 960690Draw. No.Rev.Method of Images 900547.93 3A.D-10 Infinite Pool:

Steps i) Spherical Bubble Dynamics Governed By Rayleigh Equation:ii) From R, R, And R Get Source

Strengths For Each BubbleApplication To Finite Pool:

iii) Determine "K i " Factor for Each Bubble to Account for

Finite Pool and Other Bubble Effects iv) Modify Source Strength for Each Bubble to Account for

Finite Pool and Other Bubble Effects:v) Superposition Of All Bubbles To Get Flow Field:

Velocity V, and Acceleration V Source: Reference 3A.D-9 Containment

Boundary R, R, R...R, R, R....R..R..R dR d 2 R dt 2 2 2 dt R+.R=(P-P)3 gc B...Q i (t) = R 2 2 2 Q i(t) = R+2R

.Q i (t) = R 2 RK i.V = -102i = 1 Q i (t) i V = -102i = 1 Q i (t) i.R(t)Note: R R R.....R Q i (t) = R 2 R + 2RR 2)K i...Columbia Generating StationFinal Safety Analysis Report Figure Not Available For Public Viewing C OLUMBIA G ENERATING S TATION Amendment 54 F INAL S AFETY A NALYSIS R EPORT April 2000 LDC N-9 9-0 0 0 3A.E-1 Attachment 3A.E SUPPRESSION POOL TEMPER A TURE MONITORING SYSTEM

3A.E.1 DESIGN BASES

The suppression pool temperature monitoring (S PTM) system monitors the suppression pool bulk temperature with sensors distributed around the suppression pool. This system provides the main control room operator with the information necessary to avoid the conditions which might lead to the high-temperature steam que nching vibration phenomena mentioned below and discussed in detail in Section 3.5 of Reference 3A.E-1. This phenomena is not expected to occur when using a quencher discharge de vice. However, precautions using the SPTM system are designed to furthe r make the occurrence of the vibrations impossible.

Temperatures in the suppression pool are recorded and alarmed in the main control room.

The design basis for the SPTM sy stem alarm setpoints provides the operator with adequate time to take the necessary action required to ensure that the c onditions which are postulated to lead to high-temperature steam quenching vibra tions do not occur. The design also provides the operator with the necessary information regarding localized heat-up of the pool water while the reactor vessel is being depre ssurized using the safety/relief valves (SRVs) when the SRVs are selected for actuation, they may be chosen so as to ensure mixing and uniformity of heat energy injection to the pool by monitoring the temperature sensors.

3A.E.1.1 High-Temperature Steam Quenching Vibrations

Boiling water reactor plants take advantage of the large ther mal capacity of the suppression pool during plant transients which require SR V actuation. The discharge steam from each SRV is directed through a discharge line and a quencher device to the suppression pool where it is condensed. This results in an increase in pool water temperat ure, but a negligible increase in containment pressure.

However, certain events such as small pipe break have the potential for substantial energy addition to the suppression pool and could result in a high lo cal pool temperature and the phenomenon of steam quenching vibration if timely corrective action is not taken. Suppression pool structural vibrations would occur during this condensing mode which would be forced by the periodic pulsation of the steam jet at the discharge.

The onset of the high-temperature steam quenchi ng vibration phenomenon is a function of both the local suppression pool wate r temperature and the steam ma ss flux rate. The steam mass flux in the SRV piping in turn, is a function of the reactor vessel pressure.

C OLUMBIA G ENERATING S TATION Amendment 54 F INAL S AFETY A NALYSIS R EPORT April 2000 LDC N-9 9-0 0 0 3A.E-2 3A.E.2 SYSTEM DESCRIPTION

The CGS SPTM system conformance to the criteria set by paragraph III.c of Reference 3A.E-2 is as discussed be low. Specifically, the criteria for the upper ring of the sensors are:

1. Each monitoring location has two redunda nt type thermocouples monitored in the control room;

2. There are eight monitoring locations e qually spaced about the outer containment perimeter;

3. The sensors are mounted 7 in. below the minimum technical specification water level;

4. All sensors are monitored and recorded in the control room;

5. Instrument setpoints for alarms wi ll be set at the t echnical specification temperature values such that the plant can be shutdown and depressurized prior to the water in the suppression pool reaching a temperature at which condensation instabilities are postulated to occur; and

6. The SPTM system monitors are Seis mic Category 1, Quality Class 1. The electrical power is Class 1E. Di visional separation is maintained.

In addition to the sensors described above which monitor bulk te mperature, there is a second, lower ring of sensors. The lower ring meets Cr iteria 2, 4, 5, and 6, above. The degree of conformance to Criteria 1 and 3 for the lower ring of sensors is as follows:

1. Each monitoring location has one thermocouple Division 1 and 2 at alternate locations, and

3. The sensors are mounted at el. 447 ft 10.25 in., the approxi mate elevation of the quencher discharge devices.

Since warmer water is more buoyant, the upper ring provides a more conservative value for bulk temperature than the lower ring. The lower ring of sensors is provided to allow the operator to assess if significant vertical thermal stratification occurs.

3A.E.3 REFERENCES

3A.E-1 Mark II Containment Dynamic Forc ing Function Information Report (DFFR), NEDO-21061 , Revision 3, June 1978.

C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.E-3 3A.E-2 Mark II Containment Lead Plant Program Load Evaluation and Acceptance Criteria, Nuclear Regulatory Commission, NUREG-0487, October 1978.

C OLUMBIA G ENERATING S TATION Amendment 59 F INAL S AFETY A NALYSIS R EPORT December 2007 LDCN-06-000 3A.F-1 Attachment 3A.F

COMPUTER PROGRAMS

The following are the programs referenced in this report:

3A.F.1 COMMERCIALLY AVAILABLE PROGRAMS

ANSYS ANSYS is a large scale general purpose finite element computer program used for the solution of several classes of e ngineering analysis problems. Analytical capabilities include: static and dynamic analyses; plastic, creep and swelling analyses; and st eady state and tr ansient heat transfer analyses.

NASTRAN NASTRAN is a large scale general purpose finite element co mputer program used for the solution of several classes of e ngineering analysis problems. Analytical capabilities include: static and dynamic analyses; ther mal analyses; and the determina tion of eigenvalues for use in vibration analyses.

MCAUTO-STRUDL

MCAUTO-STRUDL is a commercially available computer program with general capability for the static and dynamic analysis of structures.

The program, which is se rviced and maintained by the McDonnell Douglas Automation Company, St. Louis, Missour i, has had wide commercial usage for many years.

ADLPIPE ADLPIPE is a commercially available program us ed for the analysis of piping systems.

Analytical capabilities include:

static and dynamic analyses; and thermal analyses including thermal transient and fatigue evalua tions for Class 1 piping systems.

FLUSH FLUSH is a non-linear plain strain finite element seismic analys is program for soil-structure interaction analysis.

C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.F-2 3A.F.2 BURNS AND ROE DEVELOPED PROGRAMS

BESSEL BESSEL is a computer progra m which computes semi-analy tical hydrodynamic added masses (incompressible fluids) for cylindrical and annular geometries

HYDI-1 HYDI-1 is a finite element program which computes hydrodyna mic added masses and incident pressures for compressible fluids in three dimensional geometries.

FOX/HYDI-2

FOX/HYDI-2 is a finite element computer program used for the dynamic analysis for axisymmetric structures. The program perfor ms the analysis by determining structural displacements in the frequency domain.

SWELL This program is disc ussed in detail in A.D.

VENT This program is disc ussed in detail in A.D.

SOURCE This program is disc ussed in detail in A.B.

C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.H-1 Attachment 3A.H CONFORMANCE OF CGS DESIGN TO NRC ACC E P T ANCE CR I TERIA

Table 3A.H-1 is a summary of the CGS position for each of the pool dynamic loads. This table provides a description of each load or phenomenon, the Mark II Owner's Group load specification, the NRC evaluation reference, and the CGS position on the acceptance criteria for each load.

C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT N o v e mb e r 1998 3A.H.3 Table 3A.H-1 Conformance of CGS Design to NRC Accepta n ce Cr iter i a Load or Phenomenon

Mark II Owners Group Load Specification

NRC Evaluation CGS Positi o n on Accep t a nce Criteria I. Loss-of-coolant accident (LOCA)-related hydrodynamic loads A. Subme r ged bounda r y loads during vent clearing 24 psi over

-pressure added to local

hyd r ostatic below vent exit (walls and basemat) - linear attenu a tion to pool surface.

II.A.1 a Accep t a ble B. Pool s w ell loads

1. Pool swell analytical model (PSAM) a) Air bubble pressure Calculated by the PSAM used in calculation of submerged bounda r y loads. III.B.3.a b Accep t a ble b) Pool swell elevation Use PSAM with polytropic exponent of 1.2 to a maximum swell height which is the greater of 1.5 vent submergence or the elevation

correspond i ng to the d r ywell floor uplift P per

NUREG 0487 criteria I.A.4. The assoc i ated maximum wetwell air c o mpression is used for

design assessment.

II.A.2 c Accep t a ble C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT N o v e mb e r 1998 3A.H.4 Table 3A.H-1 Conformance of CGS Design to NRC Acceptance Criteria (Continued)

Load or Phenomenon

Mark II Owners Group Load Specification

NRC Evaluation CGS Positi o n on Accep t a nce Criteria c) Pool swell velocity Velocity history vs. pool elevation predicted by the PSAM used to compute impact l o ading on small st r u ctures a nd drag on gratings between initial pool surface and maximum

pool elevation and steady-state drag between

vent exit and maximum pool elevation.

Analytical velocity va riation is used up to maximum velocity. Maximum velocity

applies the r eafter up to maximum pool swell.

PSAM predicted veloc ities multiplied by a factor of 1.1.

III.B.3.a.3 a Accep t a ble d) Pool swell acce l eration Acceleration predicted by the PSAM. Pool

acceleration is utilized in the calculation of acceleration loads on submerged components

during pool swell.

III.B.3.a.4 b Accep t a ble e) Wetwe l l air compression Wetwell air compression is calculated by

PSAM. II.A.2 c Accep t a ble f) Drywell pressure Methods of NEDM-10320 and NEDO-20533 Appendix B. Utilized in PSAM to calculate pool swell loads.

III.B.3.a.6 b Accep t a ble C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT N o v e mb e r 1998 3A.H.5 Table 3A.H-1 Conformance of CGS Design to NRC Acceptance Criteria (Continued)

Load or Phenomenon

Mark II Owners Group Load Specification

NRC Evaluation CGS Positi o n on Accep t a nce Criteria 2. Loads on submerg e d boundaries Maximum bubble pressure predicted by the

PSAM added uniformly to local hydrostatic

below vent exit (walls and basemat) liner attenua tion to pool surface. Applied to walls up to max i mum pool elevation.

III.B.3.b b Accep t a ble 3. Impact l o ads a) Small structures 1.35 x press u re-velocity correlation for pipes and I beams based on P S TF impulse data and

flat pool assumption.

V a riable pulse

duration.

III.B.3.c.1 b Accep t a ble b) Large structures None - Plant-uni que load where applicable. III.B.3.c.6 b Criteria A.5 a Acceptable.

CGS has no large structures

in the pool swell

zone c) Grating No impact load specified. P drag vs. open area correlation and velocity vs. elevation history from the PSAM. P drag multiplied by

dynamic load factor.

III.B.3.c.3 b Criteria A.3 a Acceptable C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT N o v e mb e r 1998 3A.H.6 Table 3A.H-1 Conformance of CGS Design to NRC Acceptance Criteria (Continued)

Load or Phenomenon

Mark II Owners Group Load Specification NRC Evaluation CGS Positi o n on Accep t a nce Criteria 4. Wetwe l l air compre s s ion a) Wall loads Direct appl i cation to the PSAM calculated pressure due to wetwell compression.

III.B.3.d.1 b Accep t a ble b) Diaphragm floor upward loads 5.5 psid for diaphragm fl oor loadings only.

2.12.7 a Accep t a ble 5. Asymmetric pool LOCA Use 20% of maximum press u re statistically applied to 1/2 of t h e submerged bubble.

II.A.3 c Criteria A-4 a Accep t a ble C. Steam condensation and chugging loads

1. Downcomer lateral loads a) Single v e nt loads (24 in.) Use single vent dynamic lateral load developed in NEDE-23806.

2.3.3.2 a Criter i a B.1.a a Accep t a ble b) Multiple vent loads (24 in.) Use multivent dynamic lateral load developed

in NEDE-24106-P a nd NEDE-24794-P.

2.3.3.3 a Accep t a ble c) Single/

m u ltiple vent loads (28 in.)

Multiply basic vent loads by factor f=1.34 2.3.2.1 a B.1.b a Acceptable C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT N o v e mb e r 1998 3A.H.7 Table 3A.H-1 Conformance of CGS Design to NRC Acceptance Criteria (Continued)

Load or Phenomenon Mark II Owners Group Load Specif i cati o n NRC Evaluation CGS Positi o n on Acceptance Criteria 2. Submerged boundary loads a) High/medium steam flux condensation oscillation (CO) load Use method described in

NEDE-24288-P d 2.2.2.1.3 a CO loads are not gover n ing design condition for CGS b) Low steam flux chugging loads Representative pr e ssure fluctuation

taken from 4TCO (NEDE-24285-P)

test added to local hyd r ostatic 2.2.2.3 a Plant unique. Chugging report entitled "Chugg i ng Loads-Revised. Definition

and Application Methodology

for Mark II Containments"

submitted July 1981 - Uniform loading conditions Use method described in

NEDE-24302-P d See above

- Asymmetric loading Representative pr e ssure fluctuation

taken from 4TCO test

[NEDE-24285-P] applied as

described in NEDE-24822-P.

See above C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT N o v e mb e r 1998 3A.H.8 Table 3AH-1 Conformance of CGS Design to NRC Acceptance Criteria (Continued)

Load or Phenomenon Mark II Owners Group Load Specification NRC Evaluation CGS Positi o n on Acceptance Criteria II. S a fety/relief valve (SRV)-re l a t ed hyd r odyna m i c loads A. Pool temperature limits for X-quencher 20 o F subcooling at quencher elevation for steam mass flux of 42 lb/ft 2-sec or less.

200 oF for steam flux greater than 94 lb/ft 2-sec. 6.2.1.8.8 (5)

A (4) Acceptable B. Quencher air clearing loads Mark II plants utilizing the four arm

quencher, use quenche r load methodology described in DFFR. Criteria II.2 b CGS Plant unique SRV (x-quencher) load report

entitled "SRV Loads -

Definition and Application Methodology for Mark II

Containments" submitted August 1980 C. Quencher arm and tie-down loads Includes vertical and lateral arm load

transmitted to the basemat via the tie-down.

III.C.2.e.2 b Acceptable

1) X-quencher arm loads Vertical and lateral loads developed on the basis of bounding assump tion for air/water discharge from the quencher and

conservative combina tions of maximum/

minimum bubble pressure acting on the

quencher. III.C.2.e.1 Acceptable C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT N o v e mb e r 1998 3A.H.9 Table 3A.H-1 Conformance of CGS Design to NRC Acceptance Criteria (Continued)

Load or Phenomenon Mark II Owners Group Load Specif i cati o n NRC Evaluation CGS Positi o n on Accep t a nce Criter i a 2. X-quencher tie-down loads II.C.1 above plus vertical transient wave and thrust loa d s. Thrust load calculated using a standard

momentum balance.

V e rtical and late r a l moments for air or water

clearing are calculated based on

conservative clearing assumptions.

III.C.2.e.2 b Accep t a ble III. LOCA/SRV submerged structure loads A. SRV air bubble loads

1. Standard drag in Accelerating flow

fields Drag Coefficients are presented in

.k of the Zimmer

FSAR. Acceptable with the following modification:

1) Use C H = C M-1 in the F A formula

2) For noncylindrical structures use lift coefficient for appropriate shape or

C L = 1.6 Generic methodology acceptable.

(Amplitudes for SRV

loads verified by

CAORSO data on

submerged structures).

C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT N o v e mb e r 1998 3A.H.10 Table 3A.H-1 Conformance of CGS Design to NRC Acceptance Criteria (Continued)

Load or Phenomenon Mark II Owners Group Load Specif i cati o n NRC Evaluation CGS Positi o n on Accep t a nce Criter i a 3) The standard drag coef fic i ent f o r pool swell and SRV oscillating bubbles

should be b a sed on data for structures

with sharp edges

2. Equivalent uniform flow velocity and

acce l eration Structures are segmented into small sections such that 1.0 L/D 1.5. The loads a r e then appl i e d to the geometric center of each segment.

Acceptable See III. A.1. above 3. Interference effects A det a iled methodology is presented in Attachment 1.k of the Zimmer

FSAR. Acceptable See III. A.1 above B. LOCA j e t loads Ca l c u l a t ed by the Ring Vortex Model. 2.2.4.3 a Accep t a ble C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT N o v e mb e r 1998 3A.H.11Table 3A.H-1 Conformance of CGS Design to NRC Acceptance Criteria (Continued)

Load or Phenomenon Mark II Owners Group Load Specification NRC Evaluation CGS Position on Acceptance Criteria C. Steam condensation drag loads No generic load methodology provided CGS load specification

and NRC review is

addressed in CGS SER Generic "drag load"

methodology acceptable Plant unique flow fields are

consistent with

I.C.2.a and I.C.2.b of

this table. (See DAR A.I )

IV. Secondary loads A. Sonic wave load Negligible load - none specified Acceptable Acceptable B. Compressive wave load Negligible load - none specified Acceptable Acceptable C. Post swell wave load No generic load prov ided Plant unique load specification addressed

in CGS SER D. Seismic slosh load No generi c load provided Plant unique load specification addressed

in CGS SER

C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT N o v e mb e r 1998 3A.H.12 Table 3A.H-1 Conformance of CGS Design to NRC Acceptance Criteria (Continued)

Load or Phenomenon Mark II Owners Group Load Specification NRC Evaluation CGS Positi o n on Accep t a nce Criter i a E. Fa llback load on submerged boundary Negligible load - none spec ified Acceptable Acceptable F. Thrust l o ads Momentum balance Accep t a ble Accep t a ble G. Friction drag loads on vents Standard friction drag calculations Acceptable Acceptable H. Vent clearing loads Negligible lo ad - none specified Acceptable Acceptable a NRC Acceptance Criteria set forth in NUREG-0808.

b NRC Acceptance Criteria set forth in NUREG-0487.

c NRC Acceptance Criteria set fort h in Supplement 1 of NUREG-0487.

d NRC Acceptance Criteria set forth in WNP-2 SER NUREG (0892).

C OLUMBIA G ENERATING S TATION Amendment 54 F INAL S AFETY A NALYSIS R EPORT April 2000 3A.I-1 Attachment 3A.I SAFETY/RELIEF VALVE AND LOSS-OF-COOLANT ACCIDENT LOADS ON SUBMERGED STRUCTURES

3A.I.1 INTRODUCTION

The loss-of-coolant accident (LOCA)/safety/relief valve (SRV) discha rge devices and other submerged structures are shown in F i gures 3 A.2.1-2 , 3A.2.1-6 , 3A.2.1-7 , and 3A.2.1-8 and identified in Table 3A.I-1.

The most significant hydrodynamic load for each structure is identified in Table 3A.I-1.

3A.I.2

SUMMARY

OF METHODOLOGY USED FOR DEFINING LOSS-OF-COOLANT ACCIDENT JET/BUBBLE LOADS

Loss-of-coolant accident jet/bubble loads are defined using the ring vortex model. The pool is divided into zones and to ensure conservatism in design, the largest velocity and acceleration values seen by a submerged structure are assumed equal to the maximum calculated values anywhere in the applicable zone. The LOCA bubble charging model is us ed to verify/ensure that the design values are conservative.

3A.I.3

SUMMARY

OF METHODOLOGY USED FOR DEFINING LOSS-OF-COOLANT ACCIDENT STEAM CONDENSATION LOADS

Generic "drag load" methodology and plant unique flow fields are used for LOCA steam

condensation loads on submerged st ructures in compliance with the NRC acceptanc e criteria.

Plant unique flow fields are de fined consistently with stea m condensation boundary loads.

The generic methodology identifies three compon ents of flow induced loads on submerged structures: acceleration dependent and velocity square dependent in-line loads, velocity square dependent lift load (normal to the direction of flow).

Representative plant unique chugging flow fiel ds show that the chugging loads on submerged structures are due to accelera tion or pressure gradients esta blished in the pool during the impulsive chugging phenomen on, i.e., velocity dependent loads are small.

3A.I.4

SUMMARY

OF METHODOLOGY US ED FOR DEFINING SAFETY/RELIEF VALVE LOADS

Caorso SRV test data on subm erged structures ar e examined to supplement theoretical approaches of the acceptance criteria. The data and thei r correlation with theoretical approaches of the acceptance criteria confirm that SRV loads are primarily due to pressure C OLUMBIA G ENERATING S TATION Amendment 55 F INAL S AFETY A NALYSIS R EPORT May 2001 LDC N-0 1-0 0 0 3A.I-2 gradients established in the pool during the SRV discharge, i.e., veloc ity dependent loads are small.

The dynamic pressure gradients measured across Caorso column, vent and SRV line are used to define the peak load values (at quencher elevation), the spa tial distribution of the load and its time dependence.

The pressure time histories reco rded on submerged structures s how waveform characteristics similar to those recorded at pool boundary.

The SRV loads on submerged structures are defined consistently with the plant unique boundary loads.

The SRV loads on Columbia Generating Station structures are calculated using the following formula:

P D d 2 2 4 d P LCaorso 2WNP2 b where:

P = load on a structure (force/unit length)

D = diameter of the structure

= a load gradient factor established using Caorso SRV test data on submerged structures. The me thod to calculate () is explained in the notes for and in Figure 3A.I-1 d Caorso = horizontal distance of the structure from the nearest actuating quencher in Caorso plant

d WNP2 = horizontal distance of the structur e from the nearest actuating quencher

P b = boundary pressure load definition from Reference 3A.I-1 including any modifications agreed upon with the NRC

L = load margin = minimum value of 1.4 is used for all piping which are adequately braced and a value of 2.0 is used for the column which is the only unbraced structure and is closest to the nearest quencher

C OLUMBIA G ENERATING S TATION Amendment 55 F INAL S AFETY A NALYSIS R EPORT May 2001 LDC N-0 1-0 0 0 3A.I-3 Notes on Figure 3A.I-1

1. The SRV load gradient is obtai ned from Caorso data as follows:

A PP Dfba P 19 where:

A = measured gradient across the cylindrical structure

P f = P front P ba = P back

2. P 19 , P f , P ba waveform characteristics are similar.

3. The value of () for each set of P f (P 42 , P 41 , P 33 , P 24) and P ba (P 40 , P 39 , P 53) is obtained from Caorso SRV test data (si ngle and multiple valve actuations).

4. For miscellaneous piping which run al ong the suppression pool boundary, the load gradient factor () equal to that for the column is specified.

3A.I.5 REFERENCES

3A.I-1 "SRV Loads - Improved Definition and Application Met hodology for Mark II Containments," Technical Report (Proprietary), prepar ed by Burns and Roe, Inc. for application to Washington Public Power Supply System Nuclear Project No. 2, submitted to the Nuclear Regulatory Commission on 7/29/80.

C OLUMBIA G ENERATING S TATION Amendment 53 F INAL S AFETY A NALYSIS R EPORT November 1998 3A.I-5 Table 3A.I-1 Loss-of-Coolant Accident/Safety/Relief Valve Loads on Submerged Structures Identification of Structures Identification of Most Significant Hydrodyn a mic Load 1. (a)

S R V line SRV (due to actuation of adjacent SRV) a (b) Quencher b LOCA jet on arms (c) Quencher Support b None significant

2. Downcomer vents SRV 3. Concrete columns SRV 4. Bracing truss b at vent exit Pool swell drag

5. Platform with grating (at el. 472 ft 4 in., 78% open area)

Pool swell drag

6. Miscellaneous pipi ng, penetrations and supports along containment boundary (a) Below vent exit (el. 454 ft 4.75 in.) LOCA jet and SRV a (b) Above vent exit, below initial pool surface (el. 466 ft 4.75 in.)

Pool swell drag (c) Above initial pool surface, below maximum pool swell (el. (484 ft

4.75 in.)

a See also discussion presented in Reference 3A.3.1-8. b Loads on discharge devices a nd their supports during discha rge through the devices are addressed elsewhere.

Radial View 47.22" Figure Amendment 53November 1998 Form No. 960690Draw. No.Rev.Location of Pressure Transducerson Caorso Submerged Structures 900547.95 3A.I-1 C L C LDowncomer Vent #9 Column #7Tangential ViewSRV Line "A"SRV Line "F" El. 46.2 m El. 44.3 m P39 El. 39.2 m P40 P42 81.62" P41 24"P33P34 107.4" 150.1" Quencher "A" P19 Pool Bottom 12" P53 P24 Quencher "F" El. 40.7 m Columbia Generating StationFinal Safety Analysis Report