ML20116C951
ML20116C951 | |
Person / Time | |
---|---|
Site: | Comanche Peak ![]() |
Issue date: | 10/27/1992 |
From: | Silva H TEXAS UTILITIES ELECTRIC CO. (TU ELECTRIC) |
To: | |
Shared Package | |
ML20116C936 | List: |
References | |
RXE-92-01B, RXE-92-1B, NUDOCS 9211050107 | |
Download: ML20116C951 (314) | |
Text
{{#Wiki_filter:I l RXE-92-Oll! Individual Plant Examination Submittal: Comanche Peak Steam Electric Station Volume 11: Back-End Analysis By H. C. da Silva, Jr. October,1992 1 9211050107 921030 PDR ADOCK 05000445 p PDR
Prepared by:
~ ! d - /6-17-E2 ' 11. C. da Silva, JM Date Senior Engineer f ## ~ ~
Reviewed by: S.6.Karpfiik
/M Date Senior Engineer
_ %_ lO*5)~ V -a W. G Choe Date Supervisor, L CA Analysis Approved by: , d/Y /0/A 7/'72
- 11. G. Ham 7ehee Date Supervisor, Systems Analysis wxh A. flusain bact 10 17 ' '! L Date Director, Reactor Engineering
ACKNOWI.I DGE51ENT Ikith the front-end and back-end components of the CPSES IPE were multi-discipline ettorts. It is therefore appropriate to recogniee contributions from all members of the IPE team, LOCA Analysis Group, and support staff. Speci6cally, it is important to mention: Chris Cragg, who binned the Level I sequences into PDS and streamlined the CET computation process; Dan Tirsun, who produced the ISLOCA anhlysis and the descriptio.m ( the Level I functional sequences; Yu Shen, for his _ mathematically rigorous derivations which were used to compute several basic event prob.bilities; Dan 13rozak, who reviewed and helped prepare the M A AP Parameter Ole development iocument and compiled all the plant data for Sectioa 4.1 of this report; Steve Karpyak, who managed the report production; and Irene Reyes who did all the word processing. d
.. - - . - , . .-. -. . ~--. __-. .- - . ..- - . . - - . . .
Table of Contents -
- 1. EXECUTIVE SUhth1ARY 1.1 Background and Objectives . . . . . . . . . . . . . . . . . . . . . . ... Volume 1-1.2 Plant Familiarization . . . . . . . . . . . . . . ... Volume i 1.3 Overall hiethodology . . . . . . . . . . . . . . . . .... Volume i 1.4 Summary of Alajor Findings . . . . . . . . . . . . . . Volume I and 1-1
- 2. EXAh11 NATION DESCRIPTION 2.1 Introduction . , . . . . . . . . . . . . . . . . . .... Volur:e 1 2.2 Conformance with Generie Letter 88-20 . . . . . . . . . . . . . ,. Volume I -
2.3 General hiethodology - . . . . . . , , , . . . . . . . . . . .... Volume i 2.4 Information Assembly . . . . . . . . . . . . . . . . . . . .... Volume I
- 3. FRONT-END ANALYSIS 3.1 Accident Sequence Delineation . . . . . . . . . Volume I -
3.2 System Arialysis . . . . . . . . . . . . . . . .... Volume I
- 3.3 Sequence Quantification . . . . . . . . . . , . . . . . .... Volume I 3.4 Results and Screening Process . . . . . . . . . . . . . . .. . . . .... Volume I
- 4. BACK END ANALYSIS . . . . . . . . . . . . . . . . . . . . . . . . ~ 4- 1
- 4. l' Plant Data and Plant Description . . . . , . . . . ... . . 4-3 ii i
w m- w -w y- g- w- , w ay-w,- e e* r -v w-t-y en B7- T .
. .__ _- - . __m.__.._ . . . _ .- - - _ _ _ . _ _ . _ . - _. _ _ _ _ . _ . . . .
Reactor Core, Vessel and Primary System Data 4-3 j 4.1.1 ............... 4.1.2 Containment System . .. ... . ... ................ 4 11 4.1.2.1 Safeguards and Isolation Systems ............. .. 4-11 4.1.2.2 Containment Design and Structures . . . ..,,.. ... . 4 17 4.1.3 Emergency Core Cooling System (ECCS) ....... ...... . 4-25 4.1.4 Auxiliary Building . . . .. . .... .. ,,. .. ,, .. 4-32 . 4.2 Plant Models and Methods for Physical Processes . . . . . . . ....... ... 4-92 4.3 Bins and Plant Damage States ..... .., ,,... . 4-93 4.4 Containment Failure Characterization . . . . . . 4-98 4.4.1 Penetrations . . . ... .. ... . 4 4.4.1.1 Large Opening Penetrations .. ..... . . . . 4-99 4.4.1.2 Purge and Vent System Isolation Valves . . . . . . . . . . , , 4-102 4.4.1.3 Piping Penetrations . ...,.. .. .........., . 4-102 4.4.1.4 Electric Penetration Assemblies . . , . 4-104 4.4.2 Containment Rupture (Gross Failure) Limit ... ..... ,, . 4-104 4.4.3 Containment Liner Ductile Failure Limit . . . . . . . . . . . . ... ~ 4-l')6 3 4.4.3.1 Description of the Method , ,,.. ... .. . .. 4-103
- 4.4.3.2' Application of tn, Wthod . , . . . . . . , , . . - . ...... 4-110 4.4.4 Containment Fragility Curve . . .. ..... ....'4-113-4.4.5 Conclusion ' . . . . . . . . . . .., , . .... .,.. . 4-114 4.5 - Containment Event Trees ..... . ... . ..... . .,, . 129 4.5.1 Containment Bypass and Isolation Failures . . . .... . '4-129 l 4.5.1.1 Containment Bypass .. .......... .. .. .. . 4-129 4.5.1.2 Containment Isolation Failures , ... ... ... 4-132 iii
m 4.5.2 Containment Not-llypassed and Successfully isolated, Containment Event Trees (CETs) , . ... ......................... 4 134 4 4.6 Accident Progression and CET Quanti 0 cation . ................... 4-170 4.6.1 CET Quantification . . .. . ......... . . . . . . . . . 4 178 4.6.1.1 Determination of BE probabilities: Example, PRWCP PULT for PDS 211 = 0.42 . ......... . . . . . . . . . . . 4-178 4.6.1.2 CET Quanti 0 cation Process and Discussion of Results . . . . - 4 181 4.6.2 Ranking Accident Sequences and Accident Progression Analyses , , . 4-197 4.6.2.1 PDS 2CB (SGTR & ISGTR) . . . . . . . . . . . 4-198 4.6,2.2 PDS 411 and 4F . ... . . . . . . - 4 200-4.6.2.3 PDS 1CB (V-Sequence) and ICI (Isolation Failures) - . . . . 4 204 4.6.2.4 PDS 111 and IF (and 5}I, $F) .. . . . . . . . . . 4-207 4.6.2.5 PDS 311 and 3F , .... . . . . . . . . . . = 4-209 - 4.6.2.6 PDS 6F, 611, 2F, 211 , . .. . . . . . 4-213-4.6.2.7 PDS 3SBO and 4SBO . .. . . . . . . 4-217
-4.6.3 Treatment of Uncertainties and Sensitivity Studies . . . . . . . . 4-220 4.6.3.1 Phenomenological Uncertainties . . . . , , , . . . 4-221 4.6.3.2 System and Operational Uncertainties . . . . . . . . . -4 231-4.7 Radionuclide Release Characterization . .. . . . . . . . . ~4-269 4.7.1 'The CPSES Release Categories . ... .... . . . , , . 4-269 4.7.1.1 Isolated, Non-bypass (Intact Containment at Vessel Failure) ~
i- Release Categories . ............ . . . . . 269 L 4.7.1.2 Unisolated or Bypass Release Categories . . . . . . . . . . . 4-271 ! 4.7.2 Release Category Frequencies . ,. . . . . . . . . . . . . 271 l-iv
,_ - . . . - .. .m.. . _ , _ , ,
.1 ,
4.7.3 Release Fractions and other Characteristics for the Release Categories . 4 272 i l S. UTILITY PARTICIPATK6 AND AN1ERNAL REVIEW TEAM'.. . . . . . . Volume I i i
' 6. PLANT IMPROVGh1ENTS AND Ui'!O11E SAFEW FEATURES . . . Volume I and 61
- 7.
SUMMARY
/.ND CONCLUS10NS . ...., . . . . . . . . Volume I and 7 8. REFERENCES . . . . - . . . . . . . . . . Volume I and 81 l
l-l V .: i v r- <--W--- -' c e** -m 4 .t'^*"r--- m-? g=i- t qi#
1.ist of Tables - l u Table 4.1.1-1: Core hiaterials Weight and Volume . . ...,.. ........ ... 45 i Table 4.1.1-2: Core Geometry .. . . ... .. . ...... . . '4-6 Table 4.1.1-3: Core Performance Characteristics ,, . .. ........ . 47 Table 4.1.1-4: Rod Cluster Control Assemblies . . .. .. . . ......... . 4-7
~
, Table 4.1.1-5: Burnable Poison Rods .. . ..................... . . . .. 7 Table 4.1.1-6: Reactor Vessel (RV) hietal hiasses .. ... . ...... .... 4 Table 4.1,1-7: RV Fluid Volumes . . .... .. . .. 4-8: Table 4.1 1-8: RV Geometry , , ... . . 4-8 Table 4.1.1-9: RCS Fluid Volumes . . . , . .. . . . ... ... -48 Table 4.1.1-10: RCS Valve Data and Setpoints . . .. .. . . . .. .. 4-9 Table 4.1.1-11: RCS Normal Full Power Operating Conditions . . . .. .. ......,49 Table 4.1.1-12: RCS hietal hiasses .. ... . .... ...,,........ . 4 Table 4.1.21: Safeguards Systems .,, . . . .. , 4-21 Table 4.1.2-2: Containment Design and Structures .... , , ..,. . .. 4-22 Table 4.1.3-1: Accumulator Tanks . . ...... ... .. .. . ,... .. 4 Table 4.1.3-2: ECCS Pumps . . . . .. .. ,.. ,............ 4 r Table 4.3-1: Sequence Characteristics of Core Damage Bins . ...... .... . ,. 4-96 Table 4.3-2: Sequence Characteristics of Containment Safeguards Bins . . . . . . , 96 - Table 4.3-3: Binning of Level 1 Functional Sequences into PDS .. . . . . . . . 9 7 - Table 4.4-1: CPSES Containment Failure blodes , . . ... ..... .. . . 4 1.18. i
- Table 414-2: Summary of Various Containment Strengths . .... ........ .. 4-119 Table 4.4-3: Reference Plant Global Strains at Discontinuities . . ... . . 4-120 vi i
l
. - - _ _ - - . - . ._-._ - - -. - .. . _.. - _ _ . .. _ .-. - -~
l Table 4.4-4: Complementary Cumulailve Probability Distribution for CPSES Containment - Failure Pressures Dased on Liner Tear at Medium Penetrations at 114 psig-Assumed as the Mean and a Normal Distribution with a 7% Coefficient of ' Variation ... ....... . . . ... .. ........ 4-122 Table 4.51: ISLOCA Initiator Frequencies for Various Pathways .... . .... . 4-150 .; Taoie 4.5-2: ISLOCA initiator Frequencies for CPSES ........ ...... ... 4-151 Table 4.5-3: Core Damage Frequencies and Their Functional Sequences . . . . . . . . . 4 152 Table 4.5-4: Estimate of Debris Bed Thickness . . ................ ,,. 4-153 Table 4.5 5: Description of CET End-States .. . ... . , .. 4-154 Table 4.6-1: NUREG-1150 llPME Pressure Rise Probability Distributions for Zion . . . 4-234 Table 4.6-2: Calculation of Containment Reliability for llPME Events . ,,. 4-235 [ Table 4.6-3: End. State Probabilities for Small Break LOCA PDS . .. ....... 4 236 Table 4.6-4: End-State Probabilities for Transient PDS . . . . .. . .. .... 4-237 Table 4.6-5: End-State Probabilities for Station Blackout PDS . . . ..... - 4-238 i Table 4.6-6: End. State Probabilities for Large Break LOCA PDS .. ....... 4-239 Table 4.6-7: Summary of Conditio'nal Containment Failure Probabilities for Small Break LOCA PDS . . .. . ........ ........ ...... . 4-240 h Table 4.6-8: Summary of Conditional Containment Failure Probabilities for Transient PDS 4-240 ~ Table 4.6-9: Summary of Conditional Containment Failure ProLibilities for Station Blackout PDS - .- ... .... ... .. ............ 4-241 Table 4.6-10: Summary of Conditional Containment Failure Probabilities for Large Break-LOCA PDS . ... , -
. . . . . - . .... .. . ... 4-241 Table 4.6-11: Accident Sequence Initiator Notatioa (Prefixes) . . . . _ ... ...... .. 4-242 l Table 4.6-12: Accident Sequence Functions Notation (Suffixes) . . . . . . . ... ... . 4 242 Vii i
[
Table 4.6-13: Level ll' Characteristics of Level i Break Size Ranges . . . . . . . . . . . . 243
~
- _. Table 4.6-14: Containment Failure Probabilities Sorted by PDS Frequency . . . . . . . . 4 244 Table 4.6-15: PDS in Order of increasing Early and Total Unconditional Containment ,
Failure Probabilities . .... .... . ..... . . . . . . . . . . .. 4-245 Table 4.6-16: PDS in Order of increasing Late CCI and Steam-Induced Unconditional Containment Failure Probabilities . .,. .... . . . . . . . 4-246 Table 4.6-17: Key PDS and Their Functional Sequences Probability Composition . . . ... 4-247 Table 4.6-18: Contribution to ISGTR by PDS . . . ....... ,,.. . . . . . 4-248 Table 4.6-19: Contribution to ISGTR by Functional Sequence . . . . . . . . . . . . 4-248 Table 4.6-20: Summary of MAAP Calculations for Representative Sequences . . . . . , . 4-249 Table 4.6-21: Release Fractions for CET End-States and Respective PDS Representatives 4-251 , Table 4.6-22: Description of Calculated End-States . .. . . . . 4 252 Table 4.71: CPSES Release Category Definition and CET End-States . . . . . . . . 4-273 Table 4.7-2: Binning PDS into the CPSES Release Categories . . . . . . . . . . . . . . . . 4-274 Table 4.7-3: CPSES Release Categories in Order of Absolute Unconditional Frequency . 4-275 i Table 4.74; CPSES Release Categories in Order of Relative Conditiona'. Frequency . . 4-275. Table 4.7-5: Release Fractions for Release Categories and Respective PDS Representatives 4-276 Table 4.7-6: Principal Release Category Characteristics (Besides the Release Fractions) . 4-277 ' l l l Viii 1
.e,. .%. -._.u. , ,. - - . . - - ,_. , y. ! ,w . . . , - p
List of Ficures Figure 4.0-1: Schematic Representation of PRA Levels . ... . . ., 4-2 Figure 4.1-1: General Plant Schematie . , .. .... 4-33 Figure 4.1-2: Simplified Process Flow Diagram of the RCS 4-34 Figure 4.1-3: hiain Components of the RCS . ... 4-35 Figure 4.14: Primary Side Process Flow Diagram 4 36 Figure 4.1-5; Secondary Side Process Flav Diagram . 4-37 Figure 4.1-6: Structural i)lant Arrangement . .. . 4-38 Figure 4.17: Containment Cut View with hiain RCS Component 4-39 t Figure 4.1-8: General Reactor Vessel Assembly 4-40 Figure 4.1-9.1: Reactor Vessel Lower Internals Assembly 4-41 Figure 4.1-9.2: Lower Core Support Assembly 4 42 Figure 4.1-10: Bottom hiounted Reactor Vessel Instrumentation 4-43 Figure 4.1-11.1: Pressurizer (Cutaway) . . 4-44 Figure 4.1-11.2: Pressurizer Relief Tank 4-45 Figure 4.1-12: Steam Generator (Cutaway) 4-46 ; Figure 4.1-13: Reactor Coolant Pump C.itaway) . 4-47 Figure 4.1-14: Accumulator Tank 4-48 Figure 4.1-15: Containment Water Level vs Containment Water Volume . 4-49 Figure 4.1-16.1: Reactor Vessel Supports . . 4-50 6 Figure 4.1-16.2: Reactor Vessel Supports Details 4-51 Figure 4.1-17: Containment Cavity . .. . .. 4-52 Figure 4.1-18.1: Containment and Safeguards Building Structure (East-West Sectional) 4-53 Figure 4.1-18.2: Containment Structure and Internal Arrangement (North-South Sectional) . 4-54 ix
Figure 4.1 18.3: ' Containment Imernal Structure . . .. . . .......... . 4-55 Figure _4.1-19: Containment Spray System Header Arrangement . . . . . ...... ... 4-56
' Figure 4.1-20.1: Simplifwd Flow Diagram Containment Air Cooling and Recirculation System -(CACRS) . ... . .. ..... .. ........... . .. 4 57 Figure 4.1-20.2: Plenum and Duct Arrangement CACRS ... ....... ......... 4-58 Figure 4.121.1: Containment Layout - RCS Components . ........... .. . . . . . 4-59 ,
i Figure 4.1-21.2: Containment Layout - Floors and Cranes . ....... ...... ... 4-60 ; l Hgure 4.1-21.3: Containment Layot - Floors and Containment Spray Piping . . . . . . . . 4-61 Figurr 4.1-22.1: Containment Sump Piping and Arrangement (Structural View) . . .. 4-62 - Figure 4.1-22.2: Containment Sump Piping Schematic . , . .. .. . . 4-63 Figure 4.1-23.1: Containmer.t Shell Reinforcement Detail . . . ..... . 4 Fignre 4.1-23.2: Foundation Mat Reinforcement , . . . . .. .. 4-65 Figure 4.123.3: Containment Penetration Locations .. .. ,.. .. .... 4 ,- Figure 4.1-23.4: Containment Hatch Details . . . .. ...... .. .. . 4 67 l Figure 4.1-23.5: Personnel Airlock General Arrangement 4-68 l .. . . . .. ..,... Figure 4.1-23.6: Fuel Transfer Tube Details . . . . . . . . ... .. . ...... 4 1 Figure 4.1-23.7: Containment Penetration Details . ... . . . .. . 4-70 . Figure 4.1-24.1: Structural Plant Arrangement . . .. .. ... .. ........ , 4-71 L I ! Figure 4.1-24.2: Structural Plant Arrangement (Section A-A) ,. .. .. .. . 4-72 Figure 4.1-24.3: ~ Structural Plant Arrangement (Section B-B) . . ... .. .......... 4-73, Figure 4.1-24.4: Structural Plant Arrangement (Section C-C) . . . . . . . 4-74 Figure 4.l_-25.li Containment and Safeguards Building Plans (Els. 773,785,790) . .. . . _4-75 Figure 4.1-25.2: Safeguards Building Plan (El. 810) .,. . . , , . .... ~ 4-76 Figure 4.1-25.3: Containment and Safeguards Building Plans (El, 810) . .. . ..... 4-77 ._ x
Figure 4~ 1-25.4i Containment and Safeguards Building Plans (El. 832) . . . .
...... . 4 78 Figure 4.125.5: Containment and Safeguards' 11uilding Plans 6.i. 852) . . . . ...... 4-79 M
Figure 4.1-25.6: Containment and Safeguards Building Plans (Els. 873,880) ....... . 4-80 Figure 4.1-25.7: Coritainment and Safeguards Building Plans (El. 896,905) . .. 4-81 Figure 4.1-25.8: Containment and Safeguards Building Plans (El. 778,790) . . . . . . 4-82 Figure 4.1-25.9: Auxiliary and Electrical Control Building Plans (Els. 807,810) . .. .. 4-83 Figure 4.1-25.10: Auxiliary and Electrical Control Building Plans (El. 830) ... ....... 4-84 Figure 4.1-25.11: Auxiliary Building Plan (El. 842) ........ . . .. ... 4-85 Figure 4.1-25.12: Auxiliary and Electrical Control Building Plans (El. 852) . . . 4-86 Figure 4.1-25.13: Auxiliary and Electrical Control duilding Plans (Els. 873.886) . .. 4 87 .; Figure 4.1-26.1: ECCS Valve Alignments (Standby) . .. . , .. 4-88 Figure 4.1-26.2: ECCS Valve Alignments (Injection) . . . . .... . . 4-89 . Figure 4.1-26.3: ECCS Valve Alignments (Cold Leg Recirculation) ,. . . . 4-90 , Figure 4.1-26.4: ECCS Valve Alignments (Hot Leg Recirculation) . . .. . . . 4-91 l Figure 4.4-1: Uniaxial Strains vs Normalized Containment Pressur (Example P';mt) . 4-123 Figure 4.4-2: Comparison of CPSES Containment to Prototype ,, . .. 4-124 Figure 4.4-3.1: Biaxiality Factor for Typical Springline Geometry Reinforced Coacrete l-r Containments ,,. . . .. ..,....... . .. .. 4-125 l . L l' Figure 4.4-3.2: Strain Concentrat:an Factor for Typical Springline Geometry Reinforced-l- Concrete Containments- ,. . . . .. .., . . . 4-125 L Figure 4.4-3.3: Biaxiality Factor for Typical Equipment and Personnel Hatches Reinforced Concrete Containments . .. . . . .... 125 Figure 4 4-3.4: Strain Concentration Factor for Typical Equipment and Personnel Hatches Reinforced Concrete Containments . .. . . . ... 4-125 xi l l l:
~. - ...- . .- .- . - _ ~. - - - . . - . - - - - . - _ . . . - - - Figure 4.4 3.5: - Blaxiality Factor for Typical Steam Line and Othe; Fenetrations Reinforced - Concrete Containments . . . . ...,,..,.., ,,,.... . . . , . . . . 4-126 Figme 4.4-3.6: Strain Concentration Factor for Typical Steam Line and Other Penetrations-Reinforced Concrete Containments . . . , . . , . . , . . . . . . . . 4-126 1 1
Figure 4.4-3.7: Biaxiality Factor for Typical Wall-Basemat Juncture Reinforced Concrete - Containments . . . . . . . .... . . . . . . . . . . . . . . . . , . 4-126 Figure 4.4-3.8: Strain Concentration Factor for Typical Wall Basemat Juncture Reinforced Concrete Containments . . . . . . . . . . . . . . . . . . . . . . . 4-126 .l 1 , i Figure 4.4-4: Containment Peak Strains Compared to Uniaxial Failure Strain (Comanche Peak) . . . . . . ... . . . . . . . . . . . . . . . 4 127 Figure 4.4-5: Seal Life as a Function of Time at Temperature . . . . . . . . . 4 828 ,
. Figure 4.5-1: Containment Event Tree , , . . . . . . . . . . 4-155 'i Figure 4.5-2: hiaster Containment Fault Tree (DP) . . . . . . . . . . . 4-156 .
Figure 4.5-3: Master Containment Fault Tree (RECl) . . . . . . . . . 4 157 Figure 4.5-4: Master Containment Fault Tree (VF) . . , , , , , . . . . . 4-158' i Figure 4.5-5: Master Containment Fault Tree (CFEI) . . . . ,, , . . . , . . . . . 4-159 Figure 4.5-6: Master Containment Fault Tree (CFE2) . . ..... . . . . . . . . . . - 4 160 Figure 4.5 7: Master Containment Fault Tree (DCl) . ,
- s. . . . . . . . 4-161 Figure 4.5-8: Master Containment Fault Tree (DC2) . , , ,,,, . . . . . , , , . 4-162 Figure 4.5-9: _ Master Containment Fault Tree (DC3) . . . . . . . . . . . . . . '4-163
- Figure 4.510: Master Containment Faul't Tree (FPRO) . . . . ........ . . . . , . 4-164 Figure 4.51li Master Containment Fault Tree (FPRI) . . . . . . . . . . . . , , , 4-165 -
Figure 4.5-12: Master Containment Fault Tree (FPR2) . . .. . . . . , 4 166 Figure 4.5-13: Master Containment Fault Tree (FPR3) . . . . 4-167 t m 1 < xii 4 e,-e -, -- -
-.-s , , - ,-, ,, .c-, ,-e- - - _mn-.,-- s ,
_ . _ . - ~ _ . ...._. _. . _ _ . _ . _ _ _ . . _ . _ _ _ . _ _ . _ . . _ _ _ _ . _ . _ . . . _ . . , _ _ 1 Figure 4.514: - Master Containment Fault Tree (FPR4) ............ ......... 4-168 Figure 4.515: Master Containment Fault Tree (CFL1) .. . .......... . 4 169 Figure 4.5-16: Master Containment Fault Tree (CFL2) .... ... .... . 4-170 Figure 4.5-17: Master Containment Fault Tree (CFL3) .... .. . . . 4-171 Figure 4.5-18: Master Containment Fault Tree (CFL4) . ... ...... ..... 4-172 . Figure 4.5-19: Master Containment Fault Trw (Ct:L5) . .. . . . .... 4-173 Figure 4.5 20: Master Containment Fault Tree (CFL6) . ... . ... 4-174 Figure 4.5-21: Master Containment Fault Tree (CFM!) . . . ... .. . . . 4-175 Figure 4.5-22: Master Containment Fault Tree (CFM2) . 4 176 Figure 4.5-23: Master Containment Fault Tree (CFM3) .. . .. ., , . 4 ;77 Figure 4.6-l: Comparison of Probabilities for HPME Final Pressures with Containment Survival Probabilities for; Above: PDS lE IF,111,2E and 2F; Below: PDS 2H . .. .. . .., . . . . 4 253 Figure 4.6-2: Containment Temperatures (TG) and Pressures in (A = upper, B= tower and I l C= cavity) Compartments . . . . ... ... . . 4-253 Figure 4.6-3: Containment Temperatures (TG) and Pressures in (A = upper, B= lower and C = cavity) Compartments , . .. ... . . . . ... . 4-255 Figure 4.6 4: Masses of Hydrogen Produced m Core (CRI) and in Containment (CB1) . . 4-256 - Figure 4.6-5: Containment Temperatures (l'G) and Pressures in (A= upper, B= lower and C= cavity) Compartments . . . ..., ... ........... .. 4-257 Figure 4.6-6: Containment Temperatures (TG) and Pressures in (A=t,. B= lower and C= cavity) Compartments . .. . - . . . . .,,.... , , . . . . . - 4-258 l- Figure 4.6-7: Basemat Erosion Depth (XCNCl) and Pressures in (B= lower and C= cavity) L ! Compartments .. .... , .... .. . . . . . _4-259 xiii '
. . . . _ _ . _ , _ _ . . . _ _ . _ . . ~ _ . . _ _ _ _ . _ . _ -.__. - 't Figure 4.6-8: Containment Temperatures (TG) and Pressures in (A= upper, B= lower and l C= cavity) Compartments . .. .. . . . . . . . . . . . . . . . . . 4 260 Figure 4.6-9: Containment Tem; erature (TG) and Pressures in (A= upper, B = lower and C= cavity) Compartments . , ... . ............ . . . . . . . . 4-261-Figure 4.6-10: Containment Temperatures (TG) and Pressures in (A= upper, Belower and C= cavity) Coinpartments . .., . , . . . . . . . . . . . . 4-262 Figure 4.M1: Containment Temperatures (TG) and Pressures in (A = upper, B= lower and C= cavity) Compartments . . . .. . . . . . . . . . . . . 4-263 Figure 4.6-12: Containment of Temperatures (TG) and Pressures in (A= upper B= lower, and C= cavity) Compartments .. . . . . . . 4-264 Figure 4.6-13: Containment Temperatures (TG) and Pressures in (A= upper, B=iower and C = cavity) Compartments ... . . . . . . . . . . . . . . 4 265 Figure 4.6-14: TRAN21 (A) and TRAN22 (B) Surge Line (TSRI). Hot Leg (TUH), SG Tube (TPHSF) Temperatures . . . . , , . 4-266-Figure 4.6-15: Etfect of Fan Coolers for Dry PDS .... . . . . . . . , , . 4-267 Figure 4.616: CCI Failure Modes for CPSES Concrete: Overpressurization vs Melt through for Two Types of Concrete . . .....,, . . . . . . . . . . . . . . . 4-268-Figure 7-1: CPSES Overall Containment Performance .. . . . . . . . . . . . . . . . , 7 Figure 7 2.1: Breakdown of Early Failures by PDS . . . . . . . . 7-5 Figure 7-2.2: . Early Failurcs (except SGTR) by PDS .. . . . . . . . . . . . . . . 7-6 L
L - Figure 7-3: Breakdown of Early Failures by Cause .. . . . . . . . . . . . . 77: I
- - - Figure 7-4.1
- Breakdown of CCI-Induced Late Failures by PDS , . . . . . . 7-8 Figure 7-4.2: . Detail of ALL OTHERS (8.5%) CCI-Induced Failures . . . . . . . . . . 7-9 FiAure7-5: Breakdown of Steam-Induced Failutes by PDS . . . . . . . . . '7 107 xiv l'
I L
, , . . . . . , - , , , . . - . , , , . -,ae- ~ - , n w -+-
..-. -.-. . . . . --.- .~.. _ . - - . - . - . . . - . -- - . _ . .. ~
l_ist of Acronyms AF - Auxiliary Feedwater ARV - _ Atmospheric Relief Valve
- ASME -
American Society of Mechamcal Engineers ASTM - American Society of Testing and Materials ATU - Automatic Transfer Unit ATWS - Anticipated Transient Without Scram BATP - Boric Acid Transfer Pump - BOS - Blackout Signal CACRS - Containment Air Recirculation and Cooling System CAFTA - Computer Aided Fault Tree Analysis CC - Component Cooling Water - CCI - Core Concrete Interaction CCP - Centrifugal Charging Pump CET-- - Containment Event Tree CH - Chilled Water (Safety and Nonsafety) C1 - Instrument Air CPSES - Comanche Peak Steam Electric Station CS - Chemical and Volume Control System CST - Condensate Storage Tank L ! CT - Containment Spray CW - Circulating Water CZ - Containment Isolation System DCH - Direct Containment Heating DHR - - Decay Heat Removal ECCS - -- Emergency Core Cooling System j EHC - Electro-Hydraulic Control -
. EOP- Emergency Operating Procedure EP -
Electrical Power EPRI -
- Electric Power Research Institute L-
'- ERG - Emergency _ Response Guideline - ESl - - Reactor Protection System ESF- - - Engineered Safeguards Feature ESFAS - - Engineered Safeguards Features Actuation System ? FSAR - Final Safety Analysis Report FIT - Fuel Transfer Tube FW- - - Main Feedwater xv fr+.-- ,,w. - -ym,,, y y ,,,%_.,u ,p- ., .yg-,,,,- , .w7-.3- 4 -_ qm.y -
HPME - High Pressure Melt Ejection H RA ~- - . Hurr.an Reliability Analysis HVAC - Heating Ventillation and Air Conditioning IDCOR - Industrial Degraded Core Rulemaking Program IPE - Individual Plant Examination ISGTR - Induced Steam Generator Tune Rupture LOCA - Loss of Coolant Accident LOOP Loss of Offsite Power MAAP -- Modular Accident Analysis Program MCC - Motor Control Center MS - Main Steam MSIV - Main Steam isolation Valve MSLB - Main Steam Line Break NRC - Nuclear Regulator Conimission NSAC - Nuclear Safety Analysis Center
- j. P - Containment Hi.3 Signal PDS -
Plant Damage State -- PORV - Power Operated Relief Valve PRA - Probabilisitic Risk Assessment
~
RC/RCS - Reactor Coolant (System) RCC - Rod Cluster Control Guide RCP - Reactor Coolant Pump RCPB - Ructor Coolant Pressure Boundary RH - Residual Heat Removal RV- - Reactor Vessel RWST - Refueling Water Storage Tank S/G- - Steam Generator . S - Safety injection Signal 6 SGTR - Steam Generator Tube Rupture SI - Safety Injection SIP - Safety injection Pump -
.SRV -
Safety Relief Valve STCP - . Source Term Code Package - SW - Station Service Water - TDAFWP - Turbine Driven Auxiliary Feedwater Pump TPCW - - Turbine Plant Cooling Water xvi
., . . . - . - - ... - -i ~- - - - .
- 1. EXECUTIVE SUMM_ARY
' he CPSES IPE was prepared in two volumesi Volume i presents the front-end analysis and Volume 11 preserts the back-end analysis. The general issues common to the front-end and to the back-end were reported in Volume I along with the front-end analysis. Nevertheless, some areas require reporting from both the front-end and the back-end perspectives. In those cases, the front end perspective is presented m Volume I and the back-end perspective is presented in Volume 11. The sections which in Volume 11 contain the back-end 6ndings, and in Volume I, contain the front-end findings are:
Section 1.4 . . . . . .. .. Summary of Major Findings Section 4 .
. . 11ack-End Analysis Section 6 . ,... ... . , Plant Improvements and Unique Safety Features Section 7 . . . . . . . .. .. Summary and Conclusions Section 8 .. . . ... .. ... ... ...., Referench The deterministic part of the back-end analysis, i.e., the severe ecident sequence analyses relevant to the CPSES IPE, utilizes the Modular Accident Analysis t.ogram (MAAP 3.08, Rev.16, Ref.10), The MAAP code is a well-known, fully integrated analytical tool for the analysis of severe accidents. The logie model and the probabilistic part of the CPSES back-end analysis are based on standard methodology, specifically the EPRI back-end generic framework (Ref,1). ,
1.4 Summary of Maior Findines (Back-end) - The overall performance of the CPSES containment given a core melt accident can be summarized by 8 one of the following outcomes: (1)- The containment will remain intact . . . . ... . . . 40% (2) Late failures from steam generation ., . . . ... ... .... ..., ... , 2% (3) Late failures from core-concrete interactions ,. ,. . .. ... ... . . 49 % ,
'(4) Early bypass (mostly due to SGTR) . .. .. .... . . ... ....8%-
Early failures due to phenomena (HPME or ALPHA)
-(5) . . . . . .. 1%
TOTAL . . . .. , ... . .. ... . .. . . .. 100 % The overall core melt frequency is 5.72E-5 per year as reported in Volume I.-- 1-1
i l The main conclusion to be drawn from these findings is that the CPSES containment provides adequate protection to the public. The principal concerns in that regard are the frequency and timing af potential early containment failures. With the exception of Steam Generator Tube Ruptures (SUTR), all other possible modes of early , containment failure have frequencies nearly at or below the reporting cut-off levels of IE-7. Although classified as early failures, most SGTR events would take several hours to reach core melt, allowing that time to be used fo accident diagnosis and management. Regarding the late containment failures, aliaost all are due to overpressurization from non-condensibles originating in a post dry-out core concrete attack. This type of failure is protracted over approximately 36 hours, allowing for effective accident management, CPSES exhibits three specisie design features that are the underlying causes for the good performance of the CPSES containment. The first is the very large containment free volume, which renders the CPSES containment relatively - invulnerable to early hydrogen burn events and to the direct containment heating (DCH) phenomena associated with high pressure melt ejection sequences. Calculations performed for the CPSES containment show that the coatainment would not reach the failure pressure even if a hydrogen burn from a 100% Zirconium oxidation were postulated. Furthermore, the assessment of the direct containment. , heating phenomena for CPSES, which is based on the conservative analyses for the Zion plant under the NUREG-il50 program, has shown that the CPSES containment is unt" ely to fail as a result of a DCH overpressure transient.
- The second and third important containment features are associated with the reactor cavity configuration.
First, the reactor cavity has a large flat floor area of over 70 m (800 ft2 ). This results in a shallow debris bed of only a few inches thickness, which is coolable by an overlaying layer of water, and which would result in a slow concrete penetration rate if the debris is dry and not self cooled by convection to the atmcsphere. Second, there is no curb at the containment floor- elevation surrounding the reactor : cavity exit for the instiument guide tubes' that would prevent the return of water from the main containment floor to the reactor cavity. As' a result, all the water injected into the containment or released inside the containment has to boil off before the debris in the reactor cavity can dry out. Accident sequences with failure of RWST injection dominate the plant damage states that dominate the 1-2 ^
. . . _ . . _ _ - - ~- .- - -- - . .- - , _ , -, ,,
s containment failures. Therefore, the absence of a reactor cavity curb means that even if only the RCS water inventory is released to the containment, this entire water inventory must boll off and be in the - form of steam in the containment before the debris in the reactor cavity can dry out After that, containment pressurization proceeds very slowly, due only to non-condensible gat generation in the core- -; concrete attack. This characteristic results in the protracted containment failures.- Furthermore, any-condensation or water injected into the containment by accident management would drain back to the reactor cavity and to the debris. Finally, the back-end analysis did not reveal any vulneiabilities nor the need for any plant irnprovements. 1 i l 5 1-3 _ . _ _ _ _ _ _ _ _ _ _ _ _ . _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ . _ _ _ _ _ _ _ _ i___i-.__________._______
- 4. IIACK-END ANALYSIS The back-end (sometimes called containment) analysis of the CPSES IPE utilizes the approach of a 1.evel 11 probabilistic risk assessment (PRA), whose methods and results are summarized in this section. A :
. Level 11 analysis involves two types of considerations: (1) analyses of physical processes during severe accidents, where degraJed core and containment thermo-hydraulic variables are determined along with source terms for the accident progressi,ms, and (2) a probabilistic component where the likelihood of the Various outcomes is assessed. The relationship of a Level 11 analysis to a Level I (and a Level 111, ! sometimes called site) analysis is shown schematically in Figure 4.01, The starting point for the Level
' 11 analysis is the plant damage states (PDS). These are the bins into which the core melt sequences l determined in the Level 1, or plant, analysis are collected. The guidelines for these bins are determined such that sequences in a given bin have similar accident progressions and outcomes of similar likelihood.
For each PDS there is a containment event tree (CET). The path through each CET begins with a PDS and ends with a CET end-state which is defined by a containment failure mode, time and release ! fractions. These CET end-states are later binned into release categories. Thus, release categories represent types, quantities and timing of radioactive material releases, This section begins in Section 4.1 with the summary of plant data requested in section 2.2.2,6 of NUREG-1335. That is followed in Section 4.2 by a heription of plant models and methods utilized for the physical processes analyses. Sectior 4.3 describes the criteria and the results of the binning process whereby the Level I sequences are grouped into PDSs Section 4.4 identifies the possible challenges to. the CPSES containment, including those leading to containment bypass, early failures and late failures, it also examines the possible ways ia which the containment might fail to perform its function by evaluating penetrations, evaluating the limits for liner tear at discontinuities and evaluating the ultimate strength of the containment to catastrophic failure. Section 4.5 presents CETs and the rationale for their development. Section 4.6 discusses the CET quantification which is the assessment of probabilities, and
~
the phenomenological developments that define the accident progression. It includes an analysis of uncertainties to demonstrate the absence of blJden vulnerabilities and the overall robustness of the-findings. ' Section 4.7 groups the CET end-states into release categories and provides the frequency, the type, timing and release fractions for the release categories. 4-1 1
= , -v--.: - , ,*---,y-- ,-,-,.,,,,,,,er,+.,v , .,,.,w -
ye.wm---me-y-_..c,.,. -e ,-m-,_u,-, . . , . , , ,w- m,---, -.w ,r--,,
Figure 4.0-1: Schematic Representation et PRA Levels c- -i,,,_1! SPECIFICATION OF THE INITit', LNG EVENT i (2i RESPONSE AND ST ATUS OF SUPPORT SYSTEMS (SUCCE t.S/F AILURE COMBINATIONS) A
~
LEVEL I -~
;3) [ARLY RESPONSE AND STATUS OF FPONTLINE 6
_ _ _ .J p SYSTEMS (SUCCESS / FAILURE COM91NAllONS) L (4) 1.ONG TERM RESPONSE OF FRONTLINE SYSTEMS y (SUCCESS / FAILURES COMBINATIONS) i ___ 5i SPECIF: CATION OF PLANT DAMAGE STATE y. r-PHENOMENOLOGICAL RESPONSE OF (DEGRADED) LEVEL 11 -- (
/ 6 ')
REACTOR CORE I L , (7) RESPONSE OF CONTAINMENT STRUCTURE { A
-~7 8 T SPECIFICATION OF RELEASE CATEGORY V l.e. SEVERITY OF RADIOACTIVE RELEAbES 1
1'
) METEOROLOGICAL DISPERSION SEQUENCE LEVEL lil --
I i ('i o') EVACUATION / EMERGENCY ACTION RESPONSE 73 (1[]) s_ SPECIFICATION DAM AGE LEVELS OF PUBLIC HEALTH AND PROPERTY 4-2 \ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
m 4.1 ManLpata and Plant Description As discussed in Section 4.2, the severe accident sequences that are relevant to the CPSES IPE were , analyzed using the MA AP code (Refs, 2,7,10), in order to perform those analyses, it was necessary to collect a substantial amount of plar.t data, whica is su amarized in the co<le's parameter file. A detailed derivation of this file _ is recorded in Reference 25. The present section summarizes some of that information by identifying and highlighting most of the component, system and structure data that is of 'i significance in assessing severe accident pogressions. Additional information including the sources and - the derivation of the information presented in this section is available in Reference 25. This section is organized into the four subsections suggested by Table A.1 of NUREG 1335, viz.: , a Reactor Core, Vessel and Primary System Containment System Emergency Core Cooling Systems (ECCS) and other water injection / recirculation systems
- Auxiliary Building Each subsection contains a narntive followed by tabular data in that category. Figures that include the drawings reques ed in NUREG-1235 are at the end of the section.
r 4.1.1 Reactor Core. Vessel anJ Primary System Data
-The reactor core consists of 193 fuel assemblies, each a 17 x' 17 rod array with 264 fuel rods,24 rod
, cluster control guide (RCC) thimbles and an incore mstrumentation thimble. The fuel rods consist of slightly enriched UO2 pellets. Fuel rm! cladding is Zircaloy-4J The RCC guide thimbles and instrumentation thimbles are Zircaloy-4, ivith type-304 stainless steel sleeves positioned at each axial location of an inconel-718 spring clip grid. The sleeves are fastened to the eight grids at approximately; equal distances along the length cf the column. The RCC guide thimbles are secured to the top and - bottom nozzles to complete the assembly. , 4-3
. - . . . - = . . . . .
- , . - -... .. - . --._ ,. - . . - . - - - - ~ .- - . .- - . . . . - . - _ _ - .
1 The reactor v-ssel is sufficiently described by the data in Tables 4.1.1-6, 7 and 8.- The primary system is described in Sectian 3.2.1-7 (Volume i of this repon: Front End Submittal), .I
-Additional reactor core, vessel and primary system data is summarized in Tables 4.1.1-1 through 12 as follows:
Table 4.1.1 1 Core blaterials Weight and Volume Table 4.1 1-2 Cort Geometry Table 4.1.1-3 Core Per0rmance Characteristics Table 4.1. I-4 ' Rod Cluster Control Assemblies Table 4.1.15 Burnable Poison Rods Table 4.1.1-6 Reactor Vessel (Rm hietal hiasses Table 4.1.1-7 RV Fluid Volumes Table 4.1.1-8 RV Ge<m:etry Table 4.1.19 RCS Fluid Volumes Table 4.1.1-10 RCS_ Valve Data and Serpeints Table 4.1.1 11 RCS Normal Full Power Operating Conditions . Table 4.1.1-12 RCS Aletal htasses in addition the following drawings of the reactor coolant system are provided: Fipre 4.1-1 General Plant Schr'tatic Figure 4.1-2 Simplified Process Flow Diagram of the RCS Figure 4.1-3 hiain Components of the RCS Figure 4.1-4 - Primary Side Process Flow Diagram
- Figure 4.1-5 Secondary Side Process Flow Diagram Figure 4.1 Structural Plant Arrangement i
Figure 4.1-7 Containment Cut View with Main RCS Components . Figure 4.1-8 _ General Reactor Vessel Assembly i Figure 4.1-9.1 Reactor Vessel Lower Internals Assembly Figure 4.1-9.2 Lower Core Support Assembly Figure 4.1 10 Bottom htounted Reactor Vessel Instrumentation Figure 4.1-11.1 - Pressurizer (Cutaway) 44' , en v --9%-e m- t* + ,-- --y,- --4,-.--y.c-e. 3-~ , w =nv-r w - w -e 4-e
- . - - - - ~ - . _ . . . - . . - _ - .-. -. .. - - - - - -. - _ - . - -.- ..- - .~. -
t f I Igure 4,1 11.2 l'ressurifer Relief Tank Figure 4,1-12 Steam Generator (Cutaway) Figure 4.1 13 iteactor Coolant l' ump (Cutaway) , Figure 4.1.14 Accumulator Tank t
?
Table 4.1.1-1: Core hiateriah Weight and Volume i n-Core hiaterial Dendly Volurne Weight Active Core, Cold (Ibm /in') (in') (Ibm) l'uel, UU, 0.376 592,140 222,645
}
Zirealoy-4 0.237 198,281 46/ 93 I inconel718 0.296 6.221 1,841 ; Stainless Steel 304 0.285 4,758 1,356 i-l l 4-5 l l _ _- . _ _ ._
- t. - - ~ . - . . - . _ _ _ . _ . _ . . _ . , . _ . _ _ _ . . _ . _ _ . _ _ . . , __ _. _ . _ _ . . . _ , _ . , . . _ . _ , . , . . . _ , . . , _ . . . _ ,
- . - . - . _ . . - . . - . . - . . . ~ - - - - . - _ - - - . . . - - - - . - i i l Table 4.1.12: Core Geometry Core Aveage Active I uel licight, in 144.0 Lattice Con 0guration 17 x 17 Lattice Pitch, in 0.496 ; i Number ofl'uel Assemblies 193 Region 1 65 Region 2 64 Region 3 64 Number of Rods per Assembly 264 j Enrichments, w/o U 235 iteo gi n i 1.60 Region 2 2.40 i Region 3 3.10 Outer Fuel Rod Diameter. In 0.374 Cladding Thickness, in 0.0225 Diametral Gap, in 0.0065 , UO, Pellet Diameter, in 0,3225 Density (Percent of Theoretical) 95.0 Volume Fraction of UO: in Pellet Region 0,9880 Cladding hiaterial Zircaloy-4 Gap hiaterial llelium Guide / instrument nimble hiaterial Zircaloy-4 Structural Grid hiateriai inconel 718 Grid Sleeve hiaterial SS-304 Number of Grids over 11eight of Assembly 8 4-6 L i
=q,e .p -y- p.- , -. y y-.ws,.-g7,- g a.w-y- -ryw i% _-
p m-
. ~ . . _ _ . _ . . _ _ _ _ _ _ _ _ _ _ . _ _ -
m.____ --
~...-__.m___.--
1 i Table 4.1.1-3: Core l'erformance Characteristics l l 4 Ileat Output, MWt 3,411 l lleat Generated ir Fuel, % 97.4 Coolant Average Temperature at liFP, *F 591 Operat i ng Pressure, psia 2.250 , J - Table 4.1.1-4: Rod Cluster Control Assemblies 4 il Material Ag-80%, in 15%, Cd 5% , I Numbrr of Fuel Assemblies 53 Containing RCC Assemblies Number of Absorber Rods per RCC Assembly 24 1 Table 4.1,15: 11ornable Poison Rods Material llorosilicate Glass Content,11:0,, w/o 12.5 Number in Core 944 Table 4.1.1-6: Reactor Venel (RV) Metal Masses, Ibm Core llarrel Above the Top Elev. of the Core -39,694 llelow the Top Elev, of the Core 124,315 1_ Upper Plenum Internals 68,702 Upper Core Support Plate 39.500 R V W all Above Upper Head Flange 202,302 Ilelow Upper llead Flange 661,656 Lower Plenum Head 82,382 4-7
. - . ) .,.,..n~.,...-,-.... , ..-n. ..,,,.n.., - - - + . - . . . ~ . _ . , , - . . _ . . . - . _ . . . _ _ . . , , , . _ . . . . . - - , . - - - . . . , , -. , ~ , . , . . , - . . , - - - - . , - - - - - -
Table 4.1 '. 7: RV Fluid Volumes, it' RV Total Vrlume 4,571 Downcomer 936 Lower Plenum 889 Core + Bypass 844 Upper Plenum 999 Upper llead 903 Table 4.1.18: RV Geometry, in inner Diameter of RV Wall Above Nonles 170.88 llelow Nonles 173.00 Spherical Radius of Lower Plenum liead 88.16 Thickness of Lower Plenum IIcad 5.52 TaHe 4.1.19: RCS Fluid Volumes, ft'
'rimary System Total Water inventory Approximately 12,000 RV Total Volume 4,571 Pressurizer (steam + water) 1,800 Pressurlier Surge Line 46 Ilot Legs 317 Intermediate Legs 531 Cold Legs 332 Reactor Coolant Pumps 314 Steam Generator Inlet Plenums 670 Steam Generator Tubes 2,528 Steam Generatoc outlet Plenums - 630 Secondary System Steam Generator (steam + water) 5,954 Nominal Main Steam Line Piping from 962 SG to MSIV wer loop)-
48
- l. 1 TMy i
f Table 4.1.1 10: RCS Valve Data and Setimints Pressurlier Number of Safety Vaives 3 Safety Valve Setpoint, psia 2,500 Saf et) Valve Ratni I: low, ihm/hr 420,000 Number of l'ower Operated lleliel yalves ? PORV Setpoint, psia 2,350 PORY Rated i: low, Ibm /hr 210,(XX) Spray Setpoint, psia 2,300 Steam Generator (SG) Number of Safety Valves 5 SG Safety Valve low Setpoint, psia 1,200 SG Safety Valve liigh reipoint, psia 1,250 1 Avg SG Safety Valve Rated 1: low, Ibm /hr 911,779 Number of Power Operated itelief Valves ! PORY Setpoint, psia 1,140 PORY Rated Flow, ihm/hr 735,370 Table 4.1.1 11: RCS Normal i ull Power Operating Conditions W'rimary I S. ystem System Pressure, psia 2,250 Nominal Water Temperature, 'F 591 Total Thermal Flow Rate, Ibm /hr 142,(XX),000 Pressure Drops, psi Across Core 25.8 Across RV (including nonles) 46.3 , Secondary System SG Pressure, psia 1,000 Main Feedwater Flow per SG, Ibm /hr 3,390,000 SG Water Mass, lbm 102,500 49
Table 4.1.1 12: RCS Metal Masses, Ibm ^ llot Leg (per loop) 15,972 Intermediate Leg (per loop) 23,800 l
- Cold Leg (per loop) 14,9(4 4
Pressuriier Shell and lleaters 195,508 Pressuriier Surge Line 12,827 - SG Total Shell 310,000 t SG Lower llead & Tubesheet 180,800
)
t t 4 l l-I 4-10 (. l . _ , , _ , , . - . , , , . , . , . , _ - - , . - - , . . . . . . _ . . . - . . - - - , _ . - - - - . . . . . .
i 4.1.2 Cratainment Sntem lhe key features of the containment are: (1) safeguards and isolation systems and (2) containment design and structutes. *lhese features are summarized in Tahles 4.1.21 and 2, respectively. , 1 4.1.2.1 Safegais and.lsolatinSnlem The containment safeguards systems are the Containment Spray System (CT) and the Fan Cmlers (FC) that are part of the Containment Air Cooling and Hecirculation System (CACRS). The 8.alnment Isolation System (CZ) is a system designed to provide integrity of the containment bounoary. *1hese systems are discussed in detail in the following sections. 1 CuntaimnenLSMay Svalt!nKD ! The Containment Spray System is discussed in Section 3.2.15 (Volume 1 of this submittal: Front End) and a diagram of the system is shown there, in l'igure 3.2.1.5. Additional information on the CT system is provided here and in Tahle 4.1.2-1, Figure 4.1-19 and Figure 4.123.3. - The CT system consists of two separate, independent, and full capacity trains. Each train contains two spray pumps, one heat exchanger, two chemical eductors, spray headers, spray nonles, associated piping, valves, and instrumentation. Failure of the CT system does not result in an initiating event. The function of the CT system is to maintain the containment pressure within % dulgn limit after the-following initiating events:
- Imss Of Coolant-Accident (l.OCA)
Main Steam Line lireak (MSLil) inside containment
- Feedwater Line lireak (l'WLil) inside containment Th6 CT pumps are provided with suction lines from both the Refueling Water Storage Tank (RWST) and the containment sumps. 'Ihus, the system is capable of providing the containment with short term (injection mode) and long term (recirculation mode) cooling. Each pump train takes suction from the RWST via normally open motor-operated valve 1 HV-4758/4759 'Ihe CT system shares the RWST with 41;
the Safety injection System (SI), Residual lleat Removal System (Ril) and Chemical and Volume Coritrol System (CS). In addition, the Ril, SI, and CT systems share RWST isolation valve 151-047. Following depletion of the RWST, the suction of the CT pump train is switched over to its respective containment sump via normally closed motor-operated valve 1 ilV-4782/4783. The Ril and CT systems share the containment sumps. De design 110w rate of each CT pump is 3000 gpm at 260 psid. He design of the system is such that both pumps per trsin are required to deliver enough flow to the spray header to remove an adequate amount of heat from the containment atmosphere. The pumps are powered from separate Class 1E 6.9LV buses. Each CT pump room contains two spray pumps and two associated room cooler units to ensure that the ambient room temperature remains within equipment qualification limits. The room cooler units are powered by Class IE 480V Motor Control Centers (MCC) and are supplied chilled water by the Safety Chilled Water System (Cil). CT pump minillow protection is provided by normally open motor-operated valve 1 FV 4'.'721/4772-2/4773-l/4773 2. The pump seals ate comed by the Component Cooling Water (CC) system; the pump bearings are cooled by the Station Service Water (SW) system. The pumps are actuated by a Safety injection (*S") signal. The pumps aim receive a confirmation start 1 signal when containment pressure reaches the hi-3 ("P") setpoint. Following the "S" signal, the pumps operate in miniflow until the hi-3 setpoint is reached. At dist point, the spray header isolation valves l-l{V-4776,4777 open and the minillow valves close, Each pump is equipped with as associated chemical eductor which delivers a 28 30 weight percent solution of sodium hydroxide to the pump suction. One chemical additive tank provides gravity flow to each eductor venturi section. Success of the chemical addition system is not considered essential for system operation. Each pump discharges to a header which routes flow to its respective heat exchanger. The CC system supplies cooling to the shell side of the heat exchanger via normally closed motor-operated valve 1 ilV. 4574/4575. _ The valve is opened automatically by a "P" signal. Upon discharge from the heat exchanger, How is routed to the spray header via normally closed motor-operated isolation valve 1 ilV-4776/4777. The spray headers route flow to ring headers located in four regions of the containment. Each header contains a restriction orifice which balances the Dow to each ring. 1 4-12 wr- -r: g-e*e-tt*,11-v r. -u y- m,e e pw -.p wgmyw--c.,w 4.,--,-e,f . ,p v. . . . g-,- . is,,,,,g,,,-ge., ,, -,-.y-r- t.- m *1--,esw y ww m=--- --=,q--aryp g.n er--- y .v- ewi a wni s ey
1 l Technical specifications require the CT pumps and active valves to be operability tested quarterly. During the pump test, CT flow is recirculated back to the RWST via normally locked-closed test header isolation valve ICT-050/049. Among the valves stroke tested are RWST suction isolation valve 1 ilV-4758/4759, containment sump suction isolation valve 1 ilV-4782/4783, and spray header isolation valve 1 ilV-4776/4777. l'or the duration of the testing, the CT train remains inoperable. In addition, the CT t train is disabled prior to quarterly lingineers Safety Features Actuation System (ESFAS) slave relay actuation testing in order to prevent pump damage. CentainmcDLfan Cooltu The fan coolers are part of the Containment Air Cooling and Recirculation System (CACRS). Fan cooler 1 infortnation is summarleed in Table 4.1.2-1 and in Figuies 4.120.1 and 20.2. The CACRS for each unit consists of four 33-l/3 percent capacity cooling units and fans. The cooling unit consists of eight cooling coils. During normal operation, three out of four cooling units and fans will operate. The CACRS is 1 not required to operate following a Design liasis Accident (DilA).' Following a I.OCA, the S" signal shuts the fans down and closes the fan discharge dampers. Following a loss-of-offsite power, the lilackout Signal (110S) automatically starts the fans. The CACRS fans and dampers are each powered from two separate and independent electrical sources Train A and il of Class IE AC and DC buses, respectively. The nmesafety related chilled water system provides cooling to the CACRS cooling coils. Fan cooler operation is not credited in the CPSilS IPl!. Ilowever, the benefits of fan coolers were evaluated in section 4.6.3 for potential use in accident management. The potential impact of fan coolers !
- on the severe accident progression is twofold: (1) they can extend the RWST duration by preventing or delaying the containment pressure from reaching the spray set point; and (2) fan coolers can prevent containment failure Jue to overpressure as calculated in Section 4.6.3. These advantages notwithstanding, fan coolers were not credited because: (1) fans at CPSES are cooled by chilled water which is isolated L - on a containment isolation signal; (2) restarting the fans would require operator intervention which is not procedurallied for severe accident situations; and (3) the fans are not qualified for operation in a severe accident environment. Therefore, fans are assumed to operate only until an Si signal is generated, since - this is the expected boundary condition for the accident sequence development, Operation for this short period has little bearing on the accident progression. Neglecting the fans for the balance of the sequence 4-13 - , _.. _ _ _ _ - _ _ _ _ .. ,_ _ , , _ __ ___.__ _, _ _---m_ .-_ . _ . . _
h is conservative because when fans are not credited, a higher containment pressure is calculated, sesulting in a rnore severe challenge to the containment than it they are assumed to be in operation. This I assumption allows bounding of the scenarios in which the f ans operate with similar scenarios in which they do not, llowever, since it may be possible to restart fans, a sensitivity study is described in Section 4.6.3 demonstrating that under certain circumstances, fan operation alone, without additional ECCS or spray, can prevent containment failure. While credit for this capability is not taken in the risk assessment process as indicated above, the information was developed for incorporation into an accident management { knowledge base. C201stin!Mulh21illieRinLCHLiCZJ The design objective of the CZ is to allow normal and emergency passage of fluids through the 1 containment boundary while preserving the integrity of the boundary. 'lhe CZ logic is part of the lingineered Safety Features Actuation System. The CZ was modeled in the Front End of the IPE (Ref. 6). For completeness it should be mentioned that the CZ includes the following subsystems: 1 Steam 1.ine Isolation - closes the main steam isolation valves (MSIV) and main steam drain pot isolation valve. Once steam line isolation is initiated, the ESFAS output relays are latched and must be manually reset, Resetting the steam line isolation signal does not - cause the valves to re open.
- Main Feedwater I.ine Isolation - closes all feedwater isolation valves. Once feedwater line isolation is initiated, the ESFAS output relays are latched and must be manually reset. Resetting the feedwater isolation signal does not cause the valves to re-open.
- Containment isolation Phase A closes all non-essential process lines penetrating the containment. Containment isolat on Phase A is initiated by the Safety injection Signal or manual actuation of either of two control switches per train for Phase A lsolation on the control board. Once Containment Isolation Phase A is initiated, the ESFAS output l relays are latched and must be manually reset. Resetting the Containment isolation Phase -
l: A initiation signal does not cause the isolation valves to re-open. 4-14 F me-vn v.-r-e r- .,---..-4....+,---.re,-,...r--,wm.__.--,-wwm.--. ,,y.m..9oc,- c 'w,6% . ,w,y,,, ,,-wf r y ,-wy,., , , , , , ,,w,yc ,w- - y , -,e.g . 4. -yw,- ye-- . pey.- a,,,w
l l
- Containment isolation Phase 11 - closes all remaining process lines, with the exception of those serving Engiacered Safety Features functions penetrating the c(.ntainment.
Containment isolation Phase 11 is initiated by a *P" signal derived f rom the containment spray actuation signal or by manual activation of both of the two contrcl switches per train for Containment Spray Actuation on the control board. Once Containment isolation Phase 11 is initiated, the IISFAS output relays are latched and must be manually reset. l Resetting the Containment isolation Phase 11 initiation signal does not cause the isolation valves to re-open.
- Containment Ventilation isolation (CVI) closes all ventilation lines connected directly to the containment atmosphere. CVI is initiated by automatic or manual initiation of Containment isolation Phase A or manual initiation of Phase !! to limit radioactive emissions during ace' dent / post accident operations To hmit radioactive emissions during normal operation, the CVI is also initiated by high containment ahborne radiation Once the CVI 14 initiated, the liSFAS output relays are latched and must be manually reset.
Resetting the CVI does not cause the isolation valves to re-open. Containment penetrations and their respective h,alation schemes can be classified as:
- Type A 1.ines that form part of the reactor coolant pressure boundtry (RCPil). These penetrations are provided with one of the following isolation schemes:
One locked-closed isolation valve inside and one locked-closed valve outside the-containment, One automatic isolation valve inside and one locked-closed isolation valve outside the containment. One locked-closed isolation valve inside and one automatic isolation valve outside the containment. One automatic isolation valve inside and one automatic isolation valve outside the ! containment. 4-15 ,-em,w+,-,, .in . chmw,w. _w..., g, w ,s,,, , c,w--, ,, ,,--e--,,,,--., , w~,.. , _ pwn.,y. .ww--,,y ,y - n-n y --y - - , y p,p w - e (
Type II - Lines that connect directly to the containment atmosphere, nese penetrations are provided with isolation schemes identical to those set forth for Type A penetrations as well as the following additional isolation schemes: ne redundancy requirement is satisflat by having two isolation barriers in series, one on each side of Type A and Type 11 penetrations.
- One blind flange inside the containment and one locked closed isolation valve outside the containment.
One blind llange inside the containment and one blind flange outside the containment. Type C Lines that are part of a closed system, i.e., lines that are neither part of the RCPil nor connected to the containment atmosphere. These penetrations are providal with at least one containment isolation valve that is either automatic, locked-closed, or capable of remote-manual operation. These valves are located outside the containment and as close to it a:s practicable, l Special Containment isolation Provisions - Special provisions are provided for certain valves. Valves in lines required to operate post accident are designed to remain open or be opened following the accident, but consistent with containment isolation requirements, l they can be closed by remote-manual operation from the control room. 1 nere are four instrument lines that penetrate the containment that are required to remain functional following a LOCA or steam line break. Isolation is provided by means of sealed bellows.that are connected to a fluid filled tube. The arrangement consists of a double isolation barrier, if the instrut.ient line breaks outside the containment, leakage of the containment atmosphere is prevented by virtue of the sealed bellows. If the instrument line breaks inside the containment, leakage is prevented by a leak-tight
- diaphragm installed in the pressure instrument that is designed to withstand the full containment design -
pressure. 4 16
. _c -.-_: - __ -_ _ . . _ _ _ ._ - _ _
4.1.2.2 Containment Design and Structures The CPSES containment is a large, dry, reinforced concrete structure with approximately 3 million cu. ft. volume and a 50 psig (64.7 psia) design pressure. This section contains an overview of the structure and a description cf the :: 4reements, liner, penetrations, reactor cavity area and leakage testing. Overview of the CPSES Containment Structurr The Comanche Pen containment structure is a fully continuous, steel lined reinforced concrete r,tructure, consisting of a vertical right cylinder with a flat base and a hemispherical dome. It is supported on an essentially 11.it founJatian with a reactor cavity pit. A welded steel liner is attached to the entire inside
;urface of the containment (walls, dome and mat) with anchors to ensure a high degree ofleak-tightness. ,
The design objective is to provide vapor containment and limit leakage of radioactive material which might be released from the core during a design basis accident. It also protects the RCS from extreme environmental conditions including tornados and external missiles, ne containment structure, as shown in Figures 4.1-18.1 through 18.3, consists of the following: A cylindrical wall (internal diameter of 135 ft 0 in.), measuring 195 ft from the top of the base to the springline of the dome with a thickness of 4 ft 6 in.
- A hemispherical dome with a thickness of 2 ft 6 in. The inside radius of the dome is equal to the inside radius of the cylinder, so that the discontinuity at the springline due to the change in the thickness is on the outside surface.
A flat concrete foundation base mat with a thickness of 12 ft 0 in. An additional overall view of the containment layout is provided in Figures 4.121.1 through 21.3. \ Containment sump piping and arrangement are shown in Figure 4.122.1 and 22.2. l Reinforcements The principal reinforcement used in the containment shell (mat, walls, and dome) are No.18 bars, made-continuous at splices by the itse of cadweld connections. Shell reinforcement is illustrated in Figure 4,i- - 4-17 r r e-e--- - E --e wr c ,-r -,u t ,- ,a E .--mr vn --e , -,E.-4-<,-..-NE r ,-www e-, ve-. U. +r ,,- 4,5 = v E r-ev... Y w w 3 -r m+,v~r- -ener-b-.--.-e,:. ' .r erd y-tr E E N -m. E--r r- -w-
= ~ .---,-+, ,-m*
23.1. De reinforcing steel pattern in the cylindrical wall consists of vertical bars (inside and outside faces), horizontal hoop hats (also at each face) and 45 degree diagonal bars in each direction, near the outside face. The dome reinforcement consists of top and bottom meridional layers of rebars, extending from the cylindrical wall vertical bars. Circumferential hoop bars are provided in the top and bottom layers of the dome. The merldlonal reinforcement terminated at the apex of the dome is anchored by cadwelding the end of the rebar to a fabricated steel ring assembly. At penetration openings, reinforcing steel is generally bent around the openings; supplementary bars are provided around the opening when requhed by design. At the major penetrations (i.e., the Personnel Lock and the Equipment llatch) some of the wall reinforcement is terminated at the opening by cadwelding steel plates on the end of the bar. Additional reinforcing is provided around these openings to carry stress concentrations and make redistributions at these openings. The foundation mat is reinforced with top and bottom layers of bars as shown in Figure 4.1-23.2. 1 Uttti
%e entire inside surface is lined with welded steel 3/8 inch thick at the wall,1/2 inch in the dome. A 1/4 inch thick plate is used on top of the foundation mat and covered with a 2 ft 6 in. concrete stab, the top of which forms the floor of the containment. Typical steel liner details are provided in Figure 4.1-4 23.1 Liner chase channels are provided at liner seams which, after construction, are inaccessible for other means of leak tightness examination. Ec liner steel plates on the wall and dome are anchored into the concrete with 5/8 in by 6 3/8 in. long headed, welded stads. The studs in the cylindrical wall and dome are spaced approximately 12 inches each way. De vertical wall liner is anchored at the foundation - mat. The bottom liner is insta' led after foundation mat construction and is welded at seams to structural members embedded in the top of the mat. The embedded structural members are approximately 8 by 10 ft apart. Locally thickened liner plate sections are provided at penetrations, at major pipe and duct support attachments and at the bottom of the cylindrical wa!!'s steel liner.
- Containment Penetrations -
From the perspective of severe accidents, the CPSES containment penetrations, as with most other large, dry PWR containment penetrations, can be divided into the four categories considered in Reference 11: (1) Large Opening Penetrations, (2) Purge and Vent System isolation Valves, (3) Piping Penetrations and 4 18 . __ .. _ E_
- _ _ --t._.__ -_ . - . . . _ , _ _. 2_.2_ a _ a.u,a a m . .m _,# ,
4 , I (4) lilectrical Penetration Assemblies, the most important of which are shown in 171gures 4.123.4 through
- = 23.7. The severe accident response of each of these types of penetrations is discussed in Section 4.4.
Figure 4.123.3 shows the location of the various penetrations along the containment wall. BfELDLCAEily.Mu
; The CPSliS reactor cavity has very favorable severe accident features: (1) a thick basemat (12 ft) with t
a buried (unexpised) liner that delays basemat penetration for sequences in which core concrete interaction occurs; (2) a large area (81211.2) that is conducive to the formation of a shallow bed, which _ is highly likely to be coolable in the aftermath of violent events such as steam explosions or high pressure : r melt ejection (IIPMH) or burns in the cavity; (3) an always open but tortuous path to the containment .; lower compartment, which makes dheet containment heating (DCll) difficult because the obstacles would cause most of the debris to be de-entrained hom the blowdown gases; and (4) the absence of a curb between the cavity and the lower compartment, allowing nearly all the water in the containment to drain into the cavity and delaying the onset of core concrete interactions until cavity dry out (unless the debris falls into a non-coolable configuration, which is unlikely due to the large cavity area). The following drawings of the reactor cavity area illustrate these points: Figure 4.1-15 Containment Wa er 1.evel vs Containmer.t Water Volume Figure 4.1-16.1 Reactor Vessel Supports Figure 4.1-16.2 Reactor Vessel Sepports Details Figure 4.1 17 Containment Cavity Containment 1.eakaptIssiing Reliability is assured by conducting periodic tests to check the operability of the isolation valves, actuators, and centrols. Cor:ainment leakage tests are performed periodically to verify that containment leakage is maintained below the limits stated in the technical specifications. The leakage testing program consists of the following types of leakage tests. J
- Type A tests are those tests that are performed after the containment building has been ,
completed, prior to operatioa, and at periodic interval thereafter, to determine the overall containment integrated leakage rate. Three Type A tests are performed at approximately equal intervals during each 10 year service interval. The third test is performed while
~
4-19 4 w .E.-.I.<,_. m.,_. ._ , , . , , . ..,,,,....,,._,,.....___,,,,.,..._,,,-,_~,. [ __ ' _h , , , , , _ , ,_ , ' y _._ __, ,
l the plant is shut down for the 10 year plant inservice inspection. Type A tests are only conducted while the plant is in the shutdown condition. l
- Type 11 tests are those tests that are performed periodically to determine leakage rates for individual mechanical and electrical penetrations, air locks, and hatches. Type 11 tests ;
are performed during each reactor shutdown for refueling, or at another interval, but in no case at intervals greater than two years. The personnel airlock and emergency altlock are tested after each opening or at six month intervals if not opened for that period of time.
- Type C tests are those tests that are performed periodically to measure containment isolation valve leakage rates. Type C tests are performed during each reactor shutdown for refueling but in no case at intervals greater than 2 years.
4 Any major modification or replacement of a component that is part of the primary containment that is performed after the preoperational leakage rate test will be followed by a Type A, Type 13, or Type C test, as applicable. All other requirements for regularly scheduled leakage-rate tests apply. Furthermore, a fail safe feature is incorporated into air-operated and solepoid-operated isolation valve design, so that in the event of actuating power loss, the valve assumes the position that ensures safety, Each train of electrically activated valves is supplied from separate and independent Class IE sources. , The motor operated valves are powered from Class IE 480V AC MCCs. Pilot solenoids for the air operated valves are powered from Class IE 125V DC and ll8V AC dinribution panels. l 1-4 20 l ._ ._ __ _. ._ _ _. , . , _ . . _, _ _
I Table 4.1.21: Safeguaids Splems I i _ 1 Containment Sprays Sptern Number of Operational Trains 2 Number of Operational Purnps per Train 2 ; Differential Prenure Across Nonte, psig 40 Mass Flow Rate per Pump, gpm 2.896 Pressure Setpoint, psia 35 Fan Coolers (FC) l Number of Operational FC _ 4 Volumetric Flow Rate per FC, cfm 65,000 Inlet Cooling Water Temperature, 'F St Inlet Cooling Water Flow Rate per FC, gpm , 320 I i l l l l l 4-21 P [
l i Table 4.1.2-2: Containment Design and Structuses Total Free Volume, ft' 2,985,126 Design Pressure, psig 50r Design Temperature, 'F 280 ! Absolute Failure pressure, psia 129 l Concrete Composition Limestone / Common Sand Alass Fractions, % ! SiO, 35.80 Ca0 31.30 Al:0, MO 1.22 K:0 Na,0 0.08 Mgo, hinO, TiO2 0.69 Fe2O 3 1.44 Fe 0.00 Cr30 3 0.01 112 0 4.70 CO, 21.15 0, 0.00 Upper Compartment Outer Wall Type Reinforced Concrete Free Volume, ft' 1,984,422 hietal Equipment Volume, it' 15,835 Metal Equipment biass, ihm 7,759,150
. hietal Equipment Heat Transfer Area, ft' 22,208 Outer Wall inside Surface Area, ft' 68,821 ,
Outer Wall Total Thickness, ft 3.6 Outer Wall Liner Thickness, ft 0.036 Internal Wall Surface Area ft' 1,189 Internal Wall Thickness, ft 2.0 Internal Wall Liner Thickness, it 0.0 Deck Area, ft 2 7,038 Deck Thickness, ft 1.5 Deck Liner Thickness, ft 0.0 4 22 ,
- --. , , , ,p.
- . . . _ - . - - - _ _ --.-.-.- . . - - = _ ~ . -
l I 1 Table 4.1,2 2: Containment Design and Stnictures (continued) . 1.ower Compartment Outer Wall Type Reinforced Concrete Free Volume, ft' 836,282 -l Metal Equipment Mass, Ibm 2,133,583 , Metal Equipment lleat Transfer Area, it' 90,653 Outer Wall Inside Surf ace Area, it' 7,906 . Outer Wall Thickness, it 2,75 ! Outer Wall 1.iner Thickness, ft 0.0 Internal iVall Surface Area, ft' 166,671 Internal Wall Overall Thickness, ft 1.0 : Internal Wall 1.iner Thickness, ft 0.0 1 Floor Area, ft 7,530 Floor Thickness, it 9.3 Floor Liner Thickness, ft 0.0 Annular Compartment Outer Wall Type Reinforced Concrete Free Volume, ft' 151,146 1 Outer Wall luside Surface Area, ft' 41,457 i Outer Wall Total Thickness, ft - 4.5 . OJer Wall Liner Thickness, ft 0.03 i Floor Area, ft 2 1,546 Cavity Compartment Free Volume, ft' 13,276 j Area of Cavity Debris, ft2 812 lleight of Ilottom of Reactor Vessel above 110ttom 15 of Cavity, ft , Outer Wall inside Surface Area, ft2 - 2,896 Outer Wall Thidness, ft 14.4 Outer Wall 1.iner Thickness, ft 00 4 23 b r+w, , , - , , ,y . -,, . - - - . - , , , , - -.-,,.-.-%.,~y-v.,,-w'_y---.yv,-i,w.,.,,,w,7 . - . , , , , , , , ,m.,,,~,-m.m.7 a- ,------y , p .ns- s. g e 'r, , e , c- -
-e ,v.-,.,,.i.,...,-y..-.....+.,,m.
_._._.___.___-.____.._..-._-...___m_______.-.___~.___.
!y[r l'
l i i i Table 4.1.2 2: Containtnent Design and Structures (continued)
- Containment Sumps i- Number of Jperational Sumps 2 Area of Ove per Sump, ft 2 -81 J Surnp Depth, it 6
, Containtr:nt Operating Conditions l'rusure (Normal Full Power), psia 15 1
Temperature (Normal Full Power), 'F 117 Relative Humidity, % 100 i 1 i l i i 4-24 - l l j .- t -.;_..~ m-_.. . _ . _ _ _ _ _ . - . . _ _ _ _ _ _ _ - . _ _ . _ _ . . . . _ _ , _ _ _ . . . . - . _
4.1.3 JJnjngengLfgiglw1wg3ntVH'IECCS) The !!CCS for each unit conalsts of two separate and independent trains. 'lhe key features of the major components are:
- The Centrifugal Charging Pumps (CCPs), shown in Figure 3.2.1.6, automatically align to the RWST upon reecip'. of an "S* or "liOS* signal. The CCPs discharge directly to all four RCS cold legs and continue to maintain seal injection to the Reactor Coolant Pumps (RCPs). Emergency hotation is provided iiy redundant boric acid trensfer trains.
Suction Jmag the cold and hot leg recirculation phase is aligned to the discharge of the Ril pumps.
- The Si pumps, shown in Figure 3.2.1.8-1, start upon receipt of an "S* signal, and draw water from the RWST and discharge to all tour RCS cold legs. The S', pump suction is realigned to the discharge of the Ril pumps during the cold and he. leg reeirculation phase.
- The Ril pumps, shown in Figure 3.2.1.2, start upon receipt of an "S" signal, and draw water from the RWST. Atter RWST depletion, the Ril pump suction shifts to the containment recirculation sumps, and water that is pumped is cooled by the Ril heat exchangers and is delivered diteetly to the reactor vessel or to the suction of the CCP and Si pumps. A tross-tie is provided to ensme delivery of water to the CCP and Si pumps in the event that one train of Ril system fails or the motor operated valves in the suction line to the pumps fail. The Ril system for each unit consists of two separate, independent, and full capacity trains. Each train includes one Ril pump and heat exchanger. The heat exchanger flow control valve (1 IICV-606/607) is provided to allow the operator to control the RCS cooldown rate. Ileat is transferred from the RP., to the CC system. The Ril pumps are powered from two ,eparate and independent sets of Class 1E 6.9kV bases.
- Containment sumps provide suction for the Ril and CT pumps. One sump is provided for each train.
4 25
- One RWST is provided for each it. The RWST is shared between the 'lli and the CT systems. (
t
- The CCPs, SI, and Rif purnps are powered froin separate Class IE 6.9kV buses. The boric acid transfer pumps (llATl9 are powered from separate Class 113 480V MCCs.
- Mgures 4.126.1 through 4.126.4 show the ECCS valve alignments for the standby, _.
Injection, cold leg recirculation, and hot leg recirculation phases. The ECCS support requirements are: ;
- Each CCP, SI, and Rif pump room is equipped with an associated room cooler unit.
The room cooler units are powered by Class IE 480V MCCs and are supplied chilled water by the Cil system.
- The CCP and SI pump bearings are cooled by the SW system, and the pump i.eals are self-cooled.
- The lloric Acid Transfer Pumps (BATPs) are located in large open rooms, and the pump seals and bearings are self cooled.
- Each Ril pump room is equipped with an associated room cooler unit. The room cooler units are powered by Class IE 450V MCCs and are supplied chilled water by the Cil system.
- The R'8 pump seals are cookd by the CC system, and the pump bearings are self cooled.
4 26
, + - , . , , . , , , , . . . . , 4 .. _ _ - . . _ , . -- ,.._v,. .A L .. . . _L..
he ECCS testing acquirements ;ier Technical Specifications are:
* %e CCP5 and ilATPs are tested quarterly. During testing of the CCPs, these pumps are unavailable because the normal pump discharge flowpath is isolated and the .' umps operate in seeirculation.
- The Si putnps are tested quarterly. During testing of the Si pumps Si flow is recirculated back to the RWST via the minitlow lines. His configuration does not cause the pump to be inoperable, llowever, the pumps are inoperable during quarterly stroke testing of the minillow valves 18814A,Il and cioss connect isolation valves 1-8821 A.II.
- The Rll pumps are flow tested quanerly. Prior to the flow test, the pumps are disabled when the miniflow valves are stroke tested. During testing of the Ril pumps, Ril flow is recirculated back to the RWST via normally closed motor-operated t<:st valve 1-8890A,II. De Ril pumps remain operable during this test because the 3/4" test line does not divert enough flow to cause the pumps to be considered inoperable. Stroke testing of the RWST suction isolation valves I 8812 A Il are also completed quarterly.
Survivability concerns under conditions expected during a severe accident can be summarized as follows:
- The accident conditions are the accident and post accident environmental conditions of temperature, pressure, relative humidity, radiation doses, chemical spray, and llooding.
Rese environmental conditions are summarimi for each equipment location, and the equipment is evaluated accordingly. The requirements delineated in IEEE 323-1974 include principles, procedures and methods of qualification which, when satisfied, confirm the adequacy of the performance of the equipment under these environments.
- The plant specific equipment performance requirements during normal and a:cident environmental conditions have been identified and documented accordingly. These 4 27
_ . . . _ __.._____ _ _.___._ _._. .._._..__..._.-._.____.m _ Il performance requirements are compared to the demonstrated performance characteristics f: to show satisfactory performance and quali0 cation of equipment, j
- All vital coinponents and supports are located in buildings designed to Seismic Category I requirements. These buildings can withstand loadings due to tornado winds, i depressurization, repressurization, and esternal missiles. In addition, they protect the j components from the effects of flooding such that no Hood protection of specific components is requhed. Missile barriers within these buildings separate the trains such [
that a missile will not af feet both trains. Each train is protected fro,n the dynamic elfects from other piping associated with seismic or pipe break events.
'Ihe following quali0eation methodologies are used to qualify the electrical equipment: - Type Tests Generally used for qualification of equipment located in potentially harsh environments. This testing consists of using an identical item of equipment under similar conditions with supporting analysis to show that the equipment is quallned for its specinc application and, therefore it demonstrates quall0 cation of the installed equipment. - Partial Tests - Generally used for large equipment. A kstl0 cation for partial :
testing is provided through analysis. Sections 3.2.13,3.2.1-6, and 3.2.18 provide additional descriptions on the functions, design features, ! and success criteria of the RH, C$, and SI systems respectively. Tables 4.1.31 and 4.1.3 2 summarize
. ECCS system data. Figures 4.126.1 through 4.126.4 show the various ECCS alignments. The support systems menti ned in this section, e.g., Component Cooling Water, Station Service Water and Electrical -l Power System, are discussed in Section 3 (Volume 1 of the IPE submittal).
4-28' , 1 my t e- ww vowwr -,--w-vrv ws ir _
l Table 4.1.31: Accumulator Tanks l Number of Opert.tional Accumulator Tanks 4 - Initial Water Mas per Tank, Ibm 50,493 initial Pressure, psia 603 Liquid Temperature. 'F 130 Table 4.1.3 2: !!CCS Pumps Refueling Water Storage Tank initial Water Mass, Ibm 3,764.568 Liquid Temperature, 'F 99 Number of Operational liCCS Trains 2 Number of Operational IICCS Pumps per Train 2 Centrifugal Charging Pump i liigh llead Safety injection Pump I Residual lleat Removal Pump 1 Centrifugal Charging Pumps i Pressure Setpoint, psia 1835 RCS Pressurg&da)- Flow Rate (gtm0 2655 0 (0) 2435 200 + (138) 2115 300 + (220) 1455 450+ (325) 15 660 +.. (480) - ,
- +, ( ) Data represents 2 pumps and I pump operating, respectively 4-29 -%i-- ,-we.- ,-ev =--~#-. ,., . ...m.- c. r rw r ,w,,m-- , ,.y5-- , . , , - - + . - ,,wn,- ~.me-.,,- eye, two---- -r , - - - ,,4 --,v~., e.=,w-9 e -w v.*
h i t Table 4.1.3 2: ECCS l' umps (continued) liigh llead Safety inje: tion Pumps Pressure Setpoint, psia 1535 RCS Pressure (pila) Flow MattigPm) 1585 0 (0) 1465 175 + (125) 1215 375 + (275) 765 575 + (425) 15 800 + (625)
+, () Data represents 2 pumps and 1 pump operating, respectively Residual lieat Rernoval Pumps Pressure Setpoint, psia 210 RCS Pressure (psia) Flow Rate fenm) 215 0 (6;
200 1250 + (750) 155 3000+ (2000) 105 4250+ (3000). 15 6000 + (4150)
+, () Data represents 2 pumps and I pump operating, respectively -
4-30 _._-m. _ . . _ _ . . _ _ _ . _ - _ _ . _ . , _ . _ _ - . - -_ . _ , . _ . _ _ . , _ , . . _ , _ , . _ _ _ . _ . , , . _ - . ~ . . , _ . .
Table 4.1.3-2: !!CCS Purnps (continued) Secondary 1:eolwater Sources Number of Operational Motor-Driven 2 Auxiliary Feedwater Pumps (MDAFWP) Number of Operational Tubine-Driven 1 Auxillary Feedwater Pumps (TDAl:WP) Condensate Storage Tank Water Mass, gals 282,540 MDAFWP (per pump) SG Presnutinda) Flow Rate (com) 1$77 0 1200 568 1100 621 800 621 15 621 TDAFWP SG_fLessure (t'31aL._ Elew Rattigiull) 1608 0 1200 874 900 1123 600 1316 15 i316 4-31
4.1.4 Anillaty llnihling Figures 4.124.1 through 4.1 25.13 show the auxiliary building in relation to the control room, containment building, emergency diesel building, and turbine building Since scrubbing in the auxillary 1 building is not credited in the CPSES IPE due to the low frequency of V sequences, a detailed description of the internals of that building is not included. { s ,
,3' For a Steam d nerator Tube Rupture scenario, the release of radionuclides and non condensible gases to the outside environment will be either via the spring loaded safety valves or the atmospheric relief valves. Steam is conveyed from the steam generators to the main turbine by four steam lines. Upstream t from the MSIV's, each line is provided w!th five spring-loaded safety valves and one atmospheric relief valve. These vdves are located in the safeguards building with relief stacks located on the building roof. [
t A a d r l l 4 32-l. l l l ( ' - - --
- . . . , . , , . . _ , , . - . . . ~ . . . . - - , , - - . , , -
Figure 41-1: General Plant Schematic e s $! u
. _. , ,l mma gj g
ggj!
'- _ ;. ji ,m- j g pI g x; ;
j L..- , sj! F y "p 1 ,;.I n i rd NO o O u l ,'fI ii,uI~ itg v~~n,i,l11y oa JL
!I~ / r,//;,f -- lL; Oail!l jl s g n, . .a y l I 'is m , l j, a fgj; = c= .cl11 Illi, [yirA W% liill l
li I
.o r- E_ _ _ _ =-d -
_a_i_i llll
# 7-se m i---* j l i. . a . . .. .
ll 4 4 n. , l 3 41 4 i i 4 m ll<- l : l-
> u\ 14 4
4 =ili . lsi M \ j- I m .'S .. [ [Ya .__.M ~~" g n.
\ '< % _ -. :ij' p9 -9 _f ll_( ;,.L @Ii a , _T-
- ~,
W pf - ill l "d:- A I-
-! n jl ' I $ ) I li' 5 7 'f yipM il u
Y ' BTJL4 l .
.l o m - !1 h - - # -g 0 '
FfLPi u -m, ;>
>. i}
y se b ng ; 9 lj %&sJ j
- y. 11 i l-
- f;=-o - n lj;,O , _lih, a 3 - 4 l ll t lEf!l li - i
_$$t nic,
$1}hhUkh S -
l h AOWp ! l wag ;
. . -lU uG.O l 4-33 l
Figure 4.1-1: General P! ant Schematic i
*l ; i .i 1
a I; n I!! ,t . _.m_ I )
,_, ! . I b ,b -_ - g i L i i t- 3; g![ l n.
d Mg j
.h ,I I 7 el m .i u N3d a , ff l g
1' {fu i f ll;! -
'=] ~'q 1_Q vl r ,,i g- 3
- I j/
'V ;ust-gwy!Ijl$;
g , pg - -, ,,-= -w n O Ujj ,- j)] [ p ei:1 15-ii q); ,d,Is i : a- =a 1
- u. i g << - . ;
<h " "
a r-eil- .!! lL},i, i i 5B f i r - I r* # 31U l,
~_
g o L. *- 1 g ,_
;- "- ,.,,d ' - ny 4
jl m ,, . . . -.. _ o - 4 l
!h g3 m s l4 l3 4 4
_li i l -
>\
M s 4
'd 4
4 g,
,~ gj p!
sai sa ar= = = = ! h k W s\ i' a a l ll,' I i 1 4 %. y % ,_ t g m 11: A' .a l--- l g, %~- 9
+r ; - l H-t U
l < m f? - Ili it j' 5 i !I mav dr tjD,' j . l g g _ 1 n l l[hll i Til u Mh'N.l.A'l] l i i
-=_ , i,, u. - GS l]; if [,i,! I . Ilf'" {\ i !!
ll1 = -i h-
,em,-
pg &g; . y, g eR m ws ' ' g. 9 - j _ wad 4-33 i
Figure 4.1-2:' Simplified Prc ess Flow Diagram of the RCS li F) i ei aca:y __. 3* C s .., d,1
/5) ,n d .
l T
, 1 I w k > l p#
{!J \ gs
/
s 7
* ,i' A i
- 4 ;
i r: ,Q m~
!h! l" !:, s.u t l :J i I
L_' 1'
;;gg- 14 ! I 7)3 ,,- ,, g ,_
a blI , 3 'K;! 1'Y QIl - E' ~k c.- f'-_s c./ , a em :l t m -
, ih __ _ ni r lc fr ! c=l j I i
(
, I I
r lJ y \ 1 \ !:: " ' 4 :
- r. ,
_ _rni - l 3 L u ni !m m i l 13 nl; j lij (tb i
,l 11{ ~
ef 4-34
I
- Figure 4,13: Main Components of the RCS l
Elevation 1000' 1
\
4 r,
?\ 3 /% 906 p,4 6-g v
v 0 885 e ,1 v g , v3 ,. m ; o a' M v
. 873 ec v
j
. v g pg v v 8 61. , , Y '-
v cv d v y
, v i
s -
} %; " P,
( - h v
- v - -- (m cp 4 832 v c ;- 4 o.
v3 R L
- 812 ~
l ..
- 805-Y 4-35 4
y
l Figure 4,1-4: Primary Side Process Flow Diagram l
#\
y as 3 y-y, v. fa. - . J -- _ ., ,, i," -s n -
> Q ~
w. 4. 1 3 ;- . y ,-- 5 Q > 4 4 } }. A u
, w.%. %f ,t 4 s-*r~I- - " , %. %M,.,, - - .
5 ,W M
;o - h,,
N Cd e[ CZ'
=$$ P Db $-
3 il y y >r II m :us a== Z 3 7 7 () 3 (p ' ri l)ii %j y f, [ s a-
,, # i A'*
- p g
mm . o 4
b8$34[ d-8 '
(4f g a3
*G U 1 *****- ,., # 2- ~
9} e
\1 y^ ~,r- F ^- , ..
Q \ _ e (..
~~ - - -
4 444 l ,, ,1 L?
]" =*
E b 4 i-
**us'am9 6 -vva A
z 1
.p M*
t ,
, h *****='=y=='= '
g ._Wh ;- +m ; ' a , +: e'
?
u
?A I , i s I 4 441 i em.i 3 ? 3 .- .'
g e s
==
u (* (* 3 c
=
p k, j 4 - -- ' (. t i e . 42 b Si rv- d . i
? g Lk [ ! ; > w . - s$ c3 i "
f o4 . :: 3;
>j 54 l N'== wi ~~9 % $y ,~
qi ' m , - i 4 4 ! !
*n(J..' -Q) : '
4h{a 4 4Y jib C(Ic f h
! ' >! b >a-i 4 , - [g '"
i" ( *, d" - - - % 4" f--- w" ?ww , ,ji M I l ,i T'" cm.1'!
%s8! ? - (;
c; 3 jL iI 3 m o e : j !
- em-3 4-i *!
e 14 1 t' I t- s
<eeyi 1
- 1,1 1
? i r: )} f , l -! j r i 0 !.I.
f-l 12 ! !! *
! I i if-l4 -t 4 -
1 0 } I 46-
! i ('"" ~~ ! j# f l-I.! l}' ! } U h 5ii :== *A i I' 4 si +u5 c,... j [3 -!
0
=n
[
- k: !!l~ k:
i- -
- l. o.. .s. ,
- o 1
l
- Y'f e.* _ l .l'
[ l -
~!6 : w _ !i ;T-) :
- , { - ui Jl t 4 . 1. ;
j N 2 '
; y -7 0g 1 h
gN '? ! gE y ?- g's Y
- n y
=
tj t **** $ =, @ , h -
'i %w ._ :$ i IV"I L
y - I h 8n ....
**h l
4-36
., _ __ _ _ . _ _ _ _ _ _ ~ , - _ .
-. . - - - - . . - - = . +
1 1 g Figure 4,1-5: Secondary Side Process Flow Diagram 4.& 5 6_. fC emi
- 4. 4 ! J, Mja; ,._, l .
II ME$.J
~
l y - 1
\
(( wr: l44i 1- --t ij i 5! ! I I we j - 2 1 i ; , 4 4 4 i I
--{.
y 4 4 ;
} -- -s % - .
gWf . . d k ! k ! g
. :=- .*v*,
O:
- .. .. r--
wn I w l
- :: 7 , q, M ! "TM ! #va. )
y'
~
I L2\.d. 1 i l g - _ ;[ _. , _. p 4%+- *4_ 1 44 === i g ~ s p --_..
, = - - - - ~~}=_
4, [',"~
,l w ~,b b i . e-- 8_4_t L= , -
u ,c r _f l w v: Aj M6" 4's -
- (',
m7.
% ,_.. C C n m'" 49 7--O.)" j<1 I i 4 , 4 j 4***~
44 43 4: 41
~ --1 ..a *i % _'- '. O &-
i -
,'m i
r"U"
. 41 = - .. TrW"_' .
i
~',.* l, I"
- .:]. , W: Sv. _.." "
fv VV " e .e~"# l - ; , _ _. 4_, 4_, l e ,
+ %mi - - n n A ,
48 41-9 -U -tJ Q"4 U 4 .: A m l,i i s. m;:- D (1 g 4 4 4 i o l" s-
-41 4
i d;4 -i *88
- 4 C y * , a n t- 43 ,. - ,
k J L* W 44 } ~^3 ww r tf I. 4.! U g p .=- g- 6: .% L, 1 P - f k %; ' t M sm-- n d,. N 9*-T:} ' L. L? * ' ' *A .' 44 tv i b' C 7 lU iU , !
.,) Qumy l >P - ,
I4 4$ r - 1 , 4 a , 4, 8 - s-w .1 , id _j 7: M i "U - i
- - 19"g --' i, 6"" !-
t g 4 e ., , j 4
- i 44 44 44 44 2
4 I i. 2 PP PP PP PP .! 'I ! j , tI tt II II * +* " 44 44 44 44 '
$3 58 88 33 T i- .
I I : a i'j
; l- *83 f +83 f i n *3 y!n1 g t, n f
i
- 'c i
-m 41 8 -5lj 2
1; L "Is-* 4 +1 4' M dla4 lh 4 2js e 44 4p 8 14 v a ;- h -
l* 1 N q 8 l
i 4-37 4
~
sts. 6 l-I 1' - _ . _ - . . .
*.9 s . MdU,3 gg <Unc
- Eeg I.
.. ' -( e asm .. 7." i(/ ]e .r . o n
r+4 j 1
%./
ae a
- o. r
-w o.
tD iE I j ;- nS E
/ , . .z 7
l+l
~ ~
l4f v
, cL $, F U
E L M m C ~ U7CTO
- d. N
{T
$_'1 NCA T '
- T: N I 4 Rd u B'S EE N
Nu it A F T rL E ! OG N _ uI U OT F
. N A ~E _
2 G R UCR D S N ON NA ITTL I 4 J ~ A l
< V aay S 25 g b , ., ->!bb .N NT .MR S ~ 2EU _
NC TT
~ U R ' E i , ~ ~
J' '
~
- b. _
, B' A B _
H C'C U E A OO MN T UU I Lt X - I L D T P L R , DL I
+ I i E EO I
NA N R G _ XL GR S ^Y A R _ O _ f S O ' _ 8 6
~ _
N OT / FE UC ~R * ( b KeE E= h ; * ~ j INON TTL N A iS f: I l iT uBS .NW iR _ NdA t F IEC WU iLE r NT DG TU N I NU R E
.G A
- R
~D S
to I gSC gTO gRN
- h. Bi. yUT
*M CA , 5 m Tt N 10 3RM ' j, EE N w t ' T ~ _
_ v m mio s m s c .o s ,o,e a t e e .- 1 x s u.e e t m v
=
I5 0 e r u e .f t r 2 l0 2 s* "'
>0 . f '
_ i 3
.2 Y
r4 a
,0 sU' , 5
_ f0 0 TT".
F _ _
] -1 Figure 4.1-7: - Containment Cut View with Main RCS Component-1 4
o Elevation c _ _ 1000 906 885
~
[
=
n" = 3% 1%_ 873-
.h. AD- -8 61 ~ = ), ,; ~ ,n n
w '3 - 832 l 1 m d _i% o _< S - 812.:: g 805' L i-4-39
Figure 4,18: General Reactor Vessel Assembly l i
\{ { "l ' -. I '
l .s,.11 @ .f, @!!
,7 I i f m a l&'(. .
l l., y [ ;*, '
' ?g ( h--
f '
*fh a ,' a a gs y g L -l 1. 1 l
l l n --
,. _ ,1 1
1y , oe :
! i 4 .
iT 4
,K - , . 9, h f, L 3
jt . Y' jl; p : .h eg h ,, l t .Miq h [ "$ j 0 .it u -k [kk-]%j N' i f s i T i M ' 4 4 . l , h I \ \ l I Dh@Id w. . . i n] :
- c. ,. h 4d- h )
,7N..s , ? 't .
h m ap *( 6 gW - Y e h c [ j }l j ; l '"* 'e j _ _ _,l %6d/ .. ' I , lIi h. , : 1 s.'
.,- l V 'D w l s ;
x__ -
. r !l '
_h ! l g f [<
.. i . % l _._ _ - .!. ;i s s 3 t. 4: j- ,
J. s 1;; %- ,, ib x -l-_ ll-
- r. -
t 9 t . .i A l [ ed$mn ,
$I , +
l rIMT ij 11(./ j i Ti
- .a r
',[f hr i
g2k i ;: ::L- ,.. l m
- l h!
g J, ( w,- a lgi . I I I t i 4-40
i w 9$*EA %' {g8'[z"Erhd .=E3E ' .> . NBEM T R O P P U S N E O I EMN R O I T A O L C MTN T A A P T Tt H )M N Y ENLR E E RY E R EOO I T M A MCF IT A L D MU EL YB B REC R P GRM EF N R OE TUN ST O TDU E WE Y E ESS FSEC NB S E AA FN OIG S E mSIR NUC O BI G( L I I K T T R R T R O P U N gg'E L *". E P O P
; P P ~ Q9 P U T U G
U ,- R S N . S QN _ 0 E I K - 2 0,Mj$ >- 3". .
%g L A
R C !
]Ml2 I
O O L ~
. N D C l A
I'h l n n p
>N,'
m
/
[%p. - Q?iki7$
~
r ' N M;- - '
~ ~
s\" ,\
.M.? '
z E T
- i- N~A VEE k E h f/ - ' YI EL L K
S rD;s N TS A L d A OB m I
-,x P T N AF ~- R N EM I M E
NL& Y V RU DT L AC 3 M ' u' O C OL CO C _p iiE i P EN R ~l S SI Y L f A ~ ET SPb-A
-ri J WR OO E
R LD E 'N I LP
~ tt P I) A EI ^ A -
B UU c
< M U S - l
[ R FG EE 3'-._n l E if WR R O C _o N 7OO O LC
' ._i .
RL TE m UN EA i
~o N?
m-
; 2 E
ff i i 11 N*,- R _ O . C . _ \ RE L ET E P A R PL R UP _
- A B
L R E 3 E E R - R RR E O f . L _ C Z . Z .
.. O .
N ._
. [ h_
q q I~
y .. _ . . ._ _ . _ _ _ _ . .. . _ _ _ _ _. . . _ _.. Figure 4.1-9.2: Lower Core Support Assembly p
a , 4 = i . . ;L [
g l ! IIl,- Ix'\ Il ; lh! i ._. y'N r s i
.x E ; . ,$i .X w- .x '
k !! l , ,
. N. ;j =. x s J N
3=blust!h;.ii!UElA
' ~
p .. D
-m nw__g w y K ~
(+ +i 7 v p , , .N ;
, , I i \ .h,x ri *' -.. .e, " ?;
l;i
~, /,
x y~ .J.q\, f,,. x 6, i j' i f l N .A. I W- , _
,[ l J (Yk) .+/. dw l /,_ g sj .s- L, 7. , ~g .m n .
3 T 7. i :.
! / : i i
k , [ k' ' ; i i 3 et : ;: arr l
~EL A, l ~ - .,M_
x\N G j u.b - se r [" l y I
' !ujp :l} \ y I , g - l hQ yp_m c
442
Figure 4.1-10 Bottom Mounted Reactor Vessel Instrumentation d I .d g
'N. p a ,, .\ '/ - ; a;"
W"
' i! \":.,
i i t i \ K' l !i i i ti !i \\ - i' 'I I i' !f \\ di i 16 l} !)
+ li {j }' !i 14 i !! I i ii 6. jj. !! ! i o is 7/
II I I It !i i I T i i il ji_ (( \
) +~
I it u u . l ,' o : ,
~
t, 1 .,s. i $. _s 5 y h .,*
\
I t I l
' ! ' : . i ' my i j %. rs T k +- J t nonn 1
1
, x- l ! ~
i
+ $bh*
f sf j
, 3 -
t J l
.e ! *[a ;
g <rE! A , r.. hb f i I e l I g
~
f,
.\ ' . . . , . .. ' ? s; a
, s:1 .' j 6 . i il J
- i jf' i . 'f
{ M l gT~ I i ] 6.
- 9> k {
j m.r- ! r i s , skEE f i l l t 4-43 l
Figure 4.1-11.1: Pressurizer (Cutaway) SPRAY N0ZZLE A S AFETY N0ZZLE RELIET N0Z2LE %
/
f NANWAY
//
UPPER HEAD INSTRUMENTATION N0ZZLE __ l ff LIFTING TRUNNION l l i
, 2 l
q
\
N SHELL n -
- LOWER HEAD l
NEATER SUPP0RT PLATE . ****** " . .
. lHSTRUNENT ATION N0ZZLE w,h.$ -
( Y ,'I ELECTRICAL HEATER A sue, m S m 1 SURGE N0ZZLE 4-44
SAFETY HLADS DISCHARGE LlhE CONNECTION SPRAY WATER IHLE) i i
,__ r _ _- - _-
W ~r 1 1 - R n 1__ 1 1 0 ,~ _________9 _ _ _ _ _ _ _ _ . g
, i , . -
4 8 I 8 5 3 : imitanAt seaAv i a
% : 1 4 4 -4
- I k -
e i 4 . I 1 F i
' I Y l SPARGER PIPE L l l 2 " 4 , i 3 '
_ _t_ - I- E5 : e
. 81 g s - -
W t
'T----- - - - - - - - - - - - - - - --- ----- m - -- 6- 4 s a x
g
%J___..<- _ _ _ _ _ _ _ _ . _ _ _ _ _ _ k_ _ ._
o_ p=' - =q __ d.. _ i % T 1/ _ -
=
7'
- VESSEL SUPPORT n S ORT
_, _. _- _ . _ . . - _. . _ _ _ . . - - - - _ - - - . _ - . . . ~ - - . _ . - _ .~.. . . , Figure 4.1 12: Steam Generator (Cutawap)
~'
446 [ ./ I 's q ,- ,-- T
,, v. q s
- 7. M *,
r" / d % - [L '
--5 9
- c. g -a
/, f .g-p.. p Steam Separotors .'
D
~@ f p d-Level of Aumalsary --
N
~
Feedwater Norale 4 4 4 Trunnions
\\ i ,
l I l c Tube Bundle n r y ;
\\ \ ' lf.c k l n
Tube Supports g ggl N r Lower Shell - g; o .- /- Preheater Outlet f l 4A# Feedwater Nostte
,/ 16" Sch. 80 -
S t r -v . l 1- 'M h, 'N_ Prohoster Section
~
\ Tube Sheet c t- s. I Preheater Outlet r Primary Coolant inlet j
-p % Coolant Channet l
4-46 L
- = .. . .- .
Figure 4.1-13: Reactor Coolant Pump (Cutaway) l 9 %
" t FLYWHEEL _ /Nw- a t UPPER RADIAL ) # ING *C I THRUSl BEARING ,,.__ [ J' . I gf - '
[#f 7kfI[): k ^ 1[ MOTOR SHAFT N THRUST BEARING
.( .
7 MOTOR STATOR f - l t MOT R
# T _[ . . .
f[' ' MAIN LEAD y q p p HOTOR UNIT CONDUIT BOX % ASSEMBLY
. , q ! j +#'
j;/g I
" V LOWER RADIAL - '
BEARING , p
%/ ,f SEAL HOUSING ~ ..ille'ph H0.1 SEAL LEAK OFF NO. 3 SEAL -. ~ ' t e ~ "
g (t.'*b
~
LEAK OFF MAIN FLANGE
/ /
NO. 2 W / ' COOLING WATER
,-) I ' OUTLET PUMP SHAFT p/ '
COOLING WATER INLET
/ . . _ ?
R D AL BEARING THERMAL BARRIER AND d ' HEAT EXCHANGER k\ k. (flh' DISCHARGE ;f N0ZZLE CAs NG l ( 1 -4 IMPELLER SUCTION N0ZZLE j 4 -17
. . - . - ~. . . _. . . . ~ . .. -. - .
Figure 4.114:- Accumulator Tank 67R
/ /l cm n -on = 138.0 -
- c. >
I I
\ /
u e
\ / \ /
N / N /
'7,-
10" Outlet e l l 4 4g . ,
, t r rI
- t ,l [ t. 4 t ,. -I f r, L . !
. . ~ ' -(Mc 5 **_. L' t
u' h h n k 2~ k=8' po<"_<" qODg'~p oE m4 <1Cy 0 2 _. 1 0 _ 0' e 1 , 3
) _
t 5 2 7 7 4 4 4 _ 0 3 t f 4 3 2 7 8 6 8 ( 1 e f 1, 9 2 3, 3 9, 9, men to v 3 1 1 2 8 2 3 4 0 5 5 9 2 0 1
/_ _
0 6 o lu V a s u n r o o ito 0" 2 0 0 8 6" 6 M 0 4 ts e h T v 8' '9 0 2' 2' 6' 7' W
/
e In l 0 0 1 1 1 1 1 E 8 8 8 8 8 8 8
- 0 2
0 0 0 0 0' C' 0' u 0 _0 C 0' 6 5 4 3 2 1 v 9 8 7 6 8 8 5 8 8 B o 7 _ 7 7 l _7 r_ r
. - .u _
T _ _ J L _ L_ q J . G
.l C . / .
S 7 i [j, i1
!1 l
l 44O
Figure 4.1 16.1: Reactor Vessel Supports .
< J N
y
~- -
s> VENT OPENING 0 . F 4 0
~
g/ " p-f s,~- -;-{,
~ .
l l l 4-50
Figure 4.1 16.2: Reactor Vessel Supports Details
, l ','; . , .. j f' .. .
i . IN S'ERvlCE _ N . .,
- [lNSPECTION ,
N '
. ACCESS WAY '
7
/ c is -
R EACTOR f re - p c , s < . g,
. . [ ,A ( Nd 2l.E x
i
/f - , .N I kVENTED REACTOR / _
rw F PORT
= I .. -' , n f
p( l. .
], -
rn1 (/ t-) E:1 P hj,E 1 i, . e i ,i i i:d F:1 f . I 11 i , a e
. N li il l / * ; D' 4 i r it i ( ; .
g1 i I [/'t 1 ll i! I' % l',, F , l l , h 3 -S
.g .
- , .. i
_r I 9 y SECTION 2-2 l SECTION 1-l i i 4-51
P 1 :- y
. Figure 4.1-17: Containment Cavity w
30.0 j p,g t o'
'n / \ % , .,J % - 17.c 4 1 l L____
i" 1
- 16'1 "
t. b 4-5'
'l -_ _ _ _ _ _ _ _ = - _ _ _ _ _ _ _ _ _ _ _ . _ _ _ _ _ _ .- - _. . _ _ - =
i l l' Figure 4.1 18.1: Containment and Safeguards Building Structure (East. West Sectional)
-MMM ?_Lfl?w n h \ &
l xy
! 'N l - . rs. 54 i \
l 4' s
' ,* N ._ W y -, , --i d\ I., _rc w w * . Q-T6 -
M t Jo
,,,,,,,4_ \ -,M, i
essewomast4en.*T t
*asen s - -,. - --- ; : x b '.: 3 ..- _ a m 1 mn p, y - -A '-7*
A .o eo s ~ j ..
.'1 \ , ,
_ e7 e-une a ==s e ~ - - j i .e.- -2 m. _ ass umum esee *8.
= ,d 2p = .TF a n 4"s -vL .mcMtaw4 ua Tw cas S/,r$Iiun - , --
1 r., (~ r""*~* V . -2:.51 a-y.r..r-t-.9 -p< f 1 g
~
ir- - h' ~3, 1 a
.a - , , . ,L, _:-:. * .~m ll -'f , .-a l i., ,g F;p i ) Msi I ; ,
i i '
. _8#
m., ; g } m a _+g f . f,, U m,. - c ,- y ',,m ,, qg , .._ t
--;p e ' ' "
i p >, ; mmn i i r.r Q',1. y m .t.h. I _-E g .w M i
^
u%%,J x, - g
. ,, q1 J._ ' ' _ 4l ewy R',, I-- , w -- . . I1 .
a -. i g, , , p, w g 7 4 _._ _ n. _y. i % . J q L"'*%
.. A k y _I i g "5 : _
i F #-" u . y ; J Mr,7q 10 ,, _e-- I #~' ' 1r'ET7C ) el r: c, 2
~ = w I - ~
l l -*b. i . - 9, . y
;.w WA \\ 0 rwl l ' ~~3 j~;oo;; ~__. ,, " ' a $ $e=.4*
h b '
- k. b! g _ . <M rT, M p)
? ... 'y]sy;u.,) L Q,f .
f PWf L'*Alf?'
) .
4._. , f .,5 - .O 9: i iO O. = Ils 'l ym
,a-i '. - . s n
x-- M.l... l m . .. a IE g h ==' b fg_ i 9 .g < { l?.t. [.s ; - i a a r a e e a e e a w a .g ( ,,,L, ,,, ,,, , ,, 4-53 l
Figure 4.1 18.2: Containment Structure and Internal Arrangernent (North-South Sectional) rr%_ m\
/ N ,. %\\ ~ / / Nx / \\\ \
i \\
.c' 13 . ,1\ -
b.s nu h , t i ;
~ .n h.UL L i
e a n, n. t ',..; _ a 7= t l -a.m (..m y y_.., ' j
-- U----
l:' i - l a +4 r l h.t EE ' _C 1 hw:r I_ m .. _z_ Arj ' I - eQ. l. ' o ,I
,_ n
- ! lo.l i I i
;_r. cat u -
g , ; .,,
+
E
<3 f , ."
E' ff . - . m -A ' \ {, f I@>- i
- e. # ea. 6%' a 1 ;_u_i, ,, ,,
] [ _ u,, ,
JtA Alt t L .. y/ _7 . .
.M - * *'t u r,N- us.o ,,,,._
f.6l l I H2
~
rF
! =.5 c . - -/"~"
j j i (. '$ 'r a n, l , ( wo w --- 7 7 ; um.. , . . _ . . Yr rw my 3 .} Q .n m e 7t+ a
.m- , La.1:J:.
I
''~
b 'm -m m 4-54
Figure 4.1-18.3: Containment Internal Structure
.\= /*. , 's' f / \. 'e ' / \;\ .< - / 5, / e ?, A g \
5 M-4"/"N \ \ \
/ /. :s /\ '
l, l
- 8. .v - - - -y - ._ - .- _
'k, ,s .--._4.-. ,'* - ~ ~*
f,h\%^' 4-.a.
? .; . . ,r. , ., . m. ~? i- ; r \
I 'A:x ~ :.~;p' i
- ' ?! f ,~<gir/<;,./j /
g,/J,
.L/7 \- ' .s ss s; f ,
i
\ = / .) y? ,/" .,,% , ' / yy \, v7 8 ., ,1 5 i tr i ;L,f 3.,. # / 1:: 3 / . 'x \p \r j-I T f i i .m I
m :;
,- ' ,f $ $ / y x q . ., . /
l
, = :- y y ,f~
j,a y j
,,- : \~'O / i / /x,\ '/ egcev eK l "% L_ ,, j.d:M t'/ 7 .*' / j' f ti: i/ ii xx , '$ i .
h.- i l l.
'[ , //\\,(. / . %;f , /?/ \ , , . ---~~7' f 1 _/
t
,y , >-- ,-., -; ,i y ,
_w== ,s._3
, ,s , n ,
4 um# .
*** , '/ - /,
!- \ ( ]/.,,psx[x/?='s,nwi -{ h, g\ft . x,- ,
\, 5,f ,/ f \ /' /~ ~. J , w :q (' '%5 -
! 4-55
Figure 4;l-19: Cor;:ainment Spray System Header Arrangement 1.
- =. .
r e
>
- g l, r'.
^
- p. 'm j .!! 9 !I l!! I! . -
!!Li.' ;:
ia!L- 'L! .! L i!. s 1 lj ? I I , .tU l l. I bib n1 g
.TJ
- .ti l 14 a tt i it ii [i!t itti- u.i.!! L..
; n. s ..
1 ri 6 .. J !! n.. . f !i 9 3; nW-.! 'it-l
!; e:
ti
?ti .Il(\ ? } ,b U t2
- l O L3 .> .r: }!y;;'
l[3 c = d .(s!I t : si : : M't
=wr a
c e et 23[ " . " *I "h 0{ e 4 ed I ) YIk tjE I $'6E
- 8 f6
. 4 P.
Y 5*d
'it Do Of .I h' g- Hg y j 3 \ fid \);x9 }d \ :9;d ! j (Y p ., i a\h s.g.h) tg 3_ Gu "7r,_ N:l) \m, 0
O- Y - y
\
I l3 gd;, e Jl,l,z j
- +3 -!= = - -
j7 - ( { 3 I': , q ._ ._ _ ..- _. cah:- <
) , < ,o ' j, j, ,.p h_ $' '. _./ 'I L- e -p g
j N,fv) ni i nj 4 ;; a d M Ed .A * - m i q, 4-56 i i 1
. _, -+. > >...s,.-- n->- a --..+a. n a .~a +,nu m- u .~ .-.s .>.n -.
IFigure 4.120.1:' Simplified Flow Diagram' Containment Air Cooling and Recirculatlan System (CACKS)
'~ # %g / / / centaunent / ,^y%%%% %%%M% , .'WVA% -WM l .AvW .e WNWsI , .,
_'MJ- _ . . . _ _ _ _ . ~ . - - _ *.- _ El 90'a' z= , I T~ rm . i Im 1 1 I P6enum I l l ! Darrver @< i (m 1
, _b , , . --
Voneoxioi Fm .,
." %%A/v%,^, I '4NN Cooling Coil t - J. ~ws~s r Et S60' T N . gt 332.c" 4>
4-57
;l Figure 4.120.2: Plc.. m and Duct Arrangement CACRS Elevation g ~
1000
.M t
f--s cg -:
'~ .
02/ s=- f 75, p Qn= 3: A wJ=X vf,-g 9 , c
. . hbl , -d d
keg M i[ / lim
- 6, 9 ~Lg..
h_ - Je b ) i%- rg < 906 f3
' W f f: aus y !,dh % x a e e e, -
3 s\ F 832 [
. .s 812 805 4-58
Figure 4.121,1: Containment Layout . RCS Components y,Y y' ,, s
/ / s Elevation ,7~
N \ - 1000
~
~ [s-~[ w o ws c . b> s_ 885
, M M 7 y-. n ----
873 ~'[ [O _ 8 61 _.< L. . i 4M' fw _ v
~ 'a p-.,m (_ ) ' , -((
~ - 832 E Lqsg r
,7
" s N_ 812
~
L 805 4--59
Figure 4.121.2: Containment Layout I'loors and Cranes
- 3 ~- m -j-x g :.._-p ~~cx'Qs x w
x- . .\ s
%9 /'(/ '- e \\. ,/ - - - ", .7, 's s Elovation J ._J 1000 k . *^. g' -
W' , t /b
,/'- ' _
y '
"A p '{!{. .., , - / /- . M 9/.= , q, == ~3 /$4 , ;y, fj .f..,f'g.,. ==-; i T 37 -gM. -y.\
I
- cL f / j %,~. ,.i. 4]=_=== # i h+d"'" . j f . ; ~ '
f
!QQ /
n .< 3 y':.*% A;'$'
~ r: )& '
- 5. 4 5'
'#j,.
5[^~
? k_
i 00 L
}=r: ==4- y - - ' ~,l
_ z== ~ _', 873
%{-L v -~- . 3 ' EQ "!!.
- N .
- rr ' 71.
} .r ./I --
d f N 3C- i.&
/q 'f - . , . . . - ~
3,yg;1de * ,T'4 l
<-p-1 1 I J ,,I ==m-1 , '~ " ~~~~ N m' A,n[_'_'f(('~ -~~
S, - -ID$'
- 832 c J' I gpollb (Q Q-@ Q C 3-1 ( [g 7v3 A ' U- //
(' sx
?
Le- / 80s
- - / %.- ._..J' 44D
I Figure 4.121.3: Containment Layout Floors and Containment Spray Piping i
/ , -r - ,y - - u tw -
k._ _ - - -
- u-- - .h - . 4. 1 4 g = ' %- NI Elevation , / s- N, % n * % ,,,,,,,, ._ . - - - - - 1000 f
w, man L. y ,, _.
/ ' '
C W
) -Q.a ,g l ,$h M 'N L -
N;ENd Wg
~
M [4ym
~
soe M -
-y } '=" l g* K-gr ^
888 b1={- 873
,p 2%2s a . 1" ~J: -
gU 4
% 8 61 -4((g. s. r:- > -=: m_J
- p cW L= m.J,
['(I_r @ G ;;b h . 832
~
(g- . Q N s
~ I m
g l
);f ,
8,2
~ .., e- - -- 805 t
4-61 M* .-7 t -or-hw--mw sw-s-wwe7-. -r- ve,-.+c -w s ---3.7.mw. ems.upmm.w-++g Yr i-,y'n yT-> gy.-yy y, gg..-- y i wyg,- w gy.g'WYMr v y- g. .a,- .,--y+-.) g9.- 99.--9 g q.e g *g.,i-
Figure 4.122.1: Containtnent Surnp Piping and Arrangernent (StructurW View) g N e(\x\
'N N
j / -- !
~
i
\\ N';, , j L .L 46 M g, i (;., w.,s \ \\
A s sx s E ( 's<l. N x\ :, A3j4 *; !! ; I ,ip >h, N.! 'N \\ l2 s ${3., ss ;
.g'&p q*L.s\ x\q [hd ,q..gg l-g- .. i ss - ,y ' N .c \s &, \ \y '% A s .xl ,
j'> f 2'> k1}'{;f]> hh
~ .y'mft}wl ., \
k Wa\yAv.1 'i , 49 [{]T he
~
i 9.sBk ,f i ! 11
-dJ -
- E31 - -
-r,4 .,
lr i l [l* l l,'JL /;r i [!.i l 4 $3
~
s. e,, m ..g
;7 i 9.,
[ _ ;,, 9 *I 3,g~ ***
.' /./ 4p, j s , ,,
i 3,-
)y / '*yEiQ l. ?t'*, ,% ,l y' @
l$,,lh .
/ ~j f) ; lI , m p,y v ,y%,ge/7. ,n mam .. _m%Isl u _;
n
/ , .
1 9
~
4 . e
~ / / . b m a Im l //1 L"A;i{. c 4'
- !. L . :' :' , L (
l // / . :d . -~; .. ,e
,/ ~/ ), b an. sq ,/
Q> 14
\c 4-62
1 igure 4.122.2: Containment Sump Piping Schematic , le ! -- F1 q'- y by -G- ,
+qi +q; I; ' . I c,%r' +'Gqi a +-
f ,- , l p, , yo !u.
*Lj[N.z n h , ~~ /.T] 3) *i '{, g.14 lliu O " ,j -t-- -
c n,- 7 7.#@ Il.,A@ i,,
;~ y ,L p3 ,v t, ggg, 1r r,@
3
..e.
Il* t hr@ f It Ihr p f I I. irc ;:p # g
,} , 7.,. ll* Ac- L@ , j .,
_ h- fFi j ! , ki l l +4! "g O !
,[. .,! l +j! ] " ,p$i t
- i. i i, I
y 1! }i'- il a !il
'ii 7#ji 8
ir p., 7 l c y. f~ T (-?
+ e i
II l ' ii -
' [ ll l; . n a m.i ,A,L- i ]l ,
( lll l4I~3* a a a rs m i
" ~~ ~~
i ! !l ! .h @ 4-63
l'igure 4,1-23.1: Containment Shell Reinforcement Detail y w uf y a in t;a , vg 1 bw Mi V ' o .2 , W, :< w5 ? a!R P x-oe r-
.,. ss - -t
- s. - w:
e-r t ,a e r" x s ._ w 34 4
.f, F' n ' t [~_': %s , I y , ~
e J i y&a o 4
< ) .x 1 ., a_ _
H -
*; ! 1 ?u O
a , . .
.co r- T y w \W ff M i&
U f,
'.g , > y'{aZ :1W N O 47 U WV U J .t 1_ _ _ . - .
I *' I W r o yo I 4 y 4 4 i
.a -e M
um i t- -4
- o r e W
yss, =. l
, t m , - . m
, I J >= 0 % l 4 w y o ~h- "o 4 u 2-
> \ 8 ~
i w e
' 4 O $ h@r 4 f
a F' s o F r e e b d i +do u ! ygs u ! E w - . ,. ~ p, . . .
---4 < ,I 4 2 . - ~
W I ge ' * }, -4, 5 I gf rL.7 N n. y 3 tr z . N,
/ = Q!!!) jg! 6 (D?s9 /' f 4 i % g:;
l l
-e / 'g( -tf ~ q; s
- P : s
; e /r(' ' Ry >
u, n n\ ;; s y i n
.c i;
a n@ s d p' '
. g a
N, s ,4 t f'
,', h \ h i ! // ,q% lN% ,n_ $
a i
.)' . ye ,, ._- p l _
N E W.
**86' . *l( N at ej .J t -Al a "t $a n:7d 'T < tw 6a t y- w .u o -
6 w,g-
* *N e
( l 4 64 l
Hgure 4.123.2: Foundation Mat Reinforcernent I y___. _ . .
- ( " H O O P RC1N r \
x g / l/
>lb l'l y ?.mQ M 2 o s m
T @4A/f / j((
/
j/ . ,_ _ a + m '\ A /( m '
,l < f j} 'f , _ ~.
l . ... a
-4 -_ f ;\ .. t /(
t b g/ g j g /
's T .. , .. y g
(3~, Qy f:
't ?
ij i 2
\ L REINF EACH WAy(Typ) 's \f %:w ' 'la HOOP REjNr. = . .v, -
j
/
E L 80$'. 6" is . j
!L n, ~ ll /. . **\ ~
j*I s l I T {* $HE Ah Rc;ny' p,bE RE p AS RE0'D __ _l s
'T' ..I __
- k. .
A
'REINF. CACH WAY(TYP)
SECTION A- A 4-65 l
)
Figure 4.1-23.3: Containment l'enetration locations l w tn m - N m, tn m e C o to N O O c3 c0 !
,9 0) CO CO CO CO CD is N 3
b ru i
\ - i 4
l h (v)' l 6 ,
- I as.. , l 9 9000 l ... .... o @ ... .... I l
g ' e. o..
... l , , l @ ... .o.
I e l e l
.. .. l 0 l
1 1 1 I 6 e i 4-66
I?igure 4.123.4: Containment llatch Details
+u *i i ,-~.
i g}1
, a I xD I -% 3 11 y
l1 b;\_E k I 1 [ O_/
. , 4 t
Q ./s; jk ' 0, r 5 *
. -y ny ;x s N s '
sL is 5 l
' x- .
g g if $.s.,p t ;
, p l b k f khk e r d _.- m-i n ).
W0M _! % g
,,7% x ll l h m
_aoa @ s ,
- e ( ,, ,e -
2 a e
%, O n l ~
Q f Y
/
I h w g i.- ( \, ( + s '/ d s 9 i r 5 l
~
J' ?
~% ,, s N , .o ; ., c ,
y . - - - . t ; H -j
%- w ; + O w-4 w 2 ./e ' .
x p u z
-/,- %. v % i N'x o 3-5 4
gh a i 4 u, g eg
- l - s _
i- > r wm o i.
. g m al l .
S.N r
-- c - -c \ ~ / /
u >g x l '~
.< l!:n 4-67 i
( = E__,.._.._._.,--_ .._..._-.___,.._..._...--,_._..-._..-...._...-_.._.._._._,,.,...u,,...._,_,.....-.- -
l 4 Figure 4.FM.5: l'ersonnel Airlock General Arrangement if li
; ;; i i ! . I i i ; ~ + ~11 g f; 3 ._..~. ,w p c.
at . 'l l
'l >
ld,;i "NI$h/skk! !li
- .. . '\Q;/$ aj?
L ,, 'N *" YN. , ' h s\
.a fw <
x
\,
s t ',i l ! \r\,\ .I sy i
- -. a ?)
(O \
!i!lI .~.
ll
- C
%\ & '
D2 i
</,/ ' ^ */
I 4 ,-, . . w % - -
,--&hg[*,ihff. +." - ;'>
_- g- .. , x
,gg l '
i i mv% 1g ld llr ;! I 1! v Illi f,j , l.h . il
* -, 3- - L, E.i i w .c' ,
g.
~w- .yrst y . -m... & w y
t i b
~ % Mr W+ p?2 Mr r 8
m=1._ i
}
J 7
~
f j i i) # T
>: r, t ,oa, t / > . c, t
s
' 'y i ' l 3 r
I l o
' s*
s, g -
""*g g g. g * 'p i >= Edd }}.
\ 5 rs
% i J(g:
6!x s 3 i ., - { y 448
, ,.--a +
Figure 4.123.6: Fuel Transfer Tube Details B t a 7 =
,r--------- ,. :; o_ . - 3 .- r/c y; s-v m E> y c l % .y < a e
- ,i r, , e- i e
%v,$,-.__.y, xI . =
m, v~ ,eww=
, <e e, __Jn -
2tx - c m ai e-
~ ,_ vI n p ___ %,:; .g n\/7 NN\N -~ -
Wi C-
.a' u_ " \- u -NT , . . -
e, * ; 2 r y aj : l ',. e,a;
.e r
u
, u w a, a i ! I I, ,
i w I. i N. U t. l
. _ w' _
su .- - E e Y
,e - ~ - .Z h-W a a, , e, =
s e. s en
- o. .r Q
m
- , . : 3 ra 3 5 9 ?; m 0 E $. ,', :c.a ,_
a 9 E E y .,
- c 6 t:
.s . i _', : r-> e. , - \s .t - - - .,. 0 t .. - -
I ;' 4 = - D *I
, 5- e Nlx x.'E.-.-,: --E.: .- . x ,s w i:;.... . . . .
i 1
.. 5,0 -
w
+- yll ~/ n > :;:;. +
l .w' ...
- -y .a -
J t_
<w ., :a .:,/. =2 e :i.. :,.op. ,e m... =-
y f: * *;q ll V, :3-
..: O I, r.
1 1 .
;; ..: . * =
- n. .. , r. e: e...
... z k,. ...;. .. . .,. .g x ,,,j ^ . . . w w
{ w .W@ ! D 4 '""": , 9 U
!- - l . ' f-w ,,, / I . A >r < O w
I
*H 5g g/ ,y .~ y _..__.- .._..i- ..
g
,, ,~
5 m 6 2" e % w> ,i
? f. J % e E 4 .i > y I -v , , , . w h. = . ,.t,; .
E
- g ,e
.- _ .s .
4-69
t i ? i ! I i 1. _a*._ __a_.
,,ygg en,7, og[N - ANCHontuGs 'Nsr cesT wtNT y i [ [*'EE* E j- g N
I II (~ 7 - --
~r G ( I M ) 4_ , % ,L._ Q --.-,K
_ _ . ..= T " _ g [}' f -
\)
f--
\\ ) -p = .,wp r- c a ,7 ,y tttyg E .
7 _a II- - - i ___ , g --- u , j t'_ Y -- '
~* ' ' - -
f.iMsut A rrow l
\
I , I I. v -aNo m wss
---J ,
- n. v. ,,__.
O c 1 COLD PIPE PENETRATION MULTIPLE PIPE PENE TR ATION E o
, (CARBON STEEL) h g 1
' n
* , _u t__d_ E ttNER (_%%7 A L8NER d. x ,
ikstDE CONTAINVENT ins!DE CthTA:M&ENT [ t
,7 . steeve E i I II !sttEvEstop b i 1[ f 5 [
m j(. ; myn l l" t' i i q STC* E' i
; l -T ,t)
- r. _2 t ..
7", {P i
, \ -J f - ~ ~/ ' ; s - t,1 W -. r _ m ___ _ }i l.- } O--) . ' !_ . Sh kg M m_ __ _ _.) jl !
9:.ed-}{~ Epva ll ~
,, -- 1 y y ~
[
*?NStM. AT 804-( ,1 ,1 lj ' ANCHOR LtfGS l' INSUL ATION"l@'
I IL
- AMCHOW tt/GS -cuAe stct<t j -. s 4 ___k- m_-_
COLD . PIPE PENETRATION HOT PIPE PENETRATION (STA. H LESS STEEL) [ J i [ i i
Figure 4,124.1: Structural Plant Arrangement
'*~ $ e g..
i Iy)_3li (w Il " It 55 c: E
![ !! E . - ~
- n. < l
*1 V- a .s -'
w
$3 )
Jt \ l F 6 S
., \._._y p m i. ,
y g
- . w , y ,
-n 5t;,5 ....mne + __
r-g ir W .- F S El wy t .,.
) h -a ju uwn-womrmort-- -T 7 gge t v pug
- "t /
)t- 4 l
F l r i li .
!? I t5 -
5? g -- - ..~-- ag C g s - by b y - L
/YN O / s :
[ w- t 7
*L ,_ r ' - - - - - - sincer:wiaurmart-- aY V E- ~
7$[ ! W h 3 8: *5 r yw % r et mi!5 w$" ,
$.g Gir g 3 ; namd%ldi~ ,
K 8E 9 , I{ __,i o [ 1 , w- s _. : I N-i {. !% os k og .
,1 ; L__1_
4 k , j, .. II:$ rai st 1 __ i 4 71 1
Figure 4.124.2: Structural Plant Arrangement (Section A A)
.g ., ,, )
7 [?*d '=
.; j^ f:; c_,; w~g 1; f H= d~+
o , ,: w
- v. -
p
~w 1 ia .w bp <T --..w .-w s
k-i t, G y
*O,- gx ,__4 i P"", h 'M d *rj . E U g,
N ,I b es e
- _ w[_-- h.~} _, .+
s , ~ 49 m9_ . -6 _ . Y ~r y-gs k.m
'Y wa y ; $i,t- m" $ &ht"=*t, .
n N a ig [, ;.p h +,. * - a. l ' \Q>!J.w> ; H --
~ -b h xJ .
pa s .,
~~_ mi e . " . - L- , ~ Mi ,
g
,- - ~L *: Lijb I IL,d a' _., ?
u QOPJD_ f p!
.d dyad L t -4 ~ "'. g - ,E q=gg , 'w ; y! 'o 'o 4: } 'wf Y.! .'N Y.
d d d Y
.nu']Q ds ll :
i lw B-t-
<a p < =@ g...
l' h. u - n traD y._ n i t . < s.. 1 J-e { E sp g R~ 6 ji , {i g g E- r- , fq%iMH' p 8-
} ,, ;
- e u 3 -
*o W
E!
' E ts E g.
p=aa= d M., g_ 7 ! i t<- w 9< D l
$; _.p @$ r i !
7]
$ 'y J i Q *N,3 !
I 3 1 A n
', 71 I e 1
l i 0 l i d i
. li. t h :H t ,b N ~ ~ - - -
m s 4-72
Figure 4.174.3: Structural Plant Arrangement (Section 1111)
,. 'e 9 E1 y Ie kr r d ,
f n i
>a -_.
L;i r
- , ;; i !
_ . t6-4 . o. F
- r 4 h..~~ 3
" e J
k O bJ f O 6
! : . 'g.
E i d2 - e
.', L. ['- -- I e l u e t . R g u. t ._ 1 e 1 a 4 l "
m! 90 N . l
% % k @.
r;
-- '~s ' ~, " ,J .in a t r^'""^~~ Q = .
c. I
.s h. h + e..- \- h- w $~~
s.- d 'o Wl g,,
, - m .s<- ; j .... , .{
i
-y --w ,f , s--
w c, P~ r d~ Ya ta "r , s- . _ -._m . _m _._ _ g g,,
; w g. e ,? to i = = h, I
- 5 g o! ..-
d- L. '
- - Wi Ej ]l E -i: ~=~ ;.g_g_n; d e "6'{ r" i
e, g s
-E "j *'f x 1 .
a a. , < + a
- a .
g n lol e m g_ 5%. g
~
4
-a N - +. . ,~ = g. EE, 4
u _
..g7 y ;
c
'o, se< yt ry$
y ;.2 li d a 4 73
I M
/ \ \y CONTA!NMENT STRUCTURE - /,
I
! e i E t t h tTrndrrr nr n :rrr rr n 7 i
g
. J 4 E .=
2 l 7
, ! 5 i , _
FUEL BUILDING - ' INTERN AL S"RUCTURE ~ 6 it "?? -
, EL966 OF CONTAIBiMENT - t I I ' I M I , root's" EL 90F-f E j 'ne(5 F.
7 g gt g k-i
.[4 2 'l .. .a EL 6S? -G s
u l y h EL Sc5-I91- EL873 6* r __ 1 Tl1 i f u atz'.i u 1
- .c 1; g t,y jj tteso o- 3 .ladi l
__EtJ157 l f s [ l {
=
x m! r' 6 I
\] ,
d ,,.J_, E L.8 32'_- 6"
'"'"lI" EL 8'3 d _1.l_
b [0 (_,_h,[ llEt ( :
! i r T- n BEL s24 ci L J-[ GR ADE 5 - g Et %s- 0 ;I' Ju9 l
ps-v smogo-a 7t et:f-6 __. M i I Q ! 1 d m
~ ,typ. c; //j V Ig /// lif ' 6 o
E L 7'O*- 6*1 _ _ j ELJ71'-O" _I IIhfl _j ,, [ Mil ~e::til m ayagy o fiil=3( O BUILDING ~ _ SAFEGUARD _S _' BUILDING SE CT ION ' C- C . ne 7 rey _p po_ppgg2pp.qup:p TCatt **e ftET
Figure 4.1-25.1. Containment and Safeguards lluilding I'lans (Els. 773,785,790)
, t ;; ;ti+'t ! 4 'L..r '! li .
L *. E s u r df u( t e c c' tt l' !! tr t{r a i.f t t r 't ' 1"! .. :
,,f:i.!,,c i
i d 4 S. i s 1, ; 1. . . ! r {. J j - ida' If i .i
,6 i g iftt.i. +ei .
t#. , . . . .' a i L t. i , 'e , .'s?.'i ' ' , . >. .;;. .' r g. ,. !. f .?>' . ' " . _ - - . . >
. e . . # . a 4 .; .i *. 4 i v.3 l'i .s_
s t * ,. _.j 's .I L 1 .j?
't e ~2* < __J' ' . .? .. ?J$ ; .[ .$...E. . i . .
l}-1:,:; -
- p.: ; ? '. . 2 + u y ; 0 0 '; d 's ii . 2 T [D QC_53
~ '7 ^]'Gi}
nn Tn: !!l, * " 7 ijuUTG. -
*A- , gr, b -i-+'gJ j ' - .
i.- Q s. ) F. f,f^ r C Y? (a+I
- LJ
~
T
, .? ,.,?
a' s e . , , 9,g d'N- frm,*f WS j a,"ts % i r '1
--* -]}' M[~~ ;; ~ -
g- >a G
'1 ~,, v> . ., ,% 7 arn ' }' 4 u..
3 i= i. . s
^f~ri_ 2.lm.G- ~
i G '
+
D p
.h * ****"%f* W *"'"k r***t % [s t G' t y /y f
N( r82
- W* 'A D-a .,
e [ [ __. i e, 9or s,)( I.. . , f -d 'tf;
't +
W. , ( N.Uj O $
*sJ c ___ , , ,,03a ~~
S , s
- 9 -,* d " 2 3 - .u,. m d_ -
4 $ h = C D D 4-75
1:igure 4.125.2: Sairguards lluilding I'lan (l'l. 810) i
.6$D ,'
j .me _~ s
.-~ .i s ! - - -L ; ,
p* gg-o , , ;. ;q 3 .
* - i !;* -,~. .f , . ;
Ii . i 1l - 1 I > c - 1 I
.l . j;t; - ll e- ; _ g t ._ _ _ __ _ _ _ _ _c _ _ _. 3 ;,i *~.,~..m.
1
, i.,
q f4
*~ -L. u J. . , .- - / / ' I ' ' / ! '
Il i i. 11
',l -i;1i ~ / 1 i
- _.4 ,i
; 11 n -
t l , "t i . .
, l . I r 11 e n II ll ' as E'j J/['
l *
'l,1 -" I g l ' b{ m I; 11 ll 0 .. d .[
i ..> 3 1 il 1i - t il 11 . ll i 11 Q Il 11 .
-a
_. l l -; j - .
. - - -- Q L.
G., l' 4-76 I .
f i Figure 4,1-25.3: Containment and Safeguards lluilding Plans (El. 810) ; P i O W
. =-. . H s t O, _ 43, '- .
4 s* e . U rJ e 4 L , I ! O CJ [
- 8 4
a i .
. /, > p g m. .. w Il:co e[ pT q ,fy[> f Jn..kI; .
il'a
. .; i + Ca pF W>Q A .V ' ; .J q 'y Lj J
L ", .; g g _
~._r A}' ), 7 " = ; p' /
{h,j;:, W ((i
- 6 -'* i3 y
- p] -
m i - w iprt s -i _ e --
- 47. -=;
r 4 a 4
-? ,, y y{m _M4. ,] + -. ga.
gl.L L-@,,/j ' -~ a.
' "9 c.
i
*"' :i 2 ,o " , :) s :3 a , d _a is .-
_%Qj *gjwu. *J w gl < o h *] , y u. n .. . .; . .t J -O y, ,j3 L*l c U t a R p. T
,T..'
d 'M f *)
., j (
r' ,j-d -
~' - - ~' - ~41 4, 4r~-(( N2 - s _
w '3 g r4:# 9-.I. .:M.aA. -C ..p m Al m M C.O*W
;7 - -s
- LCre @->
ll .
= +
g 4- .,f b 'I b Nf. ,1 $ [ iW M ra o lit a w gg j-s"5 r m g 5 L._.i CJ y b- aumm t-a o a:o a g U
/ Ws +3 _ lllll H i lilll!Il , n, . . , . a\
j/i a x=- ,. v, '
,m' .
q
, ,S r-(.J.. L'g. ,
o c
./p/e,S \ ^'c ,, 9%s[-.x .83 ~ , I,\ /
w u p$ ?? S 0 - 0-yff C]c '
. g0-f g %I w ar h
to Qof{/ Q_ 6 - _g
\ - q~;Qs ,w /i 5,j a e a <t - l) w ~
O' ', D D
\ - /,f [# !\ $
NNo L bi <?/4 ' t
.m l
v.b w;/,
-) e -N' %w x C '_2M r/j l -4 77 . ? -L Y -.,.w.. -r. ..-.... , , . . . , , e- ,. . .. . , . . .. ,_,..,w__..,.....--.,.,_,_,.,.,,re,. . _ ,,..,_,..,_.,w.r.,, m,v,.., , . . , . . . , . , . - , - - . , - _ .,-,_,y.,e..,.v... .
liigure 4.125.4: Contaimnent and Safeguards 13uilding I'lans (El. 832)
# !f .r;: .. It ta a i 1{*t , ;- t$ ii n igt e ,I))!1?f'III u v;;$ At;1;l1 +
- i" p 9 it 1 a *t t :i . 3 a 4 b ) H 4 321l pi t fI"d gg7.,g! ppu iE aj;e}urf.Just j
;e, n,ith u!!;
gg !!gi / (i 2.75 ydgar
, a n , !injiu r o. .rqh ,ulm D'i b 4 h 1
t : ~. t t it u3d i H!:.t.m.q 6dn. in!!bb
!i....s??uenr:.:ro n s uiin't!? h e s ,N: } , ,
rausas7ucapo tgg
< cp Li J J _L__l l L. _J_ , "MM t---) l L~J EE' [ , f f ,,
t El D m,,$E h e. . D D o 0 O D [ ' fi , 4 [f#cfOMgs 7p I aw 3 ,~, c . . ~ 1 3, _ \ n
.s uw mm i
l L;.k = F (,-) <
.. <A : I3 o o 8 0 , s.
a l'
-.< . / (4 # ,D Q q q t- j ' .le t
o .% e
-J) ,
- s~ .
h
, e 9 _f4
- 4 Ed2-e ms*"'
b b 4 fl . 6 .mi :
}L0584 l
4., 'N [ { y i f 5
} .
s '; I ! 9 / p i i h I l' 1
^
2 i g rt. t [lgl '
.O ' , in-
- 4.*18
1 I I I I Figure 4.125.5: Containtnent and Safeguards !!uilding Plans (El. 852) j l il ; ... . a :
'*# *';**'via ,. gt; a 4"
ti..jiiHHlib.l e tia j
; j)J' }: { {nf~ih }i < !;
d;j!}!!!!tt!! t HQt d!JhNR
! ^aluu ullh[
t Nu f t L u!i.:!!!!:!n!i
.::: n !i'unv.nu:@!ami0uIe NI~[ [. _fI zMr l L _e. A " L . WoW J ~{[Fi ,
o DEEDHI!11H] !.a
-- e O O I O O D f) 8 &v :.f. l3:4 ?4FkQ,t 2u1 ,\2
- 1, :@u , l:
\/
, .y J
,,C. J !,m i ? l ,s ol[ r j /.; e >
mxm., ; fer] r l g .
/
! e y t / .c - les i
,/V*> ,/
1 6s
,, \.- -. ,~/* NJ j e / i
- t-C r
\\ es . j ! j wq t i e <
o e / .,.
$q / '
4-79 -
l l l l Figure 4.125.6: Containment and safeguards Building Plans (Els. 873,880) j [I . '1, I . I gg i
. ,y{:-ll}lw- + y i e .1s-1 v frtru mtg
- n i
n,.t-
% f k'>e %f "; li ?' !>f;k 'ti ( ** fi 7' ~ ,} l W !.< ! ?'Ifi>
Luu O 4): <5 _} -} ,f.I.f !, Y u ? I; '. f, *,1. 7 k j *-g *E E g , ,
~
J K i
, kep\st an i 4,j,,. .a c., .
A I
>J s' E l o
r , )l ijt , )
' ' t J $
I
'g f i ,j ~
QM] - 4lIp)/4., s Q.= , u
%s &J #8 .% i 1
f' - #. //)%s . c m a_
/ $ 2N \ 't ~
fib (14 ; (1) ,fi p , r_g ; L 44 ti,j i I o k(',-M) v jg/ 7/
- ..a-i Li 4 ...],_
1 h* s
, y .-
i l 4-80 l
d l Figure 4.1-25.7: Containment and Safeguards lluilding Plans (El. 896,905) 1 i i ), 1
%N /e' r -J N / / ef/ j ' j) :<';' ,, , \
- 5 ,* _ ;) 'S Q'N X'N^s:;.
L e F * .' ' i
,/-,/ \ 1, ,,4 x s 9-f' ,/h .yg w-. ,.f '\ , 9 u _ u____, - s,
, 'N ' , r" \h .. g
/ ,/ \ / ' 's P\ ~
( ._
\ % /
l ,~. / .
\
t / ; i A, s ,/ * [ ,, 1' .7 ?,] ~ m h ( , y (. I < -
; j p f \ >,/ = :
1 e r .. .1* d c", ,i 0 e
\../ F b'
f +, ' [A J i L L.
,/\
t d
)) ,
e / m s gjf].~y, p
- . 4 Ny, ?./.i \.\
s s
.,,/ . ** /
9 -
,_ J >>r 3 \
k)' 1 1
/ ./
n ,:6) ,/ e i i
~< ,' ,
i
/
ns .~.w- / q.n wm N / -
,) , /
y\*N, N
/
p\y _/ 5 ,/
/
W i I ", m 8 QL io
+-
i $ r
' r. %i i .e b .h to b .. .
chh (( [ s t s' a _ sh-z o'o+3gLt a y n m'
/ . ' 'l fl a s v u o, cwdai ., e, e
k? ($. G" f'I 5 $ % N 4 ) id I
\ \ / /
v 3e t 4-81 w.,,.. - , , __-e,,_+--r, v,%,,,-__-,,_,-,-,,,,---,m,_,.,.....w,,,__,,-c._m_,__,,,.,-y..,oe ,%.,,,,,_,, ,,,,,,,."we"mwv-v-*wrr*W=e-e*w~~=-e- -
e_ : Figure 4.b25,8: Contaltunent and Safeguards llullding Plans (El, 778,790)
- '.' h? I .r v, #. \l [ } ,'
'6 ds i 3hje i i t i = !!f v- M rJm*~MWrj tE turn e t 4tiet 6" m 00 00,- - .s- ! .lD*1 M 9aaer n at * ; " f+a}. t i .: Jh .
- E Jo--
t a-dl : 0 sa y#..Ys . aa ;+ 4 t t
; st t
4 t-
-8}t d , .- "as " "a,;, c u % lj gpp i - , [.. -
y. i g .
- :- i (k!,NS! L o b i- @g} b h. o, D' 7-I E E 1Je 9.b. m.e, r p. , v u 7, g . . s IF ,p f"r M aC" 5: -
c , o e [.
< 8 r J x{ t t bi $L b b.e Jibsm,a s. . 1 t r-- -
o ;
. s d.uir !.3; g ;6Ay Miss!g8- 8 ll "o o no' -
f'R"' V-i
'it j f fig 4 3 *, 6 b t 1,*
Eg t q ? t t ( W.' t ? ' A - ; ., , ' y t" t! h iS M QEil.%g! felis!ct k{ %g+tsU HhM!n2Ef C2 - @ sbl1 *
,. - . .~....e -.c; f e.
nt.$ 0* L a, fg_(!. s d * ** sis b 6ne . t '. .....eLtLtiE,,g, L E~FS 5 J > rrT CJ [imi8
-.fLM] .l . n J. ~ ~. u -- ,
e e , ., _ i-.--.---.
,+ + . + - $$$3 t lb - / - [go a T
a o a lj iga. a ma ' c e j ,./ m & -. -,-r.-- 4 11 ai __ ig 00 ,4i,tl . H i
, e_,.IfE a ,.mI !g , .g ., a .t d- "f;a tl fjiiff 20 jfdi. .
93 bW r
.t i
AA . !CI--c - M L ft i i,ff.. E (]J il_ a b [i )-
'" x m . i.- o, . .iaEEEb a in 3
S u E 6 t! y I g 1[*mvvyh=%v -- g D D ', 1
. E s .t ) 6 b a b < - t. .i .ta b %m r ,U td l--t-U . b h .3 (T) 00 () l ~
IEE$Ef$SilhE5$$aa : F F O C b 2 i y0 0 .
'~ $In.r:SGNUE0055EE5 'Ip 3 ())_T 3 1_ Q ) ! " Q 4 ,! t.E 0' 0) } 6 t O .OsaD [
_. = = 6 o- .,
.) E$ o i tD O p- 'I ' kJ jb ~
d M' l[ a a f
; tR 3 .; ( q_. 4, . _ _ _ _)i ====<
Ilrl"nno gN . T 3 i_ u m u ;d gg eQ w% w 'a gg fj Q.
, d T+--*- a ocri
- N x h rL e -.
lf E57 eo 4-82 l- -
_ _ . - . = _ _ _ . . - _ - _ _ ~ . . - - , _ . r r i Figure 4.125.9: Auxiliary and Electrical Control Ilullding Plans (Els. 807,810) (ff ff p eu
, k d,.y 3 ff.II .. .t . .fI b .![#.!$$y f b)9[ ir: #$ , j' E i i ! ! Mf! O i- p.M[rshditIIf"g! $ { l j y J if n t.:.
s t p.e.u Ji; g/{f f ,. - rfqftufi
- f. Init up i, p OH.ii. .i.U,b. r it[liltrr}}.t!rl.O.H!fjj>*jpij
~
f in !h n e ty nr i i t i
- ZI.T. # fMIW t e
t
]7 9::- -' 4 i I.
3 y : 1 , S 1., i .31- 4
.I l !! Ik li
_.1 N.- .b' k l. m(, _ .
+r,-i/ ' l L E l?b A h ) b * <-
g tus..] t i ,-- , hg p i
=f, _.
y ) D eJ"' ?"' .- O g i e ...> a ,-d
?;o rL q go tra io m lau.ufgo u
na Dao j J, u, c '4gQ J,' 6 g*
- 6. !
,l b C-] bqij C ID "
a o g i ,. I, a '
. 4, ,
mg E g -_I lnus;11.go 8OL jy" ra l m ni, l *~% 4
,c.: y e .
te p,. 019fnD,='<i,m . p i e ' < . o., o eN , t 7pj k
-(m i } .4 e r ]n~ Q.
tb7
-.+
__ ur i m , 4-83
Figure 4.125.10, Auxiliary and Electrical Omtrol Ilullding lians (lit. 830)
}
I k 9 k Os
#i e -,
s ,. ! ( _
, .. ,, ad" ** 6. , s , .
- . : ~. . .
A
)' {,- ?.,S 3 d ?-
- n]
a ?
;. g ,. e 3 . . j -
- e. r- 7 I
1 s a i ' l ' J- e r f;,/ f J f [ E , g N r; n .- f t-4 9 t! 7 fIl'n'!'t.J,efr
,,,., 'i ; n ,. h s: -j v J- '; j , u,m g ;.. 9tte.>1"c ~
i t [a. 4 4 2,; y at y g e t at ar6[,yt,: i 7 a.s..g u>< n> ror t ,+ 7 s tt; ' n . i u m 1. ,. . 1.t maa:.....,,4 ,, [ 3 3._ x f--f 5 1 b h i.. i, k, k. 5 blf.f. -f.[I T,.hi.ki , n. % w p[ C h 5 .5 f.'b k LE]
~ ~b .,$ 1 f Ih,!. h,..$a Nu.
n J
- 1 '00'rE Til9 Vis F1 1.i h.k f [
O7 [j[ ,Id$DfN)Ud)b ik Ef f '] lc l
,f !L.aj i-ge g . f,,...g..._-_
q ly
. . ,.. L i;
l t W (, 1l \ E / (_[ E, - _ ! %- , _ . , _ 1
,sd '7913 a.
6.4' . {l i / T _.._1 o . 5"qn,
,, f- 1 Q ,(
f Q(i.z) Q l ,6 i } , t o =' . t ,j g um r: a L w L l s t l mt;ti g [ii .=r n jO. ---) 4. i s c-C i ( O
!(t s r-1 mm b lb h ' ,.
gg O' '"t 9 r e 3 i . j l iEB i @}i C d af o < 2 .g ao ,sm ~;
!g iE]9 E- - =< 1 """
i
+ t w- I 0, *- 1 uma cmm , a g Va4 Q ,
g ~- i o G., Q 1 A 8
- E - - -o* )ll1 af }a up.3qc 3 ,,
g (/[
}) - -
g k- M \
/ '- ~>q * 'NC.,y el , (y 'ec _ -- c1 g i,
4 f, t i s 4-84
% y-, r1 .y 3.. .,---y w-------vy,.,.-w.-m-we-- y-ey---*rez--pow- .,g_.--.y... y -c +-i-g-cg.-.y-ay-71-m, - + -igrw y---y gayvy c--g.,. - ,.--.4ry y
Figure 4.1 25.11: Auxiliary !!uilding l>lan (El. 842) f i s' 1
! t.
b [ w
.: a ?
O *s
' O V) .t) [i F L
h v s ., 5 - t L e s, a<-
, <>g. .
('- il l E sJ ed SL _ g
,A. v 0
- o. o o
.h h O' - { C.
cynR h-F 6> .9 D +o ;- .: D. :- G L. +. Q (A L.
- to . . =w O V s.'s 8
f 9 n $ occ f t,* 0, ~ m 2 m- mm,4. a f g f a #4a Pi sJc( ,1 J e- 0 JI d iv t- PJId I4f4
.M., '
- p. 04
- w N,
o s j 'W, nii e r- g - g-
/ e gr w; ~
[D L: l ,x;l ' e ' O LJ=\ / g
*- N o
_ 0 O I f a h
- f. '
4
'h ) .
Id us
; r.
s EED [j rhn Ae s-i ( !,, f _ _ + v - x/ O I b -bl O I X Y O TSv 9 g o 4 1 Lal -l ^t l l : _ 8 s x v 8 n ( . p, , g,- 6J - - s f '.. ,/ . N -s a
\ x I b 'N p i x - - .) _,
s s. E 4 85
. _ - . _ . _ _ _ ~ .., _ . _ . . . _ _ _ - . . . _ . __ _ . . _ . , . _ . . . _. ,_ _,
1 I;igure 4.1 25.12: Auxiliary and Electrical Control fluilding Plans (El, 852) l i I L __jt . b7 .5
.-~ - - -
l. r:7 , ! M d IId.1 {.- w m, 4 ..4. : ; t
- ",v
>;p y { fif(f *i ?- ej t. ~
n
, g i t_ .. . <
Af Mk 1,'
^6ees wwwww
([{
. t, ; -
(
- c. -
[ e 9eTtY'.".'d4 5 ; gi fg! 4 2 s o- h q;; f e a- +F4 t, 5f 4=gc L-4 U ,1 b e. l 3 a i r s q q r .s s t c <1k e I,I$ N IIE3s } N $b h$b', s
$$c i $
1
- l g r_ t .
7 ! E E i T[,v.# B 0 0 0
- 7. Z 7,
-. 0 '9 '
in,7__Jf
--' mm O O I' Ogg * -
I e 44
$ fl M _- 'U" 7 l g [ '
l l O O ]
'N 1 l --
m c:='J d:= [ O O h ._.=
'f ]
O ; E D ! ! 2 2 , 7 ij s Dr [= a
.. J-e ' I i .' l O O 1 1 0 8 .= ] Ds* W= ~ d *+ J-y 9 L ? '}
a v D O$ l D D I I I*l _ a os .u e ild 2: I. ftdi
- s s'xA O C! D88 0 y, _
g 6-f k ;,
'yiT 7" ", - ~1 c a ..
- f. PY .;
y.e b" e at a l l l l 4-86 l l l
. --i-,, .-,e.-- , - . _ , . _ , , , , . . , , - . - .,,~-% , - , , - - _ . .,~w.r..o-,%.-v,.-,,.--m, . , .,....%....v,.. --.,,-..m, .~ww.-
1 4 e Figur 4.1 25.13: Auxiliary and lilestrical Con rol liuilding Plans (lits. 873,886) 9 i .. , 1 I I I a a a 4 4 $ a Ik i k ( l' 4 6 _u} jL
,b 4 i s 4 !. i; j '
u
'1" a .-! T 4 4 ~ nr.8e f[Ih1 . .-
h l { & W, ,i> b k
; a6 <a L 4
', t
@j 4 . -/. r/ _ _,
l 7 ; j L *. f[ ~i ----
- .; . , 'C l mm uns I L._.. ML t 1 l I L D D J '
i I I I r ~e lfi . 3 ,! ma4a o l e : ; u Wi .ib:o 4 r a, t y . Ig a .L _i ,\ _ I r t- _J ,J ; , L J._
;l p D l
- l-lI[Q.
I l
'gl +
l _
/ I y 4 ----~] { I l
e . I t i - . s - i I h- 6 i { u I il EL . c_ , g i _ . _ i I
. l .
ih o pa M'. ho i .
. 1 e.
g A- i :
- s.
- e i
~ ,
v
~ .a a E [< j l .m.
s.s . ~ 4; g __ , ..__. g . _ , 4 3 I I l l l l 4-87 i
Figure 4.1-26.1: ECCS Valve Alignments (Standby) N 510H 8
*%d 461DA A % $12A 95119 RWST -k @
W h
<> AA 65129 651tA ,3g1, 5 @
w N Id - * * * - CULD CCPs h"
$(04A h' ig LEGS IALD M
E 8812A
$% ~
RHRP
" " * ! N MA {,EGs I
CCHT. gw s1 M 8716A Q "l ' h HOI i 8811 N LEGS 4924 ,g (2, si CCNT 57168 $d' SLSAP #J " M CCAD ~
"P !- RH HA '
N X h LEGS f3 4r
! ~, i % h g e Lo 5 awe tu) exn gM e -~ 5"" in' e.g -2 P I ,o w',&,,a h. -
Ji g $;Ps 9M LEGS mA21 A
- h SM2A ma ggg-IA LEOS l f2 Si un e e sf8ai. E%A Rwst : X l i l
t-l 4-88
Figure 4.126.2: ECCS Valve Alignments (Injection) m
# 6510t % # B510A -
g e512A ett te _ RWST -k $@
-c*o--c=o- \
w#
.s,2e . tis. , , ,
f1T & n N O + tana ce , e e-p X
$ 6804A N'
6ac1B LEGS
'MI
.# I e .1: A-H-
] ea12A g cu n m as< w ' x x '*j e h > 3 =^
is'if et
@ i m.,
gy k' s'tsA
- .c o, ,, ---w-seit e .a LEGS 8 9.24 % 5 , 3 f2 3b ccur. - -M* .n o ar $typ g2 GOLD E
AtsiP $ img 464 60L I N X (3 di MM } l ,a4n sac 9e
~ , ant aa $ g=2e tui ~' el -
e j" l O st,8:'e I
;707 , - 9 5 m3 ce se. X LEGS l" aa214 g
I (EGS l l JA (2. 3) w3 g e E{sai4 A RWST X 4-89
.. _ . .. __- _ . . - _ _ _ _ _ . - _ - _ . _ . _ _ . . _ _s. ._ .._ . . . . _ _ . . . . _ _
l i
'l -i . Figure 4.126.3: ECCS Valve Alignments (Cald Leg Recirculation) 'I 'I a
95106
* # 3 8510A % 8512A 8511 8 nwst -4 .p 5 85128 6511 A .
M N #d CCPs h. GAD h ' 8804A X geo,g LkOS (ALO E -
'A-s >-
6812A I h h M N mm m I N'
x C4'4 D saow L,y - RHRP
_ CONT. 47t&A
- e -
8811 g LEGS 8924 gn f2. 3) CONT. -- g7168 &q 7 StMP #2 " d L pp . . COL.D
' GM ,,,,, 6 RH
- l g N (3. 4i M -N "
.6048 88090 " .A II $ HOT..
a)(
.g-g y l ,. l ~ .S *2a t, sos -r .
m "At .
%h hkh. , sip. i - X . Leas 4821A T S soo2A - ,<n s . ,. e e s!-. al.
_ M s 4-90
'F Figure 4,126.4: ECCS Valve Alignments (llot Leg Recirculation) i 7 &&t00 * # =
6510A
% 8592A $411B ~ '85128 SSitA' LCV,112 N TJ h 'X LEGS y --t9 .-
S
.- I e w ?
8812A i v - b com
.S. '
63A aos WN ^ U 2> A p CONT- 8716A SuuP et * $ Hot sait . X Leos G 8% 8640 CONT 87168 - StenP #2 h E -i ' an m g N x LEG.S n, a
->e-MN astre 'd beo4e soone 30, 6 88029 gX x LEG, lf) } = n .S 4 . -S J u N cow aam na218 ' 6 8835 g
SIPS M LEGS - 8821A . E anca4 x . Hos
- ,J LEcs-
- ,,,, g.sg,.g - "' "
nwsT : y i N i I 4 j,
- 4-91 w.... , , . . , , - , e - - .
- c. ,
- - - . . .- -- - - .- - - - - ...-.-~ . - - ~ . ~ . ~ ~ - - -. .
(
- 4.2 Plant Models and Methods for Physical Processes The severe accident sequences relevant to the CPSES IPE were analyzed with the Modular Accident.
Analysis Program (MAAP 3.0B, Rev.16). The MAAP code is an integrated analytical tool for the analysis of severe accidents. A detailed description of its methods for phydeal processes is available in Reference 10 and is not repeated here. The plant model utilized in M AAP for the CPSES IPE is an input to the code called a parameter file. The data used _in the preparation of the parameter file is given in Section 4.1. The actual paramater file and the derivation of each va'ne is provided in Reference 25. Many complex physical processes are often modeled in MAAP with simple formulations. In addition, the code is capable of calculating the potential range of outcomes of these processes by varying certain parameter file constants. In cases where issue re.tolution has not been achieved pending further research, parameter values that lead to more severe consequences were selected. Furthermore, sensitivity studies were carried out as described in Section 4.6. These sensitivity studies address, among other things, the issues 'isted in Table A.5 of NUREG-1335 and take into consideration the EPRI (Ref. 33) guidelines for sensitivity studies for IPEs performed using MAAP. M AAP calculations were used to quantify outcomes, but were never used to limit the number of possible outcomes. For example, with a cavity area of 70 m2 and an overlying water pool, the debris bed is coolable using base-case M AAP parameters. !n the present study, however, the issue of crust formation was recognized, and the non-coolable debris bed case was also calculated. 4 4-92
. - - ..,s - - _ . . . - - - -. . . _ , .
4,3 Bins and PlagDamage States This Section summarizes the process and the results of the binning of the Level I functional sequences leading to core damage into Plant Damage States (PDS). A detailed description of the CPSES PDS formulation process is provided in Reference 27. Results of the binning processes are discussed in detail in Reference 7. A PDS is defined as a group of core damage sequences that have similar characteristics with respect to the severe accident progression and containment response. The CPSES PDS result from combining core damage state attributes with containment safeguards systems status. The core damage states attributes and the containment safeguards status are described below: CORE DAMAGE STATES: Core Melt Timing The time of core damage determines the decay power level and directly affects the rate of core damage and energy loads to the containment. It also affects the consequence assessment, i.e., the time of release - of fission products to the environment. For the purpose of characterizing the time of release and potential off site consequences, two time periods were considered: Early: Within 3 hours from the time of shutdown. Results from unavailability of ECCS at injection and/or of AF, depending on the type of accident. Late: Later than 3 hours, Usually occurs when ECCS fails in the recirculation stage following successful injection at the required tiow rate. RCS Pressure -- The pressure of the RCS at the time of core damage affects the phenomenological events that can lead to containment challenges. 4-93
Low: These are pressures less than 200 psia. Direct Containment lleating (DCH) is assum'ed not to occur at these pressure levels. On the other hand, this is the situation where stear explosions leading to the alpha failure mode are given the highest probabilities. High: These are pressure levels near tie RCS operating pressure (denned here as more than 2000 psia) where llPME provides the highest loads to the containment and is examined closely. It is also the situation that can result in hot leg failure, pressurizer surge line
. failure, or induced steam generator tube ruptures. All of these are considered in the CET.
Medium: These are pressures in the 200 to 2000 psia range, where itPME loads to containment at vessel failure are reduced in comparison with the high pressure situation but still require investigation. All failure modes mentioned above are considered possible in this mode but have lower probabilities. Although this is a wide range, Level I results show that the range is populated almost exclusively by induced seal LOCAs of approximately 250 GPM/PMP-(l" diameter-equivalent) leading to RCS pressures at vessel failure around 1000 psi. CONTAINMENT ST/.7ES: Containment Pressure Boundary Status The containment status at the time of core damage affects the Ossion product release timing. The three cases considered are: Intact: . Normal containment leakage (0.1% volume / day)
. Un isolated: An a!r-to-air penetration is open. ,
Bypassed: The core damage release bypasses the containment as in a Steam Generator Tube Rupture sequence resulting in core damage or a V-sequence. l' I l i 4-94 ( !~ l, _ ~ , , - ,_ _
.a .-
Containment Safecuards System Statill Containment Sprays These provide substantial Ossion product mitigatior and can preclude containment failure if operating in the recirculation mode. The conditions considered are:
- sprays operate in the injection mode, ,
e sprays inject successfully but fail during recirculation, and
- sprays a:e failed.
Fan-Coolers Credit was not taken for fan coolers because they are normally shed on a safety injection signal and require an operator action that is not proceduralized to restore them. The potential benents of the fan coolers are discussed in Section 4.1.2,1 and in Section 4.6.3. PLANT D AMAGE STATES: The CPSES PDS result from combining the core damage state attributes described earlier with the containment safeguards and/or isolation status. The PDS labeling is implemented by combining the Core Damage Bin number given in Table 4.3-1 with the Containment Safeguard Bin letter given in Table 4.3-2. Thus, each of the Core Damage Bins 1 through 6 results in three PDS when combined with Containment Safeguard Bir.s, e.g.: IE,1F, lH. The station blackout (3SBO and 4SBO), the containment bypass (ICB. and 2CB), and isolation failure (ICl) Core Damage Bins are not combined with Containment Safeguard Bins because those are implied by the core damage state. The actual binning of Level I sequences is implemented in a two-step process. First, the accident-sequence event trees for all the initiating events will eventually lead to a denned end-state, which is either a stable plant condition or one of the Core Damage Bins defined above. This is seen in Figures 3.1.2-1 through 3.1.3-6. Then, after all cutsets are binned into each sequence and Core Damage Bin (Table 4.3-1), they are further segregated according to their Containment Safeguard Bins (Table 4.3-2). The frequency result of the binning is shown in Table 4.3-3, where the columns are the PDS bins, with the first character defining the Core Damage Bin and the second the Containment Safeguard Bin. The selection of representative sequences for each bin is discussed in Section 4.6. 4-95
- - . . . . . . - .- - _.. - - - - .-. - . _ ~ - ~. - - - -~.-. ..-...- - .
P Table 4.3-1: Sequence Charactefistics of Core Damage Bins CORE $0QUENCE Cil AR ACTERISTICS DAMAGE RINS Reactor Coolard System (RCS) breach wah preuure and leakage rates asuwisted with LOCAs of I 0.6 to 2 inches in diameter (includes stud open PORVs and larger acal LOCAs), with early rnciting of the core. 2 RC$ breach with preuute and leakage rates associated with LOCAs of 0.6 to 2 inches in dienwier, (includes stuck open IORVa and larger seal LOCAs), w..h late melting of the cors; Egh RC5 pressure. Leakage rates asawisted with boil-oiTof the rea.: Lor coolatu through cycling 3 pressurizer relief valves (not stud open) or snall seal LOCAs up to 60 GPM!PM (0.6 inch diameter), with early mehing of the aore, liigh RCS pressure and leakage rates asuwisted with boiloff of the coolant thruugh cycling relief 4 valves (not stud open) or nrnall seal LOCAs up to 60 GPM/PM (0.6 inch diameter), with late mehing of the core. 5 Imge rates of leakage from the RCS and low pressurcs anweiated with LOCAa greater than 2 inches in diameter and failure of soolant inicetion, resulting in early melting of the core. 6 LOCA greater than 2 mehes in dianwter conditions, with fadure of coolant recirculation and delayed mehing . ECB Bypass sequerwes (3= l interfacing LOCA, a=2 SGTR) with failure of soolant make up, ICI Any core melt acquenec where the sontainmeniis also unimtated. I SHO Station Bladout sequences (or equivalent equipment failurra), ( y =3 carly melt, y =4 late nwh). i Table 43 2: Sequence Characteristics of Containment Safeguards Bins CONTAINMENT SAFEGUARD FAN COOLERS CONTAINMENT SPRAY BIN E Failed injection only.. F Failed injection and recirculation H- Failed Failed l i i l - _ h l 4-96 1
i Table 4.3 3: Binning of Level 1 Functional Sequences into PDS , an ja -a m -a i.n .c i wm_. w on.% i , 3__ _ _m - w_a. .
,#- 4 . -iu . o.u,,_ in_ i.s.
_~ in 4 . ." - -
;s;rq q <a2 : r--r;ary t- t -
am.7 m i m 2 out,are m-- ;;g,+ m t-;{ 2 ,,;7,
, ,y.m i 4 t-tr;; g* t- j y-- -j w 7p--
m ., ggg_h-i
,,cy,
_ _4_.f__gg.._p_ .; _ q.g . =.t p_.. __q.7.g , , ,
, , , , , #__ ,_t gy .
g ,7_. -. 4-_ p- p p_. 4---q-7...pp__7_p.r_q - [ _7 ~1 . d,_7L_,irsf. _ _._ _ _- - + ~_~--t
~ ^
t- - t_._L_ L
._u_c_u. .c cu r g ,
i ,. , i ~! i i t' - ----tin _-}_iWO
, t, i
i i I g}
~ .u cus aI~~~ f~~ __.._ _. ...._ i i j . 7. l .I'~f~q ; -- . ~g~.~ P~~ iN 9_i l g ;< wgr i - b-j g cy;7 rial- tH-t-- i 1 ;
7 i - +e s.n 2ma vtycui
- i._ f
{~. -. i - Ti. ibagg . i I ~~ ~3 4bIi IJuU a ar~-~! ~ - ~~~r~ t~ .3.u i Hip WGi- 1,s s f ud *a i J2na ~~~ L_T~~ t ~~-*l -_+; ~a_u : Mj i j i j
, 1 n.+ wbs p _.
a
^1C"M U _3w- J. i nwum 1 - .L ~j -)L__.L._b.J._ L_ 4 , ,- 1 t
t ~u i6k dr~^[ -~~" "Ul i2Y 3.50 _ L_,__ _ . _ ,[_. _j. i n,_V eiI+idY, ~~'l.._ ._+ t' _-J _J _4 _1- , j_ _g g _j-_p._p 1,_g _7,4._.;_ g7_ g p H 7.g. ,; y_.77g, g _ p..y ,a ,xa
._7 ._, .g_g7_ma .
t_1, 7 _7. _7...-1 T_ ; T_ 7,'M m ie - 3a7
-_; y . _7 7._ ;__ 7 +_ ,-
7 _, , x , _
- m. q . -
_7._. 7 7 ;
, _7 7_- _ m _7_, , ,g3-__p ; _. y,__ _ , , , ,
7ggg i
;;7cy, ; +-___,.:
p ; 7-F-1 ,
+.___._ ; , ._._p___c_t r- y %p 7~-t - r r, r 7,t _,L _j -+ w,g 9
_ 4_. _.] - M , isr rwt i i ; i : t i ,.a w. T- r T T- T 7 r-r 4 t ,t r~T tM t,j ' rein c T-~[, - -~ Tunch i i i i i i T
~
escui gcu, rgirtrd- t-r i- T i-r-t t-
-tiu2 -r-4 1 t*
t i 1 t-~1 t r---d I ! i u ebut me. idi m T; --+-- T- T ~i N p -- q- gf 7- i ; ~^^m*~Y~Y WWi p 7- -~
-7 i -T~T~t~ % M sm ng myy, - +~t-+g ,nem t-9_.._f__r p.. _g.q _ , _t77-r _+p. ,p-i - m; _7t_-[-- ; i i-1 c, ,
r r [. - jia.m T i r- rl C P- r i t-t-t ;
} i 7- ~7- ;gggf j---+r p- 7 7 i
f-r
;j,cy, -t q - - ; j_ , ggnt , 3,s.,
acgr T-ar.cm ry--- hy T-;-+i 7gg i -t i , i iu si"'-^ F~-4]j i r t-i-~-~~~ - i i 2 bat
. uuoi u46.
i on .
"7 ;;g ; gi MiXViiiG:j i " *-~
usem 1L f,- 7~-
-t : t 4 i r ' s ouoi+- uu wscu2 r -t 1 ; ; i r-- ;
( y--- + 3- - .
-mcw4 :
y i L._1' .+ j itet s: 3424
. --+
i isao;
-4 I4.74E4 w3CM 3 ]r i ttFi i i i I J 1 6 uscu. , Et niEsc j f j T- i - p.co. 2.5u3 imir.
nicu2 , i i 4, 4 *ai w ! i 2 ouo iiii i n:CM t j l 4 l l f [ 2.0E.7,l f l $ .0b t2 2034 m.: n2cw2 i 1 ri ' >*.; nos: i i i i e su2 . n1CMI ~T [ ~U ! j ~f[ib 6 EE.1[ EU aru2 n.cm j *-~~i-f i i l~ tru[f i t--t tiobe:f t
, i~ I b.a.st F--t i t ~
4 7uot uu2i
=
tcU ne i a4cu2 ; t. ~-~* !2. it i.T. _i j i ._ - 1:u2l 2.64E 8 J_ . M'E2 { i i c,n'[ i
~i i~ '~l T-~~ 5.R 17 2.17E.s neCMi I } { .~[ I ( l I l nrm 2 i ~ N3 U.t ! t 2.t h.10! 8.6lEa n ,cm-t t t-i i i i i n7cu2 i;.tri i- j j l +
1 lE.i t 2.Mb7 n9CM I I I i
-.7,. _ i , !. -+ .
nucu2 nLcm j
., j ~~t. ,,,_,,a-1 i i !
2.?b 7 g ~6M.t y L- 26667 i. _.___.i I
; 1 ,
nLCM2 I I f I l { j i l
'r ea.a Ii'it i 2 ai.47i 4E41 L Jf M7M# 3. iNU E 4' $4 bbl 7.067;9 0E 75 J 3E 4ILBE.7f5'5E4 LSE+ 9969 t2b7 4 410$
4 97
. .- - , ~ . . _ . - - . - . _ . - - . - - . - - . - .. . ~ . - . . _.,
4.4 Containment Failute Characterization The first step in this task was to identify the possible CPSES containment failure modes from the point' of view of the challenges. These are listed in Table 4.4-1. Each Plant Damage State (PDS) can lead to-more than one outcome and, conversely, a failure mode can be due to various PDS, as ind:cated in Table 4.41. Although the unconditional probabilities for the various failure modes are listed in Table 4.4-1,. they were actually determined in Section 4.6. The second step in this task was to examine and summarize the possible ways by which the CPSES containment might fail to perform its function. The examination consisted of an analysis of penetrations - and an evaluation of containment s:rength from the perspective ofliner tear at discontinuities and from the perspective of ultimate strength. Section 4.4.1 of this work provides an analysis of penetrations. These penetrations were grouped into . four categories and compared to the penetrations that were examined in detail in an NRC-sponsored generic study (Ref. Il). This was done to check the structural stability and failure and leakage potential of the penetrations. The objective of this analysis was to demonstrate that the CPSES penetrations are similar to those that are used throughout the industry and that the conclusion of the generic study, that l: failure at penetrations is not expected, also holds for CPSES. l Section 4.4.2 examines the containment strength from the point of view of rupture. That is, it attempts to estimate the pressure at which gross failure is most likely to occur. This was done using the rupture to design pressure scaling method of Reference 17. The result was compared to existing finite element analysis results for similar containments and was shown to be conservative.
--Section 4.4.3 examines containment tailure from the point of view of liner tear. An approximate methodology deveioped by EPRI was used to determine leakage onset due to' liner tear at penetrations.
The approximate snethodology (Ref.13) makes use of experimental test results of peak strains applicable for steel-lined concrete containments. It also requires an assessment of the global .; trains imposed on the - containment. This was done using results obtained from detailed structural analyses of a containment of similar design. 4-98 1
<- e n -, , - - . - n 1 : ,-- ,-n n -mv e
In Section 4.4.4, a containment fragility curve was developed by assuming a normal distribution for the failure pressure, a _7% coefficient of variation and a mean equal to the lowest of the failure pressures determined above. 4.4.1 Penetrations As with any other large dry PWR containment, the CPSES containment penetrati ons were divided into the four categories considered in Reference i1: (1) Large Opening Penetrations, (2) Purge and Vent i System isolation Valves, (3) Piping Penetrations and (4) Electrical Penetration Assemblies. 4.4.1.1 Larce Openine Penetrations There are four of these in the CPSES containment structure: (1) personnel airlock, (2) emergency airlock (3) equipment hatch and (4) fuel transfer tube. All are shown in Figures 4.1-23.4 through 23.6. Egrsonnel Airlock The personnel airlock is a double door assembly approximately 9 feet in diameter located at the 832' elevation of the containment. Each door is hinged and double-gasketed, with the leakage test pressure L applied to the annulus between the gasket sealing surfaces. Both doors are interlocked such that if one l' door is open,- the other cannot be activated. The doors are designed to maintain their functional capability _ during testing with no additional requirements for locking beyond the normal locking procedure. The double lock mechanism minimizes the potential for isolation failure during normal operation. The personnel airlock, which is shown in detail in Figure 4.1-23.5, is essentially the same as that shown in l Figure 9 of Reference 11 and is similar to that of the Surry plant.- The analysis of Appendix C of f Reference 11 is appropriate for obtaining an upper bound on the flange separation under internal 1
- pressurization for.this airlock. Based on the relative similarity between the example airlock and the CPSES airlock and the similar sizes of the containment, it was assumed that these separations are also - representative for CPSES. The leak area corresponding to this separation was calculated for the CPSES 9 ft diameter airlock and is given below as a function of pressure.
l 4-99 l l 4+ w-- x-r wy b-x -- e-y e- e *---g -, w ,- m e--- - - -- - -- --
CONTAINMENT FLANGE LEAK PRESSURE SEPARATION AREA (psig) (in.) (in.2) 30 0.00017 0.06 44 0.00024 0.08 100 0.00049 0.17 119 0.00057 0.19 It should be noted that this analysis assumes that the inner door is open or tailed and the full containment pressure is placed on the outer door, it does not credit the seal which is fully compressed when the door is locked, nor does it credit the pressure-seating of the inner door or the equipment hatch described below. Elastometer seal tests discussed in Appendix A of Reference iI revealed that flange separation resulting from extremes of severe accident related pressures is not likely to be the source of significant containment leakage, in fact, for all materials and configurations tested, no leakage was observed for P.ange' separations up to 0.06in, and temperatures up to 420"F, which corresponds to the highest temperatures expected at CPSES for dry sequences. This is a bulky structure with a large thermal inertia. Therefore, it is not expected that non-uniform thermal environments or sudden temperature spikes, such as those associated with burns, would challenge its integrity. Therefore, given the small leakage area associated with these penetrations, even when improbable assumptions are made for their maximization, leaks from these penetra: ions were not credited in the analyses of containment failure times, nor were they assumed to fail at pressures lower than those i discussed in Sections 4.4.2 and 4.4.3. l^ Emercency Airlock The emergency airlock is a 5 ft 9 in, diameter double door assembly, with 2 ft 6 in. diameter doors. It has operating features similar to the personnel airlock. The interlocking mechanism prevents both doors in either personnel or emergency airlocks from being opened simultaneously and ensures that one door - is completely closed before the opposite door can be opened. Since both doors are pressure-seating, the
' discussion of equipment hatches below applies.
4-100
Eauinmenj Hatch De equipment hatch is shown in Figure 4.1~23.4 It is a 16 feet internal diameter single closure penetration. The bolted hatch cover is mounted on the inside of the containment and is double-gasketed with a leakage test tap between gaskets, it is essentially the same as the pressure-seating design snown in Figure 6 of Reference i1. The discussion on analysis methods for pressure-seating closures in Section 2.2.2 of Reference 11 is primarily a veri 0 cation for buckling stability beyond the ultimate pressure capacity of the containment, since it is felt that otherwise these hatches will not leak. An extensive non-linear analysis of Sequoya's equipment hatch was performed by Ames Laboratory to evaluate all pressure seating aspects of that pressure-seating hatch. The results of that analysis were judged in Reference 11 to indicate _ that leakage is not likely for this type of closure, nor is buckling of the hatch below containment t 'timate pressure capacity likely to occur. Hand calculations were performed to check the structural stability for Sequoya's hatch which is similar to that of CPSES, and the buckling limit is felt to be applicable to CPSES. This hand calculation method in Appendix F of Reference !! showed the probable actual minimum buckling to be around 150 psig for the Sequoya hatch, which is well above gross containment failure pressure (for CPSES - 136 psig, Section 4.4.2). Fuel Transfer Tube As shown in Figure 4.1-23.6, the fuel transfer tube is a 20 in. diameter stainless steel pipe inside a carbon steel sleeve extending from the reactor refueling cavity through the reactor cavity wall, containment _ wall and fuel building wall. The fuel transfer tube (FTF) is sealed on the containment side by a blind flange and on the fuel building side by a closure valve. The FTT is supported by anchors on the refueling cavity floor and the fuel building pool Coor. For a leak to occur between the FTT and its penetration sleeve, the leak must penetrate a belows on the containment side, the seal plate and a belows on the outside of the containment. The CPSES FTF is similar to that analyzed in Appendix F of Reference 11. That analysis has found that the FTT and closure assembly should not leak for the following reasons: l
- Rising containment pressure increases sealing force
- Hydrostatic pressure increases sealing force
- ~ Volumetric swell of o-rings from Guid environment increases sealing effectiveness 4-101
. - - .~. -__- . .- - . - -...-. . - ... . - ~ - . - - - . . _ . . . - . ~ . ~ -
4.4.1.2 Purce and Vent System holation Valves Four penetratiors are in this category: (1) hydrogen purge supply and (2) exhaust, and (3) containment purge air supply and (4) exhaust. These are the only penetrations vented to the containment atmosphere. although the containment purge air line is opened once a week for three hours during normal operation as a part of the normal ventilation program. The containment purge air lincs are 48 in, diameter butterfly valves and the hydrogen purge lines are 12 in, diameter butterfly valves. , t Reference 11 considers these butterfly valves associated with large diameter pipes as having the greatest potential for containment leakage. This is because these valves use non-metallic seals, whereas other isolation valves, such as gate, globe and check valves, use a metal-to-metal seals. Since these valves are designed to the ASME Code, they are capable of maintaining their structural integrity under severe accident conditions. The main concern is that the non-metallic seals between the body and disc might-become degraded when subjected to the high pressures and temperatures associated with severe accident - conditions. 4 Figure 4.4 5 shows seal . life as a function of time for various materials and temperatures. The materials used for pressure seals at CPSES are all silicone based (Ref.14). It is evident from the figure that
- significant purge leakage is not expected for CPSES because silicone based seals show excellent temperature resistance (over 1000 hrs at 400 F). The maximum calculated temperatures _ for dry sequences (Refs 2,7) are of that order (420"F).
4.4.1.3 Pinine Penetrations Eight different ty;% of penetrations were analyzed for the six reference plants of Reference 10. The:
--CPSES penetrations are of three types, all of which are similar to the types analyzed in Reference 11.
The first type is an embedded pipe penetration that is used at CPSES for cold pipes. It is shown in Figure 4.123.7 and is essentially the same as that shown in Figure 18 of Reference Il.1 At CPSES there is a sleeve around the process pipe, with an inner diameter slightly greater than the process pipe's outer 4-102
+w-- , , , .c , -.---.i .- , _ . . , ,%.s .-77 , - - - - - g- ,
diameter. Gusset plates or lugs are welded to the sleeve for improved load transfer to the concrete wall. The pipe is connected to the containment liner through a plate of greater thickness than the liner. The second type used at CPSES for hot piping is the Oued head type of penetration, also shown in Figure , 4.123.7 and in Figure 20 of Reference 11. The term Dued head designates the manner in which the process pipe is attached to the sleeve. This is essentially a forged piece with integral welding attachment points for sleeve and process piping. The third type of penetration used at CPSES for htt and cold piping penetrations running close to one another is the Ganged penetration shown in Figure 4.1-23.7 and in Figure 19 of Reference 11. The term also refers to the attachment between the sleeve and the process pipe: The sleeve is also embedded in the concrete with anchor lugs attached to the sleeve. Two types ofloading should be considered for all the types of piping penetrations. The first is the effect of differential pressure across the penetration. This is the same type of loading considered for the large , opening penetrations. He second type is unique to piping penetrations. DeGections of the containment wall-occur during severe accidents because of the internal pressure and because the containment experiences thermal expansion due to accident temperatures Piping that passes through the containment is typically attached to structures both interior and exterior to the containment as well as to those of the containment itself. He resulting differential support displacement induces loads in the piping and the penetration. Both types ofloading were evaluated in Reference 11 and the imdings were Judged to apply to CPSES, based on the similarity of pipe types, layouts and materials used for Westinghouse PWRs in large dry containments. The effect of differential pressure was evaluated using hand calculations. The evaluation - of the displacement loading in Reference 'lI was complex and very detailed. Nevertheless, the Undings were that for all types of piping penetrations examined for all plants, a failure in the piping penetrations L was not likely before the containment approached instability, which is past the gross failure point defined as 1% strain _(for CPSES 136 psig, Section 4.4.2). On the basis of the results of these studies, it was concluded that the piping penetrations and associated - piping were not likely to contribute to leakage before reaching the containment rupture limit discussed in Section 4.4.2 of this work. 4-103
4.4.1.4 Electric Penetration Assemblin A typical electrical penetration consists of a metal pipe passing through the containment wall, inside this pipe is a stainless steel canister containing electricai cables, as shown in Figure 4.123.6. The canister is welded on the inside and outside ends to the pipe. The welds form an hermetic barrier. The cables are passed through the stainless steel header plates located on both sides of the canister. The electrical conductors passing through these plates are sealed. These seals form critical elements of the penetration. Only one side of the penetration would be exposed to high temperatures and harsh environment. The projecting portion of the pipe could pick up a signincant amount of thermal energy by radiation and convection and its temperature could rise signincantly. However, Reference 11 concludes that even if the inboard seals fail on the inboard side of the penetration. the outboard seals will protect the penetration from leakage. As discussed in Reference 15, thermal energy transfer through the canister is low and outboard seals will never reach high temperature. 4.4.2 CuniltinUgnt Rupture (Gross Falure) 1.imil The containment rupture limit is the pressure at which gross failure of the containment structure leading to a putf release % most likely. The Comanche Peak containment is designed to be essentially leak-tight under design pressure conditions. It is a steel-lined, reinforced concrete vessel with a design pressure of 50 psig. The ultimate pressure capacity of the cylindrical wall, rounded dome and thick basemat [ configuration of typical containment structures has been extensively studied (Refs. 16,17,18,19,20). Similar containments have demonstrated structural failure limits in the range of 2.1 to 3.5 times the design pressure for reinforced concrete, as illustrated in Table 4.4-2. This is a CPSES equivalent of 105 to 175 psig. Since the radius to thickness ratio (67.5:4.6) is compatible with thin shell theory, primary internal stresses due to pressure consist of membrane meridional and hoop stresses. The meridional and hoop forces generated by the internal pressure load may be calculated as a statically determinate problem (Refs. 17,19) in which the hoop stress in the cylinder is equal to the product of the pressure and the radius divided by thickness. The results of these calculations have been compared with renned finite element analyses of the containment structure and found to be in good agreement (Refs.16,17). 4-104 L .
Therefore, based on the analysis of a similar structure (Ref.17) using the containment analysis techniques of References 19 and 21 the pressure ratios at yield (Ref.17) can be estimated by: __P. . f,y A i.fiy Ai P, R, Pa R, P, where: P = icternal pressure P. = Design pressure Ri = Containment inside radius f,, = Rebar yield strength at 1% strain A, = !.iner steel areas per unit of wall thickness 1 A, = Rebar steel area per unit of wall thickness f, = Liner yield strength at 1% strain Note that the above expression implies the failure pressure is assumed as that corresponding to a global stress state at the 1% strain level at the cylindri al wall and the dome. WASH 1400 used this conservative dennition of failure criterion for reactor containments, and it remains an imponst measure of strength that can be related to the design pressure, For the Comanche Peak containment: P. = 50 psig = 7.2ksf R, = 67.5 ft, fy = 60 ksi f, = 60 ksi For the cylindrical wall, the following areas are used: A, = 17.46 in2 /ft. A, = 4.5 in2 /ft. Similarly, for the dome, A, = 17.46 in2 /ft. A, = 6 in2 /ft. 4-105
. _ . . -. .. - - - ~ . -. . - .-.. - _. Substituting the most conservative of these numerical values yields: P_ , ,60.0 (17.46) _60.0 (4.5) P, 67.5 (7.2) 67.5 (7.2) 1 - 2.71 P, P = (50)(2.71) = 136 psig The sources of each value are: f, Yleid strength value taken from the specifications under thejurisdiction of ASTM Committee A-l' on Steel, Stainless Steel and Related Alloys (Annual Book of ASTM Standards, Vol. 01.03). The containment reinforcing steel is ASTM A 516-72 grade 60. f3 Yield strength taken from the ASTM specincation SA-537/SA 537M 86 for pressure vessel plates, heat treated carbon manganet silicon steel ASME SA 573-74 Class 2. Ai The area for the containnient liner is based on a 3/8 inch thick plate in the wall and 1/2 inch thick plate in the dome. A, ne reinforcing steel area for the vertical rebars is based on three layers of #18 bars at 10 5/8 inches, ard for the horizontal (hoop) rebars, on tour layers of #18 bars at 11 inches; Two layers-of diagonal riars are also #18 bars at an angle of 45 degreesi These represent the minimum specincations comained in Drawings 2323-SI-501 through 509. Variations in the spacing of the - vertical and horizontal reinforcements with elev:: tion are indicated in the drawings. The values selected above represent the minimum areas at elevation 856'. 4-106
, , , . , .-.s
- The calculated structural failure limit pressure, assuming 1 % uniform strain limit in the cylindrical wall, falls within the range of 105 to 175 psig (2.1 - 3.5 times the design pressure). Specifically, the value of 136 psig is 2.71 times the design pressure and is comparable to other large, dry, reinforced concrete containments with steel liner as .'ndica ed in Table 4.4-2. ~
4.4.3 Containment Liner Ductile Failure Limit Recent experiments indicate that the dominant mode of failure for steel lined, reinforced concrete domed structures is steel liner tearing and leakage. The onset oflocalized liner tearing has been shown to occur at pressure levels greater than the design pressure but lower than those needed to reach the ultimate strength of the containment. Methods for determining the onset of liner tearing, and the subsequent formation of leak pathways, have been developed in Reference 13. The correlation of test data and analysis results presented in that document are the resul; of a five-year research program for developing a test-validated methodology for ! l predicting the overpressure behavior of concrete containments. , Tests on individual containment segments and a 1:6 scale concrete containment were conducted by Construction Technology Laboratory (CTL) under contract to EPRI and by Sandia National Laboratory under contract to USNRC. The former simulate local conditions in various parts of the containment
.under overpressure loading. The latter involved internal pressures up to 145 psig, at which point the leakage rate became excessive and the test was terminated. Both segment tests and reduced model test L demonstrated the leakage mode of failure. Analysis results were correlated with the tests, and methods were developed to determine the liner Mn state at local discontinuities from global strain levels obtained -
E from a basic 2-D axisymetric finite element analysis of the containment (Ref.13). Peak strain magnincatien factors for each " type' of discontinuity were determined for the pressure range of interest and combined with the associated global strains to yield a peak liner strain versus pressure curve around each discontinuity type considered. Uniaxial ductility tests of liner materials commonly used in reactor containment construction were conducted to determine the ductile failure strain level of the materials. 4-107
4.4.3.1 Descrintion of the Method The criteria and guidelines set forth in Reference 13 are applicable to either prestressed or reinforced concrete containments. These guidelines provide an experim;ntally vertGed methodology for obtaining - peak strains at typical steel containment liner stiffness dH.ontinuities. These regions are subject to ductile - liner failure and are identified as typical leck sites. The regions are:
- Cylinder wall skirt juncture;
- Main steam / medium sized penetration (24 to 60 in. diameter);
- Equipment and personnel hatches (> 60 in. diameter);
- Springline (dome-cylinder juncture); aml
- General liner embedments (at cylinder base, midheight end springline regions). ,
These are also the regions that were considered in the analysis of the Comanche Peak containment. Figure 4.4-1 illustrates how the method is used to determine the onset ofliner tear. Liner tear (initiation of leakage) is predicted to occur at the pressure and location where the peak strain in the liner meets or exceeds the uniaxial failure strain level. The criteria assumes that liner tear, once initiated, will progresa l in a stable, incremental m mner until an equilibrium leakage rate is permitted. While it is recognized that additional pressure is requi.ed to propagate the initial tearing, leakage failure pressure is defined in this analysis to be the pressure at which onset of tearing is predicted. Therefore, it is conservative to assume that once tearing is initiated, failure of the containment occurs. The peak strains in the containment liner that occur adjacent to discontinuities are the product of a calculated global strain value at the specific discontinuity and the corresponding strain magnification factors. These are: Alnha Factor - Experimental work and very detailed calculations through the years have shown that, even - with cmsely spaced strain measurements or relatively fme computational grids, localized liner strain concentrations next to stiffness discontinuities are sharper, more localized, and have higher peaks than-what are measured or calculated. This magnincation factor has been named the strain localization factor or gage length factor (alpha).- Not enough experimental data is available to develop a functional 4-108
dependence for alpha. Consequently, an alpha is introduced in the guidelines for predicting containment, leakage as a constant magnincation factor with an experimentally determined average value of 4.0. K Factor - K produces peak effective plastic strain following global liner yielding and is denned as the peak effective strain to global strain ratio at a specifie location. K factors have been derived from test . and analytical data for each of the discontinuity types and are provided as curves of K vs. normaPN strain. These are illustrated for reinfoiced and prestressed containments in Reference 13 for each of the types discontinuity of interest. I!sta Factor - Beta is a stress biaxiality factor applied to represent the effect on ductility of the liner stress state. Over most of the cylinder wall, the hoop stress to meridional stress ratio is approximately 2 to 1 in the clastic stress range. The ratio decreases after the onset of plasticity and the biaxiality effect becomes stronger and more variable. The biaxiality conditions have been determined for all of the discontinuity types of interest and are provided in Ref erence 13 as curves of beta vs. normalized pressure. Since biaxiality effects are typically characterized as a reduction in ductility, the strain magnification approach in the criteria requires that the reciprocal of the ductility ratio be used to represent beta: beta = 1/ ductility ratio (1.0 i beta 12.0) Beta is determined from the curves in the reference document and is multiplied, along with alpha and K by the calculated global strains to obtain the peak local strains for comparison to the uniaxial failure strain for each discontinuity type at each pressure level, as illustrated in Figure 4.4-1. The global strains are defined for each discontinuity type as follows: Discontinuity Global Strain WaH-Basement Juncture Meridional strain at one wall thickness above wall-base juncture Main Steam / Medium Penetratim Hoop strain at the penetration elevation Equipment / Personnel Hatch s Hoop strain at the penetration elevation Springline Meridional strain at one wall thickness below springline 4-109 1
The results are expressed as two f ailuie pressures for the containment liner. Where the adjusted peak uniaxial liner strain intercepts epsilona, the pressure level is the lowest credible pressure at which leakage could develop for the specific location plotted; and where it intercepts epsilonw, the pressure level is the pressure at which the liner has a high probability of developing leakage. 4.4.3.2 Anpheation of the Method The example (Ref.13) of a reinforced concrete containment is a typical PWR wnn a design pressure of 40 psig. The example containment was compared to CPSES (Figure 4.4-2) and it was verified that they are indeed similar. The global strains from the finite element analysis of this containment (Table 4.4-3) were scaled by the design pressure ratios to obtain the CPSES global strains. This is justified by the similarity of the containments and by the small sensitivity of results to perturbations in these strains as stated in Reference 1. The aniaxial failure strains v'ere obtained as described in step 4 below, The method was applied according to the following 9 steps originally outlined in Reference 13. Sirp_1 Confirm that the containment is within the range of applicability of the method. The range of applicability for containment parameters specified in the reference document is sted as follows: Parameter Range of Applicability CPSES Containment Diameter 120 to 160 feet 135 Concrete Wall Thickness 3.5 to 4.5 feet 3.5-Steel Liner Thickness 0.2 to 0.4 inches 1/4-1/2 Design Pressure 30 to 60 psig 50 The Comanche Peak containment is a reinforced concrete structure, with a design pressure (50 psig) that is within the range of applicability indicated above. Other CPSES l dimensions are also noted to be within the range of rau.as and wall and cylinder liner j thickness. 4-110 l l l l
Step 2: Analyre the containment with the 2-D axisymetric finite element analysis as described in Reference 1. This step was substituted by the analysis performed in Reference 13. Dat analysis was of a typical PWR containment with a design pressure of 40 psig. That containment is quite similar to the CPSES containment as evidenced in Figure 4.4 2. De global strains computed for that plant, which are listed in Table 4.4 3 for each discontinuity type, were scaled by the ratio of design pressures (multiplied by 40/50) as an estimate for the-CPSES containment global strains. Step 3: Plot the global strains for each discontinuity type (see list in Section 4,4,3.1) as a function of internal pressure. His is done in Table 4.4-3. Strains and pressures correspond to those of the reference plant as given in Reference 22. The listed strains include the strain localization factor (alpha =4). The CPSES strains at these pressures were obtained by adjusting the values in Table 4.4-3 by the ratio of design pressures (multiply by 4/5), since the same strain levels are expected at the design pressure due to the similarity in the design of the containments. Step 4: Determine the upper and lower levels of the uniaxial failure strain. These were obtained for CPSES from the actual line material specification SA 537-74 Class 2 steel and are: epsilon a = 20% epsilonu = 22% Sup_5.; ' Perform steps 6 through 8 for each type of discontinuity. Step 6. Determine k and beta for the pressure range of interest. (Since alpha is a constant, it is already included in the global strains given in Table 4.4-3,) The k and beta factors are given as a function of normalized global strain (normalization factor epsilon = 0.002) in Figures 4.4-3.1 through 3.8. These values are from Reference 22 and correspond to
- updated values originally given in Reference 13.-
4-111
Step 7: Calculate the peak strains.- The peak strains were calculated for each discontinuity type as a function of internal pressure using: epsilong = (K) (beta) (epsilong) Step 8. Plot the peak strains together with the uniaxial failure strain. This is shown in Figure ! 4.4-4. Step 0: Determine the onset of liner tearing and leakage. Peak sttain, for each pressure increment calculated is shown in Figure 4.*-4. The peak strain curves wera plotted together with the highest and lowest credible ductility values, epsilong and epsilona. Liner tearing and leakage is predicted to occur between the pressures where epsilong intercepts the uniaxial failure lines oflowest credible ductility and highest credible ductility. The upper and lower failure pressure confidence levels were determined by calculating the two points of intersection on the uniaxial failure strain vs. pressure curve representing each of the discontinuity types The pressure levels at the intersections were calculated and presented as predicted values for containment leakage. The results indicate onset of leakage to occur first at the main steam line penetration at-
.a pressure of 114 psig. Leakage was then predicted to follow, successively, at the large penetrations and wall-skirt juncture.
The findings are as follows: r' L Medium Penetration: l Pw = 114.1 psig (Pw/P , = 2.281) Pw = 114.2 psig (Pw/P , = 2.284) l 4-112
m . _ - . _ _ _ . - _ . _ . . _ _ . _ _ . - _- . - . . .. _ - . _.. - _ . _ .. _ _ _ D l Large Penetration (equipment and persor.nel hatches): Pu = 1l'.0 psig (Pw/P , = 2.30) Pu = 117.2 psig (Pw/P ,, = 2.31) Wall-Skirt Juncture: Pw = 119.9 psig (Pw/Pu, = 2.39) Pw = 120.3 psig (Pw/Pu, = 2.41) l l Springline, i , P,, = 125.0 psig (Pw/Pa# = 2.50) Pw = 125.4 psig (Pw/P , = 2.51) 4,4.4 ' Containment Fragility Curve
.l In order to estimate the probability of containment failure due to spiked pressure loadings such as those associated with high pressure melt ejection phenomena or steam spikes associated with vessel failure or hydrogen burns, it is ..ccessary to know the containmen' failure probability as a function of the internal -
pressure' loading, The accuracy of the failure pressure predictions of the analysis results depends, to a large extent, on the accuracy.of the material nonlinearities exhibited by the reinforced concrete and reinforcing steel. The containment failure is considered to occur because of a global stress on the containment structure due to static pressure load. -_ Localized failure due to potential' construction defects or variations in the assumed ' strength of materials, or local stresses, is considered by assigning a range of containment failure _- probabilities at various containment pressures. Since failure of structures is probabilistic in nature, the actual failure level is represented by a statistical distribution about the predicted failure level as the mean. ' Typically these distributions are normal.
-_(Gaussian) (Ref. 23) and the probability density function (pdf) is:
4-113
' + - -g 4 y v n e , y + - ,- - +e- -r' - = --
_ . _ _ _ . ._ . . - . . . ~ . _. -
-l f(t; p,o)= exp- Y}
o [27 208 , where: p is the mean (also the median in this case) failure value; a is the standard deviation; and t is the independent variable (pressure failure level). such that: -m<t<m;
- m < p < m ; and a > 0.
If only one failure location is considered, the probability, P, that the containment has failed at a given pressure .t t is the cumulative noimal distribution: s exp- ~E I dx-P(r;p,c)=[2n-o[I (*20 2 L A convenient measure of variability is the coefficient of variation which is the ratio of the standard deviation of the strength (or failure pressure) to the mean value (a/p). According to Reference 24, the coefficient of variation for reici'orced concrete of typical containment structures is 1%; based on the yield strength of the reinforcing bar. Based upon the previous liner tear analysis, medium penetrations such as the steam line showed the lowest failure pressure (114 psig). For these types of failures, the complementary cumulative containment failure pressure distributions (or the probability of surviving a load) are shown in Table 4.4-4. These probabilities assume a mean value of i14 psig and 7% coefficient of variation as discussed above. 4-114
It is felt that these failure pressures are conscrvative for the following reasons:
- The liner tear pressures used as the basis are low because the global strains used in their determination were taken from a 40 psig design pre.ssure containment, whereas, CPSES has a design pressure of 50 psig. Although the global sirains were scaled, it is felt that the resulting values are still too high for CPSES. For example, if one took the 40 psig plant's global strains (Table 4.4-3) as a function of P,y/Pn.,,, for that plant, instead of renormalizing by the CPSES design pressure of 50 psig, the liner tear limits m .:Id be:
Medium Pene: ration: Pu = 141.55 psig (Pu/P , = 2.83) Pw = 141.79 psig (PJ,lPu,,, = 2.84) Large Penetration (equipment and personnel hatches): Pu = 144.59 psig (Pu/P, ,, = 2.8c.,) Pu = 145.24 psig (Pw/P., = 2.90) Wall-Skirt Juncture: Pu = 148.48 psig (Pu/Pu.., = 2.97) Fu = 149.25 psig (Pw/P , = 2.99) Springline: Pu = 1'3.20 psig (Pu/P& ,,, = 3.06) Pu = 154.20 psig (Pw/P ., = 3.08) (Here it is noted that the approach described above was not used in order to preserve an element of conservatism.) The second reason why the present failure pressure probabilities are conservative is that the rupture pressure determined in Section 3 is also too low. A more realistic failure limit could be , obtained by using actual reinforcing steel and liner plate tests. For example,. actual material-properties provided in Reference 18 indicate that the yield strength of reinforcing bars (#18) can l range from 67.3 to 74.6 ksi, with a mean value of 59 ksi. The same reference provides a l l 4-115 i l l l
- -.- .~ _- . - . - . - . - - - . _ . . - - - - - . - . - - - - -
i l specification value of 60 ksi. Similarly, the liner plate yield strength provided in the document indicates a mean value of 37 ksi, with a range of 31.3 to 48.1. The specification value is 24 kSt. Using the mean value of 69 ksi as a more realistic characteriz3 tion of the yield strength in the above equation, the fallute pressure limit obtained would be 155.8 psig, i l 4.4.5 Condualen i The f ailure pressures and failure pressure probabilities of Table 4.44 are conservative estimates that may be upuraded following a plant specific calculation of global strains. The following ways in which the CPS!!S containment might fall to perform its function have been examined: .
- Four categories of penetrations were examined, the same as those considered in Reference 11.
m .e are: (a) Large Opening Penetrations, including personnel and emergency airlock, equipment hatch and fuel transfer tube, veri 0ed to be pressure seating and their structural stability (buckling) checked where appropriate; (b) Purge and Vent System isolation Valves, examined from the stand point of temperature resistance of the seal materials; (c) Piping Penetrations, ; examined for resistance to pressure differentials and tt :oads resultir:g from differential support displacements; (d) Electrical Penetration Assemblies, examined frorn the seal exposure , perspective. ne examinations consisted of comparison of CPSES penetrations type by type with those examhed in great detail in the study of Reference 11, All CPSES penetration types have equivalents in that study and it is concluded, as in that report, that the CPSES penetrations are not likely to fall before the ultimate capacity of the containment is reached.
- The ultimate capacity of the containment including liner, rebar and concrete was determined to be P= 136 psig utilizing the scaling methodology developed in Reference 17.' The failure criteria was conservatively set at 0.01 strain in the rebar. This definition of the structural failure limit is consistent with yleiding of the containment pressure boundary that might result in rupture failure of the containment under very rapid pressurization, 4-116 L
,--~.,,,,-,-,.,nn,- ,,.,,-,,w---r,,.-v , , ,+-,,y ,-,-w r--,,--- - , , .,--v-- - , , - , -, v.--.w. - *m-
- 'lhe leakage or liner tear limit of the containment was examined at four key discontinuities: (a) the wall skirt juncture, (b) the Springline, (c) the equipment hatch and (d)large pipe penetrations such as the steam line. The methat used was recently developed by lipRI (Refs.1,13). *lhe result of the application of this method to the CpSliS containment is that ductile failure of the liner is expected around the steam line penetration at i14 psig, around large penetrations at i15 psig, at the wall skirt juncture at 120 psig and at the springline at 125 psig.
A coniamment fragility curve was developed by assuming a normal distribution for the failure pressure with a mean of 114 psig and a 7% coef fielent of variation. Also for the purposes of containment performance evaluations, it should be noted that releases from liner tear and/or gross failure of the containment were assumed to go directly to the environment. 'this is conservative because the onset of leakage is predicted to occur at a liner discontinuity near a steam line penetration; and therefore, it might be aigued that leakage would take place into the turbine building and, consequently, be subject to soine retention which could be quantified and credited. 4 117
~ . . . - - - - . - - _ . ~ . _
TaNe 4.41: CPSIIS Containment Failure Modes , l l 4 i AILUkI: MODF. PD$' UNCONDftlONAL llOW *IIC l' Alt.Ultt MODE LAbr!. IPOHAbilf Y 15 AN ALYZi.D CONT AINMr.NT BYPAS$ 1 interfacing $yeems LAsC'A 108 1,5 t?O7 lault tree {$) for prubsbilvy, I (V $eguerwe or illAnCA) MAAP for releases (7l,
- 2. Sicam 0erwretor Tee kupture (50TR 2CH ). ara 6 same as above except est, is (4) emi ISOTR'J and 15GTR prtdiabilities sonw from the CITs.
- 3. Inulation failures ICI 9M49 una as above encept ref. is 16).
E ARLY CONTAINMrKr i AILURl3 4 8'ressure load due in steem esphismn 4H,Ill,311,31.4 F,6f , 2.9tA17 CONT AINM LNT e 'pha). 1 f,45flO.6f f M ,2f,3 LVINT H,35160 TRLE 4 (CET)
- 3. Dirett Containnwnt lleating (DCil; / aanw seabove 2.2r 07 CLT toenbustion prw eines.
- 6. Vennel thrust furse (Rmkro ume as above < 104J CIT LATliCONTAINMEfff fAILUkr.$
7 Overpre.auriauan due to l it,3 H,45150,s t,$ H,4 2,161. 03 CLT nonwndensibic s (NC) f,6f,l F.35 h0,2 F,6 H 411,5 F
- 8. Overpressurinten due to steem 611, dit, 2il 9.160 07 CIT .p genetatmn (50)
- 9. llawmat meh through (hM) sana as NC IEM (irwtuded in 7 CIT merpressure eleve) 10, Lats sombustion some as NC 2.2 rAN CIT overrnseure (ireluded in 7 above)
T(Y1 AL: =========> 2 6711-05(60.5 % CMn - a t.
' Plant Damage State see Se ction 4 3 8 Induced Steam Generator Tube Rupture, 4-118 i - , ee- - + .'-+www%y-1.--,m-.~..em-Eme.,% ,w- %. uY p y,-,-,w.,ww,-Ew, wE,,,m ., a'c,o,',-fy,. ,_,,y wm,,,,v.-yrw.-. .y - ,,-,-t--.p-,-,v.py. m ,, ,y y---,,., ,,wwme ,w y9,..-,-7,,. ,,-..y
i l TaNe 4.4-2: Summary of Various Omtainment Streng:hs 1 t l Ar:alysis Failure Pw Danmant Type of Containment Free i D. Design Plant Pressure Fadute Volume Prosure P 47 psig ' NUREG-1150 !00180 png 2 3-3 3 I.eak!rv;ture m cylmder wall or f Large Dry: concrete cytmder 2 86E6 ft' 141 ft Zhn bawmsUna'l tecnccten w/steci Imer (pre-stressed) 1 17 IDCOR 101 147 psig 1 H -7 cndon stram 45 psig NUREG-1150 05155 pig 2.1-3 5 Leak!rupu-e near Am%:1 Suhat ~: .-ers:: concree with I .3 E6 ft' 126 ft Surry m crsede steel bne. ..einforted) 47 psig IDCOR 10 I 126 pur 27 11 -p reAar yrkUcylaier sheti Large Dry: ctmcrete cytmder with 2 61E6 C 135 ft. ndsn Point near sprmg hne steel liner (reinfhrted) f4 pstg Seabrmk Rok 21I pstg 35 Fadwater greraten Larre Dry: concette eyhnder weh 2.7eE6 f 140 ft. Seabrook Mansger,ent Stud,- (wet seg ) steel liner (pre-stressed)
' 1955 [ < 100 psig 3[
l (dry seg ) 59 Psig Orence PRA 1623 psig 2.7 23 x design pressure. rupure Ocrynec Unit 3 Large Dry. concrete cylinder with I .91 E6 "' I16 b (NSAC-to) of ystress tenims and hner steel liner (pre-stressed) fadu-t 4-I19
f Table 4.4-3: Reference Plant Global Strains at Discontinuities WALL IIASEMAT SPRINGLIN!! EQUIPMENT H ATCil hiAIN $ TEAM LINE i . Pressure Global Pressure Global Pressure Global Pressure Global (psig) Strain (psig) Strain (psig) Strain (psig) Strain , 10 1.2FE OG 5 1.20E-06 5 9.00E-07 5 9.00E 07 ; 58 1.84 E-04 20 1.32 E-05 10 3.22E45 10 3.22E-05 60 3.40E44 25 2.04E 05 15 4.64E-05 15 4.64 E-05 62 4.03E-04 30 2.03 E-05 20 6.06E 20 6.06E 05 64 4.45 E-04 32 2.25E-05 25 1.15E44 25 1.15E-04 66 4.46E 04 34 1.69E.J5 30 4.07E44 30 4.07E-04 68 4.74 E-04 36 2.68E-05 32 4.45 E-04 32 4.45 E-04 70 4.95E& 38 2.7111-05 34 4.71 E-04 34 4.7iHet i 72 5.21 E-04 40 3.10E-05 36 4.99E-04 36 4.99E-04 74 5.58E-04 43 3.53E-05 38 5.25E-04 38 5.25H-04 76 5.80E 04 47 4.09E-05 40 5.50E-04 40 5.50E-04 78 6.15H-0 6 50 4.96E-05 43 5.90*i-04 43 5.90E-04 80 6.50E-04 54 6.33 E-05 47 6.43E44 47 6.43 E-04 82 7.00E-04 58 7.90E-05 50 6.83 E-04 50 6.83E-04 84 7.74 E-04 60 1.1IE-04 54 7.36E-04 54 7.36E-04 l 86 8.15E-04 62 2.55E-04 58 7.89E-04 58 7.89E-04 88 8.50E-04 64 4,4811-04 60 8.16E-04 60 8.16E-04 90 8.65E-04 66 5.15E-04 62 8.43E-04 62 8.43 E-04 e 92 8.79E 34 68 5.60E-04 64 8.7 I E-04 64 8.71 E-04 , 94 9.31 E-04 70 6.61 E-04 66 9.05 E-04 66 9.05E44 96 9.41E44 72 7.30E-04 68 9.38E-04 68 9.38 E-04 98 9.66E44 74 7.91 E-04 70 9.72 E-04 70 9.72 E-04 100 1.09E 03 76 8.54 E-04 72 1.01E-03 72 1.01E43 102 1.16E-03 78 8.97E 04 74 1.04E-03 74 1.04E-03 1.07E-03 104 1.18E-03 80 9.49E-04 76 .l .07 E-03 76 106 1.17E 03 82 9.97E-04 78 1. l? E-03 78 1.12E-03 108 1.20E43 84 1,05E-03 80 1.15E-03 80 1.15E-03 110 1.25E-03 86 1.10E-03 82 1.19E-03 82 1.19E-03 110.5 1.27E-03 88 1.15E-03 84 1.24E-03 84 1.24E 03 111 1.29E-03 90 1.20E-03 86 1.26E-03 86 1.26E-03 111.5 1.31E-03 92 1.25E-03 88 1,29E 88 1.29E-03 112 1.33E-03 94 1.30E-03 90 1.33E-03 90- 1.33E-03 : 114 1.57E-03 96- 1.35E 92 1.36E-03 92 1.36E-03 114.5 1.70E-03 98 1.41E 03 94 1.40E-03 94 1.40E-03
- 115 1.80E-03 100 1.47E-03 96 1.44E-03 96 1.44 E-03 115.5 1.90E-03 102- 1.52E-03 98 1.47E-03 98 1.47E 116 - 2.01 E-03 104 1.58E-03 100 1.51 E-03 100 1.51E 116.5 2.08E-03 106 1.64E-03 102 1.54E-03 102 1.54E-03 4 120 m4 -em- - w-a----u-m-+ew w-a-eeri..---,o.s-wow-ww-g-%---r --e-m+F-m -'-a'e a-* --'+Pe.m.e--*w g-we- wp q e m %, r3- p9 .,,yy 9.,m*p9.i,.qe+ywrpr w w w mum **y-w name ey* mgy wtwit-*wnwryv='3Tyvw='ry,- yv yy g -- iv- 9-l mew 9 -t w ww y 4 y
)
l Table 4.4 3: Reference I'lant Global Strains at Discontinuities (continued) t i WAl.l. llASliht AT SPRINGLIN!! EQUIPhil!NT ll ATCil AlAIN STEAh 1.lNE 3 Pressure Global Pressure Global Pressure Global Pressure Global (psig) Strain (psig) Strain (psig) Strain (psig) Strain 117 2.16E-03 108 1.741!4)3 104 1.58E4)3 104 1.58E 03 i 117.5 2.24E-03 110 1.89E-03 106 1.63 E4)3 106 1.63E-03 118 2,33E/ 2 2.181! 03 108 1,8511-03 108 1.851!413 : 119 3.00E4 l! 3 ) 2 . " 11- 0 3 l10 2.28E-03 110 2.28E-03 120 3,71E4
- j its )?~ 93 112 3.33E-03 112 3.33E-03 121 4.90E41) ( tin 4 3 o i:s J)3 112.5 3.90E41 112.5 3.90E-03 122 6.50E 03 12 o 4.0?E-03 113 4.50E-03 113- 4.50E-03 124 8.50li 03 125 7.00E-03 113.5 5.10li-03 113.5 5.10l!-03 126 9.53E-03 130 1.50E-02 114 5.78E-03 11/. 5,78E-03 l 130 1.36E-02 13$ 3.00E4)2 114,5 6.98H413 iI4.5 6.98E-03 135 2,79E-02 140 5.00E-02 .I5 8.17E4)3 115 8.17E-03 I40 4.10E-02 150 1.00E-0 l II5.5 9.381?-03 115,5 9.38 E-03 150 8.63E4)2 160 2.00E 116 1.06E 116 1.06E-02 160 2.32E-01 116.5 1.20E-02 116.5 1.20E-02 117 1.35E-02 117 1,35 E-02 i; /.5 1.46E4)2 117.5 1.46E-02 118 1.64 E-02 118 1.64E-02 118.5 1.81 E4)2 118.5 1.81E4)2 l 119 2.00E4)2 119 2.00E-02
' i19.5 2.17E-02 119.5 2,17E-02 120 2.35E 02 120 2,35E4)2 122- 3,19E-02 122 3.19E4)2 - 124 4.02E-02 124 4.02 E-02 126 4.92 E-02 126 4.92E-02 128 5.79E-02 128 5.79E-02 l 1 4-121 ~ _ _ _ _ _ _ . _ . _ . . _ _ _ _ _ _ _ _ _ . _ _ _ _ ._;__.s_. ._, . - . . . _ _ _ . _ . _ . . _ . . , - , _ , , _ , _ . . _ . , _
O Table 4.4-4: Compiernentary Cumulative Probability Distribution for CPSliS ContMnment Failure i Picaures liain! on Liner Tear at hiedium Penetrations at 114 psig Assumni as the hiean and a Normal Distribution with a 7% Coefficient of Variation PRODAllli,lTY Oli 1.OAD ; SURVIVING LOAD _ _ (PSIA) ,
,9986 105 t .9772 113 . ,8410 121 .5000 129 .1590 137 , .0228 145 .0014 153 7
o t i I- _ I l r l 4 4-122 t,~-,v-n--,v---,.,--a,-.,,.,,, n ,-a m -,---w,,.-,ww,,.rvn-,,-v.-- v , - , , . - .-n,,, n-- -- , , , , , , , - - ,---e.....----,--,- .-- ~r- . - - -,v--a',.--w -- ,
Figure 4.41: Unlaxial Strains vs Nt r; nllie41 Containment Pressure (Example Plant) 9 n
! \
i l - ,
'< j i 6 f.;
i c2 .# es
- . . - . . - --,--_----~p.. - - . . - - . - , - - - - -
1 is \
!
- e
)
I l l1\ _.. 4 e-- . . . - . . . _ _- _ { Jg _- . _- i y m .. l \ n 2
. \ ' :P~, J l : \ c t se 4 : o g b % \ 'I \
J ce \
- ,' - p.
i.
- _3 7 \, 1 1 = '3 .\ =
A l
. e j i .___....--,._..9._.-- _,
_._.....-..._.y____..__
.. t a
b _-_,. _._ _ .p _ . k -. . _ q, . . . - . ._ m
- v. i.
=
N N R ? R R E
- uiais aussea immun 4 123
, -s- r m s- ., a.*
i n
- " ya g h gu" x5 e% $ Qh. g: 253 g=r.sa _ . $- !i 5 '
4
- D Ire , n 5- I j- U e 3 s s e yt p Ng '
u id "4
/
t N r o _ er t o f ee R c \ 1 c or nm &\ n P c Uo ~ 6 e
- D - n C t
- ny n 2 7'
/ 6 d e c ce m 1 4 i, I' -
r n r e oi a _ f n nn t u ie o - R C - s
/
3 t
.of g3 a M
r
- e I n . =
U 4_
/1 t
n A e + \
\
k m r e . n @f/. d D. ia t f,/ in f
- ~
0 C o n 3efl C r y ~ O r - 8 S ee r e 5' 0 E nm n 3 8 S Uo U 1 P " D l E C 2 " B r
/ o 1 / o - 3 l F * /- h -
e t m . l o s D e t n lo U W g i n r p S 4
- I
, e a 1 ! I ?; :l
i
._ ;_ g , ...7...__ , ,M-b E
5 ,i; , i 5';- -
, e i-1 i numF , ; r !!!l . .ii m'. .-
a 7
- ! l 'I g jir: I
!me 1 ; ;. l r- .
s~, , i' f'3] iite lI s
/
t
}
i ;. .!-ie * !! i i 4;; fi, t 9t e i 1 t e.;*!f i. I ,ti'*
- e. i s ! i i l . !. 1 f
l'ii s it
.i.
- i ,t i I
'J e : 4 a s g -- 18 , /- 4 m;f g-
- 1. --o w; .
5 ..
! 4
- u,1n n
. i ~
w:#l 1 i;* . - i
. !.n.n l. o .li ; ; .s . i, t. , , ,,n, . i ,
l' I < t al ! [. .
+
t
*4.I.. ., g' 3 . . .i . m.
i . ' . ',,,' ., . , a 1 ..i ,,imw. - t-
.. .- . t 1 . it i i n eif *i t, ! j f N .d '- - i s-1 !
h -
- f i c % i 'li
!: 1 'I pe' E ; i j i l l . . i . : , . , .n # / 1 i t i t i
i 12 . t 1 i _ m !
~ .r j:s !!n ;+;
l - j ! ;
., 1 i ,4 ,+l m,i . 1, __, l, I,. + - = ,- ~ ,I a, :ui i,mu.. ! . .i L.
_--i-,
, n.mL-- .-.-,-.
F,gt se 4 41 1: B an Jety 1 ar 'w T3&a! Sr"aC ltne Geume"y Rr*atmed G****e i egwe 4 4-12 Strem Gaemerati.= Fatue f-w Tym ni Sgvmst we Cetry shmiswed Getrete Gwitamments j Gesamments 7J % 99 l
. m ,., . .
_ _7 __ g : .. _
- r. ,,, , I ;~ u, i e I -
.l , run e u.i .. 4, i
4 1
.,,m,. i ..
1
- - 2 .; .\i u, ...t t , . i t . ... ,e .s ! j l i . a .
I, i I. i . I t i
.I I;
N y,[, t f { *2 .I. ~ g, '
, w ' f .' l F !
I
- s. a n i 4 : : :.
'l% . - ...'a' . a -p: W Ms ; ; : 7 .f j , i! .
jj g j i l 1 1 4 !, .. ; i' 9 e i . .
, I ,i i-i
- g : 1' i t
-l: { l 3 ... , , , ,i n. ; ; ; , 5 i as i . .
I ( t iig ml 4 . i { i, {-l L . __- _._i ei) f f'
'i *
- I 6'W
,i ,i ; ,i l i l I 1: - . . ...e.. !.. . a , i.. .
l.. - , _ ., . .i..2.- . - - i . . u. ,
- ~_ a _.-. h.i. - - -
F*gwre 4 4-3 3: Bienaakty Factor fw Ty;.ecel Eq= gwnent and Pen ==mel H. he. 1:gere 4 4-3.4. Saram Gwitentemum Fanw for Ty;*.; E9 er'neas ams Pe-mel Reinforced Gecrete Getamenets Hakbes Resniorust Cetrete Cos=tainments l f
_ _ . . . _ _ . _ _ _ _ . _ ..m_. _ . _ _ _ __ _ _ . ._ . . _ . _m._ ___ l l
. 1 t
- 4 e .
.. e - 3 7-.
- a. -
[ _5 : 7_..-........_..--__ i i . ,. . . - - - na
)
i i e I I
, m_ - .. g ! 1 J ; 1 , + i I'- t ...g.__.._._,... .Q p ..;..~,.--
I
. p 3 .. .. .l g , , , , ,_., __ ,g j ....._..,.~..m.
g
. E_
- j. - _ . _ _ . . .
,, 3g . a ._ g } g-t, .g W _ E. ; - . 'l w
W
,g __a 1 a l .i .- e_ . , ) . . . . . . . ~ _ . __ .. _ ...~ = x y g, , _ ,
a
, .i l ,
g
. I l ! yv l i.
_ . .. x * ~....._._ ., res. cg t-~ 1
. . . . . .. . _ . . . . . . _ . . . . . . . . . ~ . .,. .
i ea c'
...7...... g -... E, . . . . . . _ . . . . _ . . . ._ .. . g .,v ,
p-. V s Ej 8 y 7~---- g
, Ag eb . . _ , . 4 -..m. .._ { r , dE i
e. i i i 6( gU.
; g
- l.
i
. . . _ , ~ - _ .. ~... .._ ._ ,
i !
. x }
i ----.+ l l
*I \ \ u ? w$
i M le t -
- - - - it'smvauvwar.u.oa m.u -in escu esterw2mga Nrvta +
i x I
..l - :1 %*
D e bu I C i a
,t t
9 . , , , _~ 4
- e. -< 1 g
t l . --
*. A*' j .g , 3 .?'+***, i
- I._ _I I t l' l
) - ! 4 l . . __'s t
i
, 4 e .! .. .,I , y i
m.._5. j t *t g g
.'...g...-
i
- s g s -
_ . s. __.g t g g v 3 s L >. i 3 E. Ej , 1 M m - ~ is s S
. n g L;
i j't w l N ' #
.. g a E l, 4 / ,l$ ," ,
_I 3 sg. p ! 3
- p kI 2 W ]~'~~.C2 ... . . .a,.
g .? w g 1,gg-*/
, g __
e1 a 4 p ! - __ a. __.. .
% - .- n. . = U a - *- . % - - - 2
l : p-.
,q .d :
- d. g j ._ . - --
N g 82 i
, _- i_ L vt m
l 1 s- ! "5 m.y
'\ l - '{ - e 'O r - . - - .
t I *T
*b l'; -+ \+- !
- g
\: Ej ; a ! ye V LW \ ;a E .2 ;
r
\.l"*. zc
- s. .a a
- e
. .r a e .i .s'. : + *
{g g* gg pg 4-126
tL9 .!.I 1I! !:!jt1b - . .tiI' ,. il'i!ii i!;Il1 h t ,! b , mg5a *.aL aE=Eee" % 7eE mWE5 oe37 E O g._%W y=:c n m=u 3Q;E=h -(r-
] .
8 2 -
+
v r 7 j;i:id }ee
,;;i l I,; }t ;' ! ii e!
t 2 e s c s D a e e M de xa - r e r C
- c.
e W H t o
- /:I1l!;1 ' lt' l1 ti ,
6
//i 2 !l/ ! / .' wa d P .<' n I
P
--/ n e c s
l s. ,/, m ci n 5 2 P
/
w!' l,l, s iI d a / in l w
- a
/ , m r - o '* N / ;/
4/ i'// !. l/j '
/ / ) /-8 4 2 =/ / 6 h / /
9e f c / / ta H /_ /
-/ /
6q
/' ,
s / A,
/ ,! f E / ij/ - /
I- 4 ? , 2 3 e n U i f 1/+ ,8, 7-d m e
/,,g' . a te s
a f/ 1 I l
, 2 2
0 0 0 0 0 0 5 4 3 2 1
" E_23 G ea~d E TR2:c."
pUa
^ '
Figure 4.4 5: Seal Life as a Function of Time at Temperature (Ref. 26) t 1M 11 7, 12 - GENf R AL TEMPfR ATURE G .
'-LtMITS OF G AE4C 7 &lucura aW8BER ELASTOMER COMPOUND 5 . 700 .1LUCRotLASTOME R g ,ETHYLING PROPvLINE tr htOPRINE !
i;h\ NW E
. WN Ilo 3 NtTRtLE (Hl0H TEMPf R ATURE TYPfl NffRtLE ILOW TEMPE R ATURE TYPfl l l l l f I !
g 01 oa to so to u 30 too ime EXPOSURt TIUt (Hovre 5 4-128
- - .- =-
4.5 fontainment Event Tress i The CPSES IPE treats containment bypass and isolation failure cases separately from cases in which the t containment is available. Ilypass and isolation failures are discussed in Section 4.5.1, and the intact 1 containment cases and the Containment Event Trees (CETs) are discussed in Section 4.5.2. 4.5.1 Containment Ihrass_ and Isolation. Failures [ The Level I analyses identify sequences and determine probabilities of occurrence of containment bypasses (Refs. 4,5) and containment isolation failures (Ref. 6) for sequences leading to core damage. MAAP analyses are used to determine the release fractions and timing for the asso lated release categories (Ref 7). 4.5.1.1 Centainment livpass Two types of bypass situations are identified: (1) Steam Generator Tube Ruptures, both induced (Ref.
- 7) (ISGTR,1.7E-7) and as initiators (Ref. 4) (SGTR, 3.5 E-6) which are binned into PDS 2CB and, (2) interfacing systems 1.OCAs (Ref. 5) (ISLOCAs or V Sequences) which are binned into PDS ICB. r SGTR & ISGTR i The SGTR core damage frequency (3.5 E-6) was determined in the Level I analysis (Ref. 4). The ISGTR frequency was determined (Refs. 7,8) from the fraction of the non-depressurized high pressure PDS frequencies for which the SG tubes fait prior to the hot leg or the pressurizer surge line. MAAP analyses (Ref 7) were used to establish the associated release category release fractions and timing. The path for the releases is out the SG safeties or the atmospheric relief valves as described in Section 4.1, The SGTR accident progression is discussel in Section 4.6. The SGTR event tree is shown in Figure 3.1.218 and discussed in Section 3.1.2 (Volume 1 of this submittal).
4-129 " e er - .m - e* ny T-- asy t - -%ay aaw. 4y p ww aa-, ww ,eyygwwew y,w- , -q----v- 9w--ye qvg me,ey- a- wyee, ,v
ISLUCA De ISLOCA or V-sequence (PDS ICB) core melt frequency (1.5E-7) was calculated (Ref. 5) based on NUREG/CR 5102 (Ref. 8). MAAP analyses (Ref. 7) were used to establish the associated release category, release fractions and timing. The path for the releases is through the Safeguards Building as described in Reference 5. Approxirnately 70% (8.7E-08) of the total ISLOCA frequency results in releases via pump seals, which would be submerged and scrubbed. Ilowever, credit is not taken for fission product scrubbing in that building since the total ISLOC A frequency is low. Four classes of ISLOCA events are defined for CPSES: In these scenarios, the leak through the pressure Chuii Small LOCA inside containment isolation val.es (PlV) is within the capacity of the relief valves and they relieve to a tank inside containment. Chus_li.; Small LOCA outside containment - In these scenarios, the relief valves discharge to a tank outside containment. ChtsLiit Overpressurization/ LOC A inside containment -In these scenarios, either the leak through the PlVs is beyond the capacity of the relief valves or the relief valves fail to open. This results in overpressurization of the low pressure piping and a potential break in the low pressure system. Class iv: Overpressurization/LOCA outside containment - In these scenarios, either the leak through the PlVs is beyond the capacity of the relief valves or the iclief valves fail to open. This results in overpressurization of the low pressure piping and a potential break in the low pressure system. The break occurs outside the containment. The interfacing lines where ISLOCA. in any of the above categories can occur are:
- Letdown Line
- Excess Letdown Line e RH Suction Line 4-130
- Low Pressure injection to the Cold Legs
- Ixw Pressure injection to the llot Legs
- Intermediate Pressure injection to Cold Legs
- Intermediate Pressure Injection to Hot Legs
- Accumulator injection to the Cold Legs Each of these interfacing system lines was reviewed for its functional requirements, for potential small LOCA and overpressurization scenarios, and for plant parameters and/or alarms that provide indication of the event. Small LOCAs result from pressure isolation valves (PlV) leaking and lifting of the relief valves, whereas, overpressurization results from PlV rupture (or failing open) or Piv gross leakage and fal!ure of the relief valve to lift.
The determination of the initia'mr frequencies of ISLOCAs through the various pathways identified earlier was accomplished by adapting the generic system failure models found in NUREG/CR 5102, interfacing Systems LOCA: Pressurized Water Reactors", to the specific valve arrangements found at CPSES. At CPSES, as with the majority of PWRs, the low pressure injection (LPI) lines and the accumulator outlet l',aes have a common inlet header to the RCS. This inlet header is also shared with the intermediate pressure injection (IPI) lines. Therefore, based upon the result of NUREGICR 5102, the analysis of ISLOCAs involving common inlet pathways has taken into account the proneness of the accumulator outlet check valve to fall to rescat on dem:nd due to the lack of a differe- tessure across its disc. Each of the eight interfacing lines was examined for two conditions: a) Small LOCA and b) , Overpressusization, The _Small LOCA represents the case ef the low pressure piping of the interfacing _ . system caching pressures up to the relief valve setpoint leading to a Small LOCA, including pipe ruptures and failure of the RH pump seals. The Overpressurization represents the case of pressures exceeding the relief valve capability, with pressure potentially reaching those seen on the RCS. The initiating event frequency, the affected ESF equipment and the event classi0 cations are provided in Table 4.5-1 and are applicable to the Small, Medium and Large break LOCA scenarios defined in Table 4.5-2. That information was then applied to the representative LOCA sequence to ascertain the potential for core damage and its probability. The result of that step is shown in Table 4.5-3. - i 4-131
. . _ - _ . . _ . . _ _ _ _ . .._._. _ ,. _ _ - _ . . . _ _ _ _ _ . _ _ . _ . _ _ _ __. . . . ~ :
i i De initiating events of Table 4.5-2 were evaluated to detern,ine if the initiating event would lead to core darnage directly (no funher quantification required) or if additional equipment failures would be necessary l
+
to lead to core damage (additional quantification requirtx1). It was determined that event 5 would lead directly to core damage due to failure of both trains of Ril during a Large break LOCA scenario (aiso events I,4,7 and 8 have suf0ciently small initiating frequencies that further evaluation was deemed unnecessary and core melt was also assumed). Thus, any possible recovery actions were not credited for l these cases. ; The remaining initiating events (2, 3,. 6,9 and 10) were quantined using the methodology and models b defined in the front end analysis (Volume I of the submitthl), with equipment affected by the ISLOCA redefined as failed. These results include some feasible mitigation actions for recovering failed equipment or systems, but do not take credit for existing procedures for mitigating the ISLOCA initiating event prior -; to core uncovery (Ref. 5). 1 4.5.1.2 Containment isolationfallatn A fault tree model was developed for the Containment isolation System (Ref. 6). This fault tree model includes: (1) the pathways that could significantly contribute to contalament isolation failure, 9) the signals required to automatically isolate the penetrations, (3) the potential for generating the signals for each initiating event, (4) the examination of the testing and maintenance procedures aad (5) the quantification of each containment isolation failure mode (including comt vn-cause failure). The fault trer model for the containmen' isolation system was quantified independent of the accident sequence , analysis task. Then the cutsets were combined to determine for what fraction of each sequence the [ containment failed to isolate. As a practical matter, containment isolation failures were found to be independent of the support systems whose failares were present in the core melt sequences. As a result, e L L - the failure probability of the containment to isolate in those core melt sequences is low. 1 r The Containment Isolation System (CZ) consists of sleeves, pipes, valves, instrumentation and controls for piping penetrations. The system isolation valves, piping design and location ensure containment { integrity is maintained for any postulated single failure. Double isolation barriers ensure that no single failure of any active or passive component renders the CZ either partially or wholly inoperable. All system paths penetrating the containment wall were evaluated on the basis of their function at the time 4 132 ; 4 m rnwv.s.s w' w.w a y u,+y ye s--mb*-#pegrv,, V-- y +-- gn,e-- -*iy-- na.gu_a7 9 -y y
h 4 ! of the accident and classified as either essential or non-essential. Essential paths are those required to [ mitigate an accident or whose unavailability could increase the rnagnitude of an accident and, as such, , may contain remote manual isolation valves. Non-essential systems are automatically isolated by the containment isolation signal or normally closed isolation valves. e r instrumentation and controls for the CZ includes the control and indicating devices for the power operated f I valves. The containment isolation systern logic is part of the ESFAS logic and includes the following i subsystems: ,
- Steam Line Isolation
- Maln Feedwater Line Isolation
- Containment isolation Phase A ,
- Containment isolation Phase D
- Containment Ventilation Isolation Each CZ power operated valve, with 1; exception of those that serve or support ESF systems, are provided with two methods of actuation: an autornatic closure from the ESFAS and a remote-manual control switch. Each power operated valve in the CZ is also provided with a position indicating light.
Provisions have been made St each containment penetration to facilitate leak rate testing (Types A, B, and C) in accordance with 10CFR Part $0. Appendix J. The fault tree model (Ref,6) deve!oped for the containment isolation system was based on the criterion that a penetration required evaluatian if it did not fall within one of the following categories:
- Locked closed manual valves
- Valves used for ESF systems
* - Valves that are closed Prior to an accident and remain closed
- liigh pressure / Closed systems Spare and maintenance / test penetrations are normally closed and/or isolated and .were not considered due 1
to their small contribution to CZ failure. 4 133
- _ _ . - . _ - - . _ - _ _ _ _ . . _ _ _ . _ . _ . _ . _ , _ _ . _ - _ _ _ _ _ _ , _ . _ _ - _ _ _ , _ . . - , . ~ . ..
1 The remaming penetrations were quantined, independently of the accident sequence quantification effort, for each potential failuia mode, including common cause failures. A subset .:omparison of the CZ 4 quanti 0 cation cutsets with each of the accident sequence quantincation cutsets was completed to determine for what fraction of each sequence the coritainment failed to isolate. l l 4.5.2 Cgainment Not Bvpassed and_Seccessfully isolated. Containment Event Trees (CETs) A CET was developed for each PDS for which the containment is not bypassed and successfully isolated dutlag core melt. The CETS used in the CPSES IPE were based upon the approach developed by the Electric Power Hesearch Institute (EPRI) (Ref.1). Specine characteristics of CPSES were incorporated 4 through adaptation of the EpRl trees, dennition of the 11asic Event (llE) probabilities (Section 4.6) and !
- in the implementation of the CET quantification process (Section 4.6). IlE probabilities were determined by PDS k0nitions, by systems performance for core damage sequences determined from the Level I ;
4 analysis (Ref. 4), and by a substantial number of plant specific MAAP analyses (Ref. 2). The CET structure for a PDS for which the containment is initially intact is illustrated in Figure 4.51. A single CET structure is shown since the top events are the same far all these PDS. The difference between CETs for different PDS lies in the mechanisms that could challenge containment integrity and potential recovery measures associated with the RCS condition. The CETs were further developed for , each PDS using logic trees to break down the top events into phenomenological, systems, or operator human response issues. Thus, a limited number of CET event nodes were denned that convey the full spectrum of the accident progression outcomes in a single event tree. The basic events (BE) affecting the CET top event nodes are: (1) decomposed severe accident phenomena, (2) systems availability and [ performance and (3) recovery measures. The logical relationship of these 11E was modeled by a fault tree framework. The fault trees are also referred to as logic trees (LT). Figure 4.51 shows the CET top events. . The rationale and basis for each of these events is described in this section. , 4 1 4 134
, , - . . . . -- . _ . - . _ . . . . . . - . - - . - - . . , _ - - _ . - - . _ , .-..-,_w.,-.. . , - ,
1 i I J Top Event DP: RCS Not Depressurired 11efore Vessel Breath l The question asked in this top event node is related to depressurization of the RCS prior to vessel breach. Success m this branch implies that RCS pressure is low (less than 200 psla) at the time of vessel failure, reduced either through the capability of the operator to depressurire the reactor or through a phenomenological condition that could induce RCS depressurization, or because the initiating event is a l break large enough (2 inches, Ref. 2) to lead to a low pressure condition at vessel failure. This event node is relevant for high pressure PDS to indicate a potential mitigating condition during core melt prior to vessel breach. For accident sequences leading to core melt with the RCS at high ptessure, deptssurization means that the hign RCS pressure that could exacerbate containment challenges at vessel breach (such as direct c<mtainment heating) is removed. The DP event node impacts subsequent CET event nodes related to in vessel recovery and early containment challenge, Figure 4.5 2 displays the logic tree developed for this event node. The issues considered are:
- Inkial RCS state as determined by initiating event (e.g., LOCA vs. Transients).
- Active operator action to depressurize the RCS before vemt breach.
- Severe accident induced LOCA due to high RCS piping temperature.
Ieplvent REC: Coolant Not Recovered in-Vessel Hefore lireach Recovery in the back end analysis, i.e., after core damage, is not credited in the CPSES IPE with the exception of Station lilackout PDS. The question asked in this top event node is related to recovery of coolant injection and electric power after core degradadon, prior to vessel breach. The recovery is only considered possible if RCS depressurization wa successful, i.e., the question is only asked along the success path for event DP as can be seen in Figure 4.51. The logic tree for event REC is illustraw ,a Figure 4.5 3 The issues ,. considered include the following:
- Availability of coolant injection upon depressurization,
- Recovery of electric power.
l 4-135 ,
-.--- _._~ - - . -
Tep 1:vsrn VF: Vr tl Failure Qmits ne question raim this event node addresses the actual arrest of core degradation within the vessel, which prevents lower vessel head thermal attack. Yessel failure is prevented only if coolant make up ' I recovery was successful in the previous event node. Therefore, this event is also not relevant in the i CPSES IPE for PDS other than time corresponding to Station Blackout. Success at the Vessel Failure event requires that core cooling be recovered prior to core blocking (h1A AP I malel) or relocation of molten debris to the lower plenurn and thermal attack of vessel head (h1ARCil model). Therefore, the prirnary consideration for successful in vessel recovery is the time available from incipient core degradation to the point of non recovery. Core recovery in vessel is modeled as unsuccessful once molten debris starts to relocate to the lower plenum and thermal attack of the vessel penetration weldments occur. h1AAP calculations performed for the Zion plant indicate that this could , occur when approximately 25% to 50% of the core is predicted to be molten. The NRC Source Term Code package (STCP) calculations in llhil 2104, NUREG/CR-4624, and NUREG/CR-4587 show higher melt tractions (a user-specified criterion) are required before core slumping occurs. in either case, the time to vessel head attack can occur within 60 minutes (or longer) following core uncovery depending upon the sequence, based on published h1 ARCH or h1AAP calculations. More mechanistic hiELPROG 2 D calculations (discussed in NUREG 1265) indicate that progressive relocation of lower melting-point materials into the lower plenum tend to insulate the lower head penetration weldments from the higher-melting point core materials. This might support the position that the time to vessel failure could extend even longer. 1
. Dis event physically signifies that the core degra ati on processleading to vessel failure is successfully terminated, thus arresting core melt and precluding significant fission product release to the environment.
Figure 4.5 4 illustrates the issues considered for this event node. For purposes of this assessment, core . debris cooling and termination of core degradation prior to vessel head attack is a probabilistically weighted average in the plans and sequence specific time witjows of core uncovery to core melt and core melt to core slump. In the tree, the potential for terminating melt progression prior to vessel breach is either by debris cooling from within the vessel, or vessel oottom head cooling fre n outside the vessel, thus terminating head attack. The former depends on the success of the previous event node, REC and. phoomenological considerations of aoolable debris bed information, implemented by the weighted average approach mentioned above. He latter might be achieved by filling the cavity with water above 4-136 l
- - . , - , J,- - ., _-_-s..J--__.,-.,.-,,..-.m-~,,-,_ , , , , , , , , . - - . ,...,..___m_;,.--m, .m,,,. . - - . . . . , . - - - ,,,,mm._, . _ _ . ,
i l l the vessel tettom head, allowing heat transfer to be established through the vessel walls. For some vessel head (no instrument tube penetrations) and containment configurations, this option might be plausible. ; llowever, this external cooling was not credited for CPSES and the correspcmding HE probability was assigned the value of 0.0. This is felt to be a conservative assumption and some reviewers have suggested a value of 0.5. Itowever, a more detailed discussion of the rationale for probability assignments is available elsewhere (Ref. 8). Success of event VF is determined by the physical processes controlling core melting and material i relocation to the lower plenum. A good understanding is very important in order to determine l possibilities of arresting vessel head attack once significant core degradation has occurred. Exact rnodeling is not always possible; therefore, simplifying assumptions are made to approximate the anal %c treatment of the physical process involved. There is considerable uncertainty in determining the formation of a coolable configuration once significant core geometry deformation and melting has occurred. There are recognized limitations in the existing analytical models (MAAP and MARCll) in predicting coolable debris bed formation within the vessel. The limiting factor, therefore, is the time available between core vulnerability (i.e., conservatively defined as core uncovery, the end state of a core damage sequence) and core melting (i.e., peak temperatures exceeding core material eutectic temperature of 4130*F). .
. MA AP calculations conducted by IDCOR indicate that the time period between core uncovery and onset of core melting is typically one half hour to several hours (TSR, Tnx 23 Technical Reports) depending on the accident sequence. MAAP calculations also indicate that core degradation cannot be arrested once fuel melting has started and core reflood (as determined by the success branch of event REC) would not.
preclude support plate failure and vessel breach, MARCH (STC19 calculations, on the other hand, indicate that vessel breach can be arrested provided that coolant injection is recovered before significant , core melting has occurred. Because of the phenomenological uncertainty associated with this event, the limiting factor used for the conditional probability of success is the time period between core uncovery L and core melting compared with the time period between care uncovery and core collapse. , l The logic tree for this top event is shown in Figure 4.5-4. Worthy of mention in this tree is event CAV-DRY where SECCS-INJ is ANDed with SNOSPRAYl. This accounts for the fact that the cavity will not be dry if the RWST water has previously been injected into the containment, regardless of whether sprays have operated or not. The CAV DRY event appears in other logic trees as well. 4 137
, ~ . + . . . . . . , - . . , . , , . . , . + _ - , - . - m. , - - - -., .- - +.- m ......,.mm .~%_. .. ,-.<.r,.m,ww.vwy,. w - . . - ,mm- ,vrm,a
-. ~ _ . - . - _ . - - . - - - - . - - - - . . . - -
Ton Even' CFE: Early Containment Failure Occurs ; His functional event was included in the CET to signify that the containtnent integrity is maintained during the early phases of core degradation and release of fission products from the fuel up to vessel breach. De fission products released from the fuel ate contained within the primary containment system so that natural removal mechanisms can effectively act to deplete airborne concentrations in the containment, v Failure at event CFE is defined as loss of containment integrity early in the accident sequence. Several failure mechanisms were postulated for this top event node as shown in the logic tree of Figurer 4.5.$ , and 4.5-6. These include contair. ment challences resulting from:
- Pressure spikes occurring due to blowdown at reactor pressure vessel (RPV) failure with the RCS at high pressure.
- Fuel coolant interaction resulting in rapid steam generation within the vessel at core slump or in the reactor cavity at vessel breach.
- High pressure melt ejection loads such as combustion of hydrogen released prior to and at vessel breach and direct containment heating.
The failure mechanisms identified above that, individually or in combination, result in loss of cont.11nment integrity early in the accident were considered in the CFE logic trees. Uncertainties in containment loads at vessel breach arise from the non stochastic nature of some of these events (e.g., hydrogen burns), as well as a poor understanding of the phenomena governing others (e.g., direct containment heating).' 'Although more experimental and analytical information regarding direct-. containment heating has been generated, substantial uncertainties persist and the phenomenon continues to generate controversy. Pressure loads from high pressure melt ejection and fuel coolant interaction are i further discussed below, l Hich Pressure Melt Election (HPME) Imads: The potential for pressure rise as a result of the high pressure melt ejection of molten debris from the vessel to the containment atmosphere is considered in this issue. Direct Containment Heating (DCH) is used to define the phvsical process when the vessel falls 4-138 v: g., gg -ww-gr-gy--: y p- -rp-gw w a'i.ya,, p9mg99-@,-p .* w- r -wa gyvr- g,e .i g
- wrs'sa yy r, ry.-p-yy *-- v- ,m- asiav y g , q sepvi
- _ -_~_ _ __ -
1 4 at high pressure and large fraction of the molten core debris is dispersed into the containment as fine particles and a substantial portion of the core material sensible heat is directly transferred to the atmosphere. De containment pressure rise depends strongly on the reactor cavity geometry and the mass of rnaterial dispersed. De pressure rise can also be aumented by the hypergolic burn (a .orced and complete burn not limited by flammability considerations) because of the very high temperatures caused by direct heating of the containment atmosphere. De containtnent pressure rise accompanying direct containment heating depends on reactor cavity geometty, the mass of material dispersed by reactor vessel blowdown, and several other parameters described in the quantification (Section 4.6) and in IL.'erence 8. There are several parameters included in the logic tree to model the dependencies of the containment challenge resulting form DCll:
- Provided the reactor pressure prior to vessel breach is sufficiently high to transport molten material and hot gases to the upper regions of the containment, the pressure rise is probably insensitive to reactor pressure. The reactor pressure threshold below which DCil does not occur is still not clear-cut based on Sandia experiments of debris dispersal. In this assessn.ent, DCil is regarded as possible if the pressure is above 200 psi (NUREG 1150). MAAP calculations were i
performed to provide insights into the determination of the extent of debris dispersal. i
- The fraction et core melt ejected from the vessel at the time of vessel breach determines the amount of material that can participate in DCH. His is governed by the model used to represent core melting and, to some extent, the accident sequence definition.
- The size of the vessel failure, since it affects the blowdown time and debris dispersal.
- The availability of water in the reactor cavity at the time of vessel breach is considered, because it can influence DCil. The water can interrupt the pathway for debris dispersal following vessel breach as it is displaced only after a fraction of the debris is injected to the cavity, or the water can be co-dispersed, in which case the droplets can continue to quench the debris.
L l DCH resulting from the dispersal of molten core debris can induce hazards that challenge containment-- l in the short term. An additional simultaneous hypergolic burn of hydrogen can generate chemical energy ; 4 139- \ o 1 - v
that can increase the prer vre and temperature of the containment. If the amount of debris involved in this process is signincant, extremely high pressure and thermal loads can indeed fail the containment. In a llPME, combustion of sufucient hydrogen to generate a substantial pressure rise is subject to physical requirements regarding minimum hydrogen concentrations, oxygen availability, but not inerting gas concentrations, liydrogen concentrations in containment prior to vessel breach depend upon in-vessel [ core melt progression (primarily the fraction of the core Zircaloy oxidized before vessel breach), and the type Jf accident scenario being considered. liydrogen burning alone, i.e., not in combination with DCil, can for sc.me plants also induce over. pressure failure. While such burns are possible at CPSES, they lead to post burn, worst case containment pressures on the order of 80 psia if only Zircaloy oxidation generated hydrogen is considered at 100% clad reacted. Therefore, the probability that they will fail the CPSES containment is negilgible, since it has a design pressure (tested) of 65 psia and has a failure pressure of 129 psia as shown in Section 4.4 Ilowever, the possibility of these burns is included in basic events (BE) PRWCP-PULT and PRDCP-PULT (Figure 4.5-6) as a feature of the HpME challenge pressure probability distribution developed for Zion in NUREG/CR-4551, which is adjusted for each CPSES PDS as described in Section 4.6. That distribution includes the loads from blowdown, to burns alone, to DCil alone, to DCil in combination with burns, and the probability of occurrence. End. Coolant Interaction (FCIM The consequences associated with the rapid transfer of thermal energy from fuel-coolant interaction in the vessel or ex-vessel can be risk-significant as it poses a plausible threat to containment integrity. There is a possibility that in certain accident sequences, molten material can tiow into a pool of water in vessel (reacter vessel lower plenum) or ex vessel (reactor cavity) leading to steam explosion failing the containment, it is also noted that ex vessel steam explosion may result not , only in an impulse load, but also in a quasi static pressure load on the containment structures (NUREG-1150). I j in vessel steam explosion was Orst acessed in the Reactor Safety Study. WASil 1400 Since then, numerous simulant tests and related anc.ytical s.videls have been developed, flowever, some uncertainty still exists regarding this issue, much of which is id wd to the applicability of the small and intermediate-scale tests to reactor scales and geometries, in this assessment, the likelihood of in-vessel steam 4 140
. _ - - - _ _, -_ - _ _ . _ . . __ __ _ _ _ _ _ _ . _ _ _ _ _ , _ _ _ __ .~_ ,
1 explosion is dependent on the RCS pressure. The draft NUREG ll50 assessments also indicate a low likelihood of containment failure due to stearn explosions relative to other failure modes. Ex vessel steam explosions for the Zion and Surr, containments were assessed as unlikely to threaten containment integrity (NUREG 1150). The cavity walls are heavily reinforced concrete to support the reactor core and the primary shield wall and even more so for CPSES. De cavity con 0guration (if filled with water), not unlike that of Zion or Surry, would not allow water to represent a vulnerability to containment structures from impulse loads generated during ex vessel steam explosion. Although ; potential ex vessel interactions between core debris and water are of concern only for accidents scenarios during which the water covers the reactor ca/!ty door prior to vessel breach, that is always the case at l CPSES. Nevertheless, the effect of pressure spikes resulting from high pressure blowdown and rapH steam production, although subject to some uncertainty, are bound by conscrvative assumptions regarding the probability that such events would occur ani fait containment. The logic trees (Figures 4.5 5 and 4.54) developed for this event consider the time-dependenev of the various phenomenological events that contribute to early containment challenges during a severe accident, in the manner discussed above. Note that there are two logic trees, CFEl (Figure 4.7 5) and CFE2 (Figure 4.54). The difference is that CFEl is along the high pressure path, and therefore includes llPME phenomena, while CFE2 is along the low pressure path of the CET (Figure 4.51). In both cases, induced containment isolation failure is considend, (which might occur given the combustible gas control procedures requiring purging the 11, in containe:nt by opening the purge valves, although this issue was found negligible at CPSES)in addition to the phenomena discu., sed. These induced isolation failures are separate from the cases of containment isolation failures per se, which are treated in the way discussed in Section 4.5.1, Many of the phenomenological issues are developed in the determination of the probabilities for basic events PRWCP PULT r.nd PRDCP-PULT so they cannot be seen expll;itly on the trees. However, the analysis is summarized in the quantification (Section 4.6) and described in detail in Reference 8. . Ion Event DC: Debris lied Not Coolable his event is included in the CET to signify the termination of the core melt progression subsequent to vessel breach. The success branch at this CET node means that a coolable debris bed is formed, precluding concrete attack, and thus precluding ex vessel Ossion product releases from core-concrete interaction. Following the success branch also implies that containment overpressure challch;,es from { 4-141-d 7*-.amye t.,,wwww+*, w w new wreamrw w wr m -of pwti+"Twwr-us-"-MypWe-W-w w.m+vw a g w"t1'**'*W9-1-an ,,p T.*--etayist = w'estr yw- tt-_ 9qyy ew gerswwit-e-t-r---reMsew.q---wF-Ire N w+g*wpq*e*'y-m-meerg ew9%-g W de
non-condensible gas generation and from basennt melt-through are precluded. Success in this branch requires two things: (a) that there is water over the debris and (b) that the debris is in a coolable con 0guration, i.e., that it can be cooled even with water over it because it is not insulated by an impervious crust. At CPSES the debris will not be coolable when the RWST water is not injected into the containment because in those cases die debris dries out prior to vessel failure, even if it was at Orst in a coolable con 0guration. The coolable con 0guration cases are coolable when the RWST water has been injected into the containment, or for PDS where the low pressure injection systems were previously unable to inject due to high RCS pressure, these systems would start to deliver coolant when the vessel is breached. Coolant injection will quench the debris if it is in a coolable con 0guration. This last condition can also establish a heat transfer cycle from the debris to the environment in the subsequent event node if s'vitchover to recirculation is successful. In practice, these automatic LPI injections are not - probabilistically signincant at CPSES for the reas(ms discussed below in this section. Failure at this branch implies that concrete attack occurs in the cavity, the core debris remains hot and sparging of the concrete decomposition products through the melt releases the less volatile Osrion products to the containment atmosphere. This condition is considered more likely if a deep core debris bed is formed in the cavity ;.nd, absent coolant addition, the debris is not able to effectively dissipate the decay heat to the surroundings. Should an impervious crust form, coolant addition would not likely terminate concrete attack, although the released Ossion product aerosols are scrubbed by the overlying water pool. MAAP calculations for Zion (Ref. 33) indicate that dispersal and entrainment of molten core material outside the cavity region into the lower containment region occurs in most accident sequences where the vessel fails with the RCS at high pressure. The extent of debris dispersal could vary depending on the amount of core debris molten at the time of vessel failure. The CPSES cavity and instrument tunnel con 0guration is such that. the Gow path is more restrained than the sloped configuration of Zion, llowever, the STCP model used for the NUREG il50 supporting calculations for the Surry plant does not model dispersal following vessel breach at high pressure. Separate effects calculation of DCH us;ng CONTAIN (NUREG/CR-4896) indicate dispersal is likely to occur even with the restrictive design of the Surry cavity. Should debris dispersal occur, formation of a coherent. uncoolable debris bed is not likely. Conversely, formation of a more coherent o.ons bed is considered more likely for accident sequences with the RCS depressurized prior to vessel breach. The phenomenological uncertainty 4-142
-_ _ _ _ _ . . _ , _ _ _ _ _ . - . _ u_ _ - . _ _ _ . _ _ _ _ . .
associated with debris bed coolability given water injection is the formation of an impuvious crust that precludes water ingress into the debris. This event considers the formation of an uncoolable detiris geometry and/or the absence of water in the cavity implying signincant core-concrete attack that could challenge containment integrity. The ability to cool the debris after vessel breach is determined by the possiHlity of water ingress and the formation - of a coolable corium geometry. The important issues include:
- Phenomenological considerations of crust formation.
- Sequence dependencies related to corium dispersion at vessel brea.h (i.e., high RCS pressure).
- Geometric connguration dependencies allowing formation of a shallow bed (cavity con 0guration).
- Systems oriented considerations related to water availability (e.g., ECCS injection prior to or post vessel breach and/or sprays actuation).
An examination of the cut-sets indicates the availability of ECCS injection following vessel breach. The PDS dennitions (whether high or low pressure) determine debris dispersal and, whether injection or recirculation failure occurred, determine tne availability of water beyond the potential Lpl injection at 2 vessel breach. The CPSES cavity has a large spreading area (70m ) so a shallow bed is likely. His is illustrated in Table 4.5-4. Considering that the core barrel and upper plenum internals are likely to stay 2 in place, the debris depth is expected to be 2' cm (or 30 cm if an average corium density of 7000 Kg/m is used). In arv case, the possibility that the debris is not coolable cannot be uglected, particularly if the vessel fails at low pressure although, it is highly likely that it will be coolable for Ligh pressure - sequences # 3 is spread over the cavity 70m2 door with some fraction even being expelled from the cavity. C iability is at issue if an impervious crust is formed, it is recognized that there is some disagreement it this area among experts in severe accident phenomena. However, experimental information and calculations are available in the literature and are used to provide the quantl0 cation of the likelihood of crust formation for this event node. The boundary conditions are obtained from the plant-specine assessments, e.g., depth of debris and coolant injection availability. l r 4-143
,.- ,- - -_:.. _- . . _ . . - . - , . . - . - . . , . - , , .x
injection of coolant through the breached vessel provides the most effective way of preventing concrete attack (absent crust formation) and of scrubbing vaporized nssion product aerosols, if the debris is not - in a coolable configuration. The " stems-related events that contribute to the success brach depend: (a) on the PDS so the water availat .ity prerequisite is satis 0ed, if core damage resulted from ECCS recirculation failute but not for an injection failure; and (b) passive actuation of the low pressure injection systems following RPV breach for high pressure PDS involving injection failure because, if the low pressure systems are initially available, as in some PDSs involving high RCS pressure, coolant addition to the debris is possible. As a practical matter, an examination of the PDS for which this would be possible has revealed this probability to be negligible (2E-05 for PDS IH and 6E-4 for PDS 3H). Therefore, SNOLPI basic event, which is its complement, is assumed to have a 1.0 probability, and thus - it is deleted from the CFE2 and CFE3 trees which are discussed below. The reason for the low values
^
for automatic LPI injection probability at vessel failure is that CPSES Level I analy is considers the possibility of feed and bleed for success in those PDS and failure at that step occurs almost always due to failure of LPI to iniec* In other words, failure of the LPI is mostly included in the PDS already. The logic trees for this event are shown in Figures 4.5-9 for DCl,4.5-9 e - DC2 and 4.5-1i for DC3. DCl and DC2 are essentially identical. Both are along the path of succ1 t ,ful depressurization so they involve a higher likelihood that a coolable debris bed does not form than for DC3, which is along the path of a high pressure failure. The difference between DCI and DC2 is that since DCl is along the path of successful recovery, the SNOLP! event has a low probability in that tree. Conversely, it is assumed to have a 1.0 probability under the DC2 and DC3 trees for the reasons discussed in the previous - paragraph. Again, two main issues are addressed in the DC logic tree. One is the physical configuration of the debris, whether it will be coolable or not coolable, regardless of whether there is an oserlying pool. In the low pressure cases, this probability is dependent on whether a steam explosion occurs, hence the two events GDCI .:d GDC2. If it does, the debris is much more likely to be dispersed and thus coolable. In the high pressure cases, the debris is likely to be dispersed, hence GDC3. The other aspect of debris coolability involves the presence of an overlying pool. It is assumed that if there is no overlying pool in the cavity (WTRPOOL and WTRPOOL2), then the debris is also not coolable. It should be noted that this constitutes a conservative simplification for PDS IH,3H and SH where in the actual development, a pool would exist initially but would dry c,t prior to containment failure, at which point CCI would begin. The main difference between this situation and the situation where the debris is in a non-coolable configuration is that here, the debris would be coobble if water were re-introduced, 4-144
whereas in the other situation, it would not be. By lumping both cases, a higher number of late containment failures for these PDS is estimated. This conservatism "sts for PDS IH,3H and SH.
- There is one noteworthy feature of the CET regarding the DC even ..ong the early failure paths the geestion is not asked and the debris is assumed to be coolable. He reason is that the only possible causes for the early containment failure along those CET brar.ches at CPSES would be either
- (1) an alpha event or (2) HPME. If either one happens, it is further virtually certain that the debris would be so finely distributed in the containment that it would not get hot enough to cause core concrete attack, even if it were not covered by water. This is particularly obvious for CPSES when the Table 4.5-4 is
- taken into account and noting that the upper internals and core barrel should remain solid and have no associated decay heat.
Top Event CFL: Late Containment Failure Occurs This event addresses the potential loss of containment integrity in the long term, after vessel breach and Core-Cone ete Interaction (CCI) ir the cavity if the debris is not coolable, or steam overpressurization if the debris is coolable. Potential failure modes considered at this stage of the core damage sequence include overpressure failure of the containment pressure boundary or basemat concrete attack. The conditions that contribute to containment overpressurization include boilmff of steam from the reactor cavity, given loss of haat removal function, and pressure and thermal challenges from hydrogen burning resulting from the long-term combustible gas formation in containment or non-condensible gas generation. This event is included in the CET to characterize the long term behavior of the containment after core melt and vessel breach. Event CFL includes such events as overpressure failure of the primary containment, or basemat penetration. The success path here depends either on the recovery of systems that establish a complete heat transfer cycle from the core debris to the environment or on the availability of sprays as designated by containment safeguards bin "F" (see PDS definitions Section 4.3). One of the - most important considerations is related to the time taken for gradual pressure build-up in the containment following vessel breach and ultimate disposi, ion of the molten corium in the cavity and the containment thior. The long term containment pressurization is strongly influenced by the availability of decay heat removal rm ems (DHR), and this is included in th e logic trees for event CFL. This event is related directly to the long-term reliability or recovery of the containment heat rejection function of the low pressure 4-145 i I
E i systems, given that the core is recovered in vessel or the debris is quenched ex vessel. This event implies J that a direct heat transfer cycle is established from the core to the environment, such that containment pressure rise is controlled. The implication of failure at this event node is that the containment pressure would increase and reach the ultimate capacity of the containment, challenging containment integrity. The logic tree for ibis event node is shown in Figures 4.5-15 through 4.5-19. Trees CFL1, CFL3 and CFL5 apply to cm. ible debris situations where CCI does not occur and overpressure due to steam generation is likely unless sprays are operational. Trees CFL2, CFL4 and CFL6 apply where the debris is not coolable. In these trees, CCI occurs and basemat penetration is possible. Non-condensible gas overpressurization is most likely and hydrogen burns, with the additional amount generated from CCI, can fail the containment, i CFLI is along the path of successful recovery in-vessel so it does not include containment failure mechanisms associated with CCl. In addition, the fact that recovery did occur is reDected in the probability of an early burn by setting SACSPREC to 1.0. Although BE QNOFAN allows for the possibility of fan cooler operation, these are not credited in the CPSES IPE and the BE is effectively-Dagged out in the quanti 0 cation process. The bene 0ts of fan coolers are addressed as a sensitivity issue for purposes of use in future accident strategies. The impact of fan coolers on the severe accident progression is twofold: (1) by preventing or delaying the containment pressure from reaching the spray set point, they extend the RWST duration; and (2) f ?n coolers can prevent containment failure due to overpressure, if all injection has failed as calculated in Section 5 of Reference 2. These advantages-notwithstanding, fan coolers were not credited for risk abatement purposes in the base case of the CPSES IPE for the reasons discussed in Section 4.3. CFL2 is along the path of a non-coolable debris so that only the CCI-related failure mechanisms are used with SACSPREC also set to 1.0, since recovery occurred in this path as well. CFL3 is similar to CFL1 except that recovery did not occur prior to vessel failure so its probability of occurrence, SACSPREC3, must be considered for the purpose of evaluating early burn probabilities. CFL4 is similar to CFL2 except that recovery did not occur prior to vessel failure so its probability of - occurrence, SACSPREC3, must be considered for the purpose of evaluating early burn probabilities. 4-146 i
+
t _ _ - _ _ , , ,
CFL5 is similar to CFL3 except that the time window for recovery of sprays and power is different. This translates into different recovery probabilities which are S ACSPREC in CFL5 and S ACSPREC3 in CFL3. CFL6 is similar to CFL4 except that the time window for recovery of sprays and power is different. This translates into different recovery probabilities w hich are S ACSPREC in CFL6 and S ACSPREC3 in CFL4. Top Event FPR Fission Product Removcl Fails his event is included in the CET in order to characterize rcicases from the fuel (in-vessel and ex-vessel) into the containment, the fission product retention processes, and the potential release magnitudes to the environment should the containment fail. The question raised in this event node is related to the airborne fission product removal mechanisms within the containment system. Success implies reduction of the fission product release magnitudes to the environment. Failure implies that most of f. tission products are ultimately seleased to the environment from the fuel and the containment without mitigation. The release magnitudes are likely to be relatively high should the containment fail early. The issues considered in determining the success branch of this event node include mitigating the release mechanisms from the fuel (in-vessel or ex-vessel recovery) or ensuring in-containment removal processes. The capability of the containment to reduce release magnitudes is measured through availability of active systems (e.g., containment sprays), passive capabilities for natural depletion as a result of the long time period from release to containment failure, or scrubbing afforded by an overlying water pool. Success or failure of this event depends on previous event nodes in the CET and PDS boundary conditions detined by the accident sequence cut sets. The containment design mitigating features are examined to determine if fission products released from the fuel are contained within, if not permanently removed from, the containment atmosphere. These features include the containment sprays and a high ultimate pressure capacity. Tnis event node models the in-containment fission product removal process that might occur prior to containment failure. Both active and passive removal mechanisms were considered. The active systems include ser.te ag of radioactive aerosols from the containment atmosphere by the sprays. Passive removal inc. des natural processes (e.g., gravitational settling, thermospheresis, or diffusiopheresis) that act on the idioactive airborne materials. The effectiveness of these natural removal processes in 4-147
3 i reducing the fission product concentrations depends on the length of time that the containment integrity is maintained after release of fission products from the fuel. The logic tree for fission product removal translates into four trees when the path dependencies are considered, as seen in Figure 4.5 1. FPRO, Figure 4.5 10, considers that only RCS retention mechanisms - and containment sprays are available since recovery took place in this path. FPRI, Figure 4.5-11, considerr the possibility of containment revolatization, which could potentially be mitigated by late spray action and the possibility of scrubbing by an overlying pool. FPR3, Figure 4.5-13, is similar to FPRI except that an overlying pool always exists since in-vessel recovery occurred, followed by vessel failure. FPR2, Figure 4.512, is similar to FPR3 except that the pWnbility that the containment sprays operate after vessel breach is different in the two cases: SNOSPRAY2 in FPR2 and SNOSPRAY21 in FPR3. This is because in FPR2 there was no recovery of power prior to vessel failure while in FPR3 there was. FPR4, Figure 4.5-14, is similar to FPR2 except that in FPR2 an overlying water pool always exists since this branch is along the path of a coolable debris; whereas in FPR4, which is in the non-coolable debris path, an overlying pool may or may not exist and this probability is incorporated into the tree. Top Event CFM: Containment Failure Modes This event is included in the CET to characterize the impact of containment failure modes as they affect the duration and mitigation of the fission product source terms. This tree is specialized into three path-dependent trees: CFM1, Figure 4.5-21; CFM2, Figure 4.5-22; and CFM3, Figure 4.5-23. CFMi is simply the probability that the containment will fail by rupture rather than by leakage, given that it fails late. As previously discussed, late failures are highly likely to be by leakage so this event probability has a low value as discussed in the quantification Section 4.5.6. CFM2 is the probability that it will fail by rupture given: (1) an early failure and (2) the RCS failed at low pressure. CFM3 is the probability that it will fail by rupture given an early failure with the RCS having failed at high pressure. CET End-States or Release Modes The various progression paths in the CETs lead to unique release end-states. These are shown as "A";
"B", "C" and "D" release modes in the CET in Figures 4.5-1. These release mode labels are the following:
4-148
* "A" end-states are recovered in-vessel (vessel breach is precluded) * "B" end states are such that the vessel fails, but core concrete interaction is precluded and steam )
overpressure fails the containment. l
* "C" end-states include late containment failures due to CCI. * "D" end-states include early containment frilures without ex-vessel vaporization release (CCI is precluded) As discussed under CFE above, this is the most likely release for early failure.
Table 4.5-5 summarizes the possible end-states of the CET for the spectrum of core melt accident sequences. The probabilities of each of tha 44 end-states shown in Figure 4.5-1 were determined for each PDS. However, the 44 end-states were later (Section 4.7) binned into 13 release categories and the release fractions were calculated for the 13 release categories in order to make the source term calculations more tractable. 4-149
. _ . . . . -. . - . . _ - - . - . . ~ . . - . . _ - . - . _ . .. - - - . - - .
Table 4.51: ISLOCA Initiator Frequencies for Various Pa'.hways ISLOCA ISLOCA Frequency ISLOCA Overpressurization: ISLOCA INITIATOR Small(Relief Valve) Event piping failure / Ev.nt LOCA Classification pump seal failure Classification Accumulator 9.78E 03 ClassI 2.71E-05/NA Class iii Ril Suction Line 9.32E45 Class i 1.47E-07/7.08E-06 Class iv Excess Letdown 1.30E-05 Class i 1,19E-10/NA Class iv Line Normal Letdown 2.27E-03 Class i 4.77E-10/NA' Class iv Line Low Pressure 1.44E-05 Class il 2.22E-08/1.47E-07 Class iv Injection Cold Legs Low Pressure 2.22E-07 Class ii 6.70E 11/8.07E-09 Class iv Injection Hot Legs Intermediate 1.44 E-05 Class ii 123E49/1.47E 07 Class iv Pressure injection Cold Legs Intermediate 4.44E-07 Classii 7.57E-12/8.07E-09 Class iv Pressure injection Hot Legs I l l 4-150
~ -- -, - - , .
Table 4,5-2: ISLOCA initiator Frequencies for CPSES
' I Event No, LOCA Type / ESF Equipment ISLOCA initiating ISLOCA Event Line Break Size Unavailable Frequency Classification 1 Small Break N/A 5,96E 10 Class iv Less Than Or Equal To 2" 2 Small Break One Train SI 1,48E-05 Class 11 Less Than Or Equal To 2" 3 Small Break One Train RH 1.46E45 Classii Less 'Dian Or Equal To 2" 4 Medium Break Both Trains SI 1.24E-09 Class iv -
Greater Than 2", But Less Than 6" 5 Large Break Both Trains RH 3,70E-08 Class iv Greater Than Or Equal 6" 6 Large Break One Train RH 1,32E47 Class iv Greater Than Or Equal 6" 7 Medium Break RWST 9.94E-09 Class iv Greater Than 2*, But Less Than 6" 8 Large Break RWST 4.21 E-09 Class iv Greater Than Or Equal 6" 9 - Small Break Both Trains RH 7.24E4 Class iv l i (pump seals) Less Than Or Equal to 2" 10 Small Break Both Trains SI 1,55E-07 Class iv (pump seals)- Less Than Or Equal to 2" 4-151
-. .. -.-. .- . .-. -. - -- - ~ . . - . . . . . . . . . . . . - . - . -. - . . - . ..
Table 4.5 3: Core Damage Frequencies and Their Functional Sequences Event No.- ESF Equipment ISLOCA Core ISLOCA Functional Unavailable Damage Frequency Sequence i N/A 5.96E-10 SCMI 2 One Train SI 7.15E 10 SCMI 3 One Train RH 1.06E . SCMI 4 Both Trains SI 1.24E-09 MCM2 5 Both Trains RH 3.70E-08 ACM2 6 One Train RH 3.71 E-09 ACMI 7.35 E.10 ACM2 7 RWST 9.94E MCM2 8 RWST 4.21 E-09 ACM2 : 9 Both Trains RH 8.69E-08 SCMI 10 Both Trains S1 4.94E-10 SCMI-l i l l l 4-152
- , , - , , , - - .- , - . - e,-. -we y
Table 4.5-4: Estim;.te of Debris Bed Thickness Component Mass (kg) Density Volume (m') Contribution (kg/m') to Debris Bed Thickness (m) UO 100,990 10,400 9.70 0.14 2 __ 21,316 6560 3.25 0,05 Zr Support Plate 23,415 7,888 2.97 0.04 Core Barrel 1 51,756 7,888 6.56 0.09 Core Barrel 11 14,939 1,888 1.89 0.03 Upper Pl. Intern 38,630 7,888 4.89 0.07 1 TOTAL 0.42-P 4-1.53
a Table 4.5-5: Description of CET End States 1 CET END-STATES DESCRIPTION Al Recovered in vessel, inte containment failure, in-vessel fission product release not mitigated A2 Recovered in vessel, late containment failure, in-vessel fission product release not mitigated , B1 Recovered ex-vessel, late containment failure, in-vessel fission product release mitigated B2 Recovered ex-vessel, late containment failure, in-vessel fission product release not mitigated B3 No CCI, late containment failure, in vessel fission product release mitigated by sprays B4 No CCl, late containment failure, in-vessel fission product release not mitigated B5 No CCI, late containment failure, in vessel and late fission product release mitigated by sprays S6 No CCI, late containment failure, in vessel and late fission product release not mitigated Cl CCI occurs, late containment failure, ex vessel fission product release mitigated by overlying pool, in vessel release mitigated by sprays C2 CCI occurs, late containment failure, ex-vessel fission product release mitigated by
-overlying pool, in-vessel release not mitigated C3 Significant CCI occurs, late containment failure, m and ex-vessel fission product release mitigated by sprays C4 Significant CCI occurs, late contamment failure, m and ex vessel fission product -
release not mitigated. C5 Moderate CCI occurs, late containment failure, in- and ex-vessel fission product release mitigated C6 Moderate CCI occurs, late containment failure, in- and ex vessel fission product release rot mitigated D1 No CCI, early containment failure, in-vessel fission product release not mitigated D2 No CCI, early containment failure, in-vessel fission product release not mitigated D.3-D5 No CCI, early containment failure, in-vessel and late fission product release mitigated D4-D6 No CCI, early containmeut failure, in and ex vessel fission product release not mitigated
. NOTE: The cnd-states are further characterized as Leakage (L) or Rupture (R) to indicate the duration of fission product releases to the environment. NCF stands for No Containment Failure (Figure 4.5-1) -
4-154
, .e. , -- . . . . --- ,
Figure 4.51: Containment Event Tree PDS l DP l RCC l vr l trt l DC l CFL l FPR l Cru ENDSTATE NC F( A D) CFL1 Al J'fPRO I g NCr(BO)1 CFL1 roR3 i 02-L LEI"' D2-n NCr1 W1 y Cl-L-crt? Cl-R VF rppt' 'g C2-6 D1- C2-R- ' i Ot-L-Crri Ot-R rPR3 1 02*L
- !CFW2 D2-R .e NCF(80)2 i B3-L CFL3 03-R rPR2 I 88"L ICrut gg,,
NCr2 DC2 i C3-L
' C3-R Rtc1 3 _r t a rPR4 i C4-L I.c ru 1 C4+R fDs y D 3-L crt1 D3-R FPR4 1 04-L
_ICfM2 D4-R NCr(00)3 i _ er - t- - CFLS BS"R' rPR2. I B 6 -i. teru1-96-R' NCF3 DC3 , C5-L i FM t .- CS-R op - crte rPp4 i C6-L lCru1 gg_y
; DS-L -
l DS-R Crt2 + rPP2 1 . 00"l ICFu3 D6-R 4-155 q yet --.m- -
- n-w ' -- ei-i
i arS NOT DEPAESSURIZEO BEFOaE VESSEL B2EaCH I I HCS NOT DEPsESSUA!ZE OstpaTOR F a!LS TO O BEFOAE VESSEL BAEa DEPpfSSunJZE ACS CH GIVEN HIGH OR NFOILM DAFSSLAE PDS s- Iw;3-Ze ] O 'I i r HOT LEG / SURGE HOT LEG / SUAGE LINE REMa!N INYaCT LINE AEma!N INTACT GtWEN MEDIUM POS. GIVEN HI PAESSUFE PDS. I i i n j
$EGuCNCE IS a HOT LEG aND SURCE _H0Y LEG / SURGE LINE SEQUENCE IS a HIGH t WDIUw PnESSurE PDS LINE RE**alN INtaC1 REuaIN INTatt GJvEN PRESSURE PDS GivEN kED. PAESSUAE HIGH APESSUAC 90S i b
w- 1 PM 5_5 1 5 - -
.h I I HOT LEG / SUHGE HOT LEG /5 Usa 3E LINE LIFE AEwsIN INTACT INTACT GivEN *EDIUM GIVEN HIGH Ps:tESSURE EMSSURE PDS POS.
fN --w L I-~ i s i ACS PRESSunE AT SAv HOT LEG aND e m ACS PFE.SSung BELow HOT LEG apC SUAGE SE TPOINT LI'E AENaIN INTaCY SAv SETPOINT LIT AEuAIN INTACT GivEN DP-*ED
~
I N- =E O I I PM 5& I i +& l++ED I I IM 5trf ; g g' o . o . Snv s aAE NOT STUCK STE AN GENERATOA ACP SEst AfualNS ACS PRESSUAE A1 SRv OPEN YUSES 00 NOT RUPTUAE INTACT SETPCINT i i s, . -> . -, , s. MASTER CONTAINMENT FAULT TREE CP3.CAF 10-23-92 Fig 4.5-2 i
t; I l 1 1 COOLANT NOT f; At. COVERED IN-VESSEL (LOH PAESSUAE) l AE'C 1 l I I ECCS INJECTION NOT AC POsvER NOT RECOVERED IN-VESSEL RESTOAED DA AVAILABLE A . . 1 lECCSAEC 1 iSACP0aEA I U D I T I HIGH PRESSURE ECCS LOW PRESSUAE ECCS ALTERNATIVE SYSTEMS NOT RECOVEAED NOT RECOVERED NOT RECOVERED DURING CORE MELT DUAING CORE MELT DURING CORE HELT ISHP-Sisti iSLP-SISt i 1 SALT-SIS 1 l I) I) I) . MASTER--CONTAINMENT FAULT TREE CP3.CAF 10-23-92 Fig 4.5 1 i VESSEL FAILURE OCCURS ll I NO loner HEAD OEBRIS COOLING NOT COOLING VIA Ex- ESTABLISHEO IN-VESSEL VESSEL HEAT AEMOVAL I bC i v I lLwA-HEADI I I ; NO Ex-VESSEL HEAT COOLABLE DEBAIS BE0 AEACTOR CAVITY NOT NOT FORMEO IN-VESSEL j FLOODED TAANSFER ESTABLISHED
* \ IPA-Hi-1AAN j lP4COOLDOIV l ICav-DAY l
- Fig 4.5-B E !
E NO OVERFLOW INTO AwST IS NOT OISCHAAGEO INTO ll' AEACTDA CAVITY FROM l 00TER CONTAINNENT CONTAINMENT 1 l NHaSi j lPANOV-FLOW I O f' II I I i ECCS NOT INJECTED CONT. SPRAYS 00 NOT BEFORE CD & VF (ECC OPERATE BEFOAE VB FAILS ON INJECTION) l SECC5-INJ l lSNOSPAAY1 1 O A dP3.CAF 10-23-92 Fig 4.5-4 MASTER CONTAINMENT FAULT TREE
- i . . w,
EAALY CONTAlf4 MENT FAILUAE OCCUAS lCFE1 l 1 MOLTEN DEBRIS IMPING I CONTAINMENT ALPHA OA CP>PULT ES CONTAINMENT WALL,. ISOLATION FAILUAE F#ILS CONTAINMENT CAUSING FAILUAE (CA GIVEN LOW ACS V. GEOM. OEPENDEN PRESSURE E I PRIMPINGE I I PACI i 1A-OA-CPL i I I ALPHA EVENT OCCUAS CONTAINMENT PAESSUAE GIVEN ACS > ULTIMATE PRESSURE DEPRESSUAIZED GIVEN CONO. AT VES. BAEACH (LOW ACS P I PG ALPHAL l l PACP-PUL TL l 0 0 t-i
?
MASTER CONTAINMENT FAULT TREE CP3.CAF- 10-23 Fig 4.5-5 !.
EARLY CONTOINMENT FAILURE OCCURS l CFE2 I I I MOLTEN DEBRIS IkPING CONTAINMENT CONTAINMENT FAILS ES CONTAIN4ENT WALL. ISOLATION FAILLAE OUE TO ROCKET. CAUSING FAILURE (CA ALPHA. OR HIGH V. GEOH. OEPENDEN PRESSURE I PRIHUINGE I ITatTl IR-ANPH I VESSEL ACTS AS CONTAINMENT PAESSURE ALPHA EVENT OCCURS AOCKET ANO FAILS EXCEEDS ULTIMATE GI GIVEN ACS AT HIGH CONTAINMENT VEN CONO. AT VESS. PRESSURE BAEACH-HIGH ACS PA. ipr @ICwEI I I CP-PIS TH 1 1 P A A[**P A>4 I 1 CONTAINMENT FAILS CONTAINMENT FAILS g WITH SPAAYS ON WITH OUT SPAAYS ON I CF c'SP4 l lCFwb5PA I I I I I CONTAINMENT SPAAYS OPERATE CONT AIP#4NT C(WT. SPAAYS DO NOT PRESSURE > PULT. BEFORE VB PnESSURE > PULT ' OPERATE BEFORE VB GIVEN WET CAv!TY GivEN ORY CAVITY AND ACS 0 PAESS. AND ACS C PRESS t IPA *CP'-PULT I EOS$Aav 1 1 I P40CP-Pt1 Y I 1. WJSOA A v i 1 O O rn
~
CONT. SPRAYS DO NOT t OPEnATE BEFORE VB iSN055AAvi i b MASTER CONTAINMENT FAULT TREE CP3.CAF 10-23-92 Fig 4.5-6 4
i h i OESPIS BEO NOT , COOLa9LE GIVEN LOm
- PAESS. EJECTION AND E VSE $
.I COcm.a9LE DEBRIS BED NO OVEALVING MATER DOES NOT FOAN POOL IN CAVITY I
Fige,S-S
' ' Fige.5-II I
DEBAIS BED NOT O I DEBAIS BE0 NOT I MacTOR CaWITV NOT W E BOTN CONTaIw(NT 4 COOLa8LE G!vEN LOW COOLa9LE G!vEN LOm -FLOODED SPRav aNO LPI PRESS. E ECTION AND PAESS. EJECTION ANO INJECTION FAIL EWSE NO EvsE p u4 Icar-LGi 1 IL;Ijaavi i F194.5-4 w 1 i I I I I I I Ex-VESSEL STEau CDB DOESN Y FOAN COB OUESN Y FOAN Ex-VESSEL STEaN LOSS OF LPI CONT. SmavS OQ NOT EXPLOSION OCCURS GIVEN Ev5E OCCURS GIVEN NO W a*C EEPLOSION COESN Y *NACTION THAOUGN OPEpaTE AFTEA v6 ; AND NO HPNC NO EvSE OCCup FAILED VESSEL
'l i mc',cE i i acci-gst i i m:ci-g~s i Q i sst i is<s=a21 i O O O O EQ !
Ex-VESSEL STEaN , EMPLOSION OCCURS ! ! I I F0E v'E I O i'o t MASTER CONTAINMENT FAULT TREE CP3.CAF 10-23-92 Fig 4.5-7
i 4 i i DEBAIS BED NOT li i COOLABLE GIVEN LOW PRESSURE S NO - l RECOVERY l 1 052 I If L i i l COOLABLE DEBAIS BED NO OVERLYING WATER ODES NOT FORM POOL IN CAVITY
- I I IDE6dO9M i ~
I W T AP00t_2 I Fig 4.5 _a f 3 II I i AEACTOR CAVITY NOT CONT. SPRAYS DO NOT
'FLOCOED OPEAATE AFTER VB l
ICAv-ORY I I StaOSdA A v2 1 Fig 4.5-4 0 MAOTER CONTAINMENT FAULT TREE CP3.CAF 10-23-92 Fig 4.5-8
w-l I l i l l l ) DEBAIS BED NOT COOLABLE GIVEN HIGH l PPESSUAE ( I Od31 i i NO OVERLYING WATEA DEBRIS BE0 NOT POOL IN CAVITY COOLABLE GIVEN HIGH PAESS. EJECTION ( - IwiAP00L2 l l 60C31 f A Fig 4.5-8 CDG DOESN T FOAM GIVEN MPME I Iencos-up I O CP3.CAF 10-23-92 Fig 4.5-9 MASTER CONTAINMENT FAULT TREE
FISSION PAODUCT AEMOVAL FAILS GIVEN: NO VF l FPRO l l l ACS RETENTION DOES SPAAYS DO NOT NOT OCCUA OPERATE BEFOAE OR AFTER VB 104CS-PET I l NNOSP4 AY l t F 5\11 ' t (Q 1 l' CONT. SPRAYS DO NOT CONT. SPAAYS 00 NOT OPERATE BEFOAE VB OPEAATE AFTEA v3 iSNOSPRAYti ISNOSPHAv21 I S 4 1 MASTER CONTAINMENT FAULT TREE CP3.CAF 10-25-92 Fig 4.5-101
FISSION PAODUCT AEMOVAL FAILS GIVEtt REC. VF G CCI l FPA2 ; I I VOLATILE FISSION EX-VESSEL FISSION PROD AELEASE PAODUCT RELEASE LH-UNMITIGATED ' (FPA4) MITIGATED Iv0L-UN**Ii1 1 IFPN-E ] i , y 1 a IN-VESSEL FISSION LATE FISSION NO OVEALYING WATER PRODUCT RELEASE PAODUCT RELEASE U*4- POOL IN CA% ? TY
$ UNMITIGATED (FPA3) MITIGATED ' " ' F194.5-12 ' I F3gd.5-13 F19# 5-7 Fig 4.5-13 I
ii I I rr I ACS RETENTION DOES SPAAYS DO NOT ACS/ CONTAINNENT CONT. SPAAYS DO NOT NOT OCCUA OPEAATE BEFOAE OA HEATUP CAUSES OPEP. ATE AFTER VB AFTER VB AEVOLATIZATION IOACS-4EI I INN 05'AAY P I IP49E'ATUP 1 I SNOSPA AY2 l Figd.5-10 3 . MASTER CONTAINMENT FAULT TREE CP3.CAF 10-23-92 Fig 4.5-il
FISSION PRODUCT AEHOVAL FAILS GIVEN: NO REC & NO CCI IFPA2l VOLATILE FISSION PROD. RELEASE UNMITIGATED iVOL-UNMIT I F 5\14 1 I IN-VESSEL FISSION LATE FISSION PAODUCT RELE ASE JN- PRODUCT AELEASE UN-MITIGATED MITIGATED , p l FPA-IN l !FPA-LATE I T F194.5-11 II , ACS AETENTION DOES SPRAYS DON T .. j NOT OCCUA CPEAATE BEFORE OA ; AFTER VB j i IOACS'-REI I INOSPAAY12l (' (- l I I f CONT. SPAAYS DO NOT CONT. SPRAYS DO NOT . l OPEAATE BEFORE VB OPERATE AFTER VB l SNOSNAAY 1 I ISNOSNAAv2I I") il MASTER CONTAINMENT FAULT TREE CP3.CAF 10-23-92 Fig 4.5-12
FISSION PROCUCT REMOVAL FAILS GIVEN: REC. VF . NO CCI IFPA31 i 3 .' IN-VESSEL FISSION LATE FISSION $ PRODUCT RELEASE UNMITIGATED (FF /43) PRODUCT RELEASE UN-MITIGATED I FPA-IN1 l lFPA-LA1E I Figd.5-11 Fig 4.5-11 MASTER CONTAINMENT FAULT TREE CP3.CAF 10-23-92 Fig 4.5-13
I i FISSION PRODUCTS REMOVAL FAILS GIVEN; NO REC & CCI T IFPA4 1 i I I VOLATILE FISSION Ex-VESSEL FISSION PROD. RELEASE PRODUCT RELEASE UN-g UNMITIGATED MITIGATED I VOL-UNMI T l ff94-Ex2I I 4 Figd.5-12 I NO OVEALYING WATER ! POOL IN CAVITY ! l IW1HPOOL2 I f AFigo.3-o i CP3.CAF 10-23-92 Fig 4.5-1'4 MASTER CONTAINMENT FAULT TREE
- t. A ? t LON ' a ! Nw k r 8 a }tJ4 QCCVRS
(?ref aiwN! + a!L S L AT E 9.E 70 CVEn PSiC SSUAF Ii,. _ TU T kl BURN av rir. h 1 S Tt am (4 NE LA T ICW VESSEL erf *CH FAILS Fa!LS CMTa:NMENT CONT a lmMF NT D(f TO Dv!R-PRESSUAE FiE- a.. I E 1, = ; a . . I m Md $9W (Krush W r _PM shutt *ISt txA a Sttam 04 td R a T If/* W E I N 5UF F I C J EN T Citav l Civf N AC PDef p AND TD H2 e sw a T sTapf DCCUES HE a f REmev a. g
$PAA)5 McCVE Af D (# CC: F alL5 CW -
1 I- : I I- : I A sNss F rt:t N r :N- 395 F It 1E NT patav M VE SSE L Cf Cav ME AT Af e at Cs t 10 FCCS ef af RE mow at racne C tws t a w n '
@ F a it t Af
'Q I?~~e- !I 11 ' .E ' I petar mE s t Hteadwat X Lets, w ar pa rt W E DECav .atar AaTE CDaL E AS LNa k a tt aBLE E uCEED9 ACT!WE Mat EmCEEDs passive T RaNSF E R TD Meat imaNSFE R TO CONTaleemf%T CONT Ntaf S!*s=5 Ik _RI Ei - -a;I I I- -a_: I HEaI ( t NE Ha f 10N $39av5 aNQ F ANS NUT E nCE ECS Nuovat OPEmaTING a5TER wESS&L eg aCN rn .1 _ s,,I CCN1 E bPRav$ 09 NLlI h I
#AN Lt1CL f MS ps0 I OPEpavE aF TER V5 UPE Rt. 71NG I . . _ _
- c1 1 -e E) '
MASTER CONTAINMENT' FAULT TREE CP3.CAF 10-23-92 Fig 4.5-15
_ ,., - -, .,L, . . . . . , . . TO Ovt R t ATE _DUEsm TeinD.u.G.r,._F
- u. . tt,$
i:~ _- E ~
~ _ - , , ,
O _ , . E TY I BAE .CM
- A f f ,DUE 10 CEI m /SP*..5 . .W
. t m/O $8*a.r5 .seS OE .gnt S BED CD'A..m/0 WE CUmfS Ih*4 F .T.t $t.t NT msE 0S nE < a E D EFFf C TS LTBRIS et D COOL./UPM ( *l PlLE Uvr A-PAF59haE I .= . a , . ., .
W W .W,
.,~vN.*.C-... .-,_,_ ~.,..:-,..,,, .__s,. - . ,. . , , . . . . .
em ... , _ __ ,... . . ...., _v. , - < . . . _ _
.m... .
Spa 5 AEC0ve nto or CCI F alls CONT CC'A .R E BU T S**.fe5 LE G 5.t GENE.p.f! 1e . ? sus.NCue 02 twe COUL'J'sh am t.*.NO 5 *C I~' m' I I" iI UT - A _ E E_- m: I u _ _ , , . .., ~ _ . ..f- _, - . . , _ , O GEmparlom s.Its 'v t ssit 854 acM F els. $ OptC a , E .ftfe vG CONl t itemeE N T C[Pd1&!auMENT OuE TO Gvf R-PseF SSiptf O O L.'k Md 4pt*84 8. ILS CON powEm' A!Naf NYanc 75 SPR.ti Se fM F
. . . . ~ _ _ , -,_,,m-aftec G!vt'.s Spa f5AC POWE. a%
F8Eccvt8aED TO *e2 CCI BU.AN 8'UL T MASTER CONTAINMENT FAULT TREE CP3.CAF 10-23-92 Fig 4.5-16
l t ail (LWl a L NesF NY I F a [LWF. O' tl*i3 ctnt a :*est N T p a rts L a '; C54 10 OVEA PAE SSuRE IL, __- I E I N2 fM548 AF 1 t.H SIE am Gk Nt Ra I llW VE SSE L BHFaCM Fa{LS F a ]($ Ctydia jNW NT CONYa[Nt8FNY DOE YO OVEE-PAE55uMF I P- ' iAI_ I I3-~*- Al I i at Poets aNo drcaos +5- eus+N uttuph a put ssust pite Lm e 1 INsurPIcttNT cttav W i sitam bErapattoN AEC0wfAED faALv G ! vf u a t P0auf A aND 10 HP tsusaN af STaAf *f a Y F4E MOD at DCCUAS Sonav5 HCOwTAED OF CCI r aIts CaN' I -, A . t:3 s ur -' ; 1 GM I- E- E INSUF F IC IENT DE Ca v esE af RF esOdaL P AfD A CONTaINMfNT b ' Q- _ I i i utCar wtaf Haft Laraf ME A T RATE E WCEEDS ACT!wi M a7 E *Mf DS PASS!wE TR4NT#ER TD HE a f imamsrfA TO CON 1 a 1NME NT f.ON T Mfa7 $!w $ E :-' .'- A l ' E I__-- TTad I I E NEa1 LENtmailON SPR&f$ aND FAN $ NOI E NCI E DS F( e*0 vat OPE Ra11NO af fE R WESSEL Bif aCM t, m,., CtW I SPAaf5 DD NOT F AN CCD;.E RS Mt'T M Raf f aF f f R VB OPfRa7ING Cf] f_0 MASl* 1 CONTAINMENT FAULT TREE CP3.CAF 10-23-9^ Fig 4.5-17
- s. - t 1
l ? t i i
!{
l' I !
- n ._
W .
. . , . _-. .. . _ . = _. _ _. - ._ _ . _ . a ,
i W ' 7 :" 1 -~ t c 9 W 2_. ... .,. _,_.._. i a ! d m an. an wws. 3
.-.mm a m. _. , at . _.,4 ..ep a.s._-_ _A #g ......_.._..u3 , . . ew4 1.um e a
_mo .,.nsi.c. .y'am.e
- t ~_ . _ . . . .. - . --.i - i -- .
w, _e s .a.n- a_s - _ .=.=N
*-86 V
t T T U V T U j!
~-_ . -
l. A . . a m- a. . a t
~
m 6), a a .
.. ..,. . . - . - _ . . . . i . _ . . ~.
t l Y ..,_ _ _. Y ___
._., 1 9 8'S T NN M3 emst ?
d t V V V V I i.
+
T ; J-i g' - y i
< t o ' i.
t a MASTER CONTAINMENT FAULT TREE CP3.CAF 10-23-92 Fio4.5-18 ., I t y , .. .,,m - , , .. ,m . --~e .+w-w- .w.< . . -
[ L s I m at nw a:v e t C a h JCC 'X *3 ; f f i f i l tv%'s ;=ew* %r s a . . s L a *E t* * O ?w'
- Pat $$M i f I__ 1__ I !
E I W iksen a* Y t a S T E e= (4 w aa 1w l nsm nac-Farts **nscat. w CrwT a t naaE w? Ds E TO f*vF p -esar 3 gsw I - a _E I__ - *._ 1 t si. N u **C 1-* r+a
- 5 V
N resses t.u..Cz ** 5 -wM I PI'a t DJ I t h e*Ir;t%Y C*ca+ W Y 5:t ase t.e *J a
- r Ie -
\
1atC W AED (sQt.
- GI s(% aC PO*8.Fs M ? D @ e-Sam * ? 5?*37 af * ? curwm ecss SD'sa v$ E*leviIM O CF CCI F alis CCN' s4 - 3 3- --I _
d I _ 2: N + 1
, ~ . - - , < . . ;
M. e : CF*es't &L WOma
- 0% e tw as* I a
~
4. I _. I W. 1 E a;;e e .sr a r mA*1 TX , e, ear. a r sae 'E 4 E< _r.,_~, (a:EE3 ar.t.iof M a! a
- . ,EE,*15 pas 1.:w,E ~ . . s r. ,a .
3-._- a_- g .g 0 , s_a , _ . _ , . M . z m,
,- ,.a-. . . n,n as tf m wt S3EL EAE F.>*
I' -1 'M
.E I c,,1 , - . m e, _, - ;. a. E. ., -1, 63 '
MASTER CONTAINMENT FAULT TREE CP3.CAF 10-23-92 Fig 4.5-1('
- - . .*. .c e , - . . -.._w .-y -- - . .-- -.-
s i , t f
- - 1 i '
...a m w
H , e _.. ., _ . , . . . . _ _ . . t 9' Y b i
,=, , _. ~. _. - . . _ . _ ,
y y .; _. __ ._ .... ... _< __ _. ~ _ . _..
. 3 U V V V ^
T T a a
.~___ . . __. ... ... , .. .
1
. 9 Y a &
E 9
= =. :_c -
- ==-- =. .
a f
- _.~ . _ t F.-._.__ .. _ . _ ,. ~.
[ __ t
- e ,
g y y g y ; a
- 'o' T !
- 1. ,
i !* J 4 'k i i MASTER' CONTAINMENT FAULT TREE i CP3.CAF 10-2b-92 Fig 4.5-20 l t r
. . -, . . , ,. . - , , . ~ , , .
I
1 4 t ,
)[
a t
},
I l' 1
!l L i
l-l" CONTAINMENT FAILtJAE MODE GIVEN LATE F A ILt>AE ILOW PAESSUAE) [ CF M1 l
+
i CONTAINMENT FAILS t t LATE BY AUPTUAE MODE
.a r m == .,
i po-a.: m : O I! a t
, l t
1 a [ l L 1 f MASTER CONTAINMENT FAULT TREE CP3.CAF 10-23-92 Fig 4.5-21 ! I
t I t I i i CONTAIN'*ENT FAILUAE l HODE GIVEN EAALY FAILUAE ILOW ; PAESSURE1 ; 1 CFM2 l [ (- I 6 i I L CONTAINMENT FAILS ALPHA EVENT OCCURS ! EAALY BY RUPTUAE FOR LOW PRESSURE ! DUE TO OTHER THAN SEQUENCES ! ALPHA OA AOCKET ! I W -AtPwCFE 1 lAl-LOW l 2
.?_ l I
M' .I i i ALPHA EVENT OCCUAS ,? GIVEN RCS j DEPAESSUAIZED i l r ,' 1 PS ALPHat' i 1 b ; f i i I i MASTER CONTAINMENT FAULT TREE CP3.CAF 10-23-92 Fig 4.5-22 !
) 4 ECNTAINMENT FAILUAE i MODE GIVEN EAALY FAILUAE (HIGH PAESSUAE) l Ceu3 l } ' I I CONTAINMENT FAILS ROCKET OR ALPHA EARLY BY AUPTUAE EVENT FAILS " DUE TO OTHEA THAN CONTAINuENT ALPHA OA AOCKET IPR-4UPwCFE i iROCM-At I r
- 1. 2 l
~
d I I VESSEL ACTS AS ALPHA EVENT OCCURS ROCKET AND FAILS GIVEN ACS AT HIGH CONTAINMENT PAESSUAE IPAno'CxET I I postupaw 1 O O i 4 i I l h MASTER CONTAINMENT FAULT TREE CP3.CAF 10-23-92 Fig 4.5-23 ; i i
. _ ~ ._ . _ . . _ , , , _ _ . _,
1 4,6 AssMeat Progression and cliT Ouantincation J 1 Section 4.6.1 presents the conditional quantincation of the CETs for each PDS, i.e, the method and l l results for determining CET end-state probability values assuming each PDS has occurred (1.0 probability for the Orst CET event, Figure 4.51). i Section 4.6.2 presents the coupling of the PDS frequenties determined in the binning process (Section 4.3) with the conditional CET probabilities determined in Section 4.6.1. It then goes on to rank the sisk + irnportance of the PDS, determines representative accident sequences and discusses accident progressions. Section 4.6.3 describes the effect of the phenomenological uncertainties listed in NUREG+1335 Section : 2.2.2.6 as signi0 cant, and expands that list to include a discussion on other uncertainties also considered irnportant. The purpose of the uncertainties discussion is to show that the potential masking of vulnerabilities due to controversial assumptions regarding the likelihood of certain phenomena doesn't occur and that the findings of Section 4.6 are robust and insensitive to these uncertainty issues. 4,6.1 CET Ouantification ! Re basic structure of the CPSES CETs is given in l'igure 4.51 (Section 4.5) and the supporting logic trees of Figures 4.5-2 through 4,5-23. That basic structure is applied to each non-bypassed, successfully
!solat'e d containment PDS, which are listed in the top line of Table 4.3.3. Thus, there is one CET for each of those PDS. ,
Dere are two. steps in the CET quanti 0 cation process: (!) the determination of basic event (BE) probabilities and (2) the computation, interpretation and presentation of results for the conditional probabilities. 4.6.1.1 Deteimination of BE nrobabilities: Example. PRWCP PULT for PDS 2H = 0.42 The first step in CET quanti 0 cation is the determination of the basic event (BE) probabilities. These BEs 5 are the events shown in Figures 4.5-2 through 4.5-23 that are flagged with either a circle (phenonmena-related), or a pentagon (system-related), or a diamond (operator action - related). Although the - 4-178 l
. . . - - . - _ . . . - - , ,,.~,---,-._--..,,,,.-..,-,.,,-,,--,,,-.-.----,-,--,-.-~n -- - - . ,, - ,-.- . - -e , n -,.- . - ,
r,-
determination of some of these BEs is ttivial, when all PDS are considered, there are over one thousand ilE values defined. Rus, the justincation of each is not provided in this submittal but is available elsewhere (Ref. 8). Nevertheless, an example is given here in order to illusttate the process by which some BE probabilities are evaluated. This processes can be very detailed for some BE. A good example is presented below for PDS 211. It is the determination of the BE probability value for PRWCP PULT, under top event CFE2 on Figure 4.5-6. nat BE renects the probability that high pressure melt ejection (llPhtE) phenomena eccurring at the time of vessel failure at high pressure will lead to a pressure spike that is greater than the containment capacity, it includes the effects of combustion of non-condensibles and direct containment heat i ng (DCil) separately and combined, as well as their probability of occurrence. Although the rapid thermal transient at vessel failure 4 not typically calculated by h1A AP to induce the extremely high pressures and temperatures that threaten containment integrity, NUREG ll50 suggests that higher values than those computed with h1 AAP for the pressure rise are possible. In the present assessment, the NUREG ll50 methodology was combined with plant specific hiAAP analyses by adding the NUREGICR-4551 pressure rise est! mates for Zion (Table 4.61, (Ref.1)), which is similar in volume to CPSES, to the initial containment pressure at vessel failure, determined through h1 A AP analyses for PDS 211 (Ref. 2). In addition, the entry or h>ok-up conditions for the NUREG/CR-4551 pressure rise values were also based on CPSES specine htAAP analyses for PDS 211 (Ref. 2) as follows. There are four major parameters that control the pressure rise in the NUREG/CR-4551 approach, as indicated in Table 4.6-1, namely: (1) reactor vessel pressure prior to vessel breach, (2) fraction of molten , debris ejected, (3) initial slie of hole in the reactor vessel lower head and (4) presence or lack of water l l in the cavity. Although the pressure range for core damage bins 1 and 2 is 200 to 2000 psia, all Small Break LOCA PDS have RCS pressures at VF less than 1000 psia, as shown in Tables 2 6 (A and B) and Figure 2-2 of (Reference 2). This allows entry in Table 4.61 for all PDS into the hiEDIUhi RCS pressure line. The fraction of molten debris ejected is assumed to be less than 33% in all cases, so that the debris mass line entry is LOW. The RPV hole size is LOW based on the assumption that one-penetration weld wi;l fail first. Cavity status is WET in all cases as discussed in (Ref. 27) and according to htAAP CPSES-specific results, as summarized in Table 2 3 of Reference 8. 4 179
,,,weee-,-ew-.mw, -, - - - ,----.-m--p.------ ww,= , r , ,,--ew-,.-,e..-y,erm--y+-s-r,- = + - =r w**>--~ a- y e
- w- ,
Based up(m these considerations, a curve was obtained from the data in Table 4.61 for the probability that the pressure rise is less than the pressure shown, By adding to this distribution the initial value of the containment pressure, the probability that the final containment pressure is less than the value shown was estimated for each PDS. PDS 211 uses the Table 4.61 pressure rise plus 80 psi (per Table 2 3 of Reference 8). De curves obtained in this manner are shown in Figure 4.61, where they are compared with the curve for the probability of the containment surviving a given load: obtained specifically for CPSES, as described in Section 4.4 and represented in Table 4.4-4 He probability that the llPME pressure does not exceed the ultimate pressure capacity of the containment is determined from these two cumulative distributions utilizing the concept of reliability (Ref. 28). De concept of containment reliability is that, if the challenge pressure induced by an event does not exceed its capacity, the containment is reliable. Specifically, in the present study the following are given: l F i (c): Cumulative probability distribution of the challenge pressure, f F, (s): Cumulative probability distribution of the containment surviving pressure, where: c: Challenge pressure, , s: Containment surviving pressu 3. Both probability distributions are not in the form of known functions but, rather, the estimated cumulative probabilities at certain pressure points are available. He containment reliability is denned as: R (c < s): Prol, ability distribution that the challenge pressure is less than the containment surviving pressure. l l In order to determine the containment reliability, let fi (c) be the density furetion of the challenge pressure and f (s) be the density function of the containment surviving pressure. The containment reliability is then calculated as: 4-180-'
. ._ u ~ - - - .- , . - - . - - - , . - - - . . - . - - . - -. . . . - , . , ~ - . . . . -
. . i R(c<s)=[/, (c) [I-f /a(s) ds)de (4.6-1) a i
in order to compute the containment reliability directly by using the given data for both challenge i pressure and containment surviving pressure, define: 1-[f,(s)dr F,(c) (4.6-2)
- and dFi (c) f ik)de (4.6-3)
. The range of F, is obviously from 0 to I. Substituting Eq. (4.6-2) and Eq. (4.6 3) into Eq. (4.61) yields: i I R(c<s)=[F,dF (4.6 -4)
)
Equation (4.6 4) suggests that the area under an F i versus F, plot would represent the reliability of the ,
. containment, it can also be obtained numerically utilizing the following expression:
28 R(c<s)={ (Fu+Fu i)(F i,-F i,,i)/2 (4.6-5) 4.i -l 'T 9 .
. l 2H. -;
1 4 181. , v- - + o e . ---r=~. - ~ - . . , b.-.,s.,m-,~,..--.-,--w- ,r-. - - w,v. 3,- - , . e,---,--e -,e-e,.ww. ww.---rr-e,--.-,v-n-w.w,.,,e--mw -w . , wp- --t .--w
4.6.1.2 CET Ouantification Frocess and (Jiscussion of Results i > The Cl.Ts are input into the ETA Il (Ref. 29) code. The result of this stage is shown in Figure 4.51, which shows an ETA Il plot of the CPSES CET. The sarne CET structure is used for all PDS in which the containment is not bypassed or failed to isolate. ETA-Il was used to generate logic for CAITA (Ref.
- 30) by linking the various logic trees of Figures 4.5 2 through 4.5-23, each of which corresponds to a branch in the CET. This approach was taken in order to take into consideration the dependencies from earlier CET questions on later questions. There h a CAFTA file that contains all the logic trees and an
- mociated ilE file into which the PDS specific llE probabilities were input. The logic trees are the same for all PDS. Ilasic event probabilities are distinguished in the llE Ole by suffixes, e.g.,
SNOSPRAYllFi=0 vs. SNOSPRAYlllia l. The CAFTA file only contains one set of trees where the I basic event raues have no suffixes. In order to evaluate the logic trees, a flag Ole was developed that equates the unsuffixed 13E name in the linked CAFTA Ole to the suf0xed name in the IlE Ole and its i 1 associated probability. The computation was actually performed by loading for each PDS: (1) the CAFTA logic trees and its associated !!E Ole, (2) the Dag files discussed above, and (3) the Ole containing the logic connecting the logic trees, as described above. Once these files were loaded into CAITA, the actual computation of the probabilities was performed using the code GTPROB (Ref. 31). Since GTPROli does not yield cutsets but only probabilities, the logic trees were also quantified using CAITA with the logic trees top event cutsets as the product. The CAFTA probability calculation is not suf0ciently accurate for evaluating the CET end state probabilities, hence the need for GTPROll. The results of the quantification process described above are summarized in two sets of tables. The first set, Tables 4.6-3 through 4.6-6, lists the end state probabilities. The second set, Tables 4.6-7 through I 4.6-10, summarizes the same results in a more compact and descriptive form by grouping CET end- j states. The key phenomena, systems considerations and operator actions which lead to the top event probability values in the CETs, which in turn lead to the end state probability values, are discussed below for each PDS. PDSlE ] This PDS groups Small Bresk LOCA with ECCS failure at i,jection and containment spray failure at recirculation. As previously discussed, these PDS involve the RCS at medium pressure, with only a 0.05 , probability of depressurin: ion to low pressure levels due to hot leg or surge line failure. Recovery is I not considered and, therefore, vessel failure always occurs, i 4-182
t~ u-tay,,=e- & w w eqgy. n , w-b a w g s_w- , Ly,g, ,y..w.n-s,.,, w -+ Awe,'+8 w"-s- 3 -yv+m a" M3m- w seWMw= , .%w ur z-w'-W='+eDr- T1 etTh W'r**
- F"W4+--'t
h As a result, the first relevant branch probability is DP=0.95, representing the 0.95 probability that the l vessel will be at an intermediate pressure at the tirne of vessel failure. He next relevant branch probability is CFEl=8.0E-03, CONTAINhiENT FAILS EARLY, along the successful DP path. This value is due to the probability of a low pressure alpha event. Along the non-depressurization path, CFE2=1.83E-01 This result is due primarily to the chanw that llPh1E might occur and fall the containment. There is a small contribution due to the rocket event and one due to alpha at high pressure. The next relevant event is DC2=7.37E-02, DEBRIS BED NOT COOLABLE, along the successful DP path. This results from judgment probabilities:(1) that a coolable debris bed does not form given no ex vessel steam explosion (PRCDB LPNS = 0.1) times the probability that ex vessel steam explosion does not occur ( PRVSE = 0.5), ORed vith (2) the probability that an ex vessel steain explosion occurs (PRVSE = 0.5) times the probability that a coolable debris bed does not form given that (PRCDB-LPSE
= 0.05) an ex vessel steam explosion occurs. That is:
DC = l(0.1x0.5) + (0.05x0.5)] - [(0.1x0.5)x(0.05x0.5)] Alcag the nonslepressurized path, DC3=0.05. This value is due primarily to the judgement that it is very unlikely (0.05, Table 21 of Ref. 8) that a coolable debris is not formed if the RCS is not depressurized before vessel breach, which then occurs in a blowdown mode, i.e., as high pressure melt ejection. For LATE CONTAINh1ENT FAILURE OCCURS, CFL1 = CFL3 = CFL5 = 0.65, if the debris is coolable and CFL2 = CFL4 = CFL6 = 0.955, if it is not. l' For FISSION PRODUCT REh10 VAL h1ECH ANISh15 FAIL, FPR0 = FPP.1 = FPR4 = 0,95 and l FPR2 = FPR3 = 0.998 are high because of the probability that RCS heat up could cause revolatization of fission products, although RCS retention occurs due to spray operation before vessel breach. For CONTAINMENT FAILURE MODES, CFht2 =5.04E-01, CFhi3 =5.004E-01, if the failure is early and CFhil =5E-03, if it is late. These split fractions simply reflect the fact that an energetic failure is likely to result in a rupture mode failure and a puff release, while a late failure is likely to result in a leakage type failure. 4 183
. =_
p e gy 9,9y,p'
As shown in Table 4.6 7, the cornbination of these branch probabilities results in a 0.67 total probability . of no containment failure. There is a 2.lE 3 probability of early containment failure, where 22% are ; small or leakage type failures and 78% are rupture type. There is a 0.66 total probability of late I containment failure. Of this,92% are due to steam overpressure. The remaining 8% are due to non-condensible overpressure or basemat melt through or 211 burn, if the debris is not coolable and CCI occurs under the overlying pool of water. Ninety Ove percent (95%) of both late and early failures are unmitigated since the releases are mostly due to revolatization within the RCS and cannot be scrubbed by an overlying pool in the reactor cavity. Of the late failutes, 99.5% are benign, i.e., leakage as opposed to rupture. PDSIF This PDS groups Small llreak LOCA with ECCS failuie at injection and containment sprays that operate during injection and recirculation. Except for the fact that an examination of the cutsets in this PDS shows that operator depressurization as quantified via llE ilOP DP is more likely here, the CET for this event is quite similar to that discussed previously for PDS lE. The difference between the PDS is in the duration of spray operation, in this case, the containment would never fait due to overpressure, while in the previous case it can. The early failure probabilities are slightly higher than those for PDS lE, namely a 7.71E-3 probability of early containment failure, where almost all of these failures are rupture-type from the steam explosion alpha event which became more likely due to the depressurization. The , late failure probability is only 7.3E 2 due to the possibility that the debris may fall into a non-ccolable con 0guration, even if the containment is flooded, and then fait by the same mechanisms presented for the non coolable situation in PDS lE. The total containment failure probability for PDS IF is 8.lE 2. PDS 111 This PDS groups Small Break LOCA with ECCS and containment spray failure at injection. As previously discussed, these PDS involve the RCS at medium pressure, with only a 0.05 probability of depressurizrtion to low pressure levels due to hot leg or surge line failure, but with possible operator depressurization. Recovery is not considered and, therefore, vessel failure always occurs. l As a result, the Erst relevant branch probability is DP=0.0475, representing the 0.0475 probability that the vessel will be at an intermediate pressure at the time of vessel failure. 4-184
-r- *, ,,w w-in-esw..- w wwa-.-esam m- ,-e we .-,e-w, y,-mm,e+=-*vmn-,a>.c" ,u-eeg- gr WTrg--D.MMrT-- - -4 'T 7v> p W- ' -PPTy-
The next relevant branch probability is CFEl = 8.0E-03, CONTAINhtENT FAILS EARLY, along the succ.ssful DP path. This value is due to the probability of a low pressure alpha event. Along the non-depressurization path, CFE2= 1.83E-03. His result is due primarily to the chance that ilPh1E might occur and fall the containment. There is a small contribution due to the rocket event and one due to alpha at high pressure. l The next relevant event is DC2= DC3= 1.0, DEBRIS BED NOT COOLABLE. This results from the I assumption that the debris bed is not coolable if the RWST water is not injected into the containment and the debris dries out. l For LATE CONTAINhtENT FAILURE OCCURS, CFL1=CFL3=CFL5=0, if the debris is coolable i and CFL2 =0.996 and CFL4=CFL6=0.995, if it is not, if the debris is coolable, there is no containment failure because there is not enough 11: generated in the core melt to threaten the containment in a burn and the amount of steam generated will not cause containment failure without the non condensibles in the CCI. If the debris is not coolable, there is CCI with enough attendant 11; generated ; to yield a 2.49E-03 probability of a burn with enoagh energy to fall the containment, which must be con Sined with the 0.995 probability of failure due to non condensible overpressurization and with the 0.05 probability of basemat melt through. De burn probability is much lower than that for PDS lE aM IF (6.0E4)2) because for PDS Ill, the containment sprays are not in operation after ressel breach so the atmosphere is more likely to be inerted. For FISSION PRODUCT REhtOVAL h1ECilANIShtS FAIL, FPR4= 1.0 is higher than the value for PDS lE and IF because there is no overlying pool at the time of containment failure, his is because for PDS 111 the RWST water is not injected into the containment. Revolatization also occurs along with , non retention in this case, but the lack'of an overlying pool at time of containment failure dominates this probability. For'CONTAINhtENT FAILURE htODES, CFht2=5.04E-01 and CFht3=5.004E-01, if the failure is , early and CFhil =5E 03, if it is " ate. These split fractions simply reflect the fact that an energetic failure e likely to result in a runture r se failure and a puff release, while a late failure is likely to result in - a leakage type failure. 4-185
The combination of these branch probabilities results in a 0.996 probability of containment failure. his is much higher than for PDS !E and IF bec use the debris was found non-coolable due to the fact the , f RWST water was not injected into the containment. Thus, basemat melt through, steam overpressure and late hydrogen burns would fail the containment with high likelihood. There is a 7.71E 3 probability of early containment failure. All of the late failures are unmitigated since the debris is not covered at containment failure, but 99.5% of these late failures are benign, i.e., leakage as opposed to rupture. Conversely, almost all of the early failures are rupture failures leading to a puff release. I PDS 2E his PDS groups Sma'l Break LOCA with both ECCS and containment spray failure at recirculation. De results for this PDS are nearly identical to those for PDS lE. The only differences in the branch-probabilities are that for FPR2= FPR4= 1 in this case as opposed to 0.998 for PDS 'E. These differences are not signincant so that PDS lE and PDS 2E have essentially the same characteristics. ED12E TM PDS groups Small Break LOCA with ECCS failure at recirculation and containment sprays that operate during injection and recirculation. The logic tree top event probabilities for this PDS are identical to those discussed previously of PDS IF except for evert DP, which depends on the cutset cornposition of the PDS. As in PDS 1F, the containment would never fait due to overpressure. PDS IF and PDS 2F thus have similar characteristics. PDS 2H nis PDS groups Small Break LOCA with ECCS failure at recirculation and containment spray failure at injection. Because of the CT failure at injection, the RWST water lasts a very long time delaying the
; time to core damage sign 10cantly, while continuously pressurizing the containment with steam.
Therefore, the time between vessel failure and containment failure becomes less for this PDS than for- . other PDS where the RWST water was injected into the containment, and the probability of late-containment failure due to steam generation, PRSTM-OCC, is higher in this case, i.e.,0.95 as opposed to 0.65. This conclusion is conservative since the actual containment failure time for this PDS is 36 hours (vessel failure at 18 hours) which is similar to the corresponding times for PDS lE (38 hours) and 2E (37 hours), but in those cases die vessel fails in 2 to 3 hours so the time after vcssel failure is much
-larger. Thus, it appears that more containment failures than would be expected to occur in this case are p
4-186 l
t computed, so dat results for PDS 211 are felt to be somewhat more conserntive than those for other . 3 PDS. He total containment failure probability for PDS 211 is 0.95. Of this,0.03 are early failures and 0.92 are late failures. The early failures are due to liPME occurring with the containment at a relatively high c initial pressure. The easly failures are split almost 50-50 into rupture and leakage type and the late ; failures are mostly leakage. Since the RWST water is injected into the containment in this PDS, the late failure probability is 0.85, mostly due to steam overpressure, with a small contribution from non-condensible overpressurization (0.07) when the debris is founti non-coolable with a.n overlying pool. ; PDS 36 This PDS groups Transients involving loss of all feedwater with Ril and Si failure at injection, while two CCPs may or may not be available but if they are available, they do not inject until after vessel failure due to the high head and containment spray failure at recirculation. As previously discussed, these PDS involve the RCS at high pressure, with only a 0.05 probability of depressurization to low pressure levels , due to hot leg or surge line failure. Recovery is not considered and, therefore, vessel failure always occurs. As a 'esult, the first releva. branch probability is DP=0.87, representing the 0.87 probability that the vessel will be at a high pressure at the time of vessel failure. The next relevant branch probability is CFEl = 8.0E-03, CONTAINMENT FAILS EARLY, along the successful DP path. This value is due to the probability of a low pressure alpha event. Along the non-9 ;ssurization path, CFE2= 1.08E-02. This result is due primarily to the chance that liPME might occur and fail the containment. There is a small contribution due to the rocket event and one due to alpha at high pressure. l The next relevant event is DC2=7 37E-02, DEHRIS BED NOT COOLABLE, along the successful DP path. This results from judgment probabilities: (1) that a coolable debris bed does not form given no ex vessel steam explosion (PRCDB-LPNS =. 0.1) times the probability that ex-vessel steam explosion does not occur (-PRVSE = 0.5), ORed with (2) the probability that', ex-vessel steam explosion occurs - ;
-4 187 ,
4 w~pw xr --e umw u w wnmwd+a -war-e w we-e w e -wthw"%- pa wee-'w W g -w wt-we w' + e y g- ,-6,4 gw.wa-p'w- r q w wget Ws y- "' -**e- g " v t t-Wt*y' mwiy -y v vy -h+==*'F
(PRVSE = 0.5) times the probability that a coolable debris bed does not form given that (PRCDil LPSE
= 0.05) an ex vessel steam explosion occurs. That is:
DC = l(0.1x0,5) + (0.05x0.5)) - ((0.1x0.5)s(0.05x0.5)] Along the non depressurized path, DC3=0.05, This value is due primarily to the judgement that it is very unlikely (0.05, Table 21, Ref. 8) that a coolable debris is not formed if the RCS is not depressurized before vessel breach, which then occurs in a blowdown mode, i.e, as high pressure melt ejection. For LATE CONTAINMENT FAILURE OCCURS, CFL1 = CFL3 = CFL5 = 3E=0.65,if the debris is coolable and CFLt a CFL4 = CFL6 = 3E=0.996, if it is not. ; For FISSION PRODUCT REMOVAL MECllANISMS FAIL, FFR=0.175 is low because of the probability judged to be 0.175 that RCS heat up could cause revolatization of fission products, while RCS retention occurs due to spray operation before vessel breach and with a small path from the RCS to the containment. i For CONTAINMENT FAILURE MODES, CFM=5.04E-01, if the failure is early and CFM=5E-03, if it is late. These split fractions simply reflect the fact that an energetic failure d likely to result in a - rupture mode failure and a puff release, while a late failure is likely to result in a leakage type failure.
'the combination of these branch probabilities results in a 0.67 total probability of containment failure.
l- There is an 9.13E 3 probability of early containment failure, where 22% are small or leakage type > failures and 78% are large or rupture failures. There is a total 0.66 probability of late containment ; failure. Of this 0.60, are due to steam overpressure. The remaining 0.06 are due mostly to non-
- j. condensible overpressure and with small contributions from basemat melt through and to 11 2 burn, if the debris is not coolable and CCI occurs under the overlying pool of water. Only 17,5% of both late and early failures are unmitigated since the releases are mostly due to revolatization. Of the late failures, 99.5% are benign.
4 188 :i 4
r !)
)
PDS_ 3F This PDS is similar to PDS 3E except that containment sprays operate during injection and recirculation and with the exception of the potential for operator depressurization, which is cutset dependent as implemented in 11E IlOP DP. Results for this event are quite similar to those discussed previously of PDS 3E ne difference between the PDS is in the duration of spray operation and the higher probability for event DP. In this case, the containment would never f all due to overpressure, while in the previous case it could. De early f ailure probabilities are slightly less than those for PDS 3E, narnely a 8.19E 3 probability of early containment failure, where 22% are small or leakage type failures and 78% are rupture type. De late failure probabilit) is only 0.07 due to the possibility that the debris ruay fall into a non coolable configuration even if the cavity is full of water. The total contal.anent failure probability for PDS 3F is 0.08. PDS 3H his PDS groups Transients involving loss of all feedwater with Ril and Si failure at injection, while two CCPs may or may not be available but if they are a$ailable, they do not inject until after vessel failure due to the high head and containment spray failure at injection. As previously discussed, these PDS involve the RCS at high pressure, with only a 0.295 probability of not successfully deprest,urizing. , Recovery is not considered and, therefore, vessel failure always occurs. As a result, the first relevant branch probability is DP=0.295, representing the 0.295 probability that the vessel will be at low pressure at the time of vessel failure. The next relevant branch probability is CFE=8.0E-03, CONTAINMENT FAILS EARLY, along the successful DP path. This value is due to the probability of a low pressure alpha event. Along the non. depressurization path, CFE=1.08E-02. This result is due primarily to the chance that ilPME might occur and fail the containment. There is a small contribution due to the rocket event and one due to alpha at high pressure. The next relevant event is DC= 1.0, DEBRIS BED NOT COOLABLE. This results from assumption [ .. . that the debris bed is not coolable if the RWST water is not fully injected into the containment and the I debris dries out. l l 4 189 l
For LATE CONTAINh1ENT FAILURE OCCURS, CFL1 = CFL3 =CFL5 =0, if the debris is coolable and CFL2 a CFL4 = CFL6 = 0.996, if it is not. If the debris is coolable, there is no coatainment failure becaute there is not enough II, generated in the core melt to threaten the containment in a burn and the amount of s,eam generated will cause containment failure without the non condensiblu in the CCI. If the debris is not coolable, there is CCI with enough attendant 11, gene ated to yield a 2.49E-03 probabilky of a burn with enough energy to f ail the containment, which must be combined with the 0.995 probability of fallute due tc non condensible overpressuritation and with the 0.05 probability of basemat melt through he burn probability is rnuen lower than that for PDS 3E and 3F (6.0E-02) because for PDS 311, the containment sprays are not in encration alter vessel breach so the atmosphere is more likely to be inerted. For FISSION PRODUCT REh10 VAL MECil ANIShtS FAIL, FPR4 = 1.0 is higher than the 0.175 value for PDS 3E and 3F because there is no overlying pool at the time of containment failure. This is because for PDS 311, the RWST water is not injected into the containment. Revolatization also occurs along with non retention in this case, but the lack of an overlying pool at time of containment failure dominates this probability. For CONTAINh1ENT FAILURE h10 DES, CFh12 = 5.04E-01, if the failure is early and CFhil =5E-03, if it is late. Dese split fractions simply retlect the f act that an energetic failure is likely to tesult in a rupture rnode failure and a puff release, while a late failure is likely to result in a leakage type failure. He combination of these branch probabilities results in a 0.996 probability of containment failure. His is higher than for PDS 3E and 3F because the debris was found non-coolable due to the fact that the - RWST water was not injected into the containment. Thus, basemat melt through, steam overpressurc :md late hydrogen burns would fail tne containment in high likelihood. There is a 8.83E-3 probability of early containment failure hnd a 0.987 probability oflate failure. All of these late failures are viimitigated since the debris is not covered at containment failure, but 99.5% of these late failures are benign, i.e., leakage as opposed to rupture. PDS 4F Dis PDS groups Transients where the TDAFW operates for 4 hours after reactor trip and two CCPs inject on de.nand but fail at recirculation and containment spray also fails at recirculation. The results for this PDS are similar to those for PDS 3E. The main differences in the branch probabilities are: (1) 4 190 Maimm m pusu gm igaq -i - -
. . . ,,,,,,,,,,y
the CFL values for the coolable cases are higher for 4E because it takes longer to fail the containment after vessel failure and (2) FPR2 = FPR4 = 0.438 in this case as opposed to 0.0 for PDS 3E, due to the fact that, while there is an overlying pool in this case to provide scrubbing of material that ends up - in the cavity, containment sprays do not operate after vessel breach and, thus, RCS retention does not occur. De condensed end state probabilities which are listed in Table 4,6-8 are similar to those for PDS 3E. He only significant difference is in the late containtnent failure probabilities due to steam overpressurization which are .37 for PDS 4E and .65 for PDS 3E. De difference is due to fact that the containment pressure reaches its mean failure pressure later for PDS 4E than for 3E because the TDAFW delays co'e melt and extends the duration of the ECCS. PDS 4P nis PDS is similar to PDS 3F except that ECCS injects successfully in this cue and fails only at recirculation, in both 3F and 4F, containment sprays inject and switchover successfully to recirculation, so there is no containment failure due to steam overpressurization. De CET end state probabilitics for this event are identical to those discussed previously of PDS 3F except for event DP which is cutset dependent due to llE ilOP DP. As in PDS 3F, the containment would never fail due to overpressure. PDS 411 This PDS groups Transients where the TDAFW operates for 4 hours after reactor trip and two CCPs inject on demand but fail at recirculation and containment sprays fail at injection. - The cendensed end. state probabilities that are listed in Table 4.6 8 are similar to those for PDS 4E. He main difference are in the early failure probabilities,0.16 versus 9,13E-03, and in the late containment failure probabilities . due to steam overpressurization which are .37 for PDS 4E and .19 for PDS 411. The former difference is due to differences in llPME failure probabilities i od probabilities of operator action to depressurize. The latter difference is due to fact that the containn.ent pressure reaches its mean failure pressure later-for PDS 411 than for 4E because the failure of containment sprays extends the duration of the ECCS injection period. 4 191
= - , u ,..------,--n--g-, ,,.-,w, ann,n.m,,, , . , , . , + , .__w.mw,,.,,,_nm,s,..w,..--n, _,,_,,-+wn.,, , , - ,,-.--w,,_, e-- vre , , . w--,,,- r m.,--,,m,.w._,,,+a
i i i PDS 3SBO.AND 4SBO i This PDS groups Station Blackouts involving simultaneous loss of all feedwater (3SBOllR) and with 4 hours of auxiliary feedwater supplied by the TDAFW pump (4SBOllR). As previously discussed, these PDS involve the RCS at high pressure, with only a 0.068 probability of depressurimion to low pressure levels for 4SBO due to hot leg or surge line failuie and 0.98 for 3SDO due to operator action. As a result, the first relevant brrnch probability is DP=0.387 for 4 SDO and DP=0.0225 for 3SBO, representing the probability that the vessel will be at high pressure at the time of vessel failure. The next relevant branch probability is CFEl =8.0E-03, CONTAINMENT FAILS EARLY, along the successful DP path. This value i; due to the probability of a low pressure alpha event, Along the non-depressurization path, CFE2=2.706E41 for 3SBOllR and CFE2=5.078E.02 for 4SBOllR. His result is due primarily to the chance that 11PME might occur and fail the containment. Dere is a small contribution due to the rocket event and one due to alpha at high pressure.
'lhe next relevant event is DC3= 0.901 for 3SilOllR, DC3u0.091 for 4SBOllR and DC2= 0.904 for 3SilOllR, DC2= 0.991 for 4SBOllR, DEBRIS lied NOT COOLABLE. This resu!ts from assumption that the debris bed is not coolable if the RWST water is not fully injected into the containment and the debris dries out, combined with the probability of recovery of electrical power which would allow injection of the RWST into the containment to occur, if the RWST is injected into the containment, DCI = 7.38E42, because the dAls is coolable if it is geometrically coolable, Since this is along the path of successful recovery, the ECCS was injected into the containment. The value of 7.38E42 results from the probability that a low pressure steam explosion
- may occur times the probability of forming a coolable debris bed in that case, ORed with the case were the bed is formed without the occurrence of a steam explosion at low pressure (Section 4.5). Top event DCl, Figuru 4.5 7, applies only along the path of successful recovery.
For LATE CONTAINMENT FAILURE OCCURS, CFL1 = CFL3=CFL5=0,-if the debris is coolable and CFL2 = CFL4 = CFL6 = 0.996, if it is 'not, if the 'ebris is coolable there is no containment - failure because there is not enough 11 3 generated in the core melt to threaten the containment in a burn and the amount of steam generated will cause containment failure without the rion condensibles of the CCI. If the debris is not coolable, there is CCI with enough attendant lir generated to yield a 2.49E-03 1 4 192 c - -
,,,--m,r ,,, s ,I #w - ,,-e-, , , , - , - + - + wy ,%.r. --.,v.-. ,.v-.,- o.~, - ,,,y
f probability of a burn with enough energy to fail the containment, which must be combined with the 0.995 ! probability of failure due to non condensible overpressurization and with the 0.05 probability of bascmat melt through. . For FISSION PRODUCT REMOVAL MECliANISMS FAIL, the discumon in the previous Section covers the issues. For CONTAINMENT FAILURE MODES, CFM2 =5.04E-01, if the failure is early and CFMl =$E-03, if it is late. These split fractions simply tellect the fact that an energetic failure is likely to result in a rupture mode failure and a puff release, while a late failure is likely to result in a leakage type failure. l, The combination of these branch probabilities results in a 0.82 probability of containment failure for 3SDOllR and 0.98 for 4SBOllR. There is a 1.3% probability of early containment failure for 3SBOllR and a 2.5% probability of early fallr<e for 4SBOllR. The difference is due to the higher initial pressure for 3SilOHR which yields a higher chance for a llPME event, but which is compensated by_ the depressurization probability. On the other hand, the late failure probabilities are 0.G0 for 3SBOliR and 0.95 for 4SiiOl{R. The difference here is in part due to the probability of recovery which is higher for 3SilOllR because the time window begins at an earlier time (vessel failure) where the recovery probability curve is less flat and, in part, due to the deptessurization probability. All of these late failures are due to CCI related effects and are mostly unmitigated since the debris is not covered at containment failure, but 99.5% of these late failures are benign, i.e , leakage as opposed to rupture. PDSSE
'Ih PDS groups Large Break LOCA with ECCS failure at injection and containment spray failure at recirculation. As previously discussed, these PDS always involve the RCS at low pressure, recovery is ,
not considered and, therefore, vessel failure always occurs. As a result, the first relevant branch probabdity is CFEl = 8.0E-03, CONTAINMENT FAILS EARLY, This value is due to the probability of a low pressure alpha event. l
' The next relevant event is DC2=7.37E-02, DEBRIS DED NOT COOLABLE. This results from judgment probabilities: (1) that a coolable debris bed does not form given no ex vessel steam explosion ,
4 193
*'w,mry-m wwyy - rm,-y- +~--* y- 'y-py- + r wg gg7-a v- g r e
(PRCDil LPNS = 0.I) times the probability that ex vessel steam explosion does not occur ( PRVSE = 0.5), ORed with (2) the probability that an ex vessel steam explosion occurs (PRYSE = 0.5) times the probability that a coolable debris bed does not form given that (PRCDU LPSE = 0.05) an ex vusel steam explosion occurs. That is: DC = [(0.1 x 0.5)+(0.05 x 0.5)).l(0.1 x 0.5)(0.05 x 0.5)) For LATE CONTAINMENT FAILURE OCCURS, CFL3 =0.65,if the debris is coolable and CFL4 = 7.3811-02, if it is not. For FISSION PRODUCT REMOVAL MECil ANISMS FAIL, FPR2= FPR4=0.95 is high because of the probability, judged to be 0.95, that RCS heat up could cause revolatization of fission products, although RCS retention occurs due to spray operation before vessel breach. : l For CONTAINMENT FAILURE MODES, CFM2 =5.04E-01, CFM3 = 5.004E-01, if the failure is early and CFMl =5E-03, if it is late. These split fractions simply reflect the fact that an energetic failure is likely to result in a rupture mode failure and a puff release, while a late failure is likely to result in a leakage type failure. As shown in Table 4.610, the combination of these branch probabilities results in a 0.68 probability of containment failure. There is an 8.0E-3 probability of early containment failure and a 0.67 probability oflate containment failure, of which 0.60 corresponds to overpressure failure due to steam generation - and 0.07 corresponds to CCI related mechanisms: (a) overpressure due to non-condensibles, (b) basemat T melt-through and (c) !!2burn if the debris is not coolable and CCI occurs under the overlying pool of water. Ninety five percent (95%) of these late failures are unmitigated since the releases are mostly due to revolatization, but 99.5% of these late failures are benign, i.e., leakage as opposed to rupture. PDSSF This PDS groups Large Break LOCA with ECCS failure at injection and containment sprays that operate during injection and recirculation. l l 4 194 wg,y--, ag.s e-y,y- gg,4,9--.--.y..op.,, g ,, 9m,., y,,9q., p- yi .i-p. , ,-w p. y..e mg- y. mtr upp-r-wig g t' &-q,. cmp-- m.,ewgv r y- g,yga 9 .g g-w--FWy e v + m - r e g wp v- pwp-M veriMy yf*v<- eur-p+1,-+T+v+T-+- gPM-- p -p t "
i ne CET for this event is quite similar to that discussed previously of PDS SE, The difference between the PDS is in the duration ol spray operation, in this case, the containment would never fait due to overpressure while in the previous case it can. Thus, the early failure probabilities are the same as those for PDS SE, namely a 8E 3 probability of early containment failure, and all are rupture type since this is the alpha failure. De late failure probability is only 0.07 due to the possibility that the debris may fall into a non coolable configuration even if the containment is flooded and then fail by the same mechanisms presented for the non-coolable situation in PDS SE. De total containment failure probability for PDS 5F is 0.08. PDSSH This PDS groups Large 13reak LOCA with ECCS and containment spray failure at injection. As previously discussed, these PDS always involve the RCS at low pressure, recovery is not considered and therefore, vessel failure always occurs. As a result, the Drst relevant branch probability is CFE=8.0E-03, CONTAINh1ENT FAILS EARLY. This value is due to the probability of a low pressure alpha event. The next relevant event is DC2= 1.0, DEBRIS 13ED NOT COOLABLE. His results from the astumption that the debris bed is not coolable if the RWST water is not fully injected into the containment i and the debris bed dries out prior to containment failure. For LATE CONTAINhlENT FAILURE OCCURS, CFL3=0, if the debris is wolable and CFL4 = ., 0.945, if it is not, if the debris is coolable, there is no containment failure because there is not enough l H 2generated lu the core melt to threaten the containment in a burn and the amount of steam generated will not cause containment failure without the non condensibles in the CCI. If the debris is not coolable, there is CCI with enough attendant H: generated to yield a 2.49E-03 probability of a burn with enough energy to fail the containment, which must be combined with the 0.95 probability of failure due to non-condensible overpressurization and with the 0.05 probability of basemat melt through, n e burn i probability is much lower than that for PDS 5E and 5F (6.0E-02) because for PDS SH, the containment , I- sprays are not in operation after vessel breach so the atmosphere P more likely to be inerted. For FISSION PRODUCT REh10 VAL h1ECH ANISh1S FAIL, FPR4= 1.0 and FpR2=0.998 are higher i than the 0.95 value for PDS 5E and SF because there is no overlf ing pool at the time of containment l I 4-195 1 , i l' d __
failure. His is because for PDS 511 the RWST water is not injected into the containment. Revolatization also occurs along with non retention in this case, but the lack of sn overlying pool at time of containment failure dominates this probability. f For CONTAINh1ENT FAILURE h10 DES, CFht2 = 5.04E 01, CFht3 =5.004E-01, if the failure is early and CFhil=5E-03, if it is late. These split fractions redeet the fact that an energetic failure is likely to result in a rupture mode failure and a puff release, while a late failure is likely to result in a leakage type failure. The combination of these branch probabilities results in a 0.995 probability of containment failure. This is much higher than for PDS SE and 5F because the debris was found non-coolable due to the fact the RWST water was not injected into the containment. Thus, basemat melt-through, steam overpressure and late hydrogen burns would fall the containment with high likelihood. There is an 8E 3 probability of early containment failure from a low pressure alpha event. All of the late failures are unmitigated since the debris is not covered at containment failure, but 99.5% of these late failures are benign, i.e., leakage as opposed to rupture. Conversely, all the early failures are rupture failures leading to a puff release, the same as for SE and 5F. PDS 6E This PDS groups Large Break LOCA with both ECCS and containment spray failure at recirculation. The results for this PDS are nearly identical to those wr PDS SE. The only differences in the branch - probabilities are for the FPR values. These differences are not significant so that PDS SE and PDS 6E have essentially the same end-state probabilities. , PDS 6F This PDS groups Large Break LOCA with ECCS failure at recirculation and containment sprays that l operate during injection and recirculation, l ne end state probabilities for this event are identical to those for PDS 5F, As in PDS 5F, the containment would never fail due to overpressure. PDS 5F and 6F have the same characteristics. 4-196
-+s--rt- Lwac q -
PDS 611 ; This PDS groups Large Break LOCA with ECCS failure at recirculation and containment spray failure f at injection. The CET split fractions for this PDS are nearly identical to those for PDS 6E. The only signincant differencc is in the late containment failure probabilities due to steam overpressurization which 1 are .60 for PDS 6E and .64 for PDS 6H. The difference is due to fact that the failure of containment sprays extends the duration of the ECCS injection period for PDS 611 versus 6E, delaying the vessel failure time with minimal effect on the containment failure time resulting in a smaller time interval between vessel and containment failure times. . 4.6.2 Rankine Accident Sequences and Accident Procression Analyses The ranking of the accident sequences
- risk importance was done according to their unconditional probabilities of the core melt leading to: (a) early containtnent failure, (b) late CCl induced failure and (c) late steam induced failures. The unconditional probabilities were obtained by combining the PDS probabilities of occurrence (core melt frequencies or CMF) with the conditional probabilities (i.e., ghen the occurrence) of each PDS leading to: (a) early containment failure, (b) late CCI induced failure And (c) late steam-induced failure, respectively. The PDS probabilities of occurrence were taken from the bottom line of Table 4.3-3. The conditional probabilities for (a), (b) and (c) were determined in the containment event tree analysis (Ref. 2) summarized in Section 4.6.1 and in Tables 4.6-7,4.6-8,4,6-9.
and 4.6-10. The combination of those results is listed in Table 4.614, which is sorted in decreasing order of PDS occurrence probability,i.e., core melt frequency (CMF). Tables 4.6-11,4.6-12 and 4,6-13 provide the key to the notation describing the sequences, in Table 4,6-15, the PDS are sorted according to the unconditional probabilities of early containment failure, and total unconditional containment failure probabilities. _ Table 4.6-16 sorts the PDS in decreasing order of the late CCI-induced failure and late steam induced failure. - This is done by extracting the relevant columns from Table 4.614 and resorting. In Table 4.6-17, the PDS are given in decreasing order of contribut'on to early containment failure, along with their representative Level I functional sequences, which are taken from Table 4.3 3. Tables 4.6-11, 4.6-12 and 4.613 provide the key to the notation describing the sequences listed in Table 4.6-17, 4-197 . , - -- .- - ._ . - - - . ,. - - _ - - . - .-_-_a . _ - ,
l For all of the PDS listed in Table 4.3-3, the following is discussed: (1) the reasons for the PDS contribution to risk, (2) identification and description of the front-end functional sequences making up most of the PDS frequency, and (3) selected accident progressions for these sequences obtained with h1AAP. It is important to keep in mind that a particular front-end sequence may progress in various ways throughout the back-end portion of the analysis. In fact, that is what the CETs represent. De objective of the hiAAP analyses is to obtain representative and/or bounding calculations for each of the CET end-states. As a result, the front-end sequences used as the basis to construct the hiAAP analysis are not always the dominant sequence (i.e., the highest frequency contributor) in the PDS. Another point to remember is that most of the CET end-states do not correspond to the most likely outcome and require adjustrnents of htAAP parameters for the calculation to iesult in that outcome, for example - obtaining an early containment failure due to DCil combined with a hydrogen burn. Each of these points is made below again on a case-by-case basis. Then for each PDS, functional sequences and select h1AAP accident progression analyses are described in the following subsections. 4.6.2.1 PDS LCIL{SDTR & ISGTR) PDS 2CB contains all Steam Generator Tube Rupture (SGTR) functional sequences that lead to core melt. In addition, it also groups the traction of other PDS that result in induced SGTRs (ISGTR). The functional composition of this PDS is: 2CB = 95.5% RChil,3% RCA14,1.5% ISGTR The ISGTR frequency was calculated by subtracting the probability of depressurization in the CET (1-(top event DP probability)) as calculated for the base case (i.e., the case where the tubes could fail, BE PRSGOK =.982) from the case where the tubes are intact (BE PRSGOK = 1.0). This difference is the probability that he depressurization is due to the induced failure of the tube.s and is listed in the left 1,and column of Table 4.618, for every high pressure PDS (311, 3F, 411, 4F, 3SBO, 4SBO). This is multiplied by the PDS frequency to obtain the unconditionalISGTR frequency for the PDS. The results of this calculation are summarized in Table 4 6-18. By consulting the functional sequence composition of each PDS, it is possible to determine the functional sequence contributions to ISGTR. These are listed in Table 4.6-19. T1.e total frequency of ISGTR does not justify a separate PDS for this type of development, and since the release paths are identical, these sequences are binned with SGTR Although the functional composition of the ISGTh is provided in Table 4.6-19, a special analysis of any of those 4-198
__ ___ _ . . _ - . _ . _ _ _ . ___._________._m sequences is not wuranted because of the comparatively low frequencies with respect to other SGTR. i' Dus, the representative sequence for PDS 2Cil is RCM1, which accumulates the overwhelming majority I of the frequency,95.5% as shown above. Description of he Representative Functional Sequences RChil: This sequence involves a steam generato. tube rupture initialmg event. ECCS is available and safety injection is established. Following the 'S' signal, secondary heat removal is established, llowever, the operators are not able to isolate break flow. This results in RWST depletion. Following that, operators are unable to perform closed loop recirculation switchover due to their own errors or power or equipment failures. Containment isolation is lost due to failure of atmospheric dump valves to rescat properly after steam generator overfill. RCMi: This is an SGTR where ECCS failed to inject due to failure of the CS and Si equipment. Following the generation of an "S" signal, secondary heat removal is established, llowever, the operator fails to t terminate break flow within the 30 minute time window. 4 hibAP Calculations for The Representative Seouencu One calculation was perforned to determine the accident progression and source t;rm characteristics of the L9y sequence: calcu!ation 2CBI representing RCM1 described above, which represents 95.5% of the tube ruptures. Six additional tube rupture calculations are documented elsewhere (Ref. 2), and the case of RCM4 is considered there as well (SGTR03). However, since RCMi has almost all the tube ruptures, it was selected as the representative sequence for PDS 2CB. 1CD1 The calculation follows the accident description above for RCMI. Additior.al details are a. icllows: (1) MSIVs close at time of scram, (2) all AF is postulated to fail, and (3) a pre-calculation showed overfill-occurs around 2 hours so the ARVs are postulated to fail to reseat at that time. 4 The accident progression calculation plots are given in Appendix A of Reference 7. Key times and pressures are summarized in Table 4.6 20. The RWST is depleted around 10 hours. Wen, RCS - pressure drops as there is no longer CCP injection to keep it up. The accumulator set point is reached 4-199
, .i _- _ ___ . _ _ . _ , , -. _ , _ ,2 . __ _ _ _ _ _ _ , -
and they become depleted. Shortly after, RCS pressure reaches the broken SG pressure of about 2 MPa. After that, the broken SG drains into the RCS by gravity, delaying core uncovery foi about 15000 seconds. When the gravity drain is over, the primary pressure rises again due to the increased steam production. Eventually the core begins to uncover and steam production decreases, reducing the RCS pressure. % sel failure follows shortly after core uncovery and occurs at an intermediate pressure of 5 MPA (72; ,. The 2CBI calculation is similar to calculation SGTR06 done in Reference 2. In the actual sequence, operators isolate the AF to the faulted generator, but AF is available to the intact generator. Since MAAP cannot isolate AF to only one SG, AF is postulated to be lost for all steam generators. A case with A R available for all steam generators, SGTR02, was also run for comparison in Reference 2. That case led to qualitatively similar results as case SGTR06 except core uncovery occus V hours sooner when there is AF available. However, that difference is primarily due to the fact that with AF available to the faulted SG, the CST keeps being fed into it. Therefore, the gravity drain from the faulted SG lasts l 13 hours longer. Thus, case 2CBI is somewhat conservative because it neglects the benefit of having AF to the intact SG!. and, furthermore, it does not provide CST water feed to the RCS through the faulted - l SG. Table 4.6-21 summarizes the release fraction for PDS 2CB based on the 2CBI MAAP calculation. l 4.6.2.2 PDS 4H and 4F PDS 4H and 4F rank second and seventh, respectively, as contributors to early containment failures, in addition to being the second largest contributor to early containraent failures, PDS 4H is also the second largest contributor to late steam-induced overpressure failures. PDS 4F does not contribute to steam overpressurization because the containment sprays work. The cause of the early failures is primarily HPME. The high HPME delta P resulted from the use of NUREG/CR-4551 pressure rise values, as shown in Table 4.6-1 The table entry is based on a high RCS pressure at vessel failure of 2000 to 2500 psi, as described in Section 4.6.1. _Descrintion of The Renresentative Functional Sequences The functional sequence frequency composition for PDS 4H and 4F is: 4H: 9',% IVSCM4,5% VSCM4,1% IVSCMI. 4F: 83% VSCM(14),9% IVSCM4,6% X(23)CMI. 4-200
'De prefix notation for the above sequences is given in Table 4.6-11 and the suffix notation is given in Table 4.6-12.
IVSCM4: The main cutsets in this sequence involve a loss of offsite power (loss of CC and SW also contribute to this category), whkh induces a very small seal LOCA greater than 21 GPM/PMP but less than 60 GPM/PMP 45 minutes after the initiating event ECCS (CCPs and SIPS) is unavailable for establishment of safety injection due to loss of power or due to either: (a) loss of pump lube oil cooling on failure of SW or (b) failure of SW itself. Auxiliary feedwater i:. available. IVSCMl: This sequence is similar to IVSCM4 except that auxiliary feedwater is not available due to less of turbine driven AF pump discharge control leading to steam generator over fill and eventual turbine drowning. The MDAFW pumps treiost due to either the initiator directly or due to failure of components on loss of room cooling. Feed and bleed was s' ;cessfully completed, but recirculation was not established due to the initiator. The RCPs trip either at the time power is lost or on loss of AF by procedure FRH-0.1. Containment sprays failed at injection due to the loss of power or loss of supporting systems. VSCM4: The main cutsets in this sequence involve a very small break LOCA imtiating event (less than 0.6 inch). High head injection (charging) fails due to the loss of various components associated with the Chemical and Volume Control System. Auxiliary feedwater is availab!e for at least 24 hours. Containment spray is available in some cut sets. The operators fail to cool and depressurize the plant on loss of high head . injection. The RCPs will be tripped if RCS subcooling is less than 25 *F and an ECCS pump, either one CCP or SIP, is operating. (Note: CCPs are unavailable for this sequence and if the RCS stays at bigh pressure, the SIPS will be unavailable). Upon initiation of containment splay (some cutsets have sprays failed), procedure EOP-0.0A calls for the operator to trip the RCPs
.VSCM I The main cutsets in this sequence also involve a very small break LOCA initiating event (less than 0.6 inch). In this case, ECCS is available for injection. Auxiliary feedwater is available for at least 24 hours; Containment spray is av.;iable as' required. The operators fail to establish recirculation by failing 4-201 ;,-v - , - - . . , - - . - ,,-,, ,
to re-align ECCS upon RWST depletion. The RCPs will be tripped if RCS subcooling is less than 25
*F and an ECCS pump is operating, either one CCP or SlP. Upon initiation of containmeni spray, procedure EOP-0.0A calls for the operator to trip the RCPs.
X(23)CMl: This is an initiating event (loss of billed water, loss of offsite power) that did not induce a seal LOCA. . g,. Thus a 21 GPMIPMP leak rata is postulated consistent with the thermal gradient leakage associated with
- D loss of seal cooling where no RCP 3eal parts fail (Refs. 2.32). The MDAFW pumps fait due to a loss y of supports or due to heat related failures associated with a loss of room cooling. The TDAFW pump M
'(g fails due to its own faults or due to its supports which lead to an overtill condition that eventually drowns _ u 4% the turbine or fails its piping. Operators appropriately and successfully performed feed and bleed until Y the RWST is empty, but then recirculation fails due to the loss of chilled water which fails cooling to the RH pump room, or by the failure of the pumps supports. The RCPs were tripped at the loss of AF by procedure FRH-0.1. Sufficient ECCS injects on demand but fails at reeirculation due to the loss of the RH pumps caused by a loss of supports or room cooling. By procedure FRH-0.1, operators will shutdown ECCS pumps, starting with the CCPs and followed by the SIPS, to preserve the RWST water as long as subcooling is maintained in the RCS, according to the criteria provided by the procedure.
}1 A AP Calculatians for The Representative Sequense PDS 4H:
From the thermohydraulies and accident progression perspectives, IVSCM4 and VSCM4 are similar.
'lle on!" difference is that in one case, the leak is induced and in the other, it is the initiator. The IVSCMI functional sequence is more severe for early containment tailures because ECCS injection is successful and leads to a higher containment pressure at vessel tailure. Therefore. it is picked as the basis for the representative MAAP calculation.
Two calculations were carried out for PDS 4H: (1) TRAN21, corresponding to the most likely outcome of sequence IVSCMI rmt m TRAN22, where model parameters are adjusted to obtain HPME failure. IILA32_L. This base case follows the sequence description abose for IVSCMI with the following additional c.ssumptions: 4-202 l
l
- The seal leak rate is less than 60 GPM/PMP for IVSCMI and 21 GPM/PMP for the MAAP calculation. That difference is not significant but since this is a high pressure PDS, the lower leak rate is slightly more representative.
- The RCPs and AFWPs are tripped simultaneously with the reactor for simplicity.
- One CCP is the only pumped ECCS. The success criteria is I of 4 pumps. This choice delays core damage while pressurizing containment leading to a maximum containment pressure at vessel failure. Since this sequence represents PDS 4H, which is a contributor to early HPME failure, it is appropriate to consider this situation as representative since it is the mot. nducive to early containment failure and does not impact the ! ate failure time due to steam overpressurization.
h
- The PORV is closed after the RWST is dephted. Since tne operators are executing feed and bleed, after they are no longer able to feed, operators will stop the bleed.
- When the containment failure pressure is reached, the containment failure mode is assumed to be by liner tear as discussed in Section 4.4. In that failure mode, the leak rate increases to a point where containment pressure is nearly constant or decreasing slow!v, i.e., a leak type failure mode, This is accomplished in this sequence with a 0.005 m liner tear leak area, which is approximat:ly equivalent to a 3 inch diameter oritice.
The key event times and pressures for this analysis are summarized in Table 4.6-20. Calculated containmem pressure and temperature histories are presented in Figure / 6-2. Additional information on s sequence development is presented in the corresponding Figures in Appendix A of Reference 7. T_R A N22: Since PDS 4H is the number two contributor to early containment failure (HPME), MA AP model par meters were modified outside of their estimated values in order to obtain release fractions for the HPME tailure mode. The parameter modifications for TRAN22 are identical to those discussed for SB2H5 follow the guidelines suggested in References 33 and 34. The key tvent times for this analysis are >ummarized in Table 4.6-20. Calculated containment pressure and temperature histories are preseated in Figure 4.6-3. Additional information on sequence development is presented in the corresponding figures in Appendix A of Reference 7. 4-203 l
PDS 4F: -C The frequency composition of PDS 4F is: 83% VSCM(14),9% IVSCM4,6% X(23)CMI. From the descriptions provided above, it is clear that IVSCMI and VSCM1 are essentially identical from the thermohydraulle and phenomenological perspective and so are IVSCM4 and VSCM4. Since IVSCMI. was used as the basis for the representative MAAP analysis of PDS 4H, it is appropriate to examine VSCM4 for PDS 4F. This is done in VSB4Fl. M 4_Eh This case follows the sequence description above for VSCM3 with the same additional assumptions made-for TRAN22 to obtain HPME phenomena. Since PDS 4F is a contributor to early containment failure due to HPME, MAAP model parameters were modified outside of their estimated values in order to obtain release fractions for this failure mode. The parameter modifications for VSB4F1 are identical to those made for SB2H5 which folbwed the guidelines suggested in References 33 and 34 The key finding from the analysis is that it was not possible to obtain containment failure due to liPME even after the parameter adjustments because of the low initial pressure due to spray actuation. In conclusion, since the CET end state corresponding to early containment failure with sprays available is represented by the 3SBO with recovery sequence IVSCM6 (M A AP calculation SB031), and since another functional sequence which represents 4F has been analyzed as TRnN21 and TRAN22, only without-sprays, it is not necessary to have a core melt sequence MAAP calculation specifically for 4F, Table 4.6-21 summarizes the release fractions for CET end-state B6-L, represented by PDS 4H, based on the TRAN21 MAAP calculation. Table 4.6-22 describes that end-state and lists similar end-states. 4.6.2.3 PDS ICB (V-Secuence) and ICI (Isolation Failure 9 PDS 1CB ranks third as contributor to early containment failures. It is not a high contributor with a frequency of 1.19 E4)7 but is listed and analyzed for the sake of completeness. This section also includes PDS ICI representing the cere melt sequences in which the containment failed to isolate, because those frequencies are so low (9.92 E-09) that they do not warrant their own representative analyses and car, have their consequences conservatively represented by the calculation used to represent PDS ICB. The reason for the low frequency for isolation failures is the independence of the containment isolation from
.4-204 .y, -, . , _ . . - . . _ , . . . _ _ _ g.-s-,..
l I 1 the support systems failures that lead to most of the core melt sequences. As a result, the containment isolation failures are mostly the product of the conditional containment isolation failure probability as determined in Reference 6 and 11. frequency of each of the sequences leading to core melt. Thus, the
' dominant contributor to this PDS is the dominant contributor to the core melt frequency and is already -
analyzed elsewhere (PDS IH). Description of ne Representative Functional S tquences ne functional sequence frequency composition for PDS ICB is: ICB: 83% SCM1,9% MCM2,7% ACM(12). SCM I.: The cutsets in this sequence involve an interft.cing systems LOCA of 2 to 4 inches in equivalent size. Based on the analyses of References 2 and 7, this is also a low pressure PDS. Auxiliary feedwater is available for at least 21 hours, but is worth very little for this break size. The RCPs are tripped when RCS subcooling is less than 25'F (criteria denned by procedure EOP-0.0A), which is almast immediately. Suf6cient ECCS (1 of 4 pumps) injects on demand but fails at recirculation due to the operators failing to realign the systems or due to failure of components required for recirculation and because the water is not in the containment sump and, therefore, is not available for recirculation. By procedure EOS 1.1, operators shutdown ECCS pumps starting with the Si pumps followed by the RH pumps to preserve the RWST, as long as 25"F of subcooling is maintained in the RCS, but this situation is not realistically reached, The interfacing systems LOCA, by deOnition, provides a' breach of - containment leading to releases into the safeguards building. MCM2: The cutsets in this sequence involve an interfacing systems LOCA of 2-6 inches equivalent size. ECCS -
. is unavailable for establishment of safety injection due to failure of the accumulators (loss of level or- - pressure) or due to failure of ESFAS equipment to generate an SI actuation signal, followed by the operator failure to manually initiate a safety injection signal or provide for manual actuation of injection equipment. He success criteria for a L6 inch LOCA requires accumulator injection and either CCPs - (high head) or SIPS (intermediate head). The interfacing systems LOCA, by dennition, provides a breach of containment leading to releas, s into the safeguards building.
4-205
_ _ . . ~ . . _ _ _ _ _ _ _ _ . _ . . ._. a ACM2: _ , The cutsets in this sequence involve an interfacing systems LOCA initiating event of more than 6 inches equivalent size. ECCS is unavailable for e.,teblishment of safety injection due to the failure of the RH pumps directly or due to supporting systems (ie. SW, CC or Chilled Water). The success criteria for a large break LOCA requires RH (low head) ar .ccumulator injection and either CCPs (high head) or SIPS (intermediate head). The interfacing , stems LOCA, by definition, provides a breach of containment leading to releases into the safeguards building. ACMl: Same as above except ECCS is available for injection. MAAP Calculations for The lienresentative Seauences From the thermohydraulics and accident progression perspectives, the signincant differences between the sequences described above are the size of the break and availability of ECCS. In addition, the smaller break sequences are 88% RH pump seal failures where the leakage would be submerged in the safeguards building. Since the larger break and lack of ECCS makes the accident more severe, and since the overall frequency of the V-sequence is sufficiently small, the most severe sequence, namely ACM2, is selected to form the baseline for the development of the MAAP calculation: VI.
.Y_1; As described in ACM2 above, the calculation is a 10 inch large break LOCA in the cold leg into the safeguards i., jing. No ECCS injection is credited except for the accumulators. Containment sprays are also not credited since, in addition to their failure described chove, the containment pressure does not reach their set point.
This sequence leads to' the largest release traction of all due to the combination of large early vessel and ' containment failure and lack of water in the containment leading tt early CCI. The key event tunes and pressures for this analysis are summarized in Table 4.6-20. Calculated hydrogen production amounts are presented in Figure 4.64. Additional information on sequence development is presented in the corresponding Hgures in Appendix A of Reference 7. 4-206
. i
.- .- -._--= .- . - . . ._-. . . - . - - _ _ .- -_-
Table 4.6-21 summarizes the ;elease fraction for PDS 2CB based on the 2CBI htAAP calculation. Table 4.6-22 describes the end state and lists similar end-states. 'T 4.6.2.4 PDS lH and 1F (and SH. SF) PDS lH ranks fourth (2.3%) as a contributor to early continment failure (PDS 5H is sixteenth,0.1 %). However, PDS IH is the Grst (57.1 %) (5H is Ofth,1.5%) leading contributor to late CCI-induced failures and is a'so the leading contributor to total containment failures of any kind (46.7%) (5H is eighth,1.2 %). PDS 1F and 5F are not signi0 cant contributors to failures of any kind and have the same functional composition as PDS IH and 5H, resfectively. The only difference is that sprays work in PDS 1F and 5F, Since early failures with sprays operational are represented by another MA AP criculation (SBO31), it is not necessary to have an analysis for PDS IF and SF; their release fractions can be represented by that calculation as well. Since late CCl induced failures are dominated by PDS 111 and since PDS SH leads to a similar outcome, and since PDS 5H is not a significe.t contributor to early failures (0.1 %) and is only a small contributor to CCI failures (1.2%), a calcuiatmn for SH is also not necessary; it can be represented by the IH representative as well. Descrmtion of The Itentesentative Functional Seouences The functional sequence frequency composition for these PDSs is: lH: 100% ISCht2 IF: 93% ISCM2,7% SCM2 Sil: 65% ACM2,35% IMCM2 5F: 85% ACM2,12% MCM2 i 1KJd2: The majority of cutsets in this sequence involve a loss of offsite power, whici. induces a seal LOCA greater than 60 GPM/PMP (assumed 210 GPM/PMP, or approximately a 1 inch break) 45 minutes after the reactor trip. De remainder of the cutsets involve the loss of SW or CC initiators leading to the induced seal LOCA. All ECCS fails to inject on demand due to the loss of all power or supports. ! Containment sprays fail in some cutsets at injection due to loss of power, but are available in others. L 4-207 I t
SCM2* This is similar to ISCM2 except the small break LOCA is the initiating event rather than induced. ACM2. MCM2. IMCM2: , These are all large break LOCAs where the ECCS failed at injection. MAAP Calculations for De Representative Seauences ISCM2 is used as a basis for the MAAP analyses, primarily because it has 100% of PDS lH frequency. The objective of the present MAAP calculations was to obtain acc! dent progression parameters and release fractions for CCI-induced failures in which the sprays do r,at operate and in which the non-coolability of the debris is caused by lack of a water pool above, rather than a non-coolable geometry situation which was represented via PDS 3F. Two scenarios were developed using ISCM2 as a guideline for the MAAP calculations: (1) SB021,~ with a leakage-type late failure due to non-condecsible overpressurization and (2) SB022, with a rupture-type late failure due to a burn of the hydrogen generated in the CCI. SB021 This is the bcse case calculation to represent ISCM2, The sequence development is that described for ISCM2 above; no additional assumptions were needed for the period prior to vessel failure, which occurs early at around 2.5 hours. After vessel failure, the assumption is that the non-coolable configuration of the debris is the 75 mi of cavity lloor and not the unrealistic debris area of 7.5 m2 that was used for PDS 3F. In that case, the change was necessary to obtain CCI in the presence of an overlying pool. In-this case, since there is no ECCS injection, the RCS plus accumulator inventory eventually dries out and - CCI can be expected to occur with near certainty. The key event times for this analysis are summarized in Table 4.6-20. Calculated containment pressure and temperature histories are presented in Figure 4.6-5. Additionalinformation on sequence development is presented in the corresponding tigures in Appendix A of Reference 7. S.[102 2: In this calculation, a burn was forced to occur at approximately 29 hours by restarting calculation SB021 at that time with a low TAUTO = 465'K, the temperatuse for autoignition of the hydrogen. The pressure L 4-208
spike was sufficient to fail the containment. A rupture failure area was postulated for this calculation in
- order to evaluate release fractions for this case.
The key event times for this analysis are summarized in Table 4.6-20. Calculated containment pressure anu temperature histories are presented in Figure 4.6-6. Additionalinformation on sequence development-is presented in the corresponding figures in Appendix A of Reference 7. Table 4.6-21 summarizes the release fractions for CET end-states C6-L and C6-R, both represented by PDS IH and based on the SB021 and SB022 MAAP calculations, respectively. Table 4.6 22 describes that end-state and lists similar end-states. 4.6.2.5 PDS 3H and 3F PDS 3H and 3F rank fifth (1.5%) and sixth (1.0%) respectively as contribt. tors to early containment failures. PDS 3H is the second (32.2%) leading contributor to late CCI induced failures and to all failures (26.3%) as well. PDS 3F does not contribute to steam overpressurization because the containment sprays work, and it ranks fourth (1.7 %) in CCI induced late failures, which result only from the probability that the debris might fall in an non coolable configuration under a water pool, since the RWST water is injected into containment via sprays. The cause of the early failures is also primarily b HPME. The high HPME delta P results from the use of NUREG-il50 values for Zion based on a high RCS pressure at vessel failure of 7000 to 2500 psi, as tieseribed in Section 4.6.1 and illustrated in Table 4.6-1.
,Descrintion of The kenresentative Functional Seguences - The functional sequence frequency composition for PDS 3H and 3F is:
3H: 49% X(3617)CM2,45% T(146)CM2 and 5% (CVCM2 & !VSCM3). 3F: 87% ATCM(631),8% X(12)CM1 and 4%.T(6)CM2. The suftix and prefix notation for the above sequences are summarized in Tables 4.6-11 and 4.6-12.. I ATCM(36t h The cutsets in this sequence involve various initiators leading to an ATWS and can be broken inta two groups. The ATCM3 and ATCM6 group contains cutsets leading to core melt in which the iniiators-4-209
I cause, directly or indirectly, the loss of mdn feed water. The reactor fails to trip from either mechanical rod binding or the failure of the trip breakers (operator action to manually tiip the breakers also fails). For this group, AF flow is lost mainly due to component failures. For several cutsets, the combination ofloss of AF or PORVs willlead to RCS oveq)ressurization. In the second group of cutsets, (ATCht?) the reactor fails to trip due to the failure of the trip breakers X(1367: Chi 2: These are initiators (loss of a DC bus, loss of olfsite power, loss of CC, loss of SW) that did not induce a seal LOCA. Thus, a 21 GPhi/Ph1P leak rate is postulated consistent with the thermal gradient leakage associated with loss of 3eal cooling, where no RCP seal parts fail (Refs. 2,32). The hiDAFW pumps __ fail either due to loss of support systems or room cooling to either the pump rooms or to the pumps electric equipment rooms. The TD AFW fails at that time as well, due to its own faults or loss of control power. Operators fail to perform feed and bleed mostly due to operator error but some cutsets also include the fa::ure of the PORV to open. The RCP were tripped on loss of AFW by procedure FRH-0.1. There is no Si because the operators did not manually actuate Sl in this scenario. For sequences with initia.ing event X6 or X7, containment spray is not available due to supporting systems failures. A fraction of sequences with initiating events XI and X3 also have containment spray unavailable depending upon the equipment that failed due to the initiator. T(146)CM2: The initiating event for these sequences is either a general plant transient or a loss of normal feedwater. The AF pumps are lost due to their failures or for the MDAFW pumps, due to heat related failures associated with a loss of room cooling. The TDAFW pump may also fail due supports that lead to either loss of turbine steam supply or to an overfill condition and subsequent drowning the turbine or causing its piping supports to bred from static overloading. In cases of loss of support other than steam, the loss of safety chilled water (room cooling) leaJs to direct failures or indirect (electric power room cooling) failures, as safety chilled water requires manual actuation except in the case of Si actuation. Operators fail to perform feed and bleed mostly due to operator error, but some cutsets also include the failure of the PORV to open. Tia RCPs were tripped on loss of AF by procedure FRH-0.1. There is no SI because the operators did not manually actuate SI in this scenario. This bin's success / failure criteria is independent of ECCS injection. 4-210
A_ fraction of the cutsets in these sequences (those with initiating events TI, T4, or T6) also have containment spray unavailable due to failure of supporting system equipment. Another fracti< i of sequences with initiating events Tl and T6 have containment spray available. CVCM2: This is a loss of condenser vacuum initiating event. The MDAFW pumps fait either due to its own faults or due to a loss of support systems (including room cooling to the pump rooms or to the pumps electric equipment rooms). The TDAFW fails due to its own faults or loss of control power. Operators fail to-perform feed and bleed mostly due to operator error, but some cutsets also include the failure of the POF.V to open due to loss of power to the PORVs, The RCPs were tripped on loss of AF by procedu ; FRH-0.1. 'Diere is no SI because the operators did not manually actuate SI in this scenario. For this sequence containment spray will not he available due to supporting system failures. IVSCM3: The majority of cutsets in this sequence involve an induced small seal LOCA, that is, the leak rate is less than 60 GPM/PMP. ECCS was available for safety injection but feed and bleed was not implemented due mainly to loss of power to the PORV. Secondary heat iemoval is unavailable due to the initiating es ent or due to loss of AF. The MDAFW pumps are unavailable due to loss of cooling. The TDAFW cont ois are lost due to HVAC failure and battery depletion leading to overfill and eventual drowning of the turbine. M A AP Cattulations for The Representative Sequences From the thermohydraulics and accident progression perspectives, all of the functional sequences described above are similar. i I f The objective of the present MAAP calculations is to obtain accident progression parameters and release - i fractions for CCI-induced failures in the event the debris is non-coolable, since catastrophic HPME I- failures and long term steam sequences are already represented in PDS 3SBO, 4SBO,- 4H arJ 2H. Furthermore, since non coolable sequences in which the sprays do not operate are dominated by PDS IH, which is examined in another section and would have similar release fractions to the corresponding outcome for PDS 3H, it is not felt necessary to perform -!culation for PDS 3H. Thus, only three non-l coolable debris with the sprays in operation scenarios were developed using X(123)CM2 as a guideline for the MAAP calculations: VSB3FI, VSB3F2, VSB3F3. 4-21l' b
_ _______mm . . . . _ . - ._. _ _ _ _ _ . . _ _ _ _ - . . _ . _ _ - _ . , i VSB3Fl: His is the base case calculation to represent X(1367;CM2 with sprays in operation and a non-coolable debris configuration. De sequence development is that described for X(1367)CM2 above; no additional assumptions are needed for the period prior to vessel failure, which occurs early at around 2.5 hourt. After vessel failure, an assumption was made in order to obtam the non-coolable configuration of the 2 debris. That is to assume a debris area in the cavity of 7.5 m2 which is 10 times less than the 75 m of cavity door and not at all realistic. However, :.lthough the probability that the debris would fall in tnis - configuration in high pressure FDS is highly unlikely, the probability itself is factored into the CET discussed in Section 4.6.1. In this section, the accident progression if that development occurs are evaluated. The calculation was run for 220 hcurs (9 days) and stopped because the containment pressure was only 65 psig, well below the mean failure pressure of 114 psig. Although overpressurization failure did not occur, a significant amount of hydrogen and CO were generated in the CCI (about 4000 Kg of hydrogen at 220 hours or 9 days). There were some burns at vessel failure and slightly afterwards but later on, the sprays kept the temperatures too low for burns. However, due to the large amount of hydrogen, another calculation was performed to evaluste the effect of a burn if it were to occur at such a low temperature, i.e., a burn was forced, and to calculate the release fractions if such a burn were to fail the containment. That is calculation VSB3F2. i:urthermore, in the present calculation (VSB3FI), there was a signi6 cant amount of concrete attack, and the 4m basemat depth was reached at 90 hours (3.75 days). - The fast erosion depth was due to the small initial debris area. In order to estimate the release fractions if a failure were to occur at that time, ! 'another calculation was performed. That is calculation VSB3F3. VSBF2: In this calculation, a burn was forced to occur at approximately 208 hours by restarting calculation VSBFI at that time with a low TAUTO = 315'K, the temperature for autoignition of the hydrogen. The pressure spike was sutticient to fail the containment. A leak type failure area was postulated for this i calculation in order to evaluate release fractions for a leak case, since the rupture case is seen above. I l ! 4-212 i l,__ - ~, . . . - , , , . . . , __ m - - -
A rupture type failure area was postulated for VSBF3. The release fractions for this case were small due to spray operation and the leak type failure. VSB3'3: In view of the very low release fractions observed in VSB3F2 and considering the basemat erosion observed in VSB3F1, this calculation was performed to bound the release fractions for this PDS. That was done by forcing a burn just priot to the time of erosion of the 4m concrete basemat and postulating that the resulting containment failure would be of the rupture rather than leak type. The results show release fractions slightly higher than those for VSB3F2 but still very small. These valaes were then kept as bounding for this PDS for the non-coolable debris with fission product scrubbing path in the CET. - The key event times for these analyses are summarized in Table 4.6-20. Calculated containment pressure and temperature histories are presented in Figure 4.6-7 (VSB3FI), / 6-8 (VSB3F2) and 4 ' (VSB3F3). Additional information on sequence development is presented in the corresponding figures in Appendix A of Reference 7. Table 4.6-21 summarizes the release fractions for CET end-states C5.L and C5-R both represented by PDS 3F and based on the VSB3F3 and VSB3F2 hlAAP calculations, respectively. Table 4.6-22 describes that end-state and list 3 similar end-states. 4.6.2.6 PDS 6F. 6H. 2F. 2H PDS 6F,6H and 2F and 2H are unimportant as contributors to early containment failure. PDS 6F and 6H can result in early failures fram an alpha event, which is well known to be of very low probability, particulady for a cavity of the CPSES strength and contiguration (see Section 4.1). PDS 2H and 2F, being situations of mtermediate pressure at vessel failure, can result in early failures from HPhiE events. PDS 6H and 2H are important (ranked first and third respectively) as representatives of steam-induced overpressure failures. Descrit, tion of The Rcpresentative Functional Sequences The representative functional sequences are-6H: 99% lh1Chil,1% AChil 4-213
_ _ _ _ _ . _ _ _ . . . ~ , ._ ._-._._. _ _ _.m.__ . _ . _ _ _ _.-... _ . _
+
t 6F: 62% ACM1,29% MCM1,8% XLCMI ' i 211: 100% ISCMI 2F: 87% SCM1,13% ISCMI l IMCMl: The main cutsets involve loss of CC or plant transient initiating events with the loss of the CC system components. This leads to an induced medium break LOCA (2 4 inches, which leads to a low pressure PDS), due to failure of safety valves to reclose after all have been opened to relieve pressure. Safety valves are required due PORV unavailability from their own faults or due to closed block valves. ECCS injects successfully but fails at recirculation, ACMI: This has been previously described in connection with the V-sequence. It is a large break LOCA with successfui ECCS injection but failure at recirculation. MCMI This is a medium break LOC A (2-4 inches, which leads to a low pn:ssure PDS) with successful ECCS-injection but failure at recirculation due to operator error, i XLO11: This is an excessive LOCA, e.g., failure of the reactor vessel. ECCS injects but fails at recirculation. l 11G.ll: The majority of cutsets in this sequence involve a loss of CC, which induces a seal LOCA greater than 60 GPM/PMP (assumed 210 GPM/PMP, or approximately a 1 inch break) 45 minutes after the reactor trip; Auxiliary feedwater is available for at least 24 heurs. The RCPs will only be tripped as RCS-subcooling becomes less than 25"F (criteria dermed by procedure EOP-0.0A). Sufficient ECCS (1 of. 4 pumps) injects on demand but fails at recirculation due to the loss of CC directly or as a result ofloss of room cooling. By procedure EOS-1.1, operators shutdown ECCS pumps starting with the Si pumps followed by the RH pumps to preserve the RWST as long as 25'F of subcooling is rnaintained in the RCS. Containment sprays fail at injection due to the loss of CC which causes failure of the pumps. . Those sequences in which the spray pumps operate are binned into PDS 2F. Fan coolers are tripped at I Si and are not available due to the loss of CC which causes a loss of chilled water for ventilation. l
~
4-214
ScMl-This L similar to ISCMI except the small break LOCA is the initiator rather than being caused by loss of supports. As a result the containment sprays are available. MAAP Ca]s.ulations for The Representative Seouerico De importance of these sequences is as representatives of late steam overpressure failures, in that respect all the above sequences have similar accident progressions and the same most likely outcome. The representative MAAP calculations are cased oc ISCM1, Since ISCMI is a small break rather than a large break for IMCM1, MCMI, ACMI etc. ,, the vessel failure pressure tends to be somewhat higher than if the calculations were based on the I.BLOCA sequences, but for long term steam failures this is irrelevant. Two MAAP calculations were conducted for sequence ISCMI: (1) SB2H4 and (2) SB2H5. SB2114 is a base case calculation representing the most likely outcome for the sequence if no recovery actions take place. This outcome is a late containment failure due to steam overpressure.
.sibeit minor, this sequence is also a coritributor to early containment failure due to ilPME. MAAP model parameters were modified outside of their estimated values in order to obtain release fractions for this failure mode. De parameter modifications for SB2H5 followed the guidelines suggested in References 33 and 34 j SB2H4:
- This base case follows the sequence description above with the following additional assumptions
l l l '
- CCP is the only pumped ECCS. The success criteria is I of 4 pumps. This choice delays core damage while pressurizing containment leading to a maximum containment pressure at vessel failure, o When the containment failure pressure is reached, the containment is assumed to fail by ruptures T ? is not the most likely failure mode for slow steam overpressurization which is liner tear causing a leak-type failure. However, another sequence (TRAN21 for PDS-4H) is used to -
represent the leak-type failure and SB2H4 represents the rupture type. 4-215 l
SB2H5; MAAP model parameters were modified outside of their estimated values in order to obtain release - fractions for this failure mode. The parameter moditications for SB2HS follow the guidelines suggested in References 33 and 34 and are as follows:
- TTRX = 1800. Tht time to fail a bottom head penetration after support plate failure was increased from I minute in the basic calculation to 30 minutes in order to accumulate more molten material for dispersion into containment at vessel failure.
- FCDhtA = 1. The debris was assumed to be entrained directly into the upper compartment instead of the lower compartment. This leads to a greater pressure rise due to its larger volume.
= FENTR= lE-06. This is the critical superncial gas velocity for entrainment of corium. The nominal value is around 1. A very low number suggests there is no threshold.
- FCh1DCH = 1. This is the fraction of dispersed debris that contributes to DCH.
- FCDhlH= 1, This is the fraction of debris that is dispersed.
- NVP=1000. The equivalent area of 1000 bottom head penetrations was assumed to open at vessel failure.
- TAUTO=460K. This is the critical temperature for a burn to occur. It incorporates steam '
l inerting characteristics. Nominal value matching data is around 983*K. This is the most serious l contributor to the pressure spike as it causes all hydrogen to burn at HPh1E simultaneously with the DCH. 4 Without the simultaneous hydrogen burn, the pressure at vessel failure does not approach the containment failure pressure, even when all paranieters available in h1AAP to maximize the DCH spike are adjusted to the maximum in that direction. Only with the simultaneous burn can the failure occur. l The key event times for these analyses are immarized in Table 4.6-20. Calculated containment pressure and temperature histories are presented in Figures 4.6-10 (SB2H4) and 4.6-11 (SB2H5). Additions! 4-216
information on seqt.ence development is presented in the corresponding figures in Appendix A of Reference 7. Table 4.6-21 summarizes the release fractions for CET end-states B6-R and D6 R both represented by PDS 211 and based on the SB2H4 and SB2H5 MAAP calculations, respectively. Table 4.6-22 describes those end-states and lists similar end states. 4.6.2.7 PDS 3SBO and 4SQQ PDS 4SBO is the third contributor to CCI-induced failures (2.2%) and tenth (0.3%) to early failures. PDS 3SBO is ninth (0.7%) and seventeenth (0.1%), respectively. The cause of the early failures is primarily High Pressure Melt Ejection (HPME). The high HPME delta P used is from the NUREG-il50 values for Zion (Section 4.6.1 and Table 4.6-i) based on a high RCS pressure at vessel failure. Description of The Representative Functional Seouenen The representative functional sequence for PDS 3SBO is IVSCM6 which is 100% binned into PDS 3SBO and constitutes 100% of the total 3500 frequency. '"he representative functional secuence for PDS 4SBO is IVSCMS which is 100% binned into PDS 4SBO and constitt.tes 100% of the total 4SBO frequency. IVSCM6: The main cutsets in this sequence involve a loss of offsite power, which induces a very small seal LOCA, i.e., of less than 60 GPM/PMP, 45 minutes after the initiating event. All ECCS fails to inject on demand due to the loss of all power. Auxiliary feedwater is not available due to loss of power to the motor driven
~
pumps and the loss of the turuine driven AF pump disiharge control leading to steam generator overfill and eventual turbine drowning. Feed and hieed is not entered as it was not required. The RCPs will trip at the time power is lost or on the total loss of AF by procedure FRH-0.1. Containment sprays fail at injection due to the loss of power. Fan coolers trip at loss of power. ; 1 IVSChti: The main cutsets in this sequence involve a loss of offsite power, which induces a very small seal LOCA, j i.e., of less than 60 GPM/PMP,45 minutes after the initiating event. All ECCS fai!= to inject on demand l due to the loss of all power or support systems. Auxiliary feedwater is not available since the motor 4-217 l
)
driven pumps have no power and the turbine fails due to its own 1 its. Feed and bleed is not entered
~
as it was not required. The RCPs will trip at the time power is lost. Containment sprays fall at injection . due to the loss of power. Fan coolers trip at loss of power. MAAP Calculations for ne Representative Secueness PDS 3SDO and 4SBO are the only PDS for which the possibility of recovery after core uncovery and prior to vessel failure has been considered for CPSES. Although the most likely outcome for these PDS is CCI-induced failures, those failures are well represented in PDS IH. As a result, it would be most + aporopriate to develop a representative sequence which would lead to early containment failure following [ recovery of power after core uncovery but where vessel failure could not be precluded (CET end-states D1 and D2). IVSCMS was used as the base-case for that analysis. SB031. Recovery of power is considered to occur after core uncevery and prior to vessel failure but not-precluding vessel failure in order to have a calculation for the recovery CET end-state DI. Power recovery is assumed to take place after the fuel has already heated up significantly. In the simulation, the RCS pressure at time of power recovery is high so that an Si signal is not generated and injection does not occur. In practice, an Si signal would most likely be generated manually from the control room. Ilowever, as discussed in Reference 2 the injected amount would not be relevant unless the operators were to depressurize the RCS. That action is along the successful DP path and, therefore, it is not - modeled here. Furthermore Jepressurization would simulate the situation which is already modeled in SB2H5 for PDS 2H. The effect of the additional amount of hydrogen that could be generated by the ECCS injection was investigated in sensitivity runs. However, since the containment fails in a burn-DCH event at time of containment failure anyway, the extra hydrogen is redundant, Containment sprays operate since the power is recovered and mitigate the release as indicated in path D1 of the CET. Since PDS 3SBO is the greatest contributor to early containment failure due to HPME, MAAP model parameters were modified outside of their estimated values in order to obtain release fractions for this failure mode. The parameter modifications for SB031 followed the guidelines suggested in References - 33 and 34 and were as follows: 4-218
~_. - _ _ _ . . _ . .- . __ _ _ _ _ _ _ . _ . . . __ .- .. _
- TTRX=600 The time to fail a bottom head penetration after support plate failure was in:reased from I minute in the basic calculation to 10 minutes in order to accumulate more molten material for dispersion into containment at vessel failure.
- FCDM A = 1. He debris was assumed to be entrained directly ;nto the upner compartment instead of the lower compartment. His leads to a greater pressure rise due to its larger volume, i
- FENTR=3E-02. This is the critical superfielal gas velocity for entrainment of corium. De l l
nominal value is around 1. A very low number suggests there is no thrcshold. l I
= FCMDCH= 1. His is the fraction of dispersed debris which contributes to DCH. j
- FCDMH = 1. This is the fraction of debris which is dispersed.
= NVP= 100. The equivalent area of 100 bottom head penetrations was assumed to open at vessel failure, l
l TAUTO=700K. This is the critical temperature for a burn to occur. It incorporates steam inerting characteristics. Nominal value matching data is around 983 K. This is the most serious contributor to the pressure spike as it causes all hydrogen to burn at HPME simultaneously with the DCH. Without the simultaneous hydrogen burn, the pressure at vessel failure does not appmach the containment failure pressure, even when all parameters available in M AAP to maximize the DCH spike are adjusted to the maximum in that direction. Furthermore, only with a burn prior to spray actuation does the failure occur. Thus, it was necessary in the calculation to delay spray actuation until after the burn occurs following vessel failure. While this is not unrealistic given the closeness of the times involved, it should be noted that if de inerting occurs, the initial pressure is too low and the pressure spike is too low for early containment failure following the burn. It should be remembered that burn probabilities of occunence are not based on these findings but rather on the NUREG-1150 approach discussed in Section 4.6.1 and summarized in Table 4.6-1, l 4-219 l l
- . - ~ . - . - - - . - . _ . - - . . ~ . - - - - . . . _ - - - . - . - . . - . .
e SBO41: This calculation is similar to SB031. In this case, CET end-states D2 are being obtained so that there is no Ossion product removal by sprays, Therefore, in order to model that situation, it is assumed that only the RH pumps are recovered and the sprays are not.
'The key event times for these analyses are summarized in Taale 4.6 20. Hydrogen produced in-vessel and in-containment due to CCI is shown in Figure 4.6-12 for SB031. Calculated containment pressure and temperature histories are presented in Figure 4.6-13 for SBO4:. Additionalinformation on sequence development is presented in the corresponding Ogures in Appendix A of Reference 7.
Table 4.6-21 summarizes the release fractions for CET end-states Dl-R and D2-R, both represented by PDS 3SBO and based on the SB031 and SBO41 MAAP ca:culations, respectively. Table 4.6-22 describes those end-states and lists similar end-states. 4.6.3 Treatment of Uncertainties and Sensitivity Studies NUREG-1335 Section 2.2.2.6 requests a description of the methods employed for handling phenomenological uncertainties. The objective of this scuon is to comply with that request and also to describe the handling cf non-phenamenological uncertainties which are important for accident management considera' ions. The phenomenological uncertainties are addressed in Section 4.6.3.1 and are those considered in Table A.5 of NUREG-1335, Appendix A. In addition to those other uncertainties related to system characteristics and operator actions are addressed in Section +.6.3.2. The purpose of addressing uncertainties is to avoid masking the potential for vulnerabilities due to controversial assumptiens regarding the likelihood of certain phenomena. In both Sections 4.6.3,1 and
- 4.6.3c2, the uncertainties are :"pically addressed in one of three ways: (1) via the CETs, by selection of basic event (BE) probability values, based on expert opi nion probabilities from NUREG-1150 that cover the range of possible accident progression outcomes, thereby ensuring no potential vulnerabilities are masked; (2) by the NUREG-1335 Appendix A Step 8 sensitivity study approach, to address-tne consequence aspect of the uncertainties, i.e. the impact on accident progression or (3) by recalculating 4-220
--. - ._ .- ~.- . - - . - . - - .-- .- - - _- -
CET endistate probabilities with varis auis in BE probabilities, in order to demonstrate that certain existing consequence uncertainties are irrelevant. 4.6.3.1 Phenomenolecical Uncertainties The phenomenological uncertainties addressed it, this section include all of those listed in Table A.5 of NUREG-1335 Appendix A, which are re-arranged and labeled a through m below,
- a. Perfarriance of containment heat removal durine core meltdown ar.cidents ne characteristics of the available containment heat removal equipment are part of the MAAP model, and therefore, their performance is reflected in the results of the analyses performed (Refs. 2,7) for the-core meltdown accidents. The adequacy of the avail 61e equipment was thus evaluated on a case- by-case basis. Nevertheless, general conclusions can drawn from these analyses and those are summari ed below.
Containment heat removal during cote meltdown accidents is via: (1) the residual heat removal system l (RH) and/or the containment spray (CT) system, either or both gnerating in trirculation mode, or (2) the fan coolers. Fan cooler performance and availability is discussed in 4.6.3,2. MAAP calculations (VilF2 and VIIF3, Ref. 2) were performed to investigate the minimum requirements in terms of RH and/m CT trains. It was found that as long as one RH train was able to switchover successfully to recirculation, with one RH heat exchanger available, t'.a containment pressur? cation rate can be arrested. Since spray flow rates are much higher than RH tiow rates, it is concluded that one spray train is also adequate. In either case, if switchover is successfully accomplished, a subsequent interruption in the sytem is unlikely, since all key components are located in the safeguards building outside containment. An additional calevlation (VIIF5, Ref. 2) examined the case of total loss of component cooling water to the RH heat exchangers. Containment failure occured in hlmost the same time frame as of the case where switchover to recirculation failed altogether. However, when component cooling water flow to the RH
. exchangers;was not totally lost, but merely impaired (up tc 40% of its nominal value), the containment survived indefinitely (VIIF4, Ref. 2). While these findings are useful to the CPSES accident management -
knowledge base, they do not constitute success criteria for any of the front-end analyses. Those requirements are discussed in Section 3 of Volume I of this submittal and are typically more severe than the successful performance boundaries delineated above. 4-221 _-. y._ , ,- . . - . . -., N - - - ' - . _ _ - - . - - - -
sin sem Lin-vessel Lydrogen oroduction at hicii and low RCS nressures and combustion in containment. Here is considerable uncertainty regarding how much hydrogen can be produced during the core melt progression. In addressing this issue for the CPSES IPE, the objective was to insure that the effect of very high quantities of hydrogen being produced in-vessel were examined. The main concern was whether in-containment burns from hydrogen produced in-vessel could cause early containment failure. Two situations are possible: (1) hydrogen combustion alone leads to containment failure for low RCS pressures and/or (2) combustion in combination with direct containment heating (DCH) for high RCS pressures leads to containment failure. He CPSES IPE has been able to bypass much of the uncertainty regarding the problem of in vessel - hydrogen generation because the containment can resist a burn of all hydrogen produced in-vessel assuming 100% clad oxidation, as discussed in b.I belove. In an actual ses e accident of course, tu amount of hydrogen produced in vessel would be much less than that corresponding to 100% clad 3 oxidation. The NRC staffs summary of IDCOR/NRC issues 5 and 6, in-vessel Hydrogen Generation and Core Melt Progression and Vessel r ailure, reflected disagreements on the effects of degradation in core geometry and on the rate and magnitude of hydrogen production. NRC considered that the IDCOR best-estimate position, i.e., that the onset of melt would quickly cause a complete stoppage of local Zirealoy oxidatior, is not adequately substantiated by data. The applicable MAAP parameter relating to this issue is FCRBLK, a flag used to select the blockage model. In the calculations described (Ref. 7)- below, the flag was set for no blockage. Typically, this resulted in approximately 60% clad reacted and is about twice what would occur if the blockage were turned on. One other parameter can be used to .
~
increase hydrogen production further, namely FAOX, which can be set to allow for one- or two- sided oxidation of the clad. This was set to two-sided oxiddon in the cases that resulted in about 60% clad reacted. In any case, all these variations were bounded by the 100% clad reacted situation postulated for the adiabatic, isocoric, complete combustion (AICC) calculation discussed below. Thus, from a hydrogen production standpoint, there is a great deal of contidence that the amount produced in a severe accident would be somewhere between 30% and 60% of the clad reacted. From a consequence point of view, it is shown below that the combustion of even 100% clad reacted cannot fail the containment. Lt. Low RCS nressures In-containment combustion of hydrogen produced in-vessel was found not to be capable of causing early containment failure. This was concluded on the basis of an adiabatic, isocoric, complete combustion (AICC) calculation (Ref. 7), which postulated 100% Zircaloy clad reacted to maximize hydrogen 4-222 O
production, and also postulated the envir< nment composition at the onset of the flammability limit, to maximize the post burn pressure. Specifically, the pre-burn composition leading to the highest post-burn pressure is: (1) the existing air inventory of about 96000 Kg (43% molar fraction), (2) the upper bound 1000 Kg (6.5% molar fraction) of hydrogen (100% clad oxidized, i.e.,21316 Kg of Zr lead to 937 Kg of hydrogen) and (3) 69786 Kg (50.5% molar fraction) of steam (determined at the boundary of the flammability limit, given the above amounts of air and hydrogen). This composition is at the onset of the Shapiro and Moffette (Ref.12) flammability limit. The pre-burn temperature is saturation at the steam specific volume for the composition above, because spray actuation is required to reduce the steam amount, i.e., the containment would always be inerted unless the sprays had come on. The final post-burn temperature, calculated by assuming adiabatic, isocoric, complete combustion of all hydrogen, is approximately 100 psia. The containment fragility curve, developed in Section 4.4 and represented by Table 4.4-4, shows clearly that the probability of NL containment failure at 100 psia is negligible. b.2 liich RCS pressures In cases where both hydrogen burn and DCH can occur (assumed to be cases where vessel failure pressures are greater than 200 psia), the approach described in Section 4.6 (using Table 4.6-1 from NUREG/CR-4551) was used. The probability of early containment failure due to in-vessel production of hydrogen is included in the containment event tree for these cases because Table 4.6-1 includes, in the fractile pressure rises, the effects of hydrogen combustion an" DCH. By using the values of Table 4.6-1, it appears that an upper bound on the pressure rises . all high pressure melt ejection (HPME) phenomena was obtained. There are two reasons for this conclusion: (1) the Zion cavity is considerably more favorable to particle entrainment into the upper compartment than CPSES' and (2) sensitivity calculations with M A AP (Ref. 7) were performed where parameters were modified to maximize the effect of DCH and even then, the containment failure pressure was only reached when a simultaneous burn was forced to occur under inerted conditions. Although recognized as somewhat excessive, the fractile pressure rises from Table 4.6-1 were utilized for CPSES because the resulting early containment failure probabilities were still low enough to shoiv that HPME is not an important concern for CPSES, as discussed in Section 7. I
- g. Induced failure of the RCS nressure boundary at hich RCS nressures Three locations in the primary system boundary are considered in the CPSES IPE as possible locations for induced failures: (1) Reactor Coolant Pump Seals, (2) Hot Log and/or Surge Line and (3) Steam Generator Tubes. All three types of induced failures are incorporated into the CETs, as shown in Figure l
4-223 l 1
4.5 2, it should be noted that the RCP seal failures mertioned above are thosa occurring after core - ) uncovery, assuming the seals were still intact at that time. The probability of preere melt induced seal 1 LOCAs, due to loss of seal cooling, is addressed in the front-end of the IPE and is described in Volume I of the CPSES submittal. Of these three types of failures, the first two, RCP seals and hot leg and/or surge line failures, are in a sense beneficial because they cause the RCS to depressurize and effectively eliminate the possibility of direct containment heating. From an accident progression point of view, RCP seal and hot leg / surge line' failure effects de similar to the effects of operator depressurization of the RCS after core damage, The hot leg / surge line failure probability variations, shown as basic event PRHLSLOK in Figure 4.5-2, and the seal failure BE PR3EALOK, are bounded by variations in operator action to depussube, BE HOP-l>P, which is discussed in Section 4.6.3.2. Given this bounding consideration, and because the possioliity of RCP seal failure and hot leg / surge line failure are included in the CETs, it is clear that all possibilities were considered and no potential vulnerabilities were masked. Induced Steam Generator Tube Ruptures (ISGTR) are not desirable because of the containment bypass. The probability for these events is considered in BE PRSGOK, also shown in Figure 4.5-1. Thus, the potential for this vulnerability was also addressed via CETs. Furthermore, the frequency contribution j to ISGTR from each PDS, given in Table 4.6-18, shows these progressions are almost two orders of magnitude less significant than for SGTRs as initiators, so additional sensitivities on this parameter are - not warranted. The method for deriving ISGTR frequencies is described in Section 4.6 2.1. These probabilities are also bounded by the probability of operator action to depressurize the RCS, as discussed in Section 4.6.3.2. The phenomenon which has the greatest impact on induced failures of the RCS is natural circulation. l Several studies have been conducted to validate the natural circulation model in MA AP as discussed in Reference 33. The relative magnitude of the BE probabilities for hot leg / surge line failure, PRHLSLOK, versus steam generator tube failure, PRSGOK, with the hot leg / surge line failure being more likely, is. l based on the generic results reported in Reference 33 and on NUREG-il50 estimates as presented in Reference 1. In addition, CPSES-specific calculations for high pressure sequences (Figure 4.6-14) show hot leg temperatures around 900'C (1700;F), surge line temperatures around 800"C (1500 F), while the steam generator tubes are still arou ; 500'C (900*F). The differential is less (Ref. 33) if the pump bowls are cleared by operator action to bump the pumps, because d the possibility of unidirectional natural 4-224
convection flows. However, hot leg /suri'e line failures are still more likely than ISGTRs in those cases, due to the higher temperatures at those locations. In summary, these issues are all individually addressed via CETs. Furthermore, their main effeu is RCS depressurizat:an, which is investigated for a much wider probability range in Section 4.6.3.2 by - examining variations in CET end-state probabilities resulting from perturbations in the probability of operator action to depressurize the RCS, namely BE HOP-DP.
- d. Core relocation characteristics at hich RCS nressures
- e. Mode of reactor vessel melt-throuch at hich RCS nressures
- f. Direct Containment Heatine (DCJJ)
Since direct containment heating is related to the core relocation characteristics and to the mode of vessel melt-through, these three issues are discussed together. The mass of core debris released from the primary system at vessel failure, compared to that released later, can have a first-order effect on the subsequent containment response. Material released promptly from the primary system may be dispersed out of the reactor cavity and might participate in direct containment heating (DCH). Materials released later typically remain in the volume under the vessel. Dispersal processes also depend on the flow of gas from the primary system, which is affected by the mode of reactor vessel failure. In the write-up for the NRCMDCOR issue 6, the staff stated that the level of uncertainty in mechanistic calculations of core melt progres ion was high. Accordingly, they stated that parametric variations in the masses of core materials discharged at vessel failure should be used to investigate the impact of these uncertainties on the cor tainment response in future IDCOR studies. In the CPSES IPE, all the MAAP parameters related to these issues were varied from the base-case recommended values to extreme values, for the purpose of addressing these issues via sensitivity studies, More than one calculation was performed varying these parameters, as discussed in Section 4.6. However, one example is sufficient to show that the range was fully explored. For this purpose, consider "AAP calculations SB2H4 and SB2H5. SB2H4 has nominal expected values for the relevant parameters, while SB2H5 has extreme values, used to investigate the possibly of containment failure caused by DCH. 4-225
i I The parameter modifications for SB2H5 follow the guidelines suggested in Reference 33 and are as follows:
- TTRX = 1800. The time to fail a bottom head penetration after support plate failure was locreased from 1 minute in the basic calculation (SB2H4) to 30 minutes (SB2H5) in order to accumulate more molten material for dispersion into containment at vessel failure. Increasing the time any further leads to lower pressures at vessel failure because the steam production decreases as the lower plenum water boils away.
- FCDMA=1. The debris was assumed to be entrained directly into the upper compartment _ . .
(SB2H5)instead of the lower compartment (SB2H4). This leads to a greater pressure rise due to its larger volume, but is somewhat unrealistic given the CPSES containment layout (see Section 4.1).
- FENTR= IE-06. This is the critical superficial gas velocity for entrainment of corium used for SB2H5. The nominal value is around I, used for SB2H4. A very low number implies that there is no threshold and corium is entrained into the upper compartment at any gas speed. This is unrealistic but maximizes the amount of corium entrained to the upper compartment.
- FCMDCH = 1. This is the fraction of dispersed debris that contributes to DCH in SB2H5. The base-ease value in SB2H4 is 0.1.
- FCDMH = 1. This is the fraction of debris that is assumed to be disper3ed in SB2H5.
- NVP = 1000. The equivalent area of 1000 bottom head penetrations was assumed to open at vessel fai.ure in SB2!!5. The base-ease valne in SB2H4 is 1.
- TAUTO=460 K. This is the critical temperature for a burn to occur. It incorporates steam inerting characteristics. Nominal value matching data is around 983'K, which is used in SB2H4.
This is the most serious contributor to the pressure spike as it causes all hydrogen to burn at HPME simultaneously with the DCH. 4-226
.- -._ .- _ - - - .- .- ~- _ _ _ - . - . -. - -- - -
The effect of these changes in parameters on the containment pressure history can be seen by comparing Figure 4.6-10 and 4.611. (Other examples of HPME failures obtained as outlined above are seen in Figures 4.6-3,4.612 and 4.6-13). It should be notcd that in all cases, even with the extreme variations in va'ues for parameters 1 through 6 listed above, without the simultaneous hydrogen burn forced by the change in parameter 7, the pressure at vessel failure did not approach the containment failure pressure. De extreme variations in all parameters 1 through 6 available in MAAP are to maximize the DCH spike and were adjusted to the maximum in that direction.- Only with the burn can the failure
. occur. However, the burn would not be likely due to inerting conditions. If inening were not to occur, the initial pressure would also be r , low for early containment failure. These findings are part of the rationale for why, by using the values of Table 4.6-1 and the NUREG/CR-4551 HPME pressure rise fractiles for Zion, it was felt that an upper bound on the corresponding CPSES pressure rises was obtained,
- c. Core relocation characteristics at low RCS nressures -
- h. ModnLteactor vessel melt-throuch at low RCS nressures i Fuel / coolant interactions at hich and low nressures All the above issues are important because of the possibility of in-vessel and ex-vessel steam explosions with the potential for the so called alpha failure mode of the containment. This issue has been addressed I
on a generic basis in the open literature, and it is not a likely threat for large dry PWR containments. l . However, for the sake of completeness the issue i addressed in the CPSES CETs via BE PRALPAL under top event CFEl, Figure 4.5-5. Due to the small likelihood of this event, a probability of 8E-3 is - used for it, based on NUREG-1150 expert judgement as reported in Reference 1. He alpha mode failure i at high pressures is considered to be an order of magoitude less likely based on the same reference and is also considered via DE PRALPAH under top event CFE2, Figure 4.5-6. Consideration of this phenomenon in the way described assures that no vulnerabilities for it are masked. Furthermore, due - to the unlikelihood of the alpha failure mode for CPSES, particularly in view of the concrete thickness and configuration of the cavity region, as described in Section 4.1, it is felt that. additional sensitivity. studies on this issue are not warranted.
- i. Potential for early containment failure due to pressure loads in the case of CPSES, the potential for early containment failure due to pressure loads is addressed in I the CET under tcp event CFEl, Figure 4.5-5, and under top event CFE2, Figure 4.5-6. He phenomena considered are: (a) HPME, including DCH, combustion of hydrogen produced in-vessel and steam spikes 4-227
(PRWCP PULT, PRDCP PUllf), which are discussed in Section 4.6.1 and under d,e,f above; and (2) as alpha failure mode at high and low RCS failure pressores (PRAlllPil and PRAl.PL), which are discuswd in Section 4.61 and under g,h,i above. kfelculhtueLIN1ylentidn!ItenLhtikelutleitt VAlf ent6Ll'LCelfichi$ Although this esent is formally conside ed in the CIITs under logic tree top events CFill and Cl112, there is no gesibility of direct contact of the debris with the containtnent liner other than in particulate form in a DCil event, which is already addressed as discussed in Section 4.6.1 and under d,e f above. This conclusion comes from an examination of plant dtawings a~d a walldown of the plant cavity area. As a result, the lili probability for this event (PRIMPINGli) is set to rero. - lluPLtrinLdbi'eidenalteltdchh.kuuhtdem Jiulluehble) IILleDLlunLEultEURElr1LintuKtwnalvLityihl'Ilil11tudleebildlity.2Ldthi5 The issues of debris coolability and core-concrete interactiom. are mased in the CPSIIS CliTs. The treatment of these issues a comprehensive, all possibilities are considered, and thus, the potential for vulnerabilities is adequately explored via CliTs in the manner explained in this sub-section. the debris c .olability issue is addressed in 'op events DCl, DC2 and DC3, Figures 4.54,4.5-8 and 4.5-
- 9. respectively. When the debris is coolable, there is no CCI. When the debris is not coolable, there is Cr . CCI sequences dominate the containment failure modes with 49% of the CMF. There are two types of PDS which result in u related failure of the containment: (1) PDS where the debris is not coohible because it is in a noiceoolable configuration, i e., an inst lating crust has toimed which prevents heat transfer to an existing overlying water pool, and (2) PDS where the debris was originally in a coolable contiguration with water over it and the water dried out and CCI began. Top events DCl, DC2 and DC3 consider both these cases, via events DiillFORM and WTRPOOL, Distinction is +o made between low and high pressure cases, because the debris a more likay to be coolable when it is forcefully expelled f rom the vessel at h',gh pres ore, iAough the occurrence of a steam explosiors which is more likely at low pressure, also increases the chances a coolable contiguration. The approach is conserva cau : tumes that when the debris is found non-coolable due to dry-out, it then becomes ~ definitely oon-coolable and cannot be recowred, as if it had beeonie also geometrically or crust-wise non coolable. De result of this approach is that, typically, there is a 3% to 6% probability of the debris lieing non-coolable if there is an overlying pool, and a 100% probability if there is not.
,)
Given the slie of the CPSES containtnent and the amount of concrete and metal, the containment cannot f ail from steam overpressut . tion from RCS and accumulator inventories alone. Therefore, all sequences where tiCCS fails to inject lead to dry +ut prior to containment failure. All these sequences are then, after dry-out, assumed to be unrecoverable and result in CClJnduced containment failure. CCl inducal containment failure can occur in three ways: (1) from overpretsuritation due to non-condernibles generated in the CCI, (2) from basemat melt through and from (3) combustion of non-condensibles generated in the CCI. In cases where the pool above the debris dries out, i.e., when the debris is initially in a coolable con 0guration, the most likely outcome is overpressurization. Combustion of non condensibles, which is considered in the CETs under top events CF114, Figures 4.515 through 4.5 20, accounts for only a small fraction of these late failures, primarily because they are dry-out type situations with high steam content, and therefore are inerml. In order to determine whether the overpressurir.ation or the melt-through mechanism prevails at CPSES, a diy-out debris bed situation masimizing CCI we simulated with MAAP. This was done by obtaidng an early vessel failure time with minimum cavity inventory, in order to obtain an early cavity dryout as well. The early vessel failure time was obtained by a station blackout initiator with no ECCS and no auxiliary . edwater from time zero. The early dry-out of the cavity was obtained by shutting off all sprays and failing all accumulators. Results of this bounding calculation are shown in Figure 4.6-16 for two types of concrete composition: (1) high limestone (SilOCCl) and (2)- limestone / common sand (SilOCCl2) Some of the initial MAAP calculations assumed a high'limer. tone composition based on the soil composition. A subsequent investigation revealed the composition is very near the CORCON default for the ilmestone/ common sand type. Yhe two calculations illustrate that for l-either type of concrete, overpressurization precedes melt through by a large margin. In these calculations (Figure 4.6-16), it would take more than 240 hours to erode the 4.0 m thick basemat. On the other hand, l Figure 4.6-16 s' hows that overpressurization due to non-condensibles cauld accur in about 30 to 36 hours. Therefore, while there are uncertainties involved and differences due to the type of concrete. it is felt that due to the difference of a factor of about ten (based on curapolation of Figure 4.6-16) between failure times for these modes, basemat melt-thraugh would be very unlikely at CPSES for dry-out cases. The two main reasans why basemat melt-through is highly unlikely in these i are: (1) the CPSES large cavity area (70 mi) results in a low erosion depth for a given eroded e and allows for a higher radiation heat transfer area, and (2) the bolling off of the steam prior to , .y out takes the containment pressure to a high level prior to the onset of CCI. The non-coolable configuration by reason of dry out 4 229 _-. ._ u . _ _. _ n _ _ _ _ _- . .a _ . . __:.________.
-- - _ - - - - . - - _ _ _ _ . . - ~ - _ _ . . ~ . _ . - . . - -
i l applies to PDS where the ECCS failed to inject, namely, 111, 311, 511, 3513 0, 45110, and represents
- 93 7% of the CCI failures or 46% of the CMF.
in cues yhere the debris is non-coolable by reason of crtnt formation, the mode of CCI induccul failure is less clear cut. 'Diat is because ir these cases, the atmosphere may not be inerted since the debria is t insulated from its overlying water pool, and the debris raay not occupy the entire cavity area and thus j moves downward inore rapidly. This type of progression has been simulated with P 7 (e.g., VS113F1, , 2 2 , Figure 4.6-7) by reducing the cwit) area from 70 rn to 7 rn and reducing the ht: urufer toefficient between the water pool and the debris to a veiy low number. That calculation shrys w svhen the cavity area is reduced by a factor af 10, then basemat melt through (90 hours) can precede overp.murization (over 400 hours). In this case, burns prior to 90 hours would also be possible as shown in calculation VSil3F.4, Figure ..o-9. Thus, while uncertainty exists in the mode of CCI failure in cases where the debris is non coolable by reason of crust formation, these cases are not statistically significant. This is because the non coolable configuration by season of crust formation applies only to a Fmall (3% to 6%) fraction of PDS 3F, 4F,6F, IF, 2F,611,411, 5F,211, for a total frequency of only 6.3% of the CCI failures or 3.1% of the CMF. This is much less significant than the dry-out situation, which represents 93.7% of the CCI failures or 46% of the CMF. Furthermore, the magnitude of the uncertainty discussed here exists because it was assumed that: (a) the debris occupies only 10% of the cavity area and (b) there is no heat transfer to the poo;, i.e., no steam formation. Both of these assumptions are extreme and unlikely in themselves. Therefore, it is still likely that in the case of CPSES, non coolable debris CCI would also lead to overpressuritation f ailuie, as in the dry-out case. A final point to be clarilled is how debris coolabil;ty issues arc addressed in tiv: CPSES IPE in the cases of early containment failure. Dere are two groups of early failure PDSs, each of which is assumed to have a definite coolable or non coolable progression for the reasons discussed below. De first gtoup involves bypass PDSs ICl) and 2CD, namely SGTR and _ V-Sequences (containment ; isolation failures are statistically negligible, as discussed in Section 4.6). This group is non-coolable by virtue of dry out and, therefoie, all these sequences have the M A AP-calculated core-concrete interaction following vessel failure and dry-out. The second group includes all other early failures, which are either from IIPME or alpha-tyre avents, in these cases, the debris is finely dispersed and suspended in the containment atmosphere f: sete 4 230 _ _ _ . _ _ _ _ _ _ . _ . _ _ . _ _ _ _ _ _ _ _ _ _ _ _ _ _ , _ . , _ , _ _ _ _ , u. _ _ ,
_.___....._._._-.m.__.________________..___.__ it would have a very large settling a ea and, therefore, could not form a deep pool,i.e., it would have to be coolable. In order to illustrate this pomt, consider Table 4.5 4, which provides an indication of potential debris bed thicknesses. Knowing that the core barrel and upper internals would not be molten at the time of vessel failure, and would, therefore, not be apelled f om the sessel at that time, and would either stay in place or fail offlater as large solid pieces with no decay heat, it can be inferred from Table 4.5 4 that the maximum cavity debris depth would be 23 cm (30 cm if an average corium density of 7000 Kg/m' is used). Further, knowing that in the HPME events the M AAP calculations required 100% debris dispersion in order ta fail the containment, when the entire containment settling areas are considered. it is obvious that the debris bed cannot be deep enough not to be coolable. Even if ordy 50% debris dispersion were enough to fall c<mtainment, the cavity depth alone would be less than 15 cm. Th. refore, when HPME and/or alpha are the containment failure modes, the debris must ne coolable at CPSES. 4.63.2 SyngaitntOnsnttional Unmlaimics In addition to the phenomenological uncertainties discussed in Section 4.6.3.1, there are a number of system and operational uncertainties which warrant discussion particularly due to their applicability i tow"* accidem management. These are presented below. A.1:ffect of operator action to drnituurire the RCS at the ornet of care damagg Procedure FRC-0.l A instructs operators to depressurite the RCS when core exit thermocouples reach 1200 F. This action reduces the RCS pressure at vessel failure and reduces the probability of DCH. Since the action is proceduralized, the base case CET end state probabilities in this report includes credit , l for it in the CliTs, by appropriately calculating Bli HOP-DP, shown in logic tree DP, Figure 4.6-2. It should be noted that the calculation includes comideration of the availability of power and equipment (PORYS) as well as human factors, i.e.. the probability that opet. ars will act given a proceduralized action. l In spite of the foregoing, a sensitivity study was conducted to examine the impact of assuming eperators fail to depressurize the RCS. The driving force for the study is to verify that na vulnerabilities will surface in this hypothetical case, rather than concern that the action might not be taken. The key l
- concerns are increases in the early containment failure frequency due to increases in DCH and ISOTR -
probabilities. The study was conducted by setting BE HOP-DP to 1.0 and comparing the key frequencies with those obtained using the base case values for HOP-DP. It should be noted that the variation in top l 4-231
.= =
l l _ . . _ . . -._. _ . _ .__... _ _ u _ . ___ _ ,.. _ _ ,. .,_ .;_ _._ _ _______ _._____ _ .._...._ _ ._, . , _ . . , _ . _ _ _ . . . -
-.. - _ _ _ - - . - - - - - - - _ ~ _.-. - .-.-.-- -. --
e4 ! event DP, Figure 4.6 2, is much greater from varying IlOP DP, given its range, than it would be from variations in other pararneters affecting depresrurization, namely PRilLSLOK, PRSEALOK, etc... l 1herefore, this study also illustrates the sensitivity of early containment failure frequency to other the ur ertainties related to RCS depressurization, as was nointed out in Section 4.6.3.1 item c. The early containmeat failure probability for the sensitivity case is 4.2984, while the base case value shown in Table 4.6-4 h 4.12E4. This variation is small and confirms that no vulnerabilities to 11PhtE phenomena exist at CPSES. The narrow difference is in part due to the fact that depressurization increases me al, aa mode failure probability, in other words, llPME and alpha are in competitmn and 1 of similar weight as early failure contributors, j
-i bmPetenthi benefitulthe fan coolers Fan cwbra can reduce the frequency of early failures, but the attendant risk reduction is insignl0 cant, if functioning at time of core uncovery, they would reduce the llPME-induced containment failure frequency by reducing the base pressure in containment at vessel failure.110 wever, these are only 5.4%
of the early failures (which are 9.4% of CMF), i.e.,0.5% of the CMF, and the fans would improve this by a fraction at best. i an coolers can reduce the frequency of steam.inducal late containment failures, but the attendant risk redu; tion is also insignificant. Steam-induced late containment failures are 2.1% of the CMF. Again, l fans could only improve this by a fraction at best, so the potential gt.:n for these cases is also marginal. The greatest potential for Fan cooler benent is in the CCI sequences which dominate the containment. failure modes with 49% of the CMF. The key is just how effective can the fans be in them CCI cases. There are two types of PDS which l contribute to this kiad of failure: (1) PDS where the debris is not coolable because it is in a non coolable I i con 6guration and/or an insulating crust has formed which prevents heat transfer to an existing overlying - l water pool and (2) PDS where the debris was originally in a cootable configuration with water over it l and the water dried out and CCI began. The first situation cannot be helped by fan coolers because while they can stop overpressurization, they cannot preclude the basemat meh-through. The effect of fan coolers for these PDS would simply be to E 4 232 i ptf-Pp g 3 -f ag- ?-W+p-pgm ' g=.mmwFe'7aq=- *" p- W+w. -
. ~ ._ - . _-- - ____________m.... . _ _ ._ _ _ _ .- _ _ _ . .
switch the likelihood of CCI induced failure from non condensible overpressurli.ation to basemat penetration. While this is an improvement in source term, it is not an Im, rovement in containment I integrity probability. His first situation applies to a small fraction of almost all PDS, namely 3F,4F,
)
6F, IF, 2F,61{, 411, 5F. 211. Its total frequency is 6.3% of the CCI failurch or 3.1% of the CMF, which is slj,nificantly less than the second situation. The second situatian, namely PDS lH,3H,511,3SHO,40110, represent 93,7% of the CCI failures or 46% of the CMP. His is a situation where the debris was originally in a coolable configuration with water over it, the water dried out and CCI began. If a suf0cient number of fan coolers had been in ; operation, the overlying water pool would never have dried out. He potentici benefit of the fan coolers for these cases is clearly high. This is the situation for which the fan coolers are considered. While the PDS where potential benefits exist has been established, it remains to determine for each of [ t these, namely PDS 111,311,511,3SBO,4SHO: (1) how many fan coolers will keep the overlying pool from drying out, (?) for what fraction of the frequency of these PDS are the fans available, and (3) can th operating fan coolers he expected to continue to operate, if they are started and their importance is [ understcod by the operators. In order to determine the number of fan coolers required, a bounding scenario was established based on PDS 5H, which is a large break 1.OCA without ECCS. Since all the PDS under consideration are dry, , all involve ECCS tallure to inject. Thus, while 511 is bounding due to the early core melt. it is also a L representative scenario. Calculations were performed for 0,2,3, and 4 fan cases. The most significant-result is shown in Figure 4.615 which shows the mass of water in the cavity (MWCl) for each case, While the time to dry out increases, as expected, with increasing the number of fans in operation, only when all four fans are in operation does the cavity remain ' wet indefinitely. Therefore, all four fans would be req.iired to prevent'containmer failure for PDS IH,3Fl. 5H,3SBO,4SBO.
+
_t The next step is to examine PDS IH,311,5H,3SBO,4 Silo, to determine for who fraction of their - frequencies of occurrence could fans be available or recovered. A preliminary examination revealed that this fraction would be negligible, unless recovery actions are under keni Therefore, the benefits of containment failure probability reduction from fan coolers have not been included in the present IPE. Nevertheless, the present discussion is relevant as part of a knowledge base for accident management. 4233 W p- gr-- y 7 ,r ry ye- ag @ , ,y-gr-i .<- -t m ma.a2 ,.-p-i..- -g u - , ram.,ry----w-, e m--- _.em-c m--yyv--g-mn.o--,%-gy- . -4w-sq,.h .3m Ayg4-=+,g .wa i i 4 y TT"W'r 49't M F' !"' E8
4!;lii!iIt fi!lifj> f l .f[ r!l5 t<t [ .(!f;JliI[ : eti E_ _
)
i s 7578 67 8 8 S 4 01 5 539 0 p 91 39 0.3 9 2 O6 07 3 8 08 9 ( 9 S 9 74 21 761 1 6 667 651 0 4 1 R7 6 4 8 7 1 2 82 4 3 88 2 6 E 0 1 1 1 1 1 1 1 1 1 1 1 1 L I t 1 1 3 0 ;8 9 2 3 1 3 1 28 03 8 c 5 81 4. 3 9.0. 6.8 6.3 5.5 9 4.0 8 0 A 9 622 4 066 5 93 09 4 1 4 25 9 R 0 5 1 1 1 4 0 9 53 1 1 1 0 9 1 1 1 1 8 6 2 1 1 87 4 F R 4 4 34 93 7 0 44 90 02 8 O 0 6 2.2 7 %.3 2 23 84.55 3937 68 7 1 3 n F 9 01 52 1 60 0 71 00 55 76 1 0 77 4 o 0 4 2 98 1 1 4 2 0 0 1 1 1 1 1 1 1 4 i Z E S 1 3 8 r I 34 02 98 94 5 f o R 5 8 33 4 2 6. 12 63 7 1 9. 3 7.5 s 8 2758 4 s E 7 64 02187 54 8 697 4 25S o%7603 6 n R 0 1 1 1 21 1 1 98 65 l 3 i o U t u S 7272 6 8 8 4 4 0 37 8 4 2 S b n E 0 1 6 8.5 4 67 M. 2 1 8 60 0.7 0 1 7 t s R 5 8 8 6 9 8 0 8 973 66 85 3 6 7 654 8 39 1 8755 5 2 i P 0 98 65 1 D y t 428 1 8 1 8 8 968 8 4 8 05 3 h 5 2552 68 03 7064 83 51 9 t u 2 591 3 5 99 9 33 9 1 093 6 7 . a 0 7 554 8 654 54 3 3 6443 1 . b _ o r _ P 7 3 9 3 03 1 1 37 54 9 550 2 e 0 4553 4 7,2 7 J22.3 0498 5 _ 1 s 7 962 563 5 7 09 2 463 8 3 4 _ i. R 0 4 333 54 4 3 3 3 22 4 3 3 2 1 3 _ 2-e r 4 u 21 3 5 8 2 1 3 54 57 4 7 24 2 s 5 1 78 7 8 678 9593 7545 0 s 0 5 09 6 1 7 59 8 4 3 5 6004 1 e 2221 3332 r 0 3 322 4 3 3 2 1 P _ E 5 24 1 7 4 95 91 37 9742 1 M P 1 0 5263 001 7 4 1 23 3 857 6 _ 9876 8 7 7 1 7 45 3 90 45 6 H 0 1 1 1 1 2222 1 1 1 1 2221 . 0 5 1 . 1 G _ E - R _ U e e N l oz I OI O 1 OI O I OI O I OI O
- HS i
HLHL 1 1 LHL PLHL I I LHL - _ 1 6 . 4 l e s b i sr a Vb s I SI T P e a RDM I I HI LL I OO 1 1 1 1 OO i 1 LL l 1 OO. HHLL OO I 1 . PS . 1 1 LL l 1 - 0 PI S 00 P 50 2 200 0 -1 0 2 M _ ys T 00 00053 2%%M1 i t v t u TTTT EEEE YYYY RRRR TTTT-EEEE . RRRR YYYY. E 2557_320 a ta ======= CS WWWW DDDD WWWW DDDD W DDDD DDDD W D I OI O e 1 1 1 l 1 iI l EEEE EEEE O I .E O HLI L I r w 1 1 1 i 1 1 1 t 1 1 HHl i MMMM MMMM L HMLSSEE S s SSS SSLL C er 01 2 3 4 5 " 7 CCC RRRMMHH AAOO RP 1 234 567 8 9 1 1 1 1 1 1 3 8 ic 4jl! ! ia[
i Table 4.6 2: Calculation of Containment Reliability for HPME Events ; i VAL , F:h15, F }( (A'll)/2 l 0 1 0
- 80 09W 0 1.99R 0 0
; 93 0.999 0 01 1997 0.01 0 00998 98 0.999 0 05 I.997 0.04 0 0399 102 0.999 0.I 1 997 0 05 0.0499 105 0,999 0. I 6 I.997 0.06 0.0599 112 0 999 0 23 1 997 0.01 0.0698 !
1I3 0.977 0 31 1.975 0 08 0 0700 121 0.841 0.39 1.818 0 08 0.0727 , 126 07 0.5 1.541 0.11 0.0847 129 0.5 0 61 1.2 0.11 0 066 137 0.159 0.7 0,659 0.09 0.0296 138 0 04 08 0 199 0.1 0 00995 145 0 0228 0.9 0.063 0.1 0.00314 150 0 0121 0.95 0 035 0.05 00n087 I53 0 0014 0.975 0 014 0 025 0.000168 154 0 0 985 0 (014 . 0,01 0.000007 J57 __ 0 0.999 0 0 004 0 l 159 0 0 99 0 0 001 0 ' 185 0 1 0 0.01 0 CONTAINMENT RELIAll!'.lTY FOR 0.576 1 S 2It a rrl PDSs IE, 1F.III 2E.?F 33 0 999 0 1.') )9 0 0 46 0 999 0 01 1.999 0 01 0 00999 51 0 999 3 05 1.999 0 04 0 03997
$$ 0.999 0.1 I999 0.05 0.04996 65 0 999 0 25 1 998 0.15 0.14988 79 0.999 0.5 1.998 0 25- 0.24976 91 0.998 0.75 1.998 0 25 02497 90 0.998 09 1.997 0.15 0 1498 103 0 998 0.95 1.997 0 05 0 0499 105- 0.998 0 9814 1.991 0.0113 0 0313 110 0 998 0 99 1.997 0 00865 0 00863 112 0 998 I I.997 0.01 0.009985 113 0.977 l 1.976 0 0 121 0 841 1 1 818 0 0 i 129 0$ 1 1.341 0 0 137 0 159 1 0 650 0 0 ,
138 0.07 1 0 22) 0 0 145 0 0228 1 0,0928 0 0 153 0 0014 -1 0 0242 0 0 CONTAINMENT RELIAlllLITY FOR 0999l PDS 1 F,1 F,lli,2E 2F = a = > , i. 4-235
I Toble 4.6-3: End State Probabilities for Smoll Break LOCA PDS l l lit 10 IF 211 2r 2F MM i 1 1 1 1 l A1 om Om Om 0 00 O(n om l 1
/,2 0 00 0 (n om O ts) 0 (n 0 00 DI 0m O(n O(n 0 00 0 00 O N _ __
i 82 I O(O O (C Om O (n O(n 0 00 It2 R Om Ofo 0 (C 0 00 0 (v1 Om Cl.L 0 00 0 ro O (n 0 00 0 00 O(n Cl R Om 0(n 01n Om 0 00 0 00 (*2. L Om om om 0 00 0 00 0 00 (*2 R 0 (n om 0 00 0 00 0 00 0 00 Dl L 0 00 O (C 0(n Om 0 (n O (n , Dl R Om 0 00 0 00 0 (C 0 00 O (n D2 L (, m 0 00 0(0 0 00 0 00 0 00 D2.R Om 0 00 0 00 Om 0 00 Om i Bt-l. 0 00 7 42r# Om 2 fwr 03 7 42FM 0 (M H1.R Om 3 73r47 om 10404h 3 73F47 om H41 0 00 296r42 0 00 R 23F-01 2 96F 07 0 00 D4.R om I 49044 0 00 4 14 E-03 149r44 0 00 C1 1. Om 921 F 4w 7 02 F 02_ i 74r44 9 2tr4m 1 070-02 c3 R 0m 4 62r.On 3 ura. a nir-07 4 62r-05 ute 05 C4 L 936041 3 d,:r 01 0 00 6 99F4T) 3 6pF43 0 (y) C4.R 4 70E 03 1 890 05 0 (n - 3 4 t E44 1 R40-05 0% D 3. L Oto 0 00 Om ).lRF4w 0 00 0 00 D3 R ,, - O to I mr46 7 62E-0.1 20204% 100E4m ' l .16E 01 D4 t- Om om 0 (n 4 7l E-04 Om 0 00 D4 R 7 62E-03 3 99t'44 0m A 070-03 3 99E44 0 (n Hi l. 0 00 1.46043 0 00 6 IRF4% 1 46 E-0.1 0 00 B% R 0 00 7 321 -(w 0 00 3 10F-07 7 3204w 0 00 B6-L 0 00 4 8iF 01 0 00 2 47F-02 4 A10 01 0 00 86- R 0 00 2 92F-03 0 00 1.2 4 E.04 2 9?F43 0 00 C5-L Om i 17044 2 34r-03 3.4104w 1 17044 4.23 F4)2 C4 R 0 00 $ 90E 07 1,18 EM l.710 08- 4 90E-07 _ 2.120 04 06-l. 4 70F42 4 6 AE 02 0 00 1 360 03 4 680-02 0 00 ' C6 R 2 360 44 2 39044 0 00 6 R3FM 2 33E 04 0m D4-L 9 9)F OH 1.1904w '2 370-05 2 490-04 1 19F46 4 27E44 DLR l.470-07 315E46 6.30E-04 2 TOE-04 3.14 E46 1.13E 03 D6-1 2 37E45 ' 4 73E.04 0tM 9 94r413 4 73E44 0 00 D6- R 6 2 A0-09 - 126E43 0 00 9 9EF-0.1 1.26F-03 0 (n TOT 0 999 0('A 00Mw 0 934 0 668 0 056 4-236 _ _ _ _ . _ . . - . . __. _ _ _. _ _ _ .. _ , _ -___ , ~ ,_ . ._ , _
__ _ _._ .~.__._._..__._ - . _ _ _ _ . . . _ _ _ . _ . . . _ _ _ _ . _ _ _ _ _ _ l Table 4.64: End State Probabilities for Transient PDS immensus i 3H 3E 3F 4H 4E 4F PD$ i 1 i l I l Al Of0 O(O Om Om 0 00 Om A2 0 (H O fas om O(O Om Om Om 0 00 0 (n Om R1 0 in 0 fv( ft2 L 0 00 , Of0 Om o rn 6 00 Om > H2 R 0 00 Om Dro 0 01 0 00 O to C1 1. 0 00 Om OM 0 00 O (O 0 00 Cl-R OfC OfC Om om 0 00 O fu e , c2-L Om 0 00 Om Om 0 00 Om c2 R 0 00 Om 0 00 Om 0 00 Om 111.L ON O (n . 0 00 Om O (O Om DI.R 0 00 0 00 0 00 Om 0 00 0 00 D2 L 0tJ 0 00 0 00 0 00 0m 0 00 D2 R 0 00 0 00 _ 7,01 0 00 0 (0 4 00 113 L 0 00 3 $4F 01 0 (O 9(fF42 l 4Rl 41 0m til-R 0 00 1.7x r41 0m 4 s t E44 7 441:44 0 00 , fit I, ON Om 0m 4 49042 6 9%F42 0(O H4 R Om 0 00 O a0 2 26F44 340044 0 00 C3 - 1. 0 00 4 39042 6 RAF42 3 NF41 2 99F 02 4 420 02 C3 R 0m 2 ?l0 04 3 4 41'-t ad I 449.44 1.%00 44 2 22r44 C41. 6 931!4 ) ON 0 (M 1.440-07 1404'42 0 00 C4 R 3 4xr-03 0m 0m 7 29r.01 7otras Om
- DJ L 0 00 0 00 0 f ul 0 00 0 00 0(O l' D3.R 0 00 4 77F 03 7 49043 3 36F.03 3 24E41 4 80E43 j
OfM 0f6 0 00 0 00 0 (M 0 00 D41.
~
D4 R 3 64043 0 00 0m l3RE43 132F43 0 00 B M. Om 2 44F41 0m 3 $$r.02 103E 01 0in lit R 0 (O 1 230 03 0M i 80044 416E44 0 00 14 L 0 00 0 00 0 00 16RF 02 4 R2F42 0 00 l% R 0m 0m 0 00 11 All -04 2.420-04 0m l- 001 Om I 9kr42 3 3br 03 7 40E43 ! 391'.'12 196E42 COR 0 00 90404% 169F44 3 760 05 6 76rdM, 9R4F45 C61. 2.80041 0 00 0 00 3 4 t r.03 6 32F43 a 00 C6- R I 4W43 0 00 0 00 1 17E.4M 317E.0% Om-D4-L 1 I')F 43 202F43 3 43F44 - 5 48E42 1 37E-03 2 000433 DOR 1. l *E 43 2 34F-03 4 00E44 5.40E.02 160F 03 2.310-03 D6.L 4 70F 04 0 00 0 00 2.470 42 6 44F44 0 al 1% R t4AF44 0(0 0 (O 23 R F-02 7 4lF44 0 00 TOT. - 0 99% yt 00406 0 417 0.444 0 0733 4-237 ; mn,.,.- -w,,,,_---,+,,w -
,-y g y wem-,,.,,,,,- v- t, yy-e y s - . - y -w p .r n n v-e--.w ,s-em mq w
Toble 4.6 5: End-State Probabilities foi Statlan Blackout PD3 35 . 48 iM 1 1 At 0 41 O(n A7 0m 0 fo Bi 0 fn 0m
._ $17- L 0 IM _ f) OO ft7 P Om o fn Cl L 343F43 1.70r44 Cl R 177F45 844047 C7-L 6 4f.F44 3 E41 C7 R )29Fle l 79h47 pl L nm 0 00 Dl-R 3 R3 TAM i 84F.44 D7 t. 0 00 0 00 D2 R 712E44 3 87F4M ,8tt-L__ .
om 0 00 j R1 R a 00 0 fe 84 L . O(O . 0 00 . yR _ , Om __ O(0 C3 1. _ 4 60r43 3 24Q C3-R , ,
- W t F 04 l 63F44 C4.L 7 A0bOf $.76 F41 I
C4 R 197F-03 789E43 : D3-1 0.m 0 00_ D3.R 602r44 3 41F44 ! D4 - 1. O fn om i D4- R ._ 6 47E-01 4 70f*43 j R 4 1. 0 00 0m - f It4 R , 0 00: O(n IW L, 0 00 0 00 BA,R 0 On 0 00 I C41 L2Er44 I.41044 C4 R 3 14E.07 7 ORE 07 C6-1. 1 31042 3 76E 01 C6-R 6 470-04 1gaE43 a DOL i 89F.f 0 6 Rur 05 D4-R I etF43 7.10043 IM-1.- R ?? Boa ' 1.? l F 4.1 D6.R 8 34E44 3 31F43 l TOT 0 R2 - _ .
' 0 913 4-238 e *. g --.w inw. ~ . - . , . , + - _,-_y _.W y-. e'y- . , a y p w y y,p-.s-
[ Table 4.6 6: End-State Probabilitler, for Large Break LOCA PDS I r - ma ill 40 AF 6 41 60 6F PD% 1 1 i i 1 1 r A1 0 00 .._01:1 Om um 0 00 Om 0 (t) 0 00 i A2 _ 0 (a) 0 uO _ 0 to O (W) 81 _ Oro om om 0 (H ON Om (17 L 0(0 0 ft) D is) 0 00 0f6 0 00 l Orn Om n or, O (iO 0 00 0 00 DJ .tt . E cl.I, o fn 0m o ry) 0(0 0M 0 00 w __ Cl p Om 0 00 o na O (O 0 00 0 ty)
.f2-L Om om o fo 0% O (O 00)
C2.R Gm 0 (s) e ra) 0 0f) Om n (n () l . l. OM 0m Um 0 00 OfW O N) ni-u _ nm o sy) am _ om _ om Om D2 L O (W5 8 h0 010 O ini 6m 0 (O D2 R D i m) 0 (H O fp) i 0 00 Om 0 00 II) l. 0 00 14ttr 01 Om t 6nt.0) 1 4p rJ1) 0 00 [1] R 0 00 7 49F M ufo !t01F4% 7 44FJe O(W H4 ! Om i f#7F 01 - 0 'n 6 370 01 % 070 01._ 0N . 14 R 0m 2 MF41 om 1 ? @ 01 2 4't!"413. Om CLI Om i84F44 7 37f' 02 1 A40 04 1 34r- 04 7 37F (Q C) R 0to 4 2 % F .01 )70141 424007 9 74 E.07 3 7DT M C4 - 1, 9 22F UI.. 71401 0 00 7 34EI)7 L)%V . d? 0 00 C4 R 4 N F 4)3 .1 69 0 J kt 0(O 3 69DG4 )6aE4M . OJ1 D)-{. 0 (#1 0 (k) O (10 0 Or) 0 (a) 0 00 D) R 0 00 2 Jaip,04 8 (n)E O) - 2(07 0% 2 Gil?M A OnE.68 04 L 0 00 nm 0m 0 00 o (n 0to
~ ,
D4 R R Og0.) 7 OAF 03 __, Om 7 QR F-0.1 7 48F 01 Om gi 00) 0N O 00 O f *) , 0 00 O f0 t Itt R 0 fH Uto 0 00 0m 0 00 0 00 1% l. 0 00 0 00 O (w) 0 01 0 00 00gi h6 R 0N O ial om 0 (m) 0 (n nto w C4 L Oin 0 (n 0 00 0m O (0 am COM OF OM 0 (n) ON Om O (n) 06-( A 00 0 f0 - 0 00 0 00 ~ 0 00 0M , C6 R Om 0 *') 0 01 0 06 0 00 Om D4 1. 0 00 0 00 0 00 th ul 4 00 Om DOR o rmi 0 0n ' 0 00 O RT o (n O (n DA L 0 (0 0m Olu O 00- 0 00 0m DA R 0 00 0 00 0 00 Om- 0 00 0 00 2 Tfri 0 +35 0 678 0 0it21 - 0 724 o olt 0 0A21 - , 1 - i 4439-
-. , _-,,-r.--- , 44 -_..-.~.m-.,, -,, .e--- w. .,.%, ,v.-, .,v- . , , , - - . - , ,y,c, ,, , , - ..,y-. . . , , , , ,,-~.y, ,f , , -,--...-,4 ,,--,y.+-,, - , .,-.--.i-
i Table 4.6 7: Summary of Conditional Containment Failure Probabilities for Small Break LOCA PDS 4 ___ 111 lE 1F 2H 2E 2F Early, Ink 2.38 E-05 4.74E-04 2.37E-05 1.04E 02 4.74E-04 4.27E44 Early, RupNre 7.68E-03 1.66E 03 7.68E-03 1.81E 02 1.66E-03 2.29E-03
- CCl. leakage 9.8311-01 5.06E-02 7.26E-02 7,14 E-02 5.06E-02 5.30E-02 l CCl, Rupture 4.94 E-03 2.54 E44 3.65 E-04 3.59L-04 2.54 E-04 2.66E44 -
Steam, Leakage 0.00 6.12E-01 0.00 8.50E-01 6.12E-01 0.00
- Steam, Rupture 0.00 3.08E-03 0.00 4.27E-03 3.08E 03 0.00 Early, Total 7.71E-03 2.14E-03 7,71 E-03 2.85E 02 2.14E-03 2.72E43 Late, ink 9.83 E-01 6.63 E-Ol 7.26E-02 9.21 E-01 6.63E-01 5.30E 02 Late, Rupture 4.94E 0.3 3.33E43 3.65 E-04 4.63E 03 3.33R03 2.66E44
, CCI Total 9. 88 E-01 5.09E-02 7.29E-02 7.18E-02 Sa 5.33E 02
- Steam Total 0.00 6.15E 01 0.00 8.54E-01 6.15E-01 0.00 TOTAL 9.96E-01 6.68E-01 8.06E-02 9.54 E-01 6.68E41 5.60E-02 Table 4.6-8
- Sutamary of Conditional Containment Failure Probabilities for Transient PDS a.n=~ -
.- = ! 311 3E 3F 411 4E 4F .y Early, Irak l .47 E-03 202d43 3.43 E-04 8.05E-02 2.01 E-03 2.00E-03 Early, duprure 7.3%u3 7.12E-03 7.85E-03 8.57E42 7.12E-03 7.13E 03 CCI, leakage 9.82E-01 6.376-02 7.20E42 5,64E 02 6.37E-02 6.38E 02 4
CCI, Rupture 4.93E 03 3.20E 04 3.62E-04 2.83 E44 3.20E-04 3.20E-04 Steam, Ixalage 0.00 5.99ii-01 0.0') 1.93E-01 3.69E 01 0.00
- Steam, Rupture 0.00 3.0lE-03 0.00 9.71E 04 1.85 E-03 0.00-Early, Total P.83 E-03 9.14 E-03 8.19E-03 1.66E-01 9,14 E-03 ' v. i;F-03 late, leak ') 82E-01 6.63 E-01 7.20E 02 2.50E-01 e 32E-01 6.38E-02 .
Late, ilupture 4.93 E-03 3.33 E-03 3.62E-04 1.25E-03 2.17E-03 3.20E44 CCI Total 9.87E-01 6.40E-02 7.23E 02 - 5.67E-02 6.40E-02 6.4 t E.02 Steam Total- 0.00 6.02 E-01 l 0.00 1.94E-01 3.71 E-01 0.00 ' TOTAL 9.96E-01 6.75E 01 8.05 E-02 4.17E-01 4,44E-01 7.33 E-02 1 4-240
._ . - - _ . . . _ . _ _ _ . . , . _ . _ _ . _ . _ __ . . . ~ - _ _ . _ _ . . . _ _ . , . , , , . . . . _
5 Table 4.6-9: Summary of Conditional Containment Failure Probabilities for Station Birkout PDS 35 4S Early, leak 2.72E 03 1.01 E-02 Early, Rupture 1.03E 02 1.52E 02 CCI, leslage 8.03E-01 9.53E-01 CCI, Rupture 4.04E.03 4.78E 03 Steam, Leakage 0.00 0J0 Steam, Rupture 0.00 0.00 l Early. Total 1.30E 02 2.53 E-02 i Late, Leak 8.03E-01 9.53E-01 Late, Rupture 4.04E-03 4,78E-03 CCI Total 8.07E-01 9.57E 01 Steam Total 0.00 0.00 TOTAL 8.20E-01 9.8]E Ol Table 4.6-10: Summary of ConditionsI Containment Failure Probabilities for Large Break LOCA PDS l 511 SE 5F 611 6E 6F Early, Leak 0.00 0.00 0.00 0.00 0.00 l \ 0.00
~
l Early, Rupture l 8.00E-03 8.00E-03 8.00E 03 8.00E-03 8.00E-03 - 8.00E 03 l __ l CCI, Leakage 9.82E-01 7.37E42 7.37E-02 7.37E-02 7.17E-02 7.37E-02 CCl, Rupture 4.94E-03 3.10E-04 3.70E 3.70E44 3.70E-04 3.70E-04 l Steam, Leakage 0.00 5.93 E-01 0.00 6.39E-01 5.93E-01 0.00 ! Steam, Rupture 0.00 '.990-03 0.00 3.21E-03 2.99E-03 0.00 i Early, Tmal S.00E 03 8.00E-03 8.00E43 8.00E-03 8.00E 03 3.00E Late, Leak 9.82E41 6.67E-01 7.37E-02 7.12E-01 6.67E-01 7.37E-02 Lute, Rupture 4.94E-03 3.36E-03 . 3,70E-04 3.58E43 - 3.36E-03 3.70E-04 CCl Total 9.87E-01 7.41E-02 7.41 E-02 7.41E-02 7.41E 02 7.41E-02 Steam Total 0.00 5.96E-01 0.00 6.42E-01 5.96E-01 0.00 TOTAL 9.95E 01 0.79E-01 8.21 E-02 124E-01 6.79E-01 8.21E-02 4 241
*7Ntt-'+ -
y py,9 9, . - . 9,.. ,.w g.- - g._ ,,g g,g g,y,s.g. m 9yy n,z %w-m.p,y,4,,yayyg,+y,g-g,.m,.g ,gg,,.-g.yg9,.g.w4gwg p- .,yy 4. 9
. ,. .._. _ _ _ ._. _ . _ .._. _ _ ___ _ ___ _ . _ _ . _ _ .____._._m._
e=* Table 4.6-11: Accident Sequence initiator Notation (Prefixes) INITIATOR DESCRil'fl0N INITIATOR DESCRilr110N 4 XI LOSS OF 125 V DC llUS Tl GENERAL TRANSIENT l IEDI - X2 LOSS OF CilllLED WATER T6 LOSS OF FEEDWATER X3 LOSS OF OFFSITH POWER T4 MAIN STEAM LINE BREAK X4 LOSS OF 11US 1 A3 IS INDUCED SMALL HREAK (0.6 TO 2 IN) OVER 60 GPM/PMP IF SEAL LOCA XS LOSS OF IVS INDUCED VERY SMALL l PROTECTION llREAK (UP TO 0.6 IN) ! Cll ANNEL IPCI 21 TO 60 GPM/PMP SEAL LOCA , I X6 LOSS OF CC AT ATWS j I X7 LOSS OF SW R STEAM GENERATOR tulle RUPTURE X8 LOSS OF INSTRUMl!NT AIR CV LOSS OF CONDENSER I VACUUM i Table 4.6-12: Accident Sequence Functions Notation (Suffixes) , i FUNCTION DESCRIPTION l.OST (SUFFIN) - CMi Failure to switchover to recirculation after successful injection. CM2 Loss of secondary cooling and failure to perform feed and bleed (for p T,X,CV). Injection failure (for S,M,A, i.e. LOCA). I CM3 Failure of!.econdary heat removal and bleed and feed (for VS). Injection failure (for XL). CM4 Injection failure (VS) CMS Injection and Secondary heat removal faih're (VS) CM6- Same as CMS but TDAFW runs untu battery depletion (VS). 4-242 t*-rw &*-+- &z-7-' -~"T------'T *t-ew-'egF ma---- e 'W m-9P y rw wps o sw W M--yvv e- e -T-'mv'e ee'---.-- r -, w- - W'TTf- w' a ""- -^- - " ^* " ' ' ' ' ' ' '
. _ _ . _ . _ _ . ~ . _ . . ._ _ ~ _ . _ _ _ . . _ _ _ _ _ _ . . _ _ . _ _ _ . _ _ - . . _ _ .m . Table 4.613: Level 11 Characteristics of Level i Dreak Size Ranges
.u RANGE (INCilES) LEVEL I NOh1ENCLATURE LEVEL 11
[htAIN CAUSES] CliARACTERISTICS (VF PRESSURE) 0 TO O,6 VS - VERY SMALL SAME AS NO BREAK, [SMALL SEAL LOCA,60 111G11 PRESSURE AT VF GPM/PMP{ (OVER 2000 PSIA) 9.6 TO 2 VS - VERY SMALL INTERMEDIATE PRESSURES [LARGE SEAL LOCA, 250 AT VF GPM/PMP, STUCK OPEN (400 TO 1000 PSIA) PORV) 2 TO 4' S+SMALL LOW PRESSURES ISTUCK OPEN SRV] AT VF (LT 200 PSIA) 4 TO 6 M MEDIUM LOW PRESSURES [ PIPE BREAK) AT VF (LT 200 PSIA) , OVER 6 L LARGE LOW PRESSURES [ PIPE BREAK] AT VF (LT 200 PSIA)
- Breaks greater than 2 inches all lead to low pressures at vessel failure but have different Level I success criteria.
4 243 w -- --t-v- *-swa- w r- ---N->" w bw -+ "N
1 i t l i 5 i' Table 4.6-14: Containment Failure Probabilities Sorted by PDS Frequency l' i PDS PDS REP SEQ COND UNCOND COND UNCOND COND UNCOND TOTAL
- NAME FREQ SEQ PCT EARLY EARLY CCI CCI STEAM STEAM FAIL
, PROB. i I
- j. 111 1.25E45 ISCM2 100 % 7.71 E-03 9.65E-08 9.88 E-01 1.24E-05 0.00 0.00 1.25E-05 3P i 7.06E4b *CM2 99 % 8.83E43 6.23 E-OS 9.87E-01 6.97E4m 0.00 0.00 1 7.03E46 r
3F 5.05 E-06 ATCM(36) 86 % 8.19E C3 4.14E-OS 7.23E42 3.65E-07 0.00 0.00 4.07E-07 j t 4F 4.40E-06 VsCM I 74 % 9.13E-03 4.02E-OS 6.41E42 2.82E-07 0.00 0.00 3.22E-07 I l i i SGTR 3.4SE4% RCMI 97 % 1.00E + 00 3.48 E-06 0.00 0.00 0.00 0.00 3.4SE4M 1 6F' 3.29E-06 AC'.11 62 % 8.00E-03 2.63E45 7.41 E-02 2.44E47 0.00 0.00 2.70E-07
~
i i IF 2.57E4M ISCM2 93 % 7.71 E-03 1.9aE48 7.29E-02 1.8SE-07 0.00 0.00 2.07E47 l l ! 2F 1.70E-06 SCMI 87 % 2.72E43 4.63E4N 5 33E42 9.07E-08 0.no 0.00 9.54 E-08 i i 411 1.12E-06 IVSCM4 93% 1.66E-01 1.87E-07 5.07E-02 6.36E-08 I.9E41 2.2E-07 4.6SE-07 l 1 t 6H 9.00E-07 IMCM2 99 % 8.00E-03 7.20E4N 7.41 E-02 6 66E-OS 6.4E-O R 5.8E-07 6.51 E-07 I i 5F 6.96E-07 ACM2 85 % 8.00E-03 5.57E4N 7.41 E-02 5.15E-0$ 0.00 0.00 5.71E48 , 4SBO 5.02E-07 IVSCM5 100 % 2.53 E-02 1.27E-08 9.57E-01 4.81E-07 0.00 0.00 4.93E-07 i i 5H 3.27E-07 ACM2 65 % 8.00E-03 2.61E49 9.57E-01 3.22E47 0.00 0.00 3.25E-07 l ! L 3S80 1.85E47 IVSCM6 100 % 1.30E-02 2.40E-09 8.07E-Oi 1.49E-07 0.00 0.00 1.52E47 4 t 2H 1.41 E-07 ISCMI 100 % 2.85 E-02 4.02E-Os 7.1 SE-02 1 OIE-08 8.5E41 1.22-07 1.35E-07 ! 1' V-Seq 1.19E-07 SCM1 83 % l .00E + 00 1.l?E47 0.00 0.00 0 00 0.00 1.19E-07 i , Cl 9.92E-09 ISCM2 37 % l .00E + 00 9.92E-09 C.00 0.00 0.00 0.00 9.92E-09 f Total 4.41E-05 g 4 12E-06 2.16E-05 9. lE-37 2.67E 05 100 % l 9.4 % 49 % . 2.1 % 60.5 % t ==:: <
! 4-244
i Table 4.6-15: PDS in Order of increasing Early and Total Unconditional Containment Failure . Probabilities l PCT OF PDS EARLY PCT OF PDS TCTAL CONT TOTAL t NAME FAILURE EARLY NAME FAIL. PROll. Fall PROll. TOTAL SGTR 3.48E-06 84.4 % lit 1.25E45 46.7 % 411 1.87E47 4.5% 311 7.03E-06 26.3% V-Seq 1.19E-07 2.9 % SGTR 3.48E46 13.0 % 111 9.65E-08 2.3 % 6 11 6.51E-07 2.4 % 311 6.23E-08 1.5 % 4 Silo 4.93E-07 1.8 % 3F 4.14E-08 1.0 % 411 4f8E-07 1.8 % 4F 4.02E48 1.0 % 3F 4.01E-07 1.5 % , 6F 2.63E 08 0.6% 5 11 3.25E-07 1.2 % 1F 1.98E-00 0.5 % 4F 3.22E-07 1.2 % - 4 S 13 0 1.27E48 0.3 % 6F 2.70E47 1.0 % l Cl Fait. 9.92E 09 0.2 % 1F - 2.07E47 0.8 % , l l 611 7.20E-09 0.2 % 3SPO 1.52E47 0.6% 5F 5.57E49 0.1 % 2 11 1.35E-07 0. 5 % _ 2F 4.63E49 0.1 % V Seq. 1,19E 0.4 % - l 211 .4.02 E-09 0.1 % 2F 9.54E-08 0.4 % 1
$11 2.61 E49 0.1 % $F- 5.71 E-02 0.2 %
3Si3O 2.40E-09 'N 1 % Cl Fail. 9.92 E-09 - 0.0%
-TOTAL; 4.12E-06 100.0 % TOTAL 2.678-05 100,0 % ,
4 245 _ _ _ _
---i..-.-..~,.,,,-.-.---, ,_m, .,.,m ,, -. , . , , , m y_,__._.,,.,,,y..,,.-. ,r, ,.._,_.,_..,,,.p,, . _ . _ _ .. . , _ , _ ,
Table 4.6-16: PDS in Order of increasing Late CCI and Steam induced Unconditional Containtnent Failure Probabilities CCI- FCT OF STEAM- PCT OF PDS INDUCED CCI- PDS INDUC STEAM-NAME CONT Fall INDUC NAME FAIL. INDUC PROll. FAIL. PROll. FAIL. 111 1.24E 05 57.1 % 611 5.78E-07 63.0 % 311 6.971!-06 32.2 % 411 2.18E47 23.8 % 4S!!O _ 4.81 E-07 2.2 % 211 1.20E-07 13.1% 3F 3.65E47 1.7 % 311 0.00 0.0 %
$11 3.22 E-07 1.5 % 5F 0.00 0.0%
4F 2.82E-07 1.3 % 4 Silo 0.00 0.0 % 6F 2.44E47 1.1% _ 6F 0.00 0.0 % 1F 1.88E-07 0.9% Sil 0.00 - 0.0% 3 Silo 1.49fM7 0.7 % 2F 0.00 00% 2F 9.07E 08 0.4% 3 S 11 0 0 00 0.0% 6 11 6ME-08 0.3 % 3F 0.00 0.0% 411 6.36E-08 0.3 % 4F 0.00 0.0 % 5F $ 15E-08 0.2 % IF 0.00 0.0 % 211 1.01 E-03 0.0% SGTR 0.00 0.0 % SGTR 0.00 0.0% ' 111 0.00 00% V-Seq 0.00 0.0 % V Seq 0.00 0.0% CI Pall 0.00 0.0 % Cl FAIL. 0.00 0.0 % TOTAL 2.16E-05 100.0 % TOTAL 916E-07 100.0 % 4 246
. - . - . ~ . . . -. - , - . ..-. . , -.- .
i-7 !:. Table 4.6-17: Key PDS and Their Functional Sequences Probability Composition l PDS PDS PRINCIPAL SECONDAY TERTIARY NAME I-RFQUENCY FUNCTIONAL PCT FUNC110NAL PCT FUNCTIONAL PCT TOTAL. f SEWENCES
' SEQUENCES SEQUENCES l-SGTR 3.48E46 RCMI 97% RCM4 3% 100 %
l dil 1.12E o6 IVSCM4 93 % VSCM4 5% IVSCMI 1% 99 % V-Seq 1.19E-07 SCMI 83 % MCM2 9% ACMt12) 7% 09 % !. l , 1H 1.25E45 ISCM2 100 % 100 % 3H 7.06E-06 XO6171 49 % Ttt46) 45 % CVCM2 5% 99 4 CM2 CM2 IVSCM3 3F 5.05E4 ATCMt631) 87% Xt12KM2 5% ToCM2 4% 90 % 4F 4.40E-od VSCM(14) 83 % IVSCM4 9% X(23)CMI 6% 96 5 6F 3.29EM ACMI 62% MCMI 29 % XLCMI 8% 99 % IF 2.57E-06 ISCM2 93 % SCM2 7% 100 % 4SBO 5.02E47 IVSCMS 100 % 600 % i Cl 9.92E-09 ISCM2 37% ALL 63 % 100 % i 611 9.00E47 IMCM2 99 % ACMI 1% 100 % 5F 6 96E47 ACM2 85 % MCM2 12 % IMCMI 3% 100 % 2F 1.70E46 SCMI 87 % ISCMI 13 3 100 % 2 11 1.41E47 , ISCMI 100 % 100 % i
- Sil 3.27E.07 ACM2 65 % IMCMI 35 % 100 %
3SBO 1.85E47 IVSCM6 100 % ! ISGTR 5.64E-08 VSCM(14) 33% X(3617) 19 % T(146) 17% 69 % CM2 CM2 i 1 i 4-247
Table 4.6-18: Contribution to ISOTR by PDS _ = = - - - - - . - ISGTR PDS ISGTR PERCENT PDS CONDITIONAL FREQUENCY UNCONDITIONAL OFISGTR NAh1E PRO 11 ABILITY PROliAlllLITY TOTAL _. 3.00E43 7.06E-06 2.12E 08 38 % ! 311 1.00E43 5.05E46 5.05E-09 9% ' 3F 5.00E-03 1.12E-06 5.61E-(r) 10 % 4 11 5.00E-03 4.40E 06 2.20E48 39 % 4F 3.00E44 1.85 E-07 5.55E 11 0% , 3 S 11 0 5.00E-03 5.02E-07 2.51 E-09 4% 4 Silo TOTA L = = = > 5.64E-08 -100 % Table 4.6-19: Contribution to ISGTR by Functional Scquence FUNCTIONAL ISGTR SEQUENCE FREQUENC1' VSCh1(14) 1.83E-08 i X(3617)Cht2 1.04E-08 i l T(146)Cht2 9.53E-09 IVSch14 5.22 E-09 ATCht(631) 4.40E 09 l IVSch15 2.51E-09 IVSCM4 1.98E-09 X(23)CMi 1.32 E-09 CVCht2 IVSch13 1.06E-09 X(12)CM2 4.04E 10 VSCM4 2.81E 10 T6CM2 2.02E 10 IVSCMI 5.61 E-I l IVSChl6 5.55E 11 TOTAL 5.57E-08 L _= 4 24_8
i } [ l t Table 4.6-20: Summary of MAAP Calculations for Representative Sequences { - t I CONTAINMENT Com Core Mek Vessel RWST Comaammeen t j Calc. Unceery Fe.be Depleted Fa h l Labet Segwetxe Time (h6 Trme Cir) "I rw (br) Tune (hr) Tinae thr) (PDSI Deurspten Fed. Pres- @ j l MA VF, Put [RCS Presi [RCS Pxsl 1 ', SB021 5B0,250 GFWPMP seal LOCA. 4 his 1DAFW, NO ECCS. 270 3.7I i 1 , 1111l Late, neo-cendenut4e ish.ced (CCn. megressene f ,here. L 28.7 (1100p el 32 [793 p iel N/A 3834 58022 Some es SB021 etcept e t. urn men f.med to peid cweainmere 2 s .7 2.70 331 i i!HI f.aere t>y uperre rather thma !cekare. L I1300 rmal 3.2 1M5 ras! N/A 29.1 SBO31 580,60 GPMTMP seal LOCA. NO AFW. ECCS recmero 29 0 1 66 2.55 RECIRC
' 135B01 but vessel faiss. HPME fads conta,nmens eith sprays on. R [23M paal 2 08 [23m] OK 235 i I
SBO41 Same as 58031 but spreys are faded to yield umcrubbed 29 0 - I 66 2 55 RECIRC j l [35B0! release. R [2300 pmaj 2 .'4 [2300paal OK 235 2CB Sacem Generator bbe Raptum, ECCS injects. operators fed to BY- 16.76 IS 60 i [2CBil stop break flow by &preswrizatum. SG overfdis. SG mafety PASS 14 7 (2300 psal 17.73 {soo ps.el 10.15 16.7e l . sticks eyes. V! V-Sequence sinndation.10* CL LBLOCA. no pumped ECCS. BY- SiO [ {ICBI 4 Accumulsaors PASS N/A 0.04 05 0 83 ECCS 0. ! i t
?
i 4 4-249 l J i
,~.
- 1. ,-
l- Table 4.6-20: Summary of MAAP Calculations for Representative Sequences (continued...)
- CONTAINMENT Core Core vessel RWST Contamment
'f Cale. Uncovery Melt Failure Depteacd Failure t
Label Sequence m_ Time (hr) Time (hr) Taw Tune (br) Tune [PDS)- Deserrption p,;; p g (hrt Mode [RCS (hr) [RCS VF. Psb Presi Presi
' SB2114 AFW, O RilR. 4 ACC. I CCP, O SIP. O CSP inject 24 69 26.2
[2H1 but Fail @ Rec.cc., late steam-imhur.4 overpressure R 33 42 11100 psial 25.6 [435 psia] 21.46 38.1 failux. SB2115 Same as SB2114. Mode! Parameters mi4Ged to yeld 127 25.27 30.46 , [2HJ containment failure at VF une to HPME R [1100 psia) 26 4 [1100 psial 21.71 30 46 TRAN21 NO FW,1 PORV @ 20 man for F&B. , iC.1 CCP, . 94 3 2a.A ' 23.7B {411) O SIP. O CSP. O RllR. Fail @ Recire. 21 GPM/PMP, L [2300 psiaJ 22.22 [2300j 18 94 29 62 late steam-induced overpressure failure. TRAN22 . Same as TRAN22 Model Paramcrers mod 2fied to yield 94 3 21.68 23.78 [*H] containmera failure at VF due to HPME. L [2300 psial 22.22 l2300 psi.I 18.94 2338 VSB3F1 21 GFM/PMP seal LOCA. A1L ECCS & AFW FAIL 26 1.63 2 05 2.53 NO OVER [3F] 2 CSP. Non cootable debris configuration.. N/A [2304 [23001 ECCS 220 0' CT OK VSB3F2- Restan of VSB3F1 with a forced Burn at 208 hours. 26 1.63 2.53 NO E 13F] L 12300 psh! 2 05 [2300 psial ECCS 208 33 j CTOK l VSB3F3 Restast of VSB3FI with a forced Burn at 83 hours 26 1.63 2.53 NO j [3F) R (2300i 2.05 . ECCS 8334 INel 12250 psial 12:50 rsial crOK l '., '1 Basemat 4m thickness was reached at 40 hours. t-4 4-250
- Y f
I i i i Table 4.6-215 Release Fractions fivr CET End-States and Respective PDS Representatives i !
.I END-STATE' PDS MAAP NOBLE GASES CSI TEO2 SRO MOO 2 . CSOII BAO LA2O3 CEO2 SB i RUN :
t l D2-R 3SBO ~SBO41 EE-Ol S E-02 4E-02 I E-03 6E-02 4E-02 IE43 3E-06 9E-06 4E42 t i DI-R. 3580 SB031 2E-01 3E-03 0 IE-05 3E44 2E-03 9E-05 3E47 SE47 2E-03 l ! DI-L 3SBO 58032 SE-02 2E44 0 S E-07 2E-05 2E4s 56-06 IE48 3E-08 SE-05 i l B6-L 411 TRAN21 SE-01 2E-03 0 2E-05 7E45 2E-03 2E44 IE46 IE46 2E-03 ,
= B6-R ' 21i SB2114 9E-01 9E-03 0 3E44 2E43 9E-03 2E-03 3 E '-. i 4E-06 6E-03 l C5-L 3F VSB3F2 SE-01 I E-10 IE-09 2E-13 I E-10 l IE-09 IE-Il IE-Il IE-Il 4E-09 . ; t ! C5-R 3F -VSB3F3 IE + 00 7E-07 IE49 SE-13 IE-10 SE-06 4E-Il IE-11 IE-Il IE-05 C6-L" .111 SB021 9 E-01 3E-02 SE-03 7E-06 2E45 2E42 2E-03 2E46 2E-05 6E42
] C6-R ll{ 5B022 IE+ 4 2E 01 7E-01 I E-03 IE-08 2E-01 ! 3E-03 4E45 4E-03 4 E-01 i l' ! D6-L . 411 TRAN22 I E + 00 6E42 0 SE45 IE-02 6E42 6E-03 3E45 3E-05 6E-02 ^ Z? 211 SB2115 IE + 00 3E-02 0 4E45 2E-02 3E42 4E-03 3E45 6E-05 4E42 - i V-SEQ ICB VI I E + 00 EE-01 IE+0 IE-01 2E-03 SE-01 $E-02 2E42 2E-01 EE-01 1 0 , i I I' SGTR & ISGFR 2CB 2CBI 9 E-01 SE-01 3E-02 3r-03 IE41 SE-01 3E-02 4E44 7E48 SE-01 3 i CI FAIL ICI VI IE + 00 SE-01 IEA0 IE-01 2E-03 SE-01 SE-02 ' 2E-02 2E-01 SE-01 6 0 i s i , i 1 !
! i ' End-States in this column are those actually represented by the M AAP runs listed. TABLE 3-3 provides additional infiermation on the end I state characteristics and lists similar ends states which can be represented by the same MAAP calculation. ,
t 4-251 [
~
r,,e - --
i 4
- 1 d
4 Table 4.6-22: Description of Calculated End-States CAUSE OF CONT. FISSION CALC Sih11LAR CONTAINh1ENT FAIL. PRODUCY NAhiE PDS END FAILURE h10DE REh10 VAL [END -STATES STATE] Late Steam Induced Overp. L Pool TRAN21 4H B4 L Failure No Sprays IB6-L] B2-L Late Non-Cond (CCI) L No SB021 lH C4 L Induced Overp. Fall. [C6 L) C2 L Late Non Cond. (CCI) R Sprays VSB3F3 3F C3 R Induced Overp. Failure [C5 R] Cl-R Late Non-Cond. (CCl) R No SB022 111 C4 R ' Overp. Failure lC6-R] C2 R Late Steam Induced Overp. R Pool But No SB2H4 2H B4 R Failute Sprays 186-R] B2 R Recovered Early-11Ph1E R Sprays SB031 3SBO D3-R [DI R] D5 R Recovered Early-HPh1E R No SBO41 3SBO lD2 Ri Early llPh1E R No SB2115 211 D4 R ' ID6-R] Early-HPh1E L No TRAN22 4ll D4-L [D6-L] D2 L Early IIPh1E L Sprays SB032 3S00 Dl-L [D5-L) D3-L Late Non. Cond. (CCl) L Sprays VSB3F2 3F C3-L . Induced Overp. Failure .(C5 L] Cl L
- l t
4-252 ,
i i Figure 4.6-1: Comparison of Probabilities for llPME Final Pressures with Containtnent Survival Probabilities for; Above: PDS lE, IF,111,2E and 2F; Below: PDS 211 cm x ,L, v. ae. er
,4,....._,_._..._..,.._,_;_. : .. ~
a' oe. r 9 6 ..-. / i i ..
/
i ... / I, ...
/
9
; e.-
4 a e.- t 1 D
, <s. , \
f ,, . s' %. e s ---e , g-.,. , ~,)?w4 6g h, ,, et ..
-,-..,. . ..,,,o.sm,,,,,,, ., .,, , - .y ..,,,4 ,e 4 ,s..
Ses hts 64 L043% AM t see ,>(l #4 = a y LO.' va vns e e '.*s
'n
- a h ,.b [i
.e,..._.._,,_,,_m - ,_-_}}