ML20066D265
| ML20066D265 | |
| Person / Time | |
|---|---|
| Site: | Sequoyah  |
| Issue date: | 12/31/1990 |
| From: | Galyean W, Pafford D, Schroeder J EG&G IDAHO, INC. |
| To: | NRC OFFICE OF NUCLEAR REGULATORY RESEARCH (RES) |
| References | |
| CON-FIN-A-6900 EGG-2606, NUREG-CR-5602, NUDOCS 9101140284 | |
| Download: ML20066D265 (130) | |
Text
{{#Wiki_filter:. - - - .- - - -- _ NUREG/CR-5602 EGG-2606 L Simplified Containment Event Tree Analysis for the Sequoyah Ice Condenser Containment l l l V. S.$alycan, J. A. Schrmder. D. J. Pafford i Idaho Nationa Engineering Laboratory EG&G Idaho, Inc. L l Prepared far U.S. Nuclear Regulatory Commission i L l
$$R21$8 p
SEA 73S$50$v7 PDH
NUREG/CR-5602 EGO-2606 ) RG, R1, UL, l A,1S, X A ! Simplified Containment Event Tree Analysis for the Sequoyan Ice Condenser Containment Manuscript Completed: October 1990 Date Published: December 1990 Prepared by W. J. Galycan, J. A. Schroeder, D J. Pafford Idaho National lingineering I aboratory Managed by the U.S. Departrnent of I!nergy IIG&G Idaho, Inc. Idaho Falls, 10 83115 Prepared for Division of Safety Issue Resolution Office of Nuclear Regulatory Research U.S. Nuclear Regulatory Commission L Washington, DC 20555 l NRC FIN A6900 Under DOE Contract No. DE-AC07-761001570 I i
1 I NUREO/CR-5602 EGG-2606 Simplified Containment Event Tree Ana:ysis for the Sequoyah Ice Conc enser Containment .
- Prepared by W. J. Galycan, J. A. Schroeder, D. J. Pafford j'_
L hinho National Engineering Laboratory
'.EG&G Idaho, Inc.
t: I-
- l. Prepared for
. U.S. Nuclear Regulatory Commission l
l l p U Ult'
. . . - - . --. ., . - . . .~. - . - ... . - -.-..
.I AVAILABILITY NOTICE Availatiihty of Refererte Materials Cited in NRC Putilcate ns Most documents cited in NRC publications will be avaliable from one of tha following sources:
- 1. The NN Public Document Room, 2120 L Street, NW, Lower Level. Washington, DC 20555
- 2. The Superhtendent of Documents, U.S. Government Printing O fue, P.O. Box 37082. Wastington, DC 20013 7082 1
- 3. The National TechnicalInformation Scrylce, Springfield. VA 22101 Atthougn the ksting that follows represents the majority of documents cited in NGC publications, it is not htent.od to be exhaustive.
Referenced documents avaltable for hspection and copying for a fee from the NRC Public Document Room include NRC correspondence and internal NRC memoranda: NRC Othee of Inspection and Enforcement butletins, circulars, information notices, inspection and investigation notices: Licensco Event Reports! ven-dor reports and correspondence: Commission papers; and applicant and beensee documents and corre. spondence. The following documents in the NUREG series are avaltable for purchase from the GPO Sales Program; formal NRC staff and contractor reports, NRC sponsored conference proceedings, and NRC booklets and brochures. Also available are Regulatory Guides. NRC regulations in the Code of federal Regulations, and NucIcar Regulafory Commission Issuances. Documents available from the National Technical information Service hclude NURE3 series reports and technical reports prepared by other federal agencies and reports prepa'ed by the Atomic Energy Commis. s600, forerunner agency to the Nucicar Regulatory Commission. Oncuments avallable from pubho and special technical libraries include o5 open nterature items, such as books, journal and portodical articles, and transacCans. Federal Register notices, federal and state legisla. Oon, and congressional reports can usually be obtained from these libraries. Documents such as theses, d!ssertations, foreign reports and translations, and non-NRC conference pro-ceedings are available for purcMee from the organization sponsoring the publication etted. Single copies of NRC draft reports are available free, to the extent of supply, upon wiltten request to the Office of information Resources Management, Distribution Section, U.S. Nuclear ReDuiatory Commission, Washington, DC 20555. Copies of industr.y codes and standards used in a substantive manner h the NRC regulatory process are maintained at the NRC Library,7920 Norfo,tk Avenue Dethesda, Maryland, and are availabio there for refer-ence use by the public. Codes and sta'idards are usually copyrighted and may be purchased from the orighating organllation or, if they are American National Standards, from the American National Standards Institute,1430 Broadway, New York, NY 10018. DISCLAIMER NOTICE This report was prepared as an account of work sponsored by an agency of the Uruted States Govemment. Neither the United States Govemment rar any agency thereof, or any of their emp:oyees. makes any warranty, expresed or implied, or assumes any legal liability of responsibihty for any third party's use, or the resutts of such use, of any information, apparatus, product or process disclosed in this report, or represents that its tra by such third party would not infringe prt,uis '" owned rights.
4 NUREG/CR-5602 EGO-2606 RG,- R1, UL,1 A,1S, X A Simplified Containment Event Tree Analysis for the Sequoyah Ice Condenser Containment i
' Manuscript Completed: October 1990 Date Published: December 1990 Prepared by W. J. Galycan, J. A. Schroeder, D. J. Pafford Idaho National Engineering laboratory Managed by the U.S. Department of Energy EG&G Idaho, Inc.
Idaho Falls,ID 83415
' Prepared for Division of Safety Issue Resolution '
OITice of Nuclear Regulatory Research i - U.S. Nuclear Regulatory Commission ,
- Washington, DC 20545 NRC FIN A6900 Under DOE Contract No. DE-AC07-761D01570 l
l l l 1:
ABSTRACT An evaluation of a Pressurized WaterReactor(PWR) ice vdensercontainment was perfomied. In this evaluation, simplified containrr avent trees (SCETs) ' were developed that utilized the vast storehouse of infc. -,ation generated by the NRC's Draft NUREG-ll50 effort. Specifically, the computer programs and data files produced by the NUREG-ll50 analysis of Sequoyah were used to electroni-cally generate SCETs, as opposed to the NUREG-1150 accident progression event trees (APETs), This simplification was performed to allow graphic depiction of the SCEh in typical event true format, which facilitates their understanding and use. SCETs were developed for five of the seven plant damage state groups (PDSGs) ./ identified by the NUREG-1150 analyses, which includes: both short- and long- ' term station blackout sequences (SBOs), transients, loss-of-coolant accidents (LOCAs), and sticipated transient without scram (ATWS), Steam generator tube rupture (SGTR) and event-V PDSGs were not analyted because of their contain. ment bypass nature. After being benchmarked with the APETs,in terms of contain-ment fallutt mode and risk, the SCETs were used to evaluate a number of potential contaltunent modifications. The modifications were examined for their potential to mitigate or prevent containment failure from hydrogen bums or direct impinge-ment on the containment by the core,(both factors identified as significant contrib-utors to itsk in the NUREG-1150 Sequoyah analysis). However, because of the relatively low baseline risk postulated for Sequoyah (i.e.,12 person-rems per rvactor year), none of the potential modifications appear to be cost efrective. , FIN No. A6900-PWR Ice Condenser Containme.it Venting / Enhancements ill
i l EXECUTIVE SUMM ARY As presented la draft NUREG-1150, the anal- veloped utilizing the AFi Se basis for deter-ysis of the Sequoyah Nuclear Power Plant with an mining both eveni m. pendencies and Ice-condenser containment yielded the identifi- probabilities. Furthermore, the SCUT results cation of seven risk important plant damage states were benchmarked against those produced by the (PDS). These PDSs are identified as: APETs Generally, very good agreement was achieved between the SCET and the APET re-PDS-1 slow or long-term station black- sults for the containment failure mode results out (SBO-LT) (i.e., conditional containment failure probabili-ties), flowever, only a satisfactory match was PDS-2 fast or short-term station black- achieved on the risk results. Most likely, a further out (SBO-ST) refinement of the source term binning (i.e., the procedure for generating i% 14-character source PDS-3 the occurrence of a loss-of- term vector) would yic:a more precise results. coolant accident (LOCA) l PL)S-4 Once the M'ETs were available, they were
- event-V sequence used to assess the benefit associated with a num-PDS-5 transients ber of potential containment improvements, which included (a) backup power to the hydrogen PDS-6 anticipated transient without igniter system, (b) backup power to the igniters scram (ATWS) and to the containment air recirculation fan sys-tem, (c) mitigation of direct impingement con- <
PDS-7 steam generator tube rupture tainment failures, and (d) hydrogen control (SGTR), through the inerting of the containment atmosphere. Of the seven, two are containment bypass se-quences (event-V and SGTR). Because of their None of the potential containment modifica-containment bypass nature, these two are not in- tions appear to be cost effective in reducing the cluded in the present analysis, risk for Sequoyah. This is best illustrated through the use of a bounding calculation that shows a to-Simplified containment event tree (SCET) tal of $480,000 would be justifiable for backfits methodology has been applied to . the provided,100% of the population dose risk NUREG-il50 Sequoyah APET models for each (12 person-rem per reactor year) could be of the five PDSc in which risk is influenced by averted and assuming the plant life expectancy is containment performance.These SCETs were de- 40 years. t iv l l i i . ~ f
FOREWORD l SECY-88-147, dated May 25,1988, presented the NRC staff's program plan to l cvaluate generic severe accident containment vulnerabilities via the Containment Perfonnance Improvement Program (CPIP).This effort was predicated on the pre-sumption that there are generic severe accident challenges for each light water i reactor (LWR)contaimnent type that should be addressed to determine whether ad-1 ditional regulatory guidance or requirements concerning needed containment fea-tures are warranted, and to confinn the adequency of the existing Commission policy. These challenges should be addressed to determine the possible need for additional regulatory guidance or requirements related to containment features, l The ability of containments to successfully survive some severe accident chal- l i lenges is uncertain, as indicated in Draft NUREG-1150. The CPI effort is intended to focus on evaluation of hardware and procedural issues related to generic con- f tainment challenges.
'nds report documents the results of NRC-sponsored research related to severe accident challenges and potential enhancements that could improve containment performance. The purpose of this report is to provide PWR lee Condenser owners with information they may find useful in assessing their plants as part of their Indi-vidual Plant Examination (IPE) program. No requirements are contained in this re-port and it is being provided for infonnation only. Specific guidance to the industry on the use of this report, and similar reports has been given in Generic Letter 88-20, Supplement 3, dated J uly 6,1990, v
CONTENTS l A B STR ACT . . . . . ..... , ... .... ......... ........... ....... .. .... .. iii EXECUTIVE
SUMMARY
. . . . . . . . . . ...................... ........ .. .. iv FO R E W O R D . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ... .. . v ACRONYMS ............... .. . . ........... .. ...................... x
- 1. INTRODUCTION . . . . . . . . . . ....... . .. . .. .............. ... .... . I
- 2. SCET DEVELOPMENT METilODOLOG Y . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 2.1 Initial Checkout of the Draft NUREG-1150 Codes and Data . . . . ,. 3 2.2 Development of the SCETs for Each PDSG . . . . . . . . . . . . . . . . .. . . 6 2.3 Development of Containment Failure Mode Binnint ,... ....... . .... 6 2.4 Risk Development . . . . . . . . . . . . . . . . . . . . . . . . . . . . ........... ......... 6
- 3. DEVELOPMENT OF A B ASE CASE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 3.1 B ase Ca s e Sc opc . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ....... 8 3.2 Containment Failure Mode Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 3.3 Risk Results . . . . . . . . . . . . . . . ..... ............. ....... .......... 11
- 4. SCET DEVELOPMENT AND VERIFICATION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13 4.1 Top Event Selection for Each Plant Damage State Group . . . . . . . . . . . . . . . . . . , . . 13 4.1.1 SBO-ST and SBO-LT SCET Top Event Descriptions . .... ... ..... 13 4.1.2 LOCA Top Event Descriptions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19 4.1.3 Transient Top Event Descriptions . . . . . . . . . . . . . . . . . . . . . . . . . . 21 4.1.4 ATWS Top Event Descriptions . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23 4.2 Containment Failure Modes . . . . . . . . . . . . . . ......., .. .... .. ... 23 4.2.1 Containment n ..ilure Mode Binning . . . . . . . . . . . . . . . . . . . . . . . . 23 4.2.2 Containment Failure Mode Comparison With APET Results . . .. 25 4.3 Risk .. ............ ...... ..... .. . ......... . ....... ... 30 4.3.1 Release Mode Probabilities . . . . . . . . . . . . . . . . . . . . . . . ... . 31 4.3.2 Source Term Calculation for Each Release Mode . . .... ,. . 35 4.3.3 Consequence Calculation . . . . . . . . . . . . . . . .. . .. .. . . 39 4.3.4 Comparison with Draft NUREG-1150 . . .. .... . . , 43
- 5. ANALYSIS OF POTENTIAL CONTAINMENT IMPROVEMENTS ... . . . . 47 vi
? h~ ^
-.a.-..
Backup Power to the 11ydrogen Ignition System . . . . . . . . . . . . . . . . . . . . . . . . . 47 5.1 llackup Power to the lilS and Air Recirculation Fans . . . . . . . . . . . . . . . . . . . . . . 48 5.2 I liigh Pressure Melt Ejection Mitigation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49 5.3 Containment inerting . . . . . . . . . ..............,........................ 49 5.4
- 6. R ES U LTS A ND CON CL U S I ON S . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52 R E FE R E N C E S . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53 7.
APPENDIX A-lCE CONDENSER DESIGN FEATURES . . . . . . . . . . . . . . . . . . . . . . . . . . . . A-1 APPENDIX ll-APET BASED RISK ANALYSIS OF lilS AND ARFS IMPROVEMENT . . . Ibl APPENDIX C-DATA FILE LISTINGS FOR SEQUOYAll SCET DEVELOPMENT , . . . . C-1 APPENDIX D-CATALYTIC llYDROGEN IGNffERS . . . . . . . . . . . . . . , . . . . . . . . . . . . . . D-1 E-1 APPENDIX E-FRONT END ISSUES AN ALYSIS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . FIGURES NUREG-.1150 PRA code relationships and data flow requirements. . . . . . . . . . . . . . . 4 1. 14
- 2. Sequoyah short-term station blackout simplified containment event tree. . . . . . . . . . , . . .
15
- 3. Sequoyah long-term station blackout simplified containment event tree. . . . . . . . . . . . . , .
Sequoyah loss-of-coolant accident simplified containment event tree . . . . . . . . . . . . . . . 20 4. Sequoyah transient simplified containment event tree . . . . . . . . . . . . . . . . . . . . . . . . . 22 5. Sequoyah anticipated transient without scram simplified containment event tree. . . . . . 24 6.
- 7. Distribution of source terms before repooling. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36 Source term distribution after repooling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37 8.
- 9. Source term (for zero early fatality potential source tenns) distribution before and 38 af t e r r e poo l i n g . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Source t erm g roup ide n ti fie rs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39 10. A-1. General ArTangement of Containment-Sideview (from Catawba FS AR. figure 1.2.2-15) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A-4 l A-2. General Arrangement of Containment-Topview (from Catawba FS AR. figure 1.2.2-10) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A-5 TABLES
- 1. Sequoyah mean accident progression bin probabilities for PDS groups: comparison 9
with NUREG/CR-4551 results. . ... . .. ........ . .., . . . . . . . . . . . . . . . . . . vii
l 1 1
- 2. Sequoyah mean risk potentials: comparison to SNL base case results . . . . . . . . . . . . . . . I1
- 3. Sequoyah base case mean risk measures: comparison to NUREG-1150 base case results. Doses in person-tem per reactor year , . . . . . . . . . . . . . , . . . . . . . . . . 1I
- 4. Comparison of SCET and APET accident progression bin mean probabilities for SBO-LT PDS at Sequoyah . . . . . . . . . . . . . z . ................................. 26
- 5. Comparison of SCET and APET accident progression bin mean probabilities for SBO-ST PDS at Sequoyah. .. .. ,. ......,............................. . 27
- 6. Comparison of SCET and APET accident progression bin mean probabilities for LOCA PDS at Sequoyah. . . . . . . . . . ............................ . ........ 28
- 7. Comparison of SCET and APET accident progression bin mean probabilities for Transient PDS at Sequoyah. . . . . . . . . . . . . . . . . . . . ............... .... ... 29
- 8. Comparison of SCET and APET accident progression bin mean probabilities for ATWS PD S a t S e q uoy ah . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
- 9. Source term characteristic definitions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
- 10. Sequoyah source term data by PARTITION group and subgroup. . . . . , . . . . . . . . . %
- 11. Consequence data presented by release group and subgroup. . . . . . . . . . . . . . . . . . . . . . . e
- 12. Sequoyah SCET risk results for five PDSGs . . .............. ..... ......... 45
- 13. SCET risk resuits compared to NUREG-1150 risk results (FCMR and MFCR methods utilize APETs and are from NUREG/CR-4551 Vol. 5, Table 5.1-2). . . . . . . . . . . 40
- 14. Risk comparison between base case and modification #1 (backup power to igniters),
utilizing SCETs . . . . . . . . . . . . . . . . . . . . . . . . ................................ 48
- 15. Risk comparison between the base case and modification #2 (backup power to the igniters and fans), utilizing the APED . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49
- 16. Risk comparison between base case and modification #3 (high pressure melt ejection mitigation, lining the containment wall in the seal table area with refractory 1
material), utilizing SCEn . . . . . . . . . . . . . . . . . . .......... ... .. ............ -50 i
- 17. Risk comparison between base case and modification #4 (inerting containment atmosphere), utilizing SCETs . . . . . ,, . , ,.
... .... ............. . . ... 51-B-1. Conditional probability of accident progression bins at Sequoyah, with backup power t o fa ns a n d i g n i t e rs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Ib6 B-2. Comparison of station blackout weighted averages of the accident progression bin
- probabilities for Sequoyah. with backup power for fans and igniters _ . . . . . . . . . . . . . . . . B l viii
B-3. Comparison of weighted average accident progression bin probabilities for Sequoyah
.... ..... . . . .. B-10 witi' backup power for fans and igniters . . , .. .
B -l. Sequoyah mean risk potentials: comparison of base case with the backup power - supply to fans and igniters sensitivity .. ....... . . . .. . ..... .... ... .. B-12 B-5. Sequoyah mean risk measures: comparison of base case with the backup power supply to fans and igniters sensitivity case . . .. . . ............ .... .. . .
... B-12 E-1. Human Errors Dominating the Failure to Perform the Realignment from HPI to HPR for Sequoyah . . . ....... . .. ... .............................E-5 E-2. Sequoyah Risk Sensitivity Analysis to Improving Operator Performance in Realigning from HPI to HPR (Risk per reactor year) . ... ......... . .............. . E-6 E-3. Sequoyah Risk Sensitivity Analysis to Refilling the RWST for continued HPI Operation (i.e., obviating HPR failures). (Risk per reactor year.) . . . . . . . . . . . . . . . . . . . . . . . . . . . . E4 i
i I i I ix l l l
l ACRONYMS I l APET Accident Progressior. Event Tree NPP Nuclear Power Plant ARF Air Return Fans NRC Nuclear Regulatory Commission ATWS Anticipated Transient Wi0iout Scra'n PDS Plant Damage State BMT Basemat Mell-Through PDSG Plant Damage State Group CCI Core-Concrete Interaction PORY Power Operated Relief Wlve CF Chronic Fatalities PRA Probebilistic Risk Assessment CPI Containment Performance PSIA Pounds per Square Inch Absolute Improvement RCS Reactor Coolant System CPIP Containment Performance improvement Program RWST Refueling Water Storage Tank DCH Direc: Containment Heating SBO Station Blackout EF Early Fatalities SCET Simplified Containment Event Tree EVSE Ex-vessel Steam Explosion SG Steam Generator HPME liigh Pressure Melt Ejection SGTR Steam Generator Tube Rupture IVSE In-vessel Steam Explosion SNL Sandia National Laboratory LOCA Loss-of-Coolant Accident SRV Safety Relief Wlve LOSP Loss of Offsite Power VB Vessel Breach I x i
SIMPLIFIED CONTAINMENT EVENT TREE ANALYSIS FOR THE SEQUOYAH ICE CONDENSER CONTAINMENT
- 1. INTRODUCTION Because of the concem about the ability of nu- For the evaluation of potential ice condenser clear power plant (NPP) containments to survive containment improvements, the CPIP has pro-the effects of possible severe accidents, the NP.C duced a document entitled "An Assessment of initiated a program to evaluate the vulnerability lee-Condenser Containment Performance Is-of the various containment types to the chat!cnges sues."3 This document identifies several im.
posed by possible severe accidents. This pro- provements thought to have the potential for cost gram, entitled the Containment Performance Im- effective reduction in the risk resulting from se-provement Program (CPIP),ciamines each of the vere aceMents. These improvements include: containment types fotmd at U.S. NPPs. This re- (a) an improved hydrogen igniter system, port dccuments work perfomied in support of the (b) manual RCS depressurir.ation, (c) enhance-evaluation of the Ice Condenser containment ment of the air return fans,(d) enhancement of the type. containment spray system,(c) reactor cavity flooding, (f) core debris control, (g) filtered con-tainment venting, and (h) containment inerting. The analysis described in this report draws Varbus aspects of these items have already been heavily from the NRC analysis of severe accident addressed using the full weight of the 1150 meth-risks documented-in Draft NUREG-1150. odology.2This report focuses on providing back. up power to the hydrogen igniters, backup power ! NUREG-1150 represents a state-of-the-art un-derstanding of severe accident phenomenology to both the igniters and fans, prevention of direct as it existed in 1988, it provides a catalog of risk impingement failures, and containment inerting, significant accident scenarios and containment events, and is intended to provide a testbed for the The benefit of these improvements is deter-evaluation of risk-reduction measures. However, mined using the SCETs developed for this proj. NUREG-ll50 is a result of a level of effort that ect, with the exception of the backup power to cannot easily be duplicated. Furthermore, the igniters and fans improvement, which is eva-methods and data used in 1150 are not completely luated using the detailed APETs from the 1150 documented one year from the publication of the analysis of Sequoyah. The SCETs are developed main report.This makes it less likely that utilities utilizing the APETs created in i150.The method-
. will use the full 1150 methodology in their indi- ology used to construct the SCETs is outlined in vidual plant examinations. Therefore, this analy- Section 2 of this report. Section 3 describes the sis has two goals: (a) to utilize the understanding preliminary steps required before SCET develop-gained through the 1150 research to evaluate the ment can begin, namely verifying the integrity of tisk reduction potential of possible containment the 1150 computer codes and data files. Section 4 improvements and (b) to provide a less compli- then provides details of SCET construction for t
cated framework for the analysis of ice condenser five of the seven plant damage state classes or containments.The purpose of this simplified groups identified in the 1150 analysis, The ex-framework,is to provide utilities with a contain; cluded groups involve bypass scenarios that ment analysis option that could be implemented would not be affected by the aforementioned con-in the individual plant examination process, tainment improvements. The groups analyzed in-which has its roots in the NUREG-1150 - clude both short- and long-term station blackout, knowledge base. LOCA, tramients, and ATWS. The SCETs l 1 i
-- _. . . . - - - - . . - -- -- . . _ - . . ~ . - - . . - . .-- -- - - -. developed for each plant damage state group are . It should be noted that the SCETs developed then used in Section 5 to evaluate the risk reduc. for this artalysis are a condensed version of the tion achievable with the candidate improvements, 1150 APETs, and not an independent construc-The SCim were not developed in sufficient de. tion as was provided in the Mark l cpl analysis.3 tail to evaluate an improvement consisting of They are developed directly from the 1150 data. backup power to both igniters and fans (the effect The resuits of the base case containment failure of the containment air recirculation fans is not ex- mode and risk predictions obtained with the plicitly modeled it the SCETs). Therefore, this SCETs have been benchmarked against the re. improvement was evaluated directly with the sults obtained with the 1150 APETs, and general-1150 APETs. ly display good agreement with published results. 2
. -.- -. _ .- -- - .- - .-. - _ - . . ~ . - - - - _ - - - - . - -
N
- 2. SCET DEVELOPMENT METHODOLOGY
, The method of analy/ing containment re- formed using personal computer software. Spe. sponse developed for Drafi NUREG-il50 in- cifically, ETA-IIl' and Lotus 1-2-3.5This allows volves the use of large APETs, These trees for fast and inexpensive treatment of sensitivities, typically have a hundred or more events, each and provides results in a form significantly more J comprising several branches or options. Conse. scrutable than could be obtained with the more quently, the resulting event trees are too large for detailed APETs. The following sections describe graphical representation, and the endstates are so the codes, data, and procedures used to extract the numerous they are incomprehensible without SCETs from the APET data base, computer-based reduction. These factors make understanding the possibilities described by the 2.1 Initial Checkout of the Draft tree extremely difficult. Furthermore, the NUREG-1150 Codes and NUREG-il50 APETs require significant com-puter resources to process, making detailed sensi. Dala l tivity studies prohibitively expensive. The advantage of the APET methodology is that it Development of SCETs requires the use of l produces a model, complete with an uncertainty the codes and data files used in the Draft estimate,of the current knowledge of severe acci- NUREG-1150 analyses, Many of the dent progression phenomenology. What is need. NUREG-il50 level-2 codes do not yet have ed is a way to access this phenomenological data nser manuals, and of those that do, some are base that suppresses many of the details, yet pro- draft versions that do not necessarily reflect vides sufficient information to understand the what is in the working version of the code. Simi-risk significant containment events that result, larly, most of the data used in the analysis exists , on magnetic media and is not documented. Fur. The analysis described in this report relies thermore, the complexity of the data transfers re-mainly on SCETs to evaluate potential contain. quired to ensure proper information flow from ment improvements and provide a framework for one code to another makes proper use of these understanding the Draft NUREG-1150 analyses. codes very difficult for the uninitiated, These fac. Because SCETs are limited to 10-20 top events lors make verification of calculated results ex-and at most a few hundred endstates,it is possible tremely important, in this analysis, verification of to graphically display them. These event trees, the codes and data was achieved through the es. while still large, can be understood without com. tablishment of a base case calculation that could puter-based reduction. When properly bench- be ben (nmarked against published data, marked, the SCETs can reproduce many of the results obtained with the full APETs. Ilowever, The codes used in the development of the information is lost when extracting the SCETs SCETs are principally EVNTRE,6 PSTEVNT,7 from the APET data base, in most cases this is SEQSCR,s PARTITION,' and a number of un-not significant. documented translator codes. The relationship of these codes to each other and their data flow re. Once developed and benchmarked, the SCETs quirements is shown schematically in Figure 1. can be used to perform sensitivity studies on the Essentially. it is necessary to recalculate the entire !~ value of potential containment improvements, back end (level 2 and level 3) analysis to verify While not capable of duplicating results stem- that all data files are intact and all codes are ming from some of the more subtle interactions correctly used, modeled in the APETs,the results from these sen-sitivities will generally illustrate what contain- The checkout process begins with the ment response can be expected from a given EVNTRE code, which is used to evaluate the improvement Also, sensitivities can be per. APET. EVNTRE is a generalized event tree 3
J ACCIDENT SOURCE TERM CONSEQUENCE RISK PROGRESSION ANALYSIS ANALYSIS ANALYSIS ANALYSIS
- Accident Progression event tree F
- Mapping of EVNTRE ^ Id8"I a PARTITION -- -- accident progression TEMAQ g =
bin f progression DATA bins to source PSTEVNT cfiaracteristics l i term groups l Source term group , I o characteristics Probabilities of - I _ glgg accidenti progression SEOSOR STER g bins for each plant k dar. sage state j MACCS P Source term Total magnitudes and l annual MASTERK = release information i risk for eacle accident Consequence progression bin + measures for each source term group-.- i
- - -SEOFRQ ~
Figure 1. NUREG-1150 PRA code relationships and data flow requirements.
processor capable of evaluating very large trees. SEQSOR output is passed on to a reduction code c it was developed so that individual parameters . called PARTITION. could be tracked and manipulated with user-defined functions and procedures.This was nec- PARTITION idemifies an early and chronic fa-essary to evaluate the uncertainties involved in tality weight for each source term generated by the complex phenomena occurring during severe the SEQSOR program. It can optionally provide a accidents. Ilowever, a typical EVNTRE rtm for a summary of these fatality weights over all the re-single plant damage state group can take 24 hours leases, or continue with the reduction by locating on a workstation (e.g., an Apollo 3500), and even each release on a two-dimensional plot of the car. longer on a personal computer. Features are pro- ly fatality weight (EF) versus the chronic fatality vided to save EVNTRE results for later process- weight (CF). If the summary is requested, it can ing with a faster running post processor called be used to verify that the correct risk potential has PSTEVNT. The process of saving the EVNTRE been calculated. If the reduction is requested, output for later evaluation is called accident pro- PARTITION divides the fatality space into a gression binning. The results from an EVNTRE user-specified number of cells and calculates a run cannot be verified against published results, frequency weighted average source term for each so the first verification occurs after the accident cell. PARTITION also divides the releases identi-progression bins are reduced to containment fail- fied with each cell into subgroups based upon the ute modes using the PSTEVNT code. time of the releases relative to the evacuation start time. The output from PARTITION includes the averaged source term release information for The PSTEVNT code is used both to reduce the each source tenn group (cell) and subgroup. This accident progression bins to containment failure nformation, after some additional formatting, is modes,and separately, to reduce the accident pro- used in the h1ELCOR Accident Consequence gression bins to source term bins. The output Code System (htACCS) analysis.M from the containment failure mode reduction step is the first fully verifiable data produced in the The offsite consequences associated with each analysis. That is, all data produced by this step source I rm group are calculated using NIACCS. can be checked against published results. The h1ACCS mput decks provide the required site and output from the source term reduction step is meteorological data, emergency response intor-passed to the parametric source term code mahon, dose data, and odu'nclevant infonnaOn. SEQSOR' The h!ACCS consequence infomiation is the last data input required to complete the risk calcula-SEQSOR generates source term release infor. tion. After assembly in a risk matrix (e.g., utiliz-( ing Lotus 1-2-3), the final risk numbers can be l mation for each of the source term bins passed from the PSTEVNT code. The calculations in verified by comparison with published results. SEQSOR are based on parametre representa-tions of more detailed mechanistic accident pro- To summarize, risk calculations using the gression calculations (e.g., Source Term Code NUREG-1150 computer codes and data can only Package or STCP). Tbc code is also capable of be verified at three points in the process illus-representing uncertait .:s in key source term is- trated in Figure 1. These points are (a) after the sues. The source term release information in- calculation of reduced containment failure bin cludes the number of release plumes, the time of probabilities, (b) after calculation of the fatality reicase, the duration of release, the energy of the potential summary, and (c) af ter calculation of the release, the height of release, and the source tenn final risk numbers. Verification of calculated re-release fractions for each plume. Direct computa- sults at these three points is felt to provide ade-tion of consequences for each of the many source quate assurance that the codes have been properly terms generated by SEQSOR would require ex- installed, are being used correctly, and that the I cessive computer resources, therefore the correct data are being used. At completion of 5 1 1
~~n . - - - - _ . . _ - - - - . - - - - - - - - - . . - . _ _ . . . - F these checks the code system is ready for use in - the sort feature of PSTEVNT, to create the SCET
-developing the SCETs. in its final form.
2.2 DeVel0pment Of the SCETS .The sorted output from PSTEVNT is next for Each PDSG i aded into the ETA-ll pc-based event tree graphics program 3/ The PSTEYNToutput is con-verted from ASCII teu file into ETA-Il data file
, Development of the SCETs starts with defim.-
format 'using ETLOAD software." Once loaded tion of the important phenomena and containment i-failure modes. Much of this information is sum- nio ETA-!!, the SCET can be displayed graphi-Wy, Benchmarking of the SCET by containment marized in the NUREG/CR-4551 report on gg Sequoyah and/or the issues characterization re-f g3 port.' Once the critical issues are identified, the APET is reviewed to detemune if summary events 2.3 Development of have been defined for these issues. This should be . Containment Failure Mode
' the case for most SCET top events, although some . Binning of the top events will have to consist of combina-tions of existing APET events. A binner file b then The SCET development described in the pre.
created that characterizes the APET endstates in ceding section tesults in few enough endstates that ternis of the desired SCET top events. accident progression binning is not required prior to obtaining containment failure mode results. Extraction of the SCET requires an EVNTRE Containment failure mode probabilities are ob-evaluation of the APET utilizing the above bin- tained by manually applying the binning process ning definition file. The EVNTRE post processor described in Section 2.4.3 of Reference 2. This is output file contains all the reduced APET end- easily done using an ETA-Il feature that totals se-states and is used to define the SCET for the cut- quence frequencies over user defined containment rent plant damage stete group. The information is failure mode end states. For this analysis, the stored by LHS* obser vation and requires addition-SCED are initially benchmarked against the pre-al processing to complete the SCET formation, sentation bins provided in Figure 5.3 of Draft Creation of the SCETs requires additional post- NUREG-1150.i By benchmarking the SCET at processing of the APET results, which is per- this point, it is possible to verify that the choice of formed using PSTEVNT. First, the top events is adequate to represent the significant by-observation data from EVNTRE is tebinned to containment failure modes Benchmarking of the form an aggregated collection of endstates reflect-SCED by risk and risk potential is also required ing the mean response of the APET. This mean and is discussed in Section 2.4. APET response is different than would bc ob-tained by evaluating the APET in_the point esti- 2.4 Risk Development mate mode because it includes the possibility of
- containment events that only occur when samples Development of risk information for the SCET
-are taken from the tails of the uncertainty distribu-endstates requires additional analysis using the
- tions. Next, the SCETendstate identifiers and fre-NUREG-1150 tools and models. The process quencies are stripped from the PSTEVNT starts with the creation of a new PSTEVNT rebin.
aggregated output and are reformatted to create a ning definition file. The new rebinner groups the
= new PSTEVNT input file. This new input file of SCET endstates in accordance with the source the mean response is sorted by top events, using tenn binning scheme used in the i150 analyses.
Fourteen characteristics are used with caeh char-acteristic having several dimensions. Because the
- a. ' LHS refers to the limited Latin Hypercube Sam. SCET contains only a fraction of the information pling technique used by the NUREG-1150 effort to contained in the APET, a number of approxima-generate uncertainly distributions for the risk results. tions and simplifications are required to define 6
_ ._- - a
some of the source term characteristics. The a rnore manageable levelis obtained using the PSTEVNT output from this evSluation is passed PARTITION code, For the SCETs developed in to the SEQSOR program to parametrically assign this analysis, the number of source terms was re-the source term release information for each of the duced 10-17, Consequence estimates are then ob. source term bins. At this step of the analysis. the tained for these 17 source terms, using the - sampling capabilities of the S EQSOR progt:im are MACCS code. Armual risk is calculated by com-not used. Only the central estimates for distributed bining the consequence estimates with the release parameters are used. For the PDSs analyzed here, group frequencies ushig a latus 1-2-3 worksheet. 4(X)-500 source terms result from the binning pro- Benchmarking the SCETs at the final risk calcula, cess. Reduction of the number of source tenus to tion is performed to establish a base case, l 7
. . - - - . - ~ - _ - . _ - - . _ _ - - - _ _ _ - - . - - . . - -
- 3. DEVELOPMENT OF A BASE CASE The following sections describe the base case gram.- The risk comparison provides checks on calculation made to verify code use and data flow the PARTITION and MACCS codes, as discussed in Section 2.1. The data files used in s this analysis were originally constructed by 3,2 Containment Failure Mode F Sandia National Laboratory (SNL) for the .
Results Sequoyah NUREG-1150 analysis. The computer codes and data used in this analysis were obtained Table i presca the base case mean condition-directly from SNL. Others wishing access to the al probabilities of the accident progression bins computer codes, models, and data used in this (APBs) as calculated here (using the APETs) analysis should send a formal request for the Mong with those reported by the ' Draft complete suite oflevel-2 and level-3 PRA codes. NUREG-1150 effort. As presented, these bins and the Sequoyah data files to the Director, Divi' are consistent with the NUREG-il50 presenta-sion of Systems Research. Office of Nuclear Reg- tion (see Figure 2.5-3 of Reference 2), because ulatory Research, USNRC, Washington, D.C' the results are reported out to three decimal 20555. Current plans call for revision of these places. Probabilities less than 1.0E-03 are not re-codes by SNL, and for future distribution through ported. Also,Iwo separate bins were reported in the National Energy Software Center at the- NUREG/CR-45512as: no Vessel Breach (VB) Argonne National Laboratory in Argonne, w th early CF, and no VB with no CF, have been filmois, combined into one bin, no VB with early or no
- - CR These two bins were not separately reported 3.1 Base Case Scope in NUREG-il50,12 which was our primary reference before NUREG/CR-45512became The results of the base case calculations are available.
presented for three points in the process; contain-ment failure mode probabilities (presentation The mean conditional probabilities for the bins), risk potential, and risk. The presentation summary APBs of a summsry PDSG are obtained bins are a summary of the APET endstates in by weighting the conditional probabilities of the
- terms of containment failure mode. Risk potential individual PDS by their mean core damage fre-is a summary output produced by PARTITION, quency for each set of PDS that constitute a par-which estimates the potential fatalities resulting ticular summary group. Similarly, the total mean from the combined _ containment failure source conditional APB probabilities are obtained by terms.The rirt results are calculated using conse- frequency weighting the summary PDS group re-quences generated by the MACCS code. ' sults. The loss of offsite power (LOSP) PDSG consists of PDSs I and 2, the ATWS group of The base case calculation begins at core dam- PDS 6, the transient group of PDS 5, the LOCA age. Information from the core damage analysis I3 group of PDS 3, and the by-pass group of PDSs 4
-is passed to the level-2 analysis in the form of and 7. The sum of the mean conditional probabili.
TEMAC output.M These data list core damage ties over a summary PDS group will'oe slightly frequency by plant damage state group and LHS less than one because a truncation cutoff level of '
-observation. The containment failure mode com- - 1.0E-05 was used in the APET analysis.- Se- -
parison verifies the use of the EVNTRE and quences whose probability of occurrence was less -
'PSTEVNT codes, and their inputs. The risk po- than this cutoff frequency were dropped from the tential comparison verifies the SEQSOR pro- APET analysis.
- 8
Table 1. Sequoyah mean accident progression bin probabilities for PDS groups cornparison with NUREG/CR-4551 resu!!s-PDS Group (mean core damage frequency) LOSP ATWS Transients LOCAs Bypass Total Accident Progression Bin Case (I.38E-05) (2.07E-06) (232E-06) (3.52E-05) - (239E-06) (5.58E-05) CF* before VB,bearly CF Calculated:' O 014 0.003 < 1.0D-03 0.002 - 0.0n5 NUREG/CR-4551: 0.014 0.003 - 0.002 - 0.005 VB, alpha, early CF Calculated: 0.002 0.003 <1.0E-03 0.002 - 0.002 NUREG/CR-4551: 0.003 0.003 - 0.002 - 0.002 VB, RCS* > 200 psi, early CF Calculated- 0.062 0.023 0.014 0.031 - 0.036 NUREG/CR-4551: 0.064 0.023 0.014 0.031 - 0.035 o VB, RCS < 200 psi, early CF Calculated- 0.054 0.002 0.O N 0.014 - 0.023 NUREG/CR-4551: 0.054 0,002 0.O N 0.014 - 0.023 VB, H 2burn, late CF Calculated- 0.149 0.001 < 1.0E-4)3 0.001 - 0.038 NUREG/CR-4551: 0.153 0.001 - 0.001 - 0.038 VB, BMP' or very late OP* Calculated: 0.066 0.151 0.039 0.260 - 0.187 NUREG/CR-4551: 0.065 0.151 0.039 0.260 - 0.171 Bypass - Calculated: 0.001 0.134 0.006 - 0.996 0.M8 i
- NUREG/CR--4551
- 0.001 0.134 0.006 -
0.996 0.056 VB, no CF Calculated: 0.204 0.471 0.137 0301 - 0.263 NUREG/CR-4551: 0.200 0.471 0.137 0301 - 0.269 I { i
' - i Table 1. (continued)
PDS Group (mean core damage frequency) - LOSP ATWS Transients LOCAs Bypass Total Case (13SE-05) (2.07E4) (232E-06) (3.52E-05) (239E-06) (5.58E-05)
' Accident Progression Bin _
0.421 0.172 0.790 0369 - 0376 No VB, early or no CF Calculated: 0.422 0.172 0.790 0369 - 0382 NUREG/CR-4551:
- a. Containment Failure.
- b. Vessel Breach.
- c. Reactor Coolant System.
b d. , Basemat Melt-through.
- e. Overpressurization.
ii- .
As seen from Table 1 the results calculated in (h1ACCS 1.5.11) includes some corrections to this analysis compare quite.well with those the version used in the Draft NUREG-ll50 anal-reported in NUREG/CR-4551. The close agree- yses (hf ACCS 1.5.5). Ilowever, the relatively ment of the results indicates that the base case ac- large d viation in the mean early fatalities esti-cident progression analysis adequately duplicates mate a: ;ompared to the other risk measures was that done in the NUREG/CR-4551 work. surprisi. v. The cause of this deviation was fur 4 ther invedtated by determining what the contri-3.3 Risk Results butions io mean carly fatalities are by sequence. it was determined that in the current analysis, hican risk estimates were obtained using the 97.6'?e of the mean early fatality estimate (or methodology described in Section 2.1. The mean 1.8E-05 early fatalities per reactor year) is attrib. risk potentials in tenus of early and latent fatality utable to the V-sequences of PDS 4. In the estimates obtained with PARTITION are com- NUREG-1150 analyses, the V-sequence risk es-pared in Table 2, with those reported in the SNL timate is 1.8E-05 carly fatalities per reactor year, l base case results.* The mean risk measures calcu- but the fractional comribution is 68%. Clearly, lated using h!ACCS and those reported in the two analyses are in agreement for the mean l NUREG/CR-4551, Vol. 5 are shown in Table 3. early fatality risk results for the V-sequence. As seen in Table 2, the mean risk potentials calcu- llow ever, the current estimates for the balance of l lated here match those reported by SNL. Again, the sequences are much less than those calculated this agreement of the results is an indication that in the original NUREG-1150 analyses. The cur-our source term and partitioning analysis are rent analysis estimates the mean early fatalities equivalent to the original Sandia analysis. How- per reactor year for all sequences, excluding ever, the mean risk measures, given in Table 3, do not compare as well, especially the estimate of Tablo 2. Sequoyah mean risk potentials: mean early fatalities that deviate by 27% from comparison to SNL base case results the NUREG/CR-4551 estimate. Some deviation in the mean risk measures was hican Early hican Latent expected because the current analysis utilized an Fatalities Fatalities updated version of h1ACCS. This latest version (per year) (per year)
- a. Infonnation was taken from a report by J. J. Current 8.3E-05 1.lE-01 Gregory entitled "Parametrics: NUREG-1150Sens . analysis tivity Smdies for the sequoyah Pl.mt," Draft tce Con- SNL report 8.2E-05 1lE-01 denser Parametrics Letter Repoit, December 1989.
Table 3. Sequoyah base case mean risk measures: comparison to NUREG/CR-1150 base case results. Doses in person-rem per reactor year hican Early Nican Latent hiean Dose hican Dose Fatalities Fatalities 50-htile 1000-hlile h!ACCS 1.5.11 1,9E-.05 1.5E-02 1.1E+01 8.9E+01 NUREG/CR-4551 2.6E-05 1,4E-02 1.2E+0! 8.l E+01 l Percent -27% 7% -8% 10% difference l l l 11
the Wsequences, at 4.6E-07, versus 8.3E-06 re. Tables 4.3-1 and C.1, respectively, Clearly, there ported by Draft NUREG-1150. }lowever, there is are discrepancies in the results reported in Tables a discrepancy in the mean consequence results re. 4.3-1 and C.1 for source term subgroups one and ported in Draft NUREG/CR-455! Vol. 5 (specir. two, but little or no difference for sut>-group ically, Thble 4.3-1 and Table C.1 of Reference 2). three. Coincidentally, our early fatality risk esti. For example for source term subgroups one and mates for subgroups one and two are different two, many of the early fatality results are one or than those reported in the NUREG/CR-4551, but two orders of magnitude smaller in Table 4.3-1 are identical for subgroun three. than in Table C.I. None of the early fatality re. suits of Table 4.3-1 for these two subgroups are Presently, the reasons for these discrepancies larger than those reported in Thble C.I. Another in the mean early fatality risk measures and in the interesting feature of these two tables is that the mean early fatality consequence measures re-early fatality results for subgroup three are identi. ported in Draft NUREG/CR-4551, W1. 5 are not cal with two minor discrepancies. The early fatal. clear. Therefore, the results calculated in this ity results for source term groups SEQ-08-3 and analysis will be referred to as the base case and SEQ-14-3 are reported as 1.62E+00 and will be used for comparative purposes in place of 1.41 E+02, and 1,61 E+00 and 1.40E+02, in the NUREG-ll50 results. l l \; l 12 l l
I ( 4. SCET DE\ iLOPMENT AND VERIFICATION This section descrities the construction of the the SCET top events. This binner file that speci. SCETs. Also discussed is the verification of the fies the events ofinterest,is one of the input files SCETcontainment failure modes and risk results, used when running EVNTRE De binner input The SCETs were developed for five of the seven plant damage states utilized in the NUREG-il50 decks used to create the SCETs are listed in Appendix C. analysis of Sequoyah.2 The five plant damage state groups (PDSGs) analyzed here included When evaluating the APEh, the sampling op-short-term station blackout (SBO-ST), long-tion of EVNTRE was used. As in the term station blackout (SBO-LT). loss-of-<oolant NUREG-1150 Sequoyah analysis,200 samples accident (LOCA), transients, and Anticipated or observations were specified for the EVNTRE Transients Without Scram (ATWS). SRTs for runs used to create the SCEB. For each observa-the bypass PDSG, which included steam b nera. tion, the path taken through the APET is binned tor tube ruptures (SGTRs) and the event.V se-and saved in a post processing file. Using quence, were not included because of their PSTEVNT, the resulting distribution of binned containment bypass nature, results was combined to form an aggregate or mean estimate of the 200 APET observations. The events modeled on each SCET are de. PSTEVNT is again used to sort this mean esti. scribed in the section that follows along with a mate of the binned results to form the SCET comparison of the SCETs containment failure branching structure as well as the branch condi-mode (i.e., APB) probabilities with those calcu- tional probabilities. lated using the APETs. The two primary require-ments in choosing the top events were to include 4,1 Top Event Selection for all of the important containment failure modes and to provide sufficient detail to allow character- Each Plant Damage State ization of the somce term release for each acci-GTOUp dent scenario (i.e., each SCET endstate). Because the SCETs are a condensed representation of the The selection of top events and the graphical APED, simplifications and approximations were SCET models are now presented. When review-utilized so that the most significant containment ; the SCETs, the convention used here is that failure modes and source term characteristics events identified across the top of the t:ee were adequately represented while at the same (i.e., the failure events) are represented by a time, producing a relatively simple model. downward branching of the event tree. In this convention the downward branches describe the The construction of the SCETs is based on the failure (or undesired) events and are associated with the event ID listed just belcw the event de-methodology described in Chapter 2. Briefly, the EVNTRE accident progression analysis code 6 scription. These failure events (and hence the 7 and the PSTEVNT post processor code provided down-branches) typically have low probabilities the essential tools for developing the SCETs from of occurrence. The upward branches are used for the APETs used in the NUREG-1150 analyses. the successful or desired events that typically lead to an "ok" endstate. For sequences in which no branching occurs for a given event, the event does The structure of each SCET was developed uti-lizing the binning feature of the EVNTRE code. not occur (i.e., success) unless the branch is la-beled with the event ID. This feature allows the user to classify or bin each path through the APET in terms of selected APET question branches or combination of 4,1.1 SBO-ST and SBO-LT SCET Top Event Descriptions, The SCETs for the short-branches, which are of interest. The APET branches chosen for this purpose in turn define and long-term station blackout plant damage states are shown in Figures 2 and 3, respectively. 13 I __ ____ - _ - - - - - - - - - - - - - - - - - - - ' - - ' ~ ~ ' ~ ~ ~ ~ '
'itt!! Tri!P Ef:ti;n ',*'k./l
. "'E:!'"
6"'" "
" Ell!:' " W' N.;:!g '""it:::" "it:::" dib. fa
" * " * " " " " ' ' " ""' "a im:'
ts:1". ..L :/.::'h "tC' Eta m ve xine avec crvo -or soo-si vrar [ wet si ( cP i stop 1 nive dt3#
,;*ges
- 2 gatecL_J lhg[L:aL._kNs.g, we gR3eSL-vcr> i gaten gjendaM" d "
wye Ar, # gAZEC1_ 4dF y !gtaecLggg3 -- 74g.g1 g ag-t L ' ham:01 _,_,_
'ftl3*1%sse-ci_ .n
; .i*t '1h' vu fpyE:11 -
a te j,tBIg, t
'Et " ll
$$N"A g3g 412en y umr p' *p.c .
&fl$s NN[ uiKr I
' 'It'#"'-
ftIS"' te vu g2 lb,97-:Si y ye igp*e-es MG
"* 1 gjucu _ m q;Iis EtIIE*"-- tr vsr $$FGL-y pat-an N 0l",F AG ,
ffznon__ {P2^2 p g ;yo {tna a t wor:r 2 n -c2 tEP Mtr i v3 Figure 2. Sequoyah short-ternt station blackout simplified containment event tree. 11
%d ^:&
) -- Late its CePo-- Fall to Late Fall. to . Vepy Late SE40ENCE ByPa6e Concrete Pastome Cont. RE4JENCE CF tlpte ES No. j interest, sc posse Fallphe Festore Cont. PMS . CLAge as perse FalluPe 011 aftse VB very Jate LISP = CCI B2 LCF BJ VLCF
.77E-01 SEE- 2 No V9
.99E-03 SEO- 2 VS-tiCP t EW-- . 0 &E -0 4 SE9* 8 VLCF l -.. .MEM3 **I h 0 6 d'- M U 3 9. 1E-03 uw - BE9- 5..
uan g-LbP aa 8 i
' upp etu
.W, , m" -
q;;mrc- 7. 4n -0g
.ni:8.
.m n. lit:
u s-
.: No ar" ; 9 0
.4 -CS DE ECF= p .
,wi
~~N mU~
- u:81 ' :d-: : 'N :- ,g
'w- .!:gh'8l.- !':.- ff'1 6
M--
. 92-4.98E .
1 2ef- 1 9 O E 9-4 SE3 '.' H ECr y elphs NoVB
?
e I 7 SEO-u - . V9-NCF e ._' J 9eP"' Ige-ei W " ;. 4. A::n:81 .99E-QJ tee 8 SEQ-ilt i: VS-NCF nr H0
!)
Ug.gytPm- a'* ma 7.sef-Oe i:i:!:: sfa- 2'
!!t t vo-NCP-ar" 2a p H 3-:,#; NP-@ 26
- l. 2E-04
- u. 2E-04 ECF M
- 2. OE-05 9t h.d-;
p 27
,y.1 SEG
. .n -0 u. n ell- -
, E.CP i.n M. 1.$
m -O - Non 0 m lltgh-i p:lii3.g 20" il '~ uw spE-ei S. put h;gm- mc i:li 32 4 g m -
- 7. g*if-04 9E01!CR,C? LCE.- i.i.
m- -- ue -...gi giggt<2
~. :
g ;; Cr a u
- m --
-0.-2
!!La i *.T it 4:tii41
.01E-04 Ste- CFb G 46
-Mi, ,, P 99E-CP SEG.08-2
.08E-03 SEG- 0-2 1-2 NOYS 42
-M( -y h . 90E-08 SEG- $=2
; I . 8 M-04 CFb VO NOVB da 44 SEG-09-2 CFbVB 45 g yg.e g -
0.79E*02 $EG-12 , VS-W7 46 3 . 29E -0 2 SE O- 4-.. (
'h4ag.e3 ' y,3
- 9. 7M -0 { SEQ- 3 1.90E-Oc SEG- 2='
VLCP LCP VS-44CF 4'? 44 49 r,,;.1 I .m** -
- 9. 9 4E-Ca 920 .
, VLCF 90 hU $ E*UI '
CP
&TM2-qpei ~
): 03: n .
- 9. n -0, !!D.d::-r ts,-LP 4 4* Upp ~ClT
= a- DEG-05-:- ECF-LP E e
-q 1 56 6.t2E=02 45-03 Bt3-09=
u- m A:lA13' ut*17 ECF=LP Ma, e. h' g.e g QM~- t . 76E -0
- 5. Au-02 9EG-g8 SEG ,
VLCP LCF 6 m
#'Fsac a:En ::010 Ra" :.
h u m _ {,, , tds " ' Unai 1.12E-03
!!t't-.
BEG- 5-- 0'b73"L Si 66 SI ME-ca
~
m 1 Lisp GGI h. ECm 8:n 31 62E-03
!!tif:n 9EO E"?;"5 8
(*2 E CF-+f* 60 -APERTURE - m 4 t E oi GOI EMI . $ !!l3$ 2.886-08
! tot! !$3 7t2 920-00-2 ECFM - -
CARD 1r.* d g 2,17E-03 BEG-00-2 ~ uBF . GGI -- 4.33E-94 SE4-08-2 ECF M ' ECF M 3 4 V ~ un m ia-n Mgg;c g 1:'!!ai
- 4. 60E -0 9 nuit!
9EG+0S-9
- a C:
phs 7 A.Co Avaiabh On M ggg:0;:a gg, lg Aperture Card 1:!!:33 { MI !/0 9 ;I L 0 LISP GGI j _eg :L E 2E -01
- 6 V upp imp
- :' m 9s " ' @m- ::1842 llt*!! 20
4.aJE-08 SEQ- 0.j ECF-LP
- 4. 3CE-0 4 SEQ-10-2 enche l e6
. i T4 : 97 J- i ipp-- Wu : 5.76R-03 9E0-06-2 CFbVO pe 6p WW :
4 1.5sE-04 9E0-07-2 CFbV8 89 , i Mt : 1.12E=0S DEG-07-2 alpha- 90 gr Ma : 9. e'iE-0 4 SEG-06-2 CFbVB 91
## : .296-94 SEG-09-2 CFbVB 92
.59E-03 SEE-to-2 6-3 RW~-
J
.06E-05 SEG ,0-2
,.45E-04 0%V3 alche 93 94
.', ,,g.a1 1.0M-v5 BEG- ,4-2 SEG. .S-2 NcVB CFb VB 96 96 4*P 2.3dE 05 SEG-O(-2 CFbVD l
97 5.66E-04 4-2 NOVB 98 l ung 3.3M-05 SE 8EQQ. .42 CFbVB 99
- 7. 66E-05 960 B-2 NoVB '
.00 -
I k 8.
$ Q tB b une $ 4*17 *0 4 888*09-2 CFbVB Loa 9equo teh BCET -- P096-2. 380-87 (f ast? s@-P02.TRE 40/24/90
. l y$j i V Y
-e-
)
\
b [ t
w::: R,.. ww r: mu,
- m. .u...:. x-a ca.u-
~.t...
w- 1T.m.. e.m,2..i..
.. . i i .i.m..) ....... ..J. ......a 880 % 7 UTAP NCI B l _,{_ *
- _ _ E !ste . , _ W _VD
_IVet LP'
%fIF g' ,t t!-9 L._, , _._._ g g gk3K-91.--
I l g 44Er91 _ { g t-94 -
%VB jggpE-92_ gp.
f19L-9L
' ~ ~
I46VB
% F fDF~ ~ W N'03-~
ytt thIEEIE A'--4kI3EiUl~ b 17t:91 { --- D 8
.g QE-Q L.__.
EffBP 3pl8E-11 g oJ409 , , ftgena_ g,ise9t._ Itdlk'" v vs _fdP ggae-sg 6tw gattent_j. -
!I) 1.7 yn p ' --
trJL CL_
= ** 7,"L1 M 7~%'J " -
y.m_ it '8d' 1 Ws:r - igg [t-na _g"#' _ _, , Figure 3. Sequoyah long-term station blackout simplified containment event tree, vn
i ll!L4 ' " O r uM". ..T*" , EM"" I " ""' Mir.an v ...yibv . T/...'t! J'.L. ::!,M...:; ,e.M*. u. "U. nim
,.MT..L.
n_. , . . _ . . . . . . . . _
.f,,,.
. m_ g, - . . . . . . . . . . .
I (* VD DI LIDP CCI la LCF fu VLCF _ ,_gg g. g _' V 9t( _ , , ,
- --49 t u g3:3 gg: q gg I, qM8b" ZN;F ~" u . ": m,.;J;* U,,g:E nr
_ . _ myen { _ 1,4 ., l, ftyc4' %._ nen - 7:n - tin 1 Ut: :::, l.: Pr ! l
@ s, t qleP"".. - .
1:m:l:l gg:. ! ft ,.p n - (tpp - g g OEil . ..._ , I- - - ~' 8f DF g
?a ' F 0' 4 0 4 l g#gDC 2 _... _.__ J. $E M -i~ -~"gglH1' ~ _'~ , )[F l*j -0 4j g*@
L&t
- QE l M2' .. _ _ _ u gp _
..c ~n or a :t,y:q ;r,:: : , ,y,y !
gg g3
,, , g g ,
J } Ve-Ncr a
}4 g}_g^g - ,.
- g. Af M i et . t.
y yat-n. .y ypt y; e gy ; _..-.{t .- , s . ,,e n ,. - n.e. m OtME* - pg y- H g t41:T g:,j lft:, j ; [* l'. "; unn m _nh t.n_. n,. , g e-a_
. mr :Hg:ji lj:g:' lgy" g :: h; j
@m _=.en
= a,,:a. _ s . ,,,g:.r:
. m,c, m
-- H pH -. y *';.l gG* l*, $gy" g ;
y 29 _ = := a :n ut:p, na : .<, g , gr- o.- _
... ._ ~ 1 . _. ~_Z.l ZZ ' _l' _.._.l..i N:$ gl l
= =a a
- g. u,
== _:_ _ -._- :: !
g!:]..t _ _r nr r_ :c .ruc___ nz _. Ol!:t : : l ;g?, ;; em - mgg:1 gg:c,:; ;gg,
- rrr :Qrn:r .- rrrr :
--.__{ g.g } Z l g] g p {
DJ" * - ;Jg: gg: VDg", yo, p,
-_ . - . . .. .. .__.-,--.4 g7g{x ,
l
>{::
; y" ]g 14Pu -4 g 41-- .jlg 8 Kf- l-g7 4t-c L ..{-~ yt:st:; i,.<,, :;
Styeu~ma nectn.3"'," _. ._ Ed "u mcrx_ rrr t0:*l
. .. _ _ .. gr_ ._ u_ _ __ . .- _ _. . _ _ _ ru _ _ . . . . .
_ r .._ um:.g gmt
;gt; ;g Sc 41 - _ _ _ . . . .Ji' h6 b 4
__ - g g . , .. _ ,._.. - - _ .g { --+ g., { V gNCF
._ p?R11 - ~ hMD9l-4 [-Q17 f;,U , gl* ( *, f'y" h
.R
^
11 E.T r.r M .* l wp g gg g47E-01._.gjbH._CCg-. . _ . _ . - . _. _._. gg.g p; , 7 , _ _ -- 7 ; ,-- ; 7 _ g *, a g ,
. , e gpn_._ i tge-" ecr - "[ici:n_ 7;W7'7""rcxxr n 7)) z: a:'g:e! pa:j :1 d' , 4 g2lE11~ ...._._gg..... gyg.g{7y- - 7;';7,__7 .
g I Jl 7j g. 7_....
._ g3g gi7 Z ZZ_ - n g ZZ;- 7 ,7, g g9[ [g it'"~" minn cet -- - ~4 .a:M1 - - - - - - - - . - 4:
yN$ gl: l g h:L ::R: !{ {ld"{fg{"~ ~~~~ - hf SE2 l: j;:g gl: h4: h,6,
~
[ la l; _ _ _ _ ,_ ,n..-._._
, ;;rZa:l y
!! EL U !
h"_""- ---ig}een_ ,_. . _ . _ . . . . ._. . _ g:. . d c .up ~ [:N i Ul: 6 p'g:g $ 4,p:cl,.' : dm q y, _ . . gEit_1 -.__ xr 1r- ... ..__ _ E. lkp._ ._ ..a ;;u;__xx..xrn _rxx- - ll UE4IZ f ~l I, Z.Z _ .1. ' Z_lT_ i ..ZZZT~ ZZZZ ; hI l $* f l4
--_ c x u d $ n r. m:rcurrn_ 1:n1;rx _ rx gg; gl:'1: ggE g ,
t
. ~ . , - .-. -. -
_....{ g.y3_3- 3 ;3;; _ _;;;3 yy~~ _3 3 xx ':- umn_ rrn:r : ; i .: ;gt, q run __ . . _ g:lg,,. t 4 $! ) g *, ! ! [N$ _.x xxx r..cag c kr._
. . _ _ . - ._ . LDP_ _ ._
___ .. . _.__.. ._n_ r .. _;n.x.xxcur.:-
- ty:n m:g,: ; ggl j!g j
- 9. Neveh BCET +* PD96 1, 900 4.T le t o w) BQ.P04,tRR 40/34/po
~
fyym M l g #1 p' 'n' [
~
15
}
_b.-A pedm e ( J
4 The events for both station blutout trees are the cos er) and s essel brea;h. Benda .,f techng the same. Ilowever, the resulting branch structure succeeding events in the accident proytc shion, the and frequencies are dtflerent. source tenn release analpis utih/es this es ent in idenuf)ing w hether the sprays and f ans e oper. Loss of Allac Power.This fast esent iden. aunp. Note that, w hile f ailure to restore ac pow er lifies the PDS input to the SCET. For the SBO carly alwap results in vessel breac h. contaimnent trees. the PDS is either the long- or short-tenn failure can be still be avened. This top event is ad-version of the sequence (SBO-lT or SBO-ST dremd by Questhn 22 of the APET. respectively). Hydrogen Burn Before Vessel Brenh. The SBO4T PDS is characterited by a loss of This event questions if the contaminent fails > . lor all at power (both offute prid connection and the to or at vessel breach (i.e..early) because of a hy. onsite emergency diesel generators). Ilow ever, drogen detonation or denagration. Question $8 of de power is initially available and consequently the APET was used to identify those APET sce-ausiliary feedwater (AIM) and mstrumentation narios that resulted in an early containment fail. and control systems are initially available. When ure caused by hydrogen burns, i the batteries drain down (after approximately four hours), these systems are lost and core un- Early ice Bypass.This top event questions covery would begin about three hours later, whether a large bypass of the ice condenser oe. cuts prior to vessel breach. The ice condenser is The SBO-ST PDS is also characterited by a unponant for its pressure suppression capability loss of all ac power. For the short-term or fast and for removing Assion proJuets from the con. SBO, either de pow er or AFW is also lost initial- tainment atmosphere. This event is used in the ly. Therefore. as its name implies, it progresses to source term analysts to determine the decontami-core uncovery much faster (Le., a shorter tenn to nation factor up to the time of vessel breach, core uncovery) than the SBO-1T sequence. Question 59 of the APET addresses the status of the ice condenser before vessel breach and in. Uncovery of Top of Active Fuel (Core cludes three options. Branch 1 identifies those se-Darnage Begins). By definition, entry into the quences in which there is a large bypass of the ice SCET begins with the beginning of core uncov. condenser. Branch 2 identifies sequences in cry. This is the moment that core damage is as- which the bypass of the ice condenser is minimal. sumed to start and is used in the core damage The final branch identifies those sequences in frequency analysis" as the time at which recov- which there is no early ice bypass and the ice con-ery is impossible. Therefore, this event identifies denser is totally effective. In creating the SCET, the beginning of core damage for all sequences these latter two branches were combined to iden-and is implicit in the accident progression lify those sequences for which a large ice bypass analysis, did not occur. No Containtnent Isolation, Ttus event Failure to Depressurize the RCS. This questions if the containment leaks at a rate ugnif- event questions whether the reastor coolant sys. icantly larger than the design leak rate. If the con- tem has been depressurized before vessel breach, tainment is not isolated then the scenario is either intentionally or as a result of the accident binned as an early containment failure mode. progression This event affects the potential for This issue is addressed by Question 12 of the subsequent containment failure events such as the APET. occurrence of DCil,in-vessel steam explosions, ex-vessel steam explosions, and type of core-Failure to Restore ac Powerbafore Ves. concrete interaction (CCI). Therefore, thh event set Breach, This event questions it ac power is used in deiermining containment failurt modes has been restored between the time of core un. and in the source term analysis. 16
Question 25 of the APET was used m dehmny Ex-Vessel Storm Explosion. This event this event for the SCl.T. T he APl T question questions il an ex-venel steam esplosion addresses w hat the s essel pressure n just before WYSE) f ails the containtnent at sessel breach. vessel breac h and involves leur branc hes identify. Question 7I of the APET addresses whether an ing four vessel pressure levels. The four preuure !!YSl! at vessel breach occurs. There are three levels are the system set pint pressure-approsi- brans hes for this question. The first branch iden-rnately 2$fKi psia; high pressure level -preater tifies those sequences in w hich an EYSE occurs than 1(KKipsia but usually less than 2(KK) psia; the but does not result in containment Iailure. The intermediate pressure level-about 200 to NKipsia; second branch identifies those sequences that in-and the low pressure level-2(K) psta or less. Al- vohe an !!VSE that does result in containment though, f or the station blackout PDSs, all fom failure. The third branch identifies those se-pressure levels are possible.the majonty of the se- quences for which no EVSE ocents. In creating quences are at low pressure. In simplifymg this the SCET, the second branch was used to specify event to a bmary branch structure, depressurita- the sequences that involve an liVSE failing con-tion of the RCS was identined as those sequenses tainment and the first and third branches were in which the sequences are at low pressure. All combined to identify those cases in which con-other sequences were regarded to involve Iailure tainment f ailure does not occur from an EVS!!. to depressuri/c the reactor. Containment Falluto (O ver-Pressurization) at Vossol Breach. This Reactor Pressure Vessel Falls (Vessel escut questions if the containment fails from Breach) This top event questions whether or over-pressuritation at vessel breach. Question 82 not the vessel fails. For those sequences in whic h of the APET summari/cs the containment fail-the vessel f ails, this event combined with other urcs at vessel breach, but includes both over-events (such as vessel prenure ptior to breac h) is pressure failures and f ailures from steam used to identify the mode of vessel breach (and explonons and rocket nmde failures. ( A;though it hence subsequent containment failure mecha- s an insignificant contributor or containment nisms) and to characteri/c the type of CCI. ladute probability, the roelet mode was included here for completeness.) Rocket nate failures oe-Question 26 of the APET was w define this cut when the vessel f ails and is accelerated up-event for the SCHT and determines whether or ward at high speed and fails the contaimnent. not core damage is arrested. The Al ET question 1herefore, in creating the SCl!T event, the over-has two options: no vessel breach and vessel pressure failures were explicitly identified by breach. Itecause the APET question is binary, no birming those sequences that failed the contain-simplification was necessary in creating the ment as specified by Question 82, but did not in-SCET event. volve an alpha, EVSil, or rocket mode of containmel;t lailure. In-VesscI Steam Explosion, This event Direct Impingement on Seal Tabic Wall. questions whether an in-vessel steam explosion This ennt questions if the containment fails by (IVSE) fails the containment at vessel breach. Au the direct contact of the molten core with the con-in-vessel steam explosion that fails the contain- tainment wall. Question 78 of the APET address-ment is conunonly referred to as an alpha mode es this issue. The question contains two branehes, failure. Question M of the APET addresses this the first afGrmative and the second negative. Ile-issue by asking: "if an IVSE occurs, does it fail cause of this binary structure, no simplification the containment as well as the reactor pressure was necessary in creating the SCET top esent, vessel (11PV)?" This APET question has two branches: alpha and no alpha. Again, because of Lato Ice Bypass. This event questions this binary structure, no nmplification was neces- whether a large bypass of the ice condenser oc sary in creating 'he SCl'T es ent, euts alter vessel breach and is similar to the Early 17
m im u Ice D> pass esent discussed previously. The only Failure to Rostoro ac Powcr After Vos-dif ference in the two events is the omehame in sel Breach. This n ent questions if ac pm er is w hich they occur. This es ent is used in the source recovered late. Like the esent that questions ac term binning to estimate the decentamination fac- pow er before s essel breach, this event aflects ter 5 to 30 minutes af ter sessel breach. Question succeeding events in the accident propiession and 83 of the APET addresses the status of the ice is used in identilying w hether the sprays and fans condenser immediately af ter vessel breach and are operatmp for the source term telease analysis, includes three possibilities. liranch 1 identifies Question 90 of the APET addresses the availabil-those sequences in whi i there is a large ice by- ity of ac power m the late timeirame. The ques-pass, and therefore the ice condenser is ineffec- tion has three branches. The first branch lise, tiranch 2 identifies sequences in which the identifies sequences in w hich power is available. bypass of the ice condenser is minimal. The final The second and third branches involve sequences branch identifies those sequences in which there in which a70wer is not available. For the second is no ice bypass and the ice condenser is totally branch, ac pown may be recovered in the future effective. In creating the SCET, these latter two and cannot be recovered for the third branch. In branches were combined to identify those se- simplifying this branch structure for the SCET, quences for which a large ice bypaa dW not late ac power was identiGed with the first branch occur, and failure to restore ac power late with the re-maining branches. Prompt core-concreto Interaction Oc- Late Containment Falluto. This event cuts. This event questions if promp core con
- models the oceanence of containment failure crete interaction occurs after vcssel breach. This caused by the pressure nse resulting from a late information is used to determine nature of the hydrogen deflagration. Question 103 of the CCI for the source term analysis. Question 89 of APET sununarites the late containment failures the APET addresses the nature of a prompt CCI. and includes six branches. While the first branch but insolves live branches llowever, for the sla-identifies those APET scenarios that do not in-tion blackout PDSs, only four of the branches are clude a late containment failure, the remaining applicable, These four branches are described as Ove branches identily the site of the failure and follows: The first three branches allinvolve the area of containment in which the failure oe.
prompt CCI after vessel breach. The fourth curs. Therefore,in creating the SCET, the five branch is not applicable and the Ofth branch in-containment failure branches were combined. volves those sequences for which prompt CCI does not occur. Of the three branches that involve Fallure To Rosfore ac Power Very Lato. , prompt CCI, the Grst one occurs in a dry cavity, This event questions if ac power is restored in the the second occurs with limitM water from the very late timef rame. This event is used to deter-reactor coolant system (RCS : inventory and the mine if the sprays are operating during this time emergency core cooling system (ECCS) accumu- period. The status of the sprays is used in the lators, and the third occurs in a wet or deeply source term estimation. Question 105 of the flooded cavity (i.e. water depth is at least 10 ft) APET identifies if ac power is available in the that would occur if the contents of the RWST very late timeframe. The question is binary with were injected. In simplifying this branching an athrmative and negative branch. Therefore, no structure for the SCET, the three prompt CCl simplification was required in creating the SCET branches were grouped together to characterite event. the prompt CCI event and the fourth and fif th branches were combined to form the no prompt Very late Containment Fallure. This CCI event. The nature of the prompt CCI for the event questions whether the containment fails source term estimation can be determined when very late from overpressuritation or basemat this event is considered in combmed with other snell-through, in creating the SCET top event, events in the SCET. the llMTs and very late os erpressuri/.ations were IS
sc? cined using Question 107 and 109 of the Reactor Pressure Vessel Falls (Vessel APE 1, which address the DMT and very late Breach). This event is identical to the station containment iallure events, respectively, blackout SCET event previously described. Question 107 has two branches, and Question 109 has six. For Question 109, the first branch Early Conta!nment Heat Removal Un. identines those sequences for which the contain- available. This event questions if the various ment does not fall very late. The remaining modes of containment heat removal are available branches all involve containment failures due to before VB. The heat removal systems include the overptessurization. These various branches serve containment sprays, the containment fans, and to characterize the failure size and location. the ice condenser. This event affects succeeding events in the accident progression and is used to
" E"#" " # "# ##' E "I '
4.1.2 LOCA Top Event Descriptions. The "# SCET for the LOCA plant damage state is shown
#I E" "" "E # '
Questions 27,28,29, and 30 were utilized in in Figure 4. Some of the events for the LOCA creating this event for the SCET. Question 27 ad. SCET are identical to those for the station black-dresses the status of the early sprays; Question 28 out SCET and will not be repeated here. examines the status of the air return fans; Question 29 ask if the ice has melted out of the ice LOCA. This first event identifies the PDS that condenser before vessel breach; and Question 30 dennes the entry conditions for this particular inquires if the ice condenser has been bypassed SCET. prior to vessel breach. in-Vessel Steam Explosion. This event is No Containment isolatlon. This event is identical to that described for the station blackout identical to the station blackout SCET event pre- PDSs except for the inclusion of the rocket mode viously described. of containtnent failure. Question 64 of the APET addresses the alpha mode event and Question 70 Hydrogen Burn Before Vessel Breach. . questions are binary with Branch 1 identifying his event questions if the containment falls from that the event occurs. These two APET questions a hydrogen detonation or deflagratio2, before were combined in creating the SCET event, vessel breach. Question 58 of the APITT address-es whether the contairunent falls prior to vessel breach and includes six branches. While the first Ex-Vesso/ Steam Explosion.nis event is identical to the station blackout SCET event pre-branch identifies those APET scenarios that do vi usly described, not inchK!c an early containment failure, the re. maining five branches identify the size and loca-tion of the containment failure.nerefore,for the Contalnmenf Failure (Over-SCET, early containment failure was determined Pressurlzaf ton) at Vessel Breach. This by combining the five failure branches of eveni questions if the containment fails duc to ov-Question 58. It should be poted that because of er-pressurization at vessel breach Question 82 of the availability of ac power (and therefore the hy- the APET summarizes the containment failures at drogen igniters), early containment failures are vessel breach, but includes both over-pressure two orders of magnitude less likely for the LOCA failures and failures attributable to steam explo. PDS than for the station blackout PDSs. sions and failures prior to vessel breach.There-fore,in creating the SCET event, an overpressure failure was determined by identifying those se. l Fallure to Depressurlze the RCS.This quences that failed the containtnent as specified event is identical to the station blackout SCET by Question 82, but did not involve an alpha, event previously described. EVSE, or earlier containment failure. 19
l
~ . ,
LDCA No Cont. Hydf coon Failure HPV takin ifarly cent Ifi-iv s p e l l5'swannel Cent. D1 PD$G-3 Inolttien twr n ta (Vessel Heat Steam Stsam Falluno f ru LDfDPW n'D *t e IP b 9 9 . bPSOLhl 46&DVal F AO )L 510f t E E p l O 910ft W6') at VU O f) U"*#*II' t "I'
. . _ _ . . _. . . . . . _ . . . .__ . . . _ _ _ _ _ I. l' . .." '
_ . ,' PEA .. ._ N._ y { . - W 60 - n'U FHHJ I nW E bW WU g-I r-
---A j ! !
j- ,
.jgm
{7. __, q igt -v e . ._
~.
g,yE-V ._.- g t-93 _ _ . _ . . _ _ _ _ . I
! ..g R:01 y l _ . _ .
l-Phg
.{b 4 qt .m01. O Ne tcr --"-~ MjD#--- - - ~ ~ - - - - - ---
_ .. [ ' , Ji{' I g eca_ 0&ug.gg_~ ~ ~ ~ tt ~~~
'C t - - ' "-~ ~~
~~~
gNe9L.q,7..__.___
,, 4
. . . _ . . _ . - - ,9 W C-D2 _ LI.
.b.59Er0L_., 3, cot:ps .. _ cNtt ._ ___
i4J .01L-O L _EGE
,.l
' gf
- 0 0. _
- - - yM- - - -
f hg 9 4 ' ~I.. _ . __.{p, 1.H0L-01 Io ut q r---- lZ ~."~TZTl[ZTi.~Z~T1._. TZ1.i. ._!_.. ' _ ._.
.v,._.._.i_..._.
s Qf yg- ~~~--'{ IlM. . ...._.. - .- . _ . . - - .
. , . _ . . , . . . ~~.~ T T p--- .,._ _ .___. _ _. _ _ . . _ . _. . . . , _
g 100-O L y2tirQ1. . . . . _ _ hl[jE701 - - --- --- -- - g51E:0 L _ .
- g. 4]C-01. _ . . _ . . . . _ . . . . _ _ _ . .
Figure 4. Sequoyah loss +f--coolant accident simplifkd containmere event tree. 20 m i .m
_3 7d q .- _.- L. ate cent Late Venf Late Sf0JENCF *LFWJFNCF CF Moco fB No, 'ect Debels thg. bed is Heat Cont. Coht. 5HDD . C.ASG seal Not 5telto v a l Fallbet Fallues e call C o o l ab it, Unavati, VLCF _DI__ m .. _DNC_._ . _LH4J _ ,_ _LCF. . . B.75F-0t EFQ-t2-t NoVD 1 g,g2F-Qa grg.12-1 VD-NCF 2
$fQ-12-1 J.W 09
-{L4(gg,)601-_ _-. _--- _ _ . (( 1.110-02 VD-NCF 3
. . _--- g,g7p.g3 ggg.13-1 VLCF 4 r 2.oa -r 02 srQ-t2-t VD-NCr s 8.19E-03 SFQ-13-1 6
MLI~91-- L I'eE- - 6 + S- VLCF J p1 - 2.05E-05 SEQ-tB-1 LCF 7 _ g,p p ,99E-03 0FQ 12-t vD-NCF e
-A D I- 0.$0E-03 DfQ-13-1 VLCF 9 JLh4J . 4 4kQ l- -.f'
_QL _ _ .._ tS.'._OL- .._p . _ . _,3,3ap_o4 1.03F-04 SFu-14-1 10 PD3 _ _ . . gro.g4_2 ;.CF,r.Lp (C gg I 2.30E-03 SE u -O t -2 FCF-LP (2
. % -Q E _._ _ _ ..._.... _ __ ___._ .._._ _.J 2.370-03 UEQ-03-2 CCF-LP 13
- -{ %--Iy O L__ _ _ _ _._ _ ..__.- -
H 0.42E-05 p,39g_g4 SFQ-01-2 FCr-LP ggg 03-D ECF-LP ta t$ 1 03F-04 erQ-01-2 alpha 6
.. D- L . _ _ _ __ _ _ _ _ ~_..._. - - M 3,05F-04
-- - 1
, , w ffk: ._._..._..... SEQ-03+2 alpha L7
- 8. e4E -04 DEQ-12-1 VS-NCF LB 1
g DD.1_ ._._. _ .~ q - .Q1 y- g e V)jCF i ____ _{fqcy_- y 1.7f 4.ps'E-03r-04 S00-12-1 9Fu-1B-1 VD-NCF VLCF 21 22 Mg nt M iD.7BC.-Ot _ g d2. pe.g. g _ . .- g.99F-044
,ggp-02 Sru-12-1 grg.g3 1 VD-NCF VLCF 23 24 TJt4J JyhkDL Lv
. L 3.50E-04 DEG-ts-2 LCF 20
- f. -Oil -.
._.~
1.etF-04 SFQ-0$-2 FCF-LP 20
. OIS3 -'
7.40E-03 900-09-2 E C 5' -LP 27
- T H.90E-04 SEQ-09-2 ECF-LP 2H 1 9. 63E 4 4 SEG-09-2 alpha 20 h U 09F-02 9P0-12-1 VD-NCF 30
'k- 4,7bE-02 SEQ-12-1 VD-NCF 3:
L4. Lh liJ1601_ . _ _ 1' nr:D3'-- a 03E-03 SEu-13-1 y- VLCF 32 33 ed U _ 2.17E-02 sEu-12-1
, b,69E-02 VD-NCF 34 hD9L LE:_-
_ _ , _ ,,f a SFu-13-1 VLCF q 7 ppg t- 6.22F-05 0E0-13-1 LCF 35 MC - 2 ow -Qa SEQ-i2-t vD-NCF 36 4 - 2.7M -02 9Cg.13-1 .' E Q - YLCF 37 3Lftt]idR 201_._.T IDg UkOL. __L -.' 4,$08-04 g3-1 LCF 38 5 W -- b + 2.2tE-03 DEL. 2 FCF-HP ap 40 q 2,17E-02 s DEu-c6-2 FCF-HP 8"Q-06-? ECF-Ho 41 2.760-04 SFu-00-2 alpha
],9,49E-04 42 i 0.70E-04 SEQ-14-1 VB -WF 43 SEQ-12-1 VD-N T 44 d!<4u74601 Lh b- [.g:M*--i3,02E 3.itE-06 SEQ-tB*l V 6.CF 45 1,p7t'-03 SEQ-ibt VD-NCF 40 kg4ggg bp 6
Na p. U.B9E-04 0,54E-04 SEQ-13-1 BEQ-12-1 VLCF VD-NCF 47 de
.h; , _b--- 0.05E DEQ-tB-t r
VLCF 40 b9A
.9,dtJNIDQ1-.J.,
LV Igik0u,,' d-- - - - - - - 1.47F-04 SE Q - t D-1 5.09E-04 SEu-03-2 LCF ECF-HP
*O St
- ~
d 0.42F-03 SEua08-2 ECF4F B2 i 4,13F-04 SEQ-06-2 ECF-HP 03
'J SEQ-CP-2 alphe 54 d B.67E-05
- - + 5,15E-04 SEQ-06-2 NOVD B5 N, -- . . . . _ . . _ . . - - - - 1,89E-04 SEQ-08-2 t,72H-04 UEQ-00-2 CFbVD CFbVD
$0 57 t.500-03 UEQ-14-2 NOVD BB PT 6.75E-05 SEO-00-2 NoVD 59 di 6.77U-05 SE G 2 CFbVD 60 r- n 7.89E-00 SEQ-01-2 CFbVD 6i / td)b. , N,.!, UM
.%f$E291 ffMb&
6
- h. :b- -
1, 7 4 f* -C B 1,94E-05 SEQ-09-2 SEQ-Ot-2 CFbVD CFbVD 02 63 hNO 5.30E-05 SEQ-0B*2 CFbVD 64 _{hjk* E~91 - g-- Unid
-U* h<I---
l - 1,20E-04 3.40E-04 SEQ-09-2 SFu-09-2 CFbVB CFbVD 65 60 07
<-w 4,00E-04
; - ---j 2. 5 9E-0 4 SEQ-14-2 HFQ-14-2 CFbVD CFbVD 0B Abu 'h&W OU LQSEQi f' 7.10E-0B SEQ-Cli-2 CFbv0 00 A perftjrp (;;g g DNC 7' 95kC 1._ B.7bE-05 DEU-14+2 CFbVD 70 LNid
._ - B.05E-05 SEQ-09-2 CFbVD 71 1.30E-04 SEG-04-2 CFbVD 72 gt
[g g gp - - 7_ -+ gg J,76E-04 SEQ-00-2 CFbVD 73 SCET -- PDS6-3, LOCA SQ-PDB.TRE 10/24/00 (& D i W 'e
@T -
3
i 1
; Oltect impingament on Seal Table Wall. to trip the reactor, but does include a requirement i This event is identical to the station blackout for plant shutdown. The sequences cornprised by ; SCET event previously described. this PDSO include situations where Mme of the systems required for safe shutdown nase failed Debris Sed Is Not Coolable. This event and core damage is imminent.1his condition dr.
models the probability of the debris bed fonning f nes the entry contition for the innsient SNr. a coolable configuration. This information is used to determine nature of the CCI for the source tenn No Containment isolation. This event is analysis. Question 88 of the APET examines if identical to the station blackout SCET event pre-
; the debris bed is in a coolable configuration and viously described.
4 involves two branches. Because of this binary structure, no simplification was necessary in Temperature Induced Steam Generator l creating the SCET top event. Tube Rupture. This event models the occur. ] rence of a temperature induced steam generator l Late Confalnmeni Heat Memoval Un. tube rupture (SOTP.), which is used in estimating avallable. This event questions if the various the source tenn. Quertion 20 of the APET, which
- j. rnodes of containment heat removal have fal!cd in includes Iwo branches,is used to define this event the late or very late tirneframes.The heat removal for the SCET. Because of its binary structure, no
- systems include the contairunent sprays and fans' s mplification was necessary.
This event affects succeeding events in the acci-dent progression and is used in determining the Reactor Pressure Vessel Depressur. operation of the sprays, which has an impact on lied by MCS Fallure. This event models the 3 the source term estimation. Questions 91,92 and occurrence of a rupture in the RCS, which resuhs 106 were utilized in creating this event for the in depressuriting the RPV before vessel breach. SCEl, Question 91 addresses the status of the This event is used to identify if vessel f ailure oc. sprays late; Question 92 examines the status of ces M hi@ or low pressum the air return fans; and Question 106 addresses the status of sprays very late. g,,,,,,p,,,,y,,y,,,,,y,g,,(y,,,,, late Confalnment Failure. This event is Breach). This event is identical to the station identical to the station blackout SCET cvent pre. blackout SCET event previously described. viously described. Ex-VesselSteam Explosion.This event is Very late Confalnment Failure. This identical to the station blackout SCET event pre-event is identical to the station blackout SCET viously described. event previously described. Containment Fallure (O ver-4.1.3 Transient Top Event Descriptions. Pressurlistlon) at Vessel Brasch. This The SCET for the transient PDSO is shown in event questions if the containment is failed from Figure 5. Some of the events for the Transient over-pressurization at vessel breach. Question 82 SCET are identical to those for the station black- of the APET summarizes the containment fall-out and/or LOCA SCElk and in such instances ures at vessel breach, but includes both over-the descriptions wili not be repeated here, pressure failures and failures duc to steam explosions and failures prior to vessel breach. , Transient. The transient plant damage state Therefore, in creating the SCET event, an o.cr. group is used to represent those sequences that pressure failme was determined by identifying cannot be categorized in one of the other seven those sequences that failed the containment as PDSOs. That is, a transient sequence does not in- specified by Question 82. but did not involve an , volve a loss of coolant (including SGTR and by. EYSE induced failure or earlier containment pass sequences), a loss of all ne power, or a f ailure failure. 21
/
. , , - . , , , - - -- .n . n,-..
. ~ ~
re .,.i.nt w .. in v
- c. ng n w v.,e a u.g ,,.3
. .s i..>. i,r.p,.u, g. ,p.rg;. ,
e.no.
"d 1
";I 5 6 "' _ _
V8 d g , g, _ ._ .. . w-l Pg Hkt L-worm ,,g, {pp1Ed 1_
.y
.g4 LEG 1_
ReVD te,
,gErG&__
NiVD GtLL:9L_ Qt2L$R yottsc._ 'RbVD
'ist 46*'"'" '-
%)sTR E~V6
{pCLL:91_ gS9hSA_ gskwL. Evo W g,not:91_.
.gEG L.-
s miVD Figure 5. Sequoyah transient simplified containment event tree, '(p:ugL7e a8Eca-
~-
45** l 25-V37531 C ht. DlPost De tiP ) e Late cent. Ver'y Late St @Jt H we,'NGE 909JFNCE CF Nede l'8 No. S t en Fe luce Ie+ 1h g t.ed wet Heat Cont. class Esplesteh tw at 4 on esel Lcolehle 56e su v e l F o t br e totale nell Onevell. EV11E CFVtl DI DNC L44J VLCF
. . ~ . . _ _ . _ . . . . _ . . _ . _ _ _ . , . _ . . . . _ _ . _ _ . _ p,43g pa grg,t2-1 NeVD 1 4,p40-02 9E0-12-1 @ -N CF 2 2.03E-02 SFQ-12-1 VS *CF 3
, g k[-E A- '
- 9. 09E -03 SE Q-18 -1 VLCF 4
-~
Mw 1.10E-02 DEG-12-1 @wCF 5 g' pSL _ .(dpD91- t . it.E-02 9E0-13-1 VLCF 6 g50kCA- j1.7tE-06 SEQ-14-2 ECF40 7
.QEGA 1.17E-02 SEQ-04-2 FCF49 8 4.27E-04 9E0-04-2 ECF49 9
}UM
--- 7.00E-Ot DEG la-1 NeVB 10 1.34E-02 BEG-t2-1 VD-NCF 11 g_gg ,5.07E-03 SEQ-12-1 VB-NCF 12
% DO 2.t0E-03 DEQ-13-1 VLCF 13 j 2.42E-02 SEG-12-1 YB-NCF 14 Aw 9g khbI 7.86E +08 SE Q-18
- 1 VLCF 15 8.06E-08 SE9-12-1 VB-NCF 16 b dlUS A-- e.2tE-Os DEo ta-t vLCF $7
{g90DE4 - 4.17E-05 BEG-14-2 ECF-LP (D
- 3. DOE -05 DEG-03-2 ECF-LP 19 ME"$A
%,fD08 3.00E-04 REG-03-2 FCF-Lp 20 3.27E-08 DEQ-07 2 bypees 21 P.32E-04 BEG-07-2 bypese 22 h [ IA- 8. 99F -05 9EG-09-2 bypees 2,8 Mjk94- g --
e.37E-On BE s-0 9-2 bypees 24
- 2. t.6E -03 UEG-07-2 bypees 25 r- 2.4BE-04 BE9-07-2 bypese 26 h[EL -
- 3. 90t'-06 SE9-07-2 Lvpose 27 hDN gg 1.52E-07 4,8t>E-0 4 SEE 2 DEG-14-2 bypees NoVD 29 29 2.570-04 SE9-14-2 CFb VD 80
, 1.12E-04 BEG-14-2 CFbyB 81 g_g t,_ kMU9k 3.tae-05 sEu-06-2 CFbVD a2 4.0tE-05 sr0-14-2 CFbVD 83 jD91--- 2.6sE-05 DE9-06 -2 CrtVD a4
{tC 8- D.2HE-06 DEG-le-2 CFb 4 35 ME~0i ogg 2.ase-07 see-Co-2 CrbVD a6 2.95e-Os SEg-te-2 NeVD 87 - 4.04E-05 e.24e-0e BEu-l5-2 UEG-ts-a CrbVD CFbVD as 30 g N Jtt-O L- e . 32e -07 sEG-09-2 Crbv0 40 #4 p ., N M I$
~ ~
, 9.06E-05 SEO-01-2 CPbVD di bdkb g .9 khDEA- 2.04E-05 3EG-09-2 CFbVD 42 2.04E-05 SEQ-01-2 CFbVD 43 fl pk9 L~ 7.47E-06 SEQ-09-2 CFbVD . 44 4 Il 1.54E-05 SE9-07-2 bypese 45 "IC I'd
- 1. 8 7E -07 BE9-07-2 bypees 46 r---- - 8. 9 7E -07 BEG-07-2 bypees 47 ggg g_ hE~E A- 2.46E-07 SI'Q-09-2 bypees 40 2.46E-07 SE0 -0 7 -2 bypese 49 T
dp-El- 5.36E-08 SEQ-09-2 bypees DO 9.2tE-06 SEQ-07-2 bypees 51 2.itE-07 SEG 2 bytese 52
& 4 Seguoyah BCET - PDSE-5 Transient SO4*DS . DIE 10/24/90 C '
1/Ol L 40 Oh
Direct Impingement on Seal Table Wall. Fallure to Depressurize the RCS. This This event is identical to the station blackout SCET event previously described. event is identical to the station blackout SCET event previously described. Debris Bed is Not Coolable. This event is Reactor Pressure Vessel Falls (Vessel identical to the LOCA SCET event previously Breach). This cvent is identical to the station described. bbckout SCET event previously described. Late Containment Heat Removal Un- . In-Ves el Stoam Explosion. This event is available. His event is identical to the LOCA identical to the LOCA SCET event previously SCET event previously described. desenbed. Very late Conta/nment Fallure. This Ex-Vessel Steam Explosion. nis event is event is identical to the station blackout SCET dentical to the station blackout SCET event pre-event previously described. viously described. 4.1.4 ATWS Top Event Descriptions. The Containment Failure (Over-Pressuriza. SCET for the ATWS plant damage state is shown lion) at Vessel Breach. This event is identical in Figure 6. Some of the events for the ATWS to the LOCA SCET event previously described. SCET are identical to those for the station black. DirectImpingement on Seal Table Wall, out and/or LOCA SCETs. in which case they will not be repeated here. This event is identical to the station blackout SCET event previously described. ATWS. The anticipated transients without Debris Bed is Not Coolable. This event is scram (ATWS) PDSO refresents those sequences identical to the LOCA SCET event previously that include a failure to shutdown the reactor. described. Again, this first event identifies the entry condi-tion for the ATWS SCET, which implies immi. Late Conta/nment Heat Removal Un-nent core damage. available. This event is identical to the LOCA SCET event previously described. Steam Generator Tube Rupture Initlally Present. This event segregated those scenarios Late Containment Fallure. This event is in which SGTR is initially present. For the ATWS identical to the station blackout SCET event pre-PDS, the size and location of the RCS break at the viously described. time of core uncovery can be either a very small break in the RCS piping, coolant loss through the Very late Containment Fallure. This cycling PORVs or SRVs, or a SGTR. Question 1, event is identical to the station blackout SCET event previously described. Branch 5 of the APETidentifies the SGTR initial-ing event. ne SGTR event is a key parameter in the source term estimate (i.e., containment 4.2 Contalnnient Failure Modes bypass), in order to suppon the use of the SCETs in esti-mating risk and for evaluating the benefits of No Containment isolation. This event is potential containment improvements, the con ~ identical to the station blackout SCET event nre-viously described. tainment failure mode probabihties were calcu. lated and compared with the base case APET results. Hydrogen Burn Before Vessel Breach. This event is identical to the LOCA SCET event 4.2.1 Containment Fallure Mode Binning. previously described. After loading the SCETs into the ETA-!! event 23
.,n fo!!-c No$t!on "ENI ENI
- FE m,Ib.iVu o 1y innr i t.[.a$.n t "#0,$,'"
- e. or vn og g to
. N.e c n r* oio.A'
"'(er. ion n o w.e ion ,
W I IVCE CF Vit ATWS lGTH NCI ECF VB F V r.0 REV0~ LF Y6 ~~{gt-93_fdbbE Mgrcr ,- l "CI }:' yar=v . idP** VD
, Q t-ga _ f g. u n
~
yggyn feb 0 ifgDDL:ai k WT'~~ ' gERE-Il bl NoTCY k ' " Vo gt.,,_ -- rgg, . Lv I h![E293 kh ~ ggggp k0 pt:Da c1 hfU #~# VV J If gC-03.- jg{&E'Q1 jghEk03 72E-G E fu VD 7 f j~dgaE-Da_ N670F # (p 73RQ1 , Figure 6. Sequoyah anticipated transient without ser:un simphfied contailunent event tree. 24 (
-s-
- - - _ _ ~ _ . - - _ _ - - _ - - _ - _ . _ _ _ _ __ _ _ _ _ _ _ _ _ _ _ _ _ _
DtMct Debole Late cont. Lets Very 6.sts SFQOFNCE SEQOFNCE CF Mocs ES No. Impino, boo is 61s e t Corit . Cont. Niuo . CLABB on o;al Not listo v e l Fallues Fellues 1ble call Coolable unavit. _D1 DNC L6TK) LCF VLCF t.76F-01 SFQ-12-1 NoVD 1
, D.22F-02 SFQ-{2-1 VD-NCF 2 4
O. L - 4.36F-02 960-12-1 VD-NCF 3 If(,0'4 E-Q 1 5
- 1.23E-02 900-13-1 VLCF 4 L t.20F-Ot sFQ-te-t vD-NCF i I [~D1 B.98F-02 SFQ-th-t VLCF 5
D E~E 1.44E-04 SEQ-18-( LCF 7 f48'kDL~ & w 4. 20F -0 2 SFQ-(2-1 VD-NCF G EE E3 Ef[0E 'l---{7 ' git-Qa _37 * !;$[~8] }l[81:l V$F g 0.440-04 900- 4-2 hCF-LP 11 1 0F0-03-2 FCF-LP 12 2 77F-02 900-0$-2 LCF-LP $3 2 00F-03 D00-01-2 alpha
. 7 pF -03 A4
, 5.00F-02 nFQ-t2-1 VD-NCF E5 e 3 L 4.15F-02 prQ-ta-L VD-NCF La U
14./kJ%h01 5 4-P_OL-~ B fi?F-04 90Q VLCF '7, e 4 7.49F-02 HF Q -' 2-- VD-NCF 10 j tt pE:DI- 2.40E-02 sF Q- - a- VLCF 19 o h,4&pt-Qt {gf4E QA 5.ahr-05 0F0- 3-L LCF 20 2.73F-02 VD-NCF 2t {fgiE:01--- {MDE*23 5 h: 1 - 2.00e-02 a.(OF-04 SEQ-12-L. 9FQ-ta-8t 0-15% VLCF LCr 22 23 fW D 2. fiOF -03 1 SFQ-14-2 SEQ-06-p FCF-671 F CF-6p> 24 25 1.140-03 1%F-02 BF Q g E CP -6W8 26 a.5tF-04 SFQ-05-2 alpha 27 1.77F-04 BEQ-08-2 NpVD 20 5.62F-04 BF0-09-2 CFbVD 29 1.BBF-04 BEQ-OH-2 CFbVD DO M tBE-04 tiFQ-14-2 NoVD at 5.97F-04 HFu-15-2 CFbVD 32
~
,MSN '
[:sc_ 1::!$:n if8:a:! Er!%il !! e U2 L 5./4F-04 BFQ-01-2 CFbVD 85 0 QUC-Q1 1 5
'.63F-04 DFQ-09-2 CFbyD S6
,DAL .7M-04 900-01-2 CFbVD $7
[U"Fid79F-Al 5' ( fE-01 . 00F-04 BFQ-00-2 CFb Vin SB 3.03F-04 9E0-14-2 CFbVD 89 / 3.53E-04 9F0-14-2 CFbVil 40 6 2DbQ 5
- 9. fi6F -0 D DEG-08-2 CFb VD 41 DAC Wiv DQt-Q L 1pE-01 1.08F-04 7.10F-05 0FQ-14-2 BEQ-OH -2 CFbVD CFbVD 42 43 1.00F-01 SFQ-Q7-2 bypass 44 9.510-03 SFQ-07-2 bypass 45 e g, 5f wd N- - - - 2.82F-05 DEQ-07-2 bypass 40 6.7tE-05 900-09-2 bypass 47 5 M 1.6tE-05 800-09-2 bypass 40 g ")t - 2.74E-OS DFQ-09-2 bypass 40 2.3BF-02 SFQ-07-2 bypass '.0 '
7.98E-04 SFQ-09-2 bypass fit g- 7.0BE-05 DEQ-09-2 bypses 62 1.27E-04 DEG-09-2 bypens US g- -- gg 1.12F-00 BFQ-09-2 bypass 54 2.70r-04 DFQ-07-2 bypass 55 2.64E-05 0F0-07-2 bypass 56 r gg 4.32F-05 BE Q-07-g: bypass 57 1 4.2E.01 I ' DEG-09 a bypass 58 D4C l'U[DDkg1 r g l.22F-05
.26F-05 B00-07-2 bypass 59 40 L p[~0g 9.90E-06 DEQ-09-2 bypass 60 DCET -- PDSB-0 ATWD DQ-PD6.TRE 10/25/90 I
SI APERTURE i e cm D Y[I Ilfilp ()y Apu ture Card 6 hI ,-
& )
l
) . _
a i tree roftware pael.afe, the event tree endstates appropriate (with respect to source term defini-j' were binned according to containment failme lion) bin. mode. In binning the endstates, the
- NUREG-1150 presentation bins were used. For 4.2.2 Containment Fallure Mode Comparl-j Sequoyah, there are ten bins. The ten accident son With APET Results. For each SCET, the progression summary bins are
ominmut hilm m@ mh a pad
" # "@" " "#8" * '^ """
VB, very early (during CD) CF or iso-
- E * "I' " "EI
lation failures
- VB, carly (al VB) CF or alpha mode SBO-LT Containment Fallure Mode Re-
- VB with the RCS pressure > 200 psia, sults. Tab!c 4 compares the accident progression early (at VB) CF smumary bin probabilities for the SBO-LT PDS, l
- VB with the RCS pressure < 200 psia, as predicted by the SCET and the APET. The 4
carly (at VB)CF SBO-LT SCET results compare quite well with
- VB, late CF the full APET results. The one exception is the
- VB, BMT or very late CF bypass bin. For the SCET, the probability of by.
l
- Bypass pass is zero because this event was not included in j
- VB, no CF the simplified tree. For the SBO PDSs, the bypass I
- No VB but with very early (during bin includes only those scenarios involving a temperature-induced SGTR. This event was not I CD) CF or isolation failures Included in the SCET because of the low proba-
- No VB, no CF.
bility of occurrence as estimated by the APET, Overall, these results indicate that the SCET ac. in presenting the results, the last two bins were
- combined into a single bin and remuned No Ves- curately models the containment failure modes ;
j for the SBO-LT PDS. As a final note, the sum of i sel Breach. For the actual process of assigning 4 each endstate to one of these summary bins, the the bin probabilities do not necessarily add to uni-bins were considered in a hierarchical order as ty because of sounding off error for the SCET re. i rotiowst suits and because of truncation in the APET results.
- Bypass -
S80-STContainment Fallute Mode Re-
*- VB, early (at VB) CF or alpha mode
- suits. Table 5 compares the accident progression
- No VB suminary bin probabilities for the SBO-LT PDS,
- VB, very early (during CD)CF or iso- as predicted by the SCET and the APET. The lation failures SBO-LT SCET results compare quite well with
- VB with RCS pressure > 200 psia, car- the full APET results. As is the case for the SBO-ly (at VB) CF (at VB) ST PDS, the one exception for the SBO-ST,is the
- VB with RCS pressure < 200 psia, car
- bypass bin. Again, this event was not included in ly (at VB)CF(at VB) the SCET because of the low probability of oc.
l
- VB, late CF currence as predicted by the APET, -
*- VB, BMT or very late CF e VB, no CF. LOCA Containment Failure Mode Re-suits. Thble 6 compares the accident progression This ordering of the summary bins was re- smnmary bin probabilities for the LOCA PDS, as quired because some endstates satisfy the criteria predicted by the SCET and the APET.11e LOCA for more than one containment failure mode. For SCET results compare quite well with the full these cases, the sequence is placed into the most \ PET results.
25 s _ . . . - _ . _ . . , . _ . _ _ _ - - . _ . - . - . . . - .i
i j i 1 Table 4. Comparison of SCET and APET accident progression bin mean probabilities for SBO-LT PDS at Sequoyah 1 l Accident Progression Din SCET APErr a CF* before Vil.h early CF (CFbVB) 1.2E-02 1.2E-02 VD, alpha, early CF (alpha) 6.6tM4 6.9E-(4 Vil, RCS' > 200 psi, early CF (ECF-HP) 4.2E-02 5.5B-02 Vil, RCS < 200 psi, early CF (ECF-LP) 3.3 E-02 3.2FAl2 VB, late CF (LCF) 9.7E-02 9.6E-02 V B, D M'id or very late 01" (VLCF) 4.5E-02 4.5 E-02 i Dypass 0.0 1,3tMM VD, no CF (VD-NCF) 1,7E-01 1.6E-01 No VD, early or no CF (NoVD) 5.7E-01 5.78-01 d
- a. Containment failure.
- b. VesselDreacts
- c. Reactor coolant system.
- d. Basemat mell-through.
- c. Overpressurization.
I 26
1 1 a T3bie 5. Comparison of SCET and APET accident progression bin snean probabilities for SBO-ST i PDS at Sequoyah
- . c-.
Accident Progression Illn SCET APET CP before Vil,6 carly CF (CFbVil) 1.5E4)2 1.5E-02 l VB, alpha, early CF (alpha) 2.$FM)3 2.5FA13 VB, RCS' > 200 psi, carly CF (ECF-1IP) 6.6E412 6.6E-02 VII, RCS < 200 psi, early CF (ECF-LP) 6.$lh02 6.5 E-02 VD, late CF (LCF) 1.8E-01 f.8E-01 VB, DMTdor very Iatc OI" (VI.CF) 7.7E-02 7.7FAl2 Uypass 0.0 1.71kO3 VB, no CF (VD-NCF) 2.3fMil 2.3E-01 No VB, early or no CF (NoVB) 3.5FA)1 3.51bO!
- a. Containment failure.
- b. Vessellireach.
! c. Reactor Coolant System. l-I d, Basemat melt-through.
- c. Overpressurization, t
27
T;ble 6. Comparison of SCET End APET tecident proptession bin mean probatulities for LOCA PDS at Sequoyah Accident Progression Bin ,SCET APET CP before VB,6 carly CF (CFbVB) 2.3E43 2.3 E-03 VB, alpha, early CF (alpha) 1.7 E-03 1.7 E-03 YB, RCS' > 200 psi, early CF (ECF-llP) 3.0E--02 3.0E-02 VB, RCS < 200 psi, early CF (ECF-LP) 1.3 E-02 1.3E-02 VB, late CF (LCF) 1.1 E-03 1.1E-03 ? VB, BM'i dor very late OP' (VLCF) 2.6E-01 2.6E-Ol Bypass - - VB, no CF (VB-NCF) 3.0E-01 3.0E-01 No VB, early or no CF (NoVB) 3.7E-01 3.7E-01
- s. Contairunent failure.
- b. Vessel Dreach.
- c. Reactor coolant system.
- d. Basemat mell-through.
- c. Overpressurization.
Translent Containment Failure Mode APET. Overall, these results indicate that the Results. Table 7 compares the accident SCET accurately models the containment failure progression summary bin probabilities for the modes for the Transient PDS. Transient PDS, as predicted by the SCET and the ' APET. The Transient SCET results compare quite ATWS Containment Failure Mode well with the full APET results, Two exceptions 80sults. Table 8 compares the accident progres-are the alpha and late CF bins. For the SCET, the sion summary bin probabilities for the ATWS probability of these two bins is zero.These events PDS, as predicted by the SCET and the APET. were not included in the SCET bectuse of their The ATWS SCET results compare quite well with low probability of occurrence as estimated by the the full APET results. 28
Tablo 7. Comparison of SCET and APET accident progression bin mean probabilities for Transient PDS at Sequoyah Accident Progression Bin SCET APET CF* before VB,6 early CF (CFbVB) 6.5E-04 7.3E-(kl VB, alpha, early CF (alpha) 0.0 3.9fMM VB, RCSS > 200 psi, early CF (ECF-IIP) 1.4 E-02 1.4E-02 VB, RCS < 200 psi, early CF (ECF-LP) 4.1 E-03 3.8E-03 VB, late CF (LCF) 0.0 8.2E43 VB,11M'l* or very late OI" (VLCF) 3.8E-02 3.8E-02 Ilypass 6.4fM)3 6.4E-03 VB, no CF (VB-NCF) 1.4E-01 1.4EM)I No VB, early or no CF (NoVB) 7.9E-01 7.9E-01
- a. Contairunent failure.
- b. Yessel Breactt
- c. Reactor Coolant System.
- d. Basemat melt-through,
- c. Overpressurization.
29
Table 8. Comparison of SCET and APET aniJent progression bin rucan probabilities for ATWS PI)$ at Sequoyah I Accident hogssion Din , SCET APET CF" bef ore VB,l' carly CF (CFt,VD) 3.2 E-03 3.2E-03 Vil, alpha, early CF (alpha) 3.l E-03 3.l E-03 VD, RCS' > 200 psi, eatly CF (ECF-IIP) 2.2E4)2 2.3E-02 VB, RCS < 200 psi, early CF (ECF-LP) 2.0E-02 2.0E-02 VD, late CF (LCF) 1,lE-03 1.1FA)3 Vil, BM'I or very late 01" (VLCF) 1.$E-01 1.5 E-01 Dypass 1.3E-01 1.3E-01 VD, no CF (VD-NCF) 4.7 E--01 4.7 E4)I No Yll, early or no CF (NoVD) 1.7E-01 1.7FA)1
- n. Conn.inment failure.
- h. Yessel Dicach.
- c. Reactor Coolani System.
- d. Itasemat meh-through.
- c. Overpressurization.
4.3 Risk where R1 s
= the risk associated with Risk is calculated by Membling the results c nsequence measure k from the plant damage frequency (level-1) analy.
sis, the containment failure (level-2) analysis,
= the frequency f plant dam-and thc offsite consequence (levci-3) analysis.
age state group i The equation representing the assembly of these three parts of a complete risk analysis can be ex- the conditional probability CRMPg ,: pressed as follows: of containment release mode j, given plant damage RISKg c L VgFREQ,
- CRMPy state group i
"* * "" E' "# * "'# # ""
- CONS t(FPD) (1) f or containment release 30
mode j of plant damage the SCETs described in SNtion 4.1,into source state Froup i term release groups or bins.The, rebinning relies on the framework established in the CONSg = meanmagnitudeof conse- NUREG-1150 analysis of Sequoyah. Each quence L. given fission SCET endstr.c is assigned a source term vector product source term (FPn). that comr.ises fourteen characteristics. These charactrtistics, which are defined in Table 9. are 'the frequency of each plant damage state group then used to delme the source term resulung from is obtained Irom the Sequoyah level-1 PRA.13 a prtticular path through the event tree. The containment failure mode probabilities are obtained from the SCETs presented in Section 4.1 Each of the ateve characteri tier, and its possi-nnd the binning procedure described in Section ble values, are defined in Sectit i 2.4.2 of Refer-4.2.1. The consequence data are obtained by de- ence 2. A4 an aid to understandi1g the following termining a source term for each release mode discussions, commented listingt of the and then calculating a consequence for each PSTEVNT binning data files are provided in source term. This is a tw& step process and is de- Appendix C. These listings provide a detailed re-scrited below in Sections 4.3.2 and 4.3.3. cord of how the source term bins were created for this analysis,llowever,the data files are difficult 4.3.1 Release Modo Probabilltles. A condi- to inte pret without some f amiliarity with tional probability for each containment release PSTEVNT. Therefore, the following example is mode is obtained by rebinning the endstates of provided: Table 9. Sowce term characteristic definitions Characteristic hinemonic _ Description 1 CF-Time Time of coriteinment tailure 2 Sprays Pericxis in which sprays operate 3 CCI Occurrence of core-concrete interactions 4 Rf .-Pres RCS pressure before vessel breach
$ VB-Mode hiode of vessel breach 6 SOTR Steam generator tube rupture 7 AhtT-CCI Amount of core ava!!able for CCI 8 Zr-Ox Fraction of Zr oxidized in--vessel 9 !!Ph1R Fraction of core in llPhtE 10 CF-Size Size of containment failure 11 RCS-Ilole Number oflarge holes in the RCS after VB 12 E2-lC Early lee condenser function 13 12-lC Late ice condenser function 14 ARFans Status of air return fans 31
.-. .-. ~___ -- - . _. - - - - ._-.- _- -
Sequoyah Source Term Rebinning - PDSU-l and 2. St O 14 CF-Time Spray.s CCI RCS-Pies VD-Male SGTR Ami-CCI Zr-Ox IIPME CILSize RCS-Ilole E2-IC. 12-IC ARFans 7 7 V-Dry V-Wet CFJarly CF-atVB CF-Late CF-V1. ate NoCF 2 1 1 1 $ A. Event V, not scrubbed i l' /1 V-Dry 2 2 1 1 $11 Event V, scrubbed l' /1 V-Wet 2 3 1 3 $ C. CF during core degradation 2+ 2 CILEarly 4 4 7 8 9 10 $ D. CF at vessel breach 242+2+2 CFLatVB 1 5 14 $E. Late CF 2 CILLate l 1 6 16 $ F. Very late CF ; 2 CF-Vlate 8 7 1 3 7 8 9 10 14 16 5 G. No contaimnent failure 1*1*1*1*1*1*1*1 NoCF Those familiar with EVNTRE and PSTEVNT PSTEVNT manuals). The following discussions input will recognize that a portion of this data explain the application of the NUREG-1150 bin-fragment (rpecifically, lines 2,3, and 4) identifies ning scheme to the SCEh. the 14 characteristics of the source term vector listed in Thble 9.The balance of the data fragment Containment Failure Time. The time at then defines the rules for detenntning the fitst which evnralnmentfalls is described in terms of characteristic of the vectori in this case, time of seven attributes or options. The first two charac. containment failure. There are seven options this terite the nature of V-events, and therefore never characteristic can assmne, they are (abbreviated): appear in SCET source term binning results. The V-Dry, V-Wet, CF-Early, etc. The selection log 4 remaining options define the time of containment ic for each attribute follows. For example the , failure as occurring during core degradation, at logic statements indicate option A (representing vessel breach, late, very late, or never. V-Dry) cannot occur for this PDSO. The com-ment (indicated by the "$" character) explains During SHO, LOCA, and ATWS sequences, that attribute A is assigned only for Y sequences early containment failures (i.e., during core deg-
' with a dry release (i.e., not scrubbed). The num- radation) are composed almost entirely of failure bers used in the logic statements refer to the to isolate, and hydrogen burn events. Both of l SCET top event number and bianch number. For these appear on the SBO, LOCA, and ATWS
! a complete description of the EVNTRE and SCETs and are used by the binner to identity see- ! PSTEVNT data requirements, the reader should natios that include early containment failures, refer to References 6 and 7 (EVNTRE and The transient SCET does not include a 32
, -----.--n,- --n--- . . -.----r - - - - - - - ~
hydrogen- burn-caused containnu al f ailure in the LOCA SCET, containtnent spiny func-event duriny core degradation because of its rela- tion during core degradation and during the peri-tis ely low probabihty. Theteloic, ca,1y contaln- od following vessel breach are included as top ment tailures for the transient PDSG are events.'lhis allows nmdeling of spray options A, deleimmed from just the failure to isolate event. D, F, and it with reasonable accuracy. The re-mainit.g attributes are not explicitly accounted for because they are relatively unlikely and could Containment failure at vessel breach inchtdes not be modeled properly with the level of detail in-venel steam explosions, ex-vessel steam ex, available in the SCET. plosions, overpressmitation at vessel breach, and direct impingement events. Rese events, except in the transient and ATWS trees,the sprays are for in-vessel steam explosion, appear in every assumed to be available for all pathways up to the SCET. In-vessel steam explosion does not appear time of vessel breach. After vessel breach, con-in the u ansient tree because of its low probability tainment spray is questioned explicitly becaur,e in this PDSO. its continued availability depends on the occur-rence of energetic events at the time of vessel Late containment failures nomintdly occur breach. Only options A and D were accounted for during the initial part of CCl, and are explicitly in the source term binning. As above, the remain-represented on each SCET except transient, ing attributes were excluded from consideration again, because of the extremely low probability because of their relative insignificance (conse-of occurrence for this group. quently, the SCETs for these plant damage state groups lack sufficient detail to model them). Wry late failures occur 12 to 24 hours after vessel breach and are explicitly represented in all Cote--Concreto Interactions. Characteris-SCETs. Therefore, the binner references this tic 3, the occurrence of core-concrctc interac-event directly, tion, was explicitly represented by a top event in all five SCETs. llowever,'he SCETs were not de-vel ped in the same detail as the APED in this re-The last posubility for the containment failure sput, so smne appmximation was necessary to time characteristic is no containment failure. This capture the necessary CCI attributes, event collects all SCET endstates not assigned one of the previous dimensions, in the SBO SCET, attributes A, B, C, and D were modeled. Attribute A, dry CCl that starts , Sprays. The second source term characteris- immediately, was modeled as dependent on fall-tie is the operability of containment sprays. To ure to restore ne power early, on failure to depres. Leep the trees simple, containment spray funelion surize, on vessel breach, and on the prompt CCI was not included as a SCET event when it was event in the SB tree. Attribute B,CCI that occurs possible to capture the timeframe for spray opera- undcr .5 ft of water, was assumed to occur under tion through binning assumptions, or through the same conditions as attribute A, with the addi-correlation with other SCET events, in the SBO tional requirement that the reactor is depressu-trees, spray operability is assumed to follow the rized before vessel breach. Attribute C, no CCl, availability of ac power llowever, this assump- was considered to occur when vessel breach is lion did not allow the modeling of option D- avoided,or when vessel breach is followed by the spray operation in the early and intermediate no CCI event in the SCET. Attribute D, CCI un-timeframe. Conclating spray operation with ac der 10ft of water, picks up any pathways that in-power availability introduces some error because ciude vessel breach, and CCl, but were not a small number of scenarios exist where hydro- collected by the previous attributes. pen detonations or steam explosions damage the spray system and prevent its operation when ne CCI attributes assigned to the LOCA, tran-power is later recovered. sient, and ATWS SCET pathways included only 33
I C and D. '!he understanding here is that if CCI oc. Attribute D was assigned to paths that included cuts at all,it will likely occur under a signincant alpha mooe failmes (event IVSE). and attribute F amount of water. 'this is because a large numtvr was assigned to the remaining pathways. The i of LOCA PDSG endstates went to core damage transient and ATWS SCETs were treated similar-after a significant amount of watu had been in- ly, except aat IVSE was not accounted for. jected into the RCS. The logic for assigning tran- . sient and ATWS SCET endstates was similar. Steam Generator Tube Rupture.Charac-teristh six, srcam generator rube rupture,is ac-RCS Pressure before Vessel Bleach. counted for direct;y in the trees were it occurs The fourth characteristic, RCS pressure before with a significant probability. Ilowever, this char-vessel breach, appears explicitly in all SCETs. acteristic has three dimensions defining whether The SEQSOR program actually recognizes four SGTR occurs with or without stuck-open SG pressure range attributes, more than can be ac- SRVs. '!he binner assumes that if SGTR occurs, counted for using the simplified approach, llow. the SRVs will not stick open. This is nonconser-ever, in developing the SCETs, the RCS pressure vative with respect to release timing, but consis-is characterized as low,if it is below 200 psia, and tent with the fact that the conditional probability hign otherwise. f a stuck-open SRV is quite low. SOTR does not occur at allin the SBO and Station blackout, LOCA, and ATWS SCET LOCA SCETs, therefore all pathways are as-pathways were assigned attribute B, pressure be- signed attribute C. The transient and ATWS tween 1000.-2000 psia,if there was a failure to SCETS include SGTR events, and pathe.nys depressurize the RCS. If depressurization did oc. where this event occurs are assigned atuibute A, cur, then attribute D, pressure less than 200 psia SG7R with secondary RVs reclosing. Otherwise was assigned. Transient SCET pathways were as- attribute C, no SGTR',is assigned. signed atuibute A-pressure at the system set-pointpressure,if depressuritation falls. Arnount of Core Not in HPME Avallable for CCI. Characteristic seven, the amount ofcore Mode of Vossol Breach. The ncnt charac. not invoh cd in HPME that is availabicfor CCI, is teristic, mode of vessel breach, accounts for: not directly addressed in any of the SCEB. Ex-IIPME, molten debris pouring through a hole in amination of the APET output shows that most of the bottom of the vessel, complete bottom head the pathways through the APET that have CClin-failure, alpha or rocket failur:s, and no contain. volve 0-30% of the core, ment failure. Because alpha mode failures are ac. counted for explicitly in all the SCETS cxcept SBO, LOCA, and ATWS SCEh are given a transient, where it is a very low probability event, consuvauve ucannent of We amount of core the binner can account for this attribute directly, avaHable [or CCI. That is, if CCI occurs, it is as. Of the remaining possibilities, the assumption is sumed to myolve 70-100% of the core. Transient made that if vessel failure occurs,it will be a SCET endstates are assumed to have option-B, pour-type event.The f>EQSOR program does not M0% of core involved,if CCI occurs, refer to this characteristic in calculating a release, so the approximations made here have no effect Zr Ox/dallon. Characteristic cight, the amount of Zr oAidation in-vessel, was also not on the consequence analysis. This characteristic addressed by the SCET liowever the majority of is required as a place holder, and for comparison scenarios in the APETs experienced low (0-409, with published source team bin data, nominal value of 25%) oxidation, so all SCET endstates were assigned this dimension for char. In SBO and LOCA, attribute A was assigned if acteristic eight. depressurization failed and vessel breach oc. curred, but IVSE did not. Attribute B was as- High Pressure Melt Ejection. Characteris-signed if depressurization was successful, tic nine, thefraction of corc that was ejected 34
under pres 3nrc at ressel birach, was not ad- way. in the LOCA and ATWS trees, all paths are dressed in the SCETs. The assumption, based on assigned attnbute A, no ice hypes. frequency output data from the APET, was that when the vesselis breached at hiFh pressure, Status of Air Return Fans. Charactuistic
- 14. status ofair returnfans,is correlated with the greater than 401 is involved in llPME. Other.
wise llPME is assumed not to occur, availability of ac power in the blackout trees. In the LOCA SCET, attribute A, operation only in the early timeframe,is assigned if there are no Containment Fallure Sizo. Characteristic hydrogen- burn-induced early failures, and the 10, containmentfailure size, is not explicitly ad. vesselis later breached. Attribute B is assigned if dressed by the SCETs, but is closely related to there are no burns, and there is no vessel breach. containment failure mode, which is addressed Attribute C is never assigned, and attribute D is Attribute A, catastrophic rupturc,is assigned to assigned to the remaining paths, paths with hydrogen burns, failures at vessel breach, and late overpressure failures. Attrib- in the transient SCET, air return fans are as-ute 11, ruptures charactcrl:cd by a hole larger sumed to always be operable in the early time-than 7ft 2,is assigned to paths with ex-vessel frame. After vessel breach, they are assurned to steam explosions and very late containment fall-be operable unless an ex-vessel steam explosion utes. Attribute C,Icalagc,is associated with fail-occurs, t unless overpressure at VB occurs. In ute to isolate, direct impingement failures, and the ATWS SCET, similar logic is applied. The very late containment failures (when dominated fans an: assunwd to be operable unless disabled by basemat melt-through instead of overpres- by a hydrogen burn during the early timefrarne, sure). Attribute D, no containment failure, is as- , signed to pathways not covered by one of the 4.3.2 Source Term Calculation for Each atxwe assignments. Roloats Mode, Applying the binning scheme described alme to each of the SCETs endstates Holes in the RCS. Charactenstic 11, the resulted in the definition of 464 unique source numbcr ofhvIrs in the RCS,is strongly correlated term bins. These bins, each described by a with alpha or rocket failures. Should a pathway 14-character vector, are passed to the SEQSOR through the SCET include an IVSE event it is as- program and it calculates a source term for each, signed the 2-ilote attribute of this characteristic, These release source terms are then processed otherwise it is assigned the one-hole attribute. w th the PARTITION code, which generates and plots each source tenn on a consequence grid. In Early leo Condenser Function. Charac- addition to calculating an estimate of the early tcristic 12, carly ice condenscrfamction, is ad* and latent fatalities for each source term, dressed directly in the station blackout SCETs If PARTITION also divides the source terms into early ice bypass occurs,it is assigned attribute C, two release categories. totalice bypass or melting, Attribute B is not uti-liied in the SCET source tenu binning. In the re- The first release category includes those re-tanining SCETs, early ice bypass is so unlikely leases with both an early and latent fatality poten-that attribute A, no ice bypass,is assigned to all tial. The second includes those with only latent pathways, fatality potential. The PARTITION grid chosen for the SCET analysis is shown in Figure 7. Both Lato leo Condonsor Function, Character- the number of source terms and the combined fre. istic 13, late lcc condenscrfunction, is addressed quency for each cell on the grid are presented. directly in the station blackout SCETs. If late ice Grid locations associated with a very low fre-bypass occurs,it is assigned attribute C, total ice quency are repooled with adjacent grids, resulting bypass, in the transient SCET, late ice bypass is in even fewer source terms that must be eva-correlated with ex-vessel steam explosion. If luated. After repooling, the arrangement in Fig-IVSE occurs, attribute C is assigned to the path- ute 8 results. Note that there are only 11 nonzero 35 l
CELL COUNTS WITHIN THE GRID FOR A TOTAL COUNT OF 294: 1 2 3 4 5 6 7 8 1
+......+......+............. ......+......+......+......+
l l l l l l 12 2
+......+......+.............+......+......+l......+l.......l l l l l l 58 51 1 4......+......+......+......+......+l.....+l......+l......l 3 l l l l l 1l 40 l l
+......+.....4..................+......+.............+l 4 l l l l 8l 36l 4l
..................................+......+......+l......+l 5 l l l 3l 24l 7
+......+......+......+..............l......+l.......l......+l 6 l 4l 23l 22l l
+......+......+.............+......+l......+l......+l......l PERCENTAGE OF WEIGHTED FREQUENCIES CONTAINED IN EACH CELL:
1 2 3 4 5 6 7 8 1
+......+......+...........................+......+......+
l l l l l 1.83l
.....................+.............l......l.......l.....4 2 l l l l l l6.73l8.69l2.80l . ...... ......
4
+......+......+......+.............+......+..... 4......+
l l l l0.63l34.77 0.41
+......+......+......+......+......+l......+l......+l......+l 5 l l l1.69l1.02l0.45
+......+......+......+......+......+l.....+l......+l......+l 6
l0.08l7.31l6.17l l
+......+......+......+......+l......+......+l......+l......+l Figure 7. Distribetion of source terms before repooling, 36 l
i
k 1 CELL COUNTS WITHIN THE GRID FOR A TOTAL COUNT OF 294: 1 2 3 4 5 6 7 8 4.............+..................................+....... 1 l l l l l l l l 12l
....... ...... ............. ....................+.......
2 l l l 58 51
......1 1l
.I.......................................1 l l
......+
3 i I 45
+......I......I.......I ......+.....4.......l......I.......I I
4 l l l l 9l 42l l l l
+....................+.............+.............+.......
5 l l l 3l 24l l l l l
..............+...... .............+.....................
6 l l 27l 22l l l l l l
+......+......+......+................... 4..............
PERCENTAGE OF WElGHTED FREQUENCIES CONTAINED IN EACH CELL: 1 2 3 4 5 6 7 8
+......+......+...... ...... ...... ......+...... .......
1 l l l l l l l l1.83l
+......+......+......+,.....+......+......+......+......+
2 l l l l l l6.73l8.69l2.80l
+...........................+......+...... ..... 4....... "
3 I I I 127.811 I I
+......+.......I ......+.......l .............+..............
4 l l l l0.74 l l l
+......+....................135.12l ......+...... ......+.......
5 l l l1.69l1.02l l l l l 4......+...... ......+...... ...... ......+......+....... 6 l l7.39l6.17l l l l l l
+.....+......+......+......+......+......+......+......+
Figure 8. Source term distribution after repooling. groups with leth early and latent fatality potential segment is released. Individuals evacuating in this inat require source tenu analysis. timeframe are assumed to escape radiation expo. sure from the release. The second subgroup in-A similat process is done for source tenns that cludes releases where evacuation is started too late have the potential for producing only latent can. for individuals to completely escape the first re. ctrs. The result before and after repooling is lease segment, but within one hour of the begin. shown in Figure 9.The process creates four addi. ning of the release. The final group includes tional source term groups. Figure 10 shows how releases where evacuation is started later than one cach source tenn group resulting from the parti. hout after the beginning of the first release tioning process is identified in the remainder of segment. the analysis. The assignment of releases to subgroups is The 15 groups are each further divided into made on the basis of the evacuation waming time tha ce subgroups on the basis of evacuation timing. (TW) calculated by SEQSOR, and on the evacua. The ftrst subgroup includes releases in which tion delay time (2.3 hours). The evacuation warn-evacuation is started 30 min before the first plume ing time is generally the time at which core 37
- =- -
CELL COUNTS WITHIN THE GRID FOR A TOTAL COUNT OF 170: BEFORE POOLING 1 2 3 4 5
+......+............. ..... 4.......
1 l 59l 1l 48l 50l 12l
+......+...... ......+......+.......
PERCENTAGE OF WEIGHTED FREQUENCIES CONTAINED IN EACH CELL: 1 2 3 4 5
+......+......+......+...... ......+
1 l73.28l0.02l25.85l0.71l0.14l
+......+......+......+......+......+
AFTER POOLING 1 2 3 4 5
+......+......+..... 4...... .......
1 l 59 l l 49 l 50 1 12l
.+......+......+......+......+.......y l
PERCENTAGE OF WEIGHTED FREQUENCIES CONTAINED IN EACH CELL: 1 2 3 4 5
+..... +......+......+..... +.......
1 l73.28l l25.8Gl0.71l0.14l
+......+......+......+......+......+
Figure 9. Source term (for zero early fatality potential source terms) distribution before and after re[mling. i { l 38
SOURCE TERM GROUP IDENTIFIERS 1 2 3 4 5 6 7 8
+......+......+......+......+......+......+......+......+
1 l l l l l l l l SEQ.10l
+......+......+......+........ ....+......+......+......+
2 l l l l l l SEQ.07l SEQ.09l SEQ-11l
+......+.............+......+...... ......+......+......+
3 l l l l l. l SEQ.08l l l
.......+......+......+......+......+......+......+......+
4 l l l lSE0-04l SEQ.06l l l l
+......+....................+..... +......+......+......+
5 l l l SEQ.02l SEQ.05l l l l l
+.............+......+......+......+.............+......+
6 l } SEQ-01l SEQ.03l l l l l l
+......+......+......+......+......+......+......+.,. ..+
1 2 3 4 5
..............+......+......+......+
1 l SEQ.12] l SEQ-13l SEQ-14l SEQ-151
.......+...... ......+......+......+
Figure 10. Source term group identifiers. Y collapse occurs, in V-sequences, and sequences lities, latent cancer fatalities, and 50-mile with contairiment failure before core damage, the population dose. A PC-based version of the coce evacuation warning time is the time of core un. is available and was used in Ac present analysis.* covery. Using the PARTITION logic for separat- The MACCS input files that contain the site char. y ing subgroups, and the warning and release times acteristic data and dose data were those used in calculated in the SEQSOR program, provides the the NUREG-1150 Sequoyah analysis. These files following assignment of release bins to release were modified only as necessary to be compatible subgroups. Bins with no containment failure, with the PC version of the code, which is a later with containment failure in the late or very late release than was used in the NUREG-1150 analy. periods, with steam generator tube rupture and ses. The fission product releases were provided secondary SRVs stuck open all go to subgroup by the SEQSOR program described in the pre. one. Steam generator tube ruptures and contain. vious section, with format translation provided by ment failures at or before vessel breach are as. the STER program) signed to subgroup two. Only V-sequences are assigned to subgroup three (therefon there are no group-three assignments resulting from SCET a. Information was taken from " Documentation of analysis), Ti,e source term data resulting from INEL PC Version of MACCS 1.5.ll,1NEL Calcula-this step of the process is presented in Table 10. tion Package," dated November 3,1989, and done by K. R. Jones, EG&G Idaho, Inc. 4.3.3 Consequence Calculation. The source b. The STER program is in undocumented trarmla-terms produced from the PARTITION runs are tion code that is not strictly necessary to the analysis. used as inputs to the MACCS code to generate It can be ot/aiced from Sandia National Laboratory consequence measures. For the SCET develop- (SNL) along with the other level 2 and 3 analysis ment, the consequence vectors include early fata. codes as described in Section 3. 39 l
Table 10. Sequoyah source term data by PARTITION group and subgroup i Warning Evac Release Fractions Source Time Time . xvarion Energy Stast Duration Term (s) (s) (m) (w) (s) (s) NG I CS T, Sr Ru La Ce Ba SEQ.O! 2.2E+04 -2 6E+03 1.0E+01 1.7E+07 2.8E45 6.SE+02 83E-01 3.9603 3.95-03 1.7E-03 5.5E4M 4.0E4M 1.1E 31 1.1538 6OFAM S OE+05 33E+05 3.2E+05 1.7E-01 7.2E4M 71FA4 5.5E4M 4.2E-os 1.4E-G1 5.5FA5 5 05-05 3.1EJM SEQ-01-1 . - - - - - - - - - - - - - - - L SEQ-CI-2 2.2E+0* -.2.6E+03 1.0E+01 1.7E+07 2.8E+0* 6.8E+02 83FAI 3.9E-03 3.9FA3 1. ,FA3 5.5E41 4DE-os 1.IE44 1.1FAM 61E4M 5.0E+05 33E405 3.2E+05 1.7FA)1 7.2E-0* 7.2E-o* 5.5E41 41FAS 1.4E-ot 5.5FA5 5.05-05 3.1E44 SEQ-01-3 - - - - - - - - - - - - - - - SEQ 2.2E44M -2.6E+03 3.0E+01 5.2E+05 2.8E+04 1.0E+03 7.5FAI 3.9E-03 3.9FA3 1.7FA3 5.2E-os 6.2FAM 1.7E o* 1.7E4M 6.0E4M 1.6E+06 2.9E+0* 2.2E+04 2.5FAi 13FA3 13E-03 6.8E4)3 13E-02 2.1FA4 8.0E-31 6.15 44 7.7E 03 SEQ-02-1 - - - - - - - - - - - - - - - SEQ-02-2 2.2Est -2.6E+03 1.0E+01 5.2E+05 2.8E+04 1.0E+03 7.5E-OI 3.9E-03 3.9E-03 1.7E-03 5.2FAM 625-O* 1.7E-Ot 1.7E-o* EUE4M 1.6E+06 2.9E+0* 2.2E+04 2 5E4)! 13FA3 13E-03 6.SE4)3 13 &O2 2.150s 8.0FAM 6.lE4M 7.7E-03 SEQ-02-3 - - - - - - - - - - - - - - - SEQ.03 2.2E+0* -2.6E+03 1.0E+01 33E+07 2.8EMM 7.9E+02 9.0E4)1 7.lE-03 7.1E4)3 2.5E-03 83E-at 4.0E-.05 1.1E-05 1.1E-05 8.4EAM M 7.4E405 13E+05 1.2E+05 9.7E-02 1.2E-03 1.2E-03 1.85-03 7.6E-04 2.9E-4h 3.7FAs 2.SEM)5 SJE-OS SEQ-03-1 - - - - - - - - - - - - - - - SEQ-03-2 2.2E+0* -2 6E+03 1.0E+01 33E+07 2.8E404 7.9E+02 9.0FAI 7.1E-03 7.1FA)3 2.5E-03 83E4M 4.0FAS 1.1FA5 1.lE-05 8.4FAM 7.4E+05 1.3E+05 1.2E+05 9.7E-02 1.2E-03 1.2403 1.8E-03 7.6E-OS 2.9E4M 3.7E-05 2.8FAS 5.0EAM SEQ-03-3 - - - - - - - - - - - - - - - SEQ-os 2.2E+0* -2.6E+03 1.0E+01 4.8E+07 - 2.8Ed4 1.5E+03 1.0E+00 3.0FA2 2.7E-02 2.4E-02 3.6E413 6.7E-03 1.8 5-03 1.8E03 4.4E-03 1.4E+06 7.8E+05 7.8E+05 0.0E+00 0.0E+00 0.0E+00 1.5E-05 1.9E-05 8.1 5-10 1.1E4% 8.5G07 1.!E-05 SEQ St-1 - - - - - - - - - - - - - - - SEQ-05-2 2 2E+04 -2.6E+03 1.0E+01 4.8E+07 2.8E+Gt 1.5E403 1.0E+00 3.0EA12 2.7&O2 2.4E.-02 3.6FA3 6.7FA)3 1.8FA3 1.8E-03 4.4FA3 1.4E+06 7.8E+05 7.8E+05 .0.0E+00 0.0E+00 0 0E+00 1.5FA5 1.9FA5 8.lE-10 1.1EJE 8.5E-U7 1.1F A 5 SEQ-o!-3 - - - - - - - - - - - - - - - SEQ-05 2.2E+04 -2.6E+03 10E+01 2.0E+06 2.8E+04 13E+03 7.9E-01 1.4E412 1.4E-02 5.0E. 03 1.9E-03 9.6FAM 2.6E-os 2.6E-OS 2.0E-03 1.2E+06 8.0E+04 73E+C4 2.1E-01 32FA3 3.2E-03 1.3E-02 2.8E-03 1.8EAM 1.9E-01 1.5E4u 1.9E-03 SEQ-05-1 - - - - - - - - - - - - - - SEQ-05-2 2.2E+G4 -2.6E+03 1.0E+01 2.0E+06 2.8E+02 13E+03 7.9E-01 1.4E 02 1.4E-02 5.0E-03 1.9E-03 9.6E-o* 2.6E-05 2.6GO4 2 0E-03 1.2E+06 8.0E+04 73E+0* 2.1E-01 3.2E-03 3.2FA3 1.3FA)2 2.8E-03 1.8FAM 1.9E-GS 1.55 38 1.9E-03 SEQ-05-3 - -- - - - - - - - - - -
Table 10. (continued) -. Releam Fractions Warning Evac. sourte Time Time Elevation Enern Stan Duration Ru La Ce Ha I CS _ Te Sr Term (s) (s) (m) (w) r s) (s) NG_ SEQ-46 2.2E+0* -2AE+03 1.0E+01 2.2E+08 2.8E+0* 3.5E +01 1.0E+00 3.6FAf2 3AE.42 3.5E-02 5.8&O3 6.5E-01 1.RFAJ3 1.8Go3 6.5FA3 8.1E+05 4.7E+05 4.7E+05 2.2E-03 9.9605 9.9E-05 7.1E-os 1.0&O3 2.7E4E 58&O5 43 &o5 5.9Edu SEQ 441 - - 2.8E+0* 3.5E+01 1.0E+00 3.6&O2 3AFA2 3.5602 5.8FA3 6.5E-03 1.RFA)3 1.8G03 6.5FA13 SEQ 442 2.2EdM -2.6E+03 1.0E+01 2.2E+08
- 8. lE+05 4.7E405 4.7E+05 2.2FA3 9.9405 9.9&o5 7.lFAM 1.0FAs3 2.7FAE 5.8&OS 43&OS SEAM SEQ 443 - -
SEQ 4U 13E+0* -1.1Ed8 1.0E+01 3.9E+06 2.0E+01 3.1E+03 4.2ER)1 13601 1.5FAI 2.3901 22E-02 1.8 EAU 1.8E46 1.RFA)8 2.2E-02 1AE+03 83E+05 8.2E+05 2.2E-02 2.9FA3 2.9403 5.2FA3 5.5FAM 1.1FAE 7.1E4% 53&JM 5.1FAu SEQfE1 - - - - - 1.8 EAR 2.2FA2 SEQ 4 %2 13E+04 -1.1E+03 1.OE+01 3.9E406 2E+Gt 3.lE+03 4.2FA)I 1.5E01 1.5FAI 2JRon 2.2E4C 1.NE4R 1.RFAE 1AE+03 83E+05 8.2E+05 2.2 fan 2.9E4)3 2.9E03 5.2403 5.5FAM 1.ILM 7.1 E46 53FAW 5.lFAM SEQ 4%3 - SEQ 4* 2.2E+04 -2.6E+03 1.0E+01 4AE+08 2.8E+04 3.1E+01 1.0E+00 1.0FA11 IDFA)I 3.4E-02 1.3FA2 1.8FA3 48F of 4.8FAM 13&o2 3 6AE+05 33E+05 33E+05 0.0Ed10 0.0Edo 00EW) 9 E A)3 5.0M13 1.7FAU 2MAM 2.2FAM 2_9FA3 l SEQ 4A1 - - SEQ 442 2.2E+0* -16E+03 IE+01 4AEME 2.8EMM 3.1E+0i 10Edo 1.0E-01 1.0FAI 3ASO2 13FAE I .8FA3 4.RFAM 4 R 6AE+05 33E+05 33E+05 0 0E+00 0.0E+00 0.0ERW) 96603 5.0FA3 1.7FAU 2 4FAM 2.2Ent 2.9 F SE(FE-3 - - 2.2Edu -2E+03 1.0E+01 4.8E+08 2.8E+ 0* 3.2E+01 IE+00 2.5&ol 2.5E-01 6.7&o2 2.9FA2 13FAM 3AF475 3AEW15 2EA2 SE(MN 3.8E+05 3.8E+05 23 &O3 3.2FAM 3.2E44 2.0&O3 49FAM 3AFA% 2.7E05 2.!&O5 3 0FA4 1..tEd)6 - - - - - - - 16FA5 2EA>2 2.8EMM 3.2E+01 IE+00 2.5FAI 2.5FA1 6. TFA2 2.9EfC LtFAM SEQJA1 - - 3Ai'.'" SEQJ42 2.2E+04 -2AE+03 LOE+01 4.8E+0S 1AE+06 3.8E+05 3_8E+05 23FA)3 3.2FAM 3.2FAM 2.0FA3 4.9E o? 3AEM 25t 05 2.lFAO SEAM SEQ 443 - - SEQ-10 2.2 E+nt -2.6E403 1.0E+01 4.1E+08 2.8E+0* 1.9E+0! 1.0E+00 7.6FAI 7AEstl 2EAI 8.9E.4G 68 EMU 6 RE4r MF49 8E.4 2 3AE+05 1.(E+05 93Edu 00E+00 00E+00 00Edo 4.1E02 8.6E03 4 EAU 5.1FA)2 1RFA M 5.1E 03 SEQ-1G-1 SEO-10-2 2.2E+Gt -2.6E+03 1.0E+01 4.lF48 2.8E+0! 1.9E+01 1.0E 4V) 7A601 7.6E4)I 2.06-01 8.9FAC 6 R&4U 6.RFAE 3.4E+05 IE+05 93E+01 00Edo 00E+00 00E+9) 4.1E-472 RAE.03 4.EE fC 5.1E4x 3 3E44 5 IFA3 SEQ-10-3 - 2.2E401 -2AE+r13 1.UE+01 5AE+07 2.8E404 IE+01 10E+00 4DFAI 4DFA1 1.0FAI 4.7 FAG 3AE 'n 3.6F, (E 3AFAE 4.7E 02 SEQ-11 L6E+05 1E+06 1.0E+06 00E+(w) ODE +00 0(E+00 OEco 0.0E+0) 0.0E dw) 0.orAvi DDEnt) onEd
l - _ L Table 10. . (continued)- ' i
- Waming Evac.
Soune Reles*e Fractms Time Time Elevation Energy Start Duration , Tern -(s) (s) (m) -(w) (s)_ (s) NG I CS Te Sr ' Ru La Ce Ba I SEQ-11-1_ - t SEQ-11.-2 2.2EH14 -2E+03 ~ 1.0E+01 16E+07. 2.8E+04 - 1.0E+01 1E+00 4.0&ol 4.0&O1 1.0&01. 4.7E-02 3.6E-07 3.6&o8 3.6&O8 47&O2 SEQ-11-3 . - - thE+05 IE+06 IE+06 DE+00 OE+00 OE+00 OE+00 OEdo OEd10 OM+00 O E do O E +00
- -- - - - -- -- - - - - - I SEQ-12 2.2E+04 1.6E+0! . 0.0E+00 OLE +00 . 4.7E+04 0.0E4C OM+00 0.0Ed10 OE+00 OE+00 0.0EWO OE+00 0.0E+00 O E do 0.06+00 OE40 43E+0* 8hE+0S 105-03 1EG09 18&OG 1.4&O9 7.7510 SA&l5 4AE 82 335-12 7AL10 '
SEQ-12-1 '2.2E+04 ' 1.6E+0* . 0.0E+00 OE+00 4.7E+04 0.0E+00 0.UE+00 0.0E+00 OE+00 OE+00 0.0E+00 OMdo OE+00 OEdo <wE4C ' 0.0E+00 ' 4JE+04 8.6E+41 ' 5 OS03 185-09 18& O9 1.4E-09 . 7.7510 8.4&l5 4AE-12 . 33&l2 7AGIO ; SEQ-12-2 --- - - - - - - - - - - - - - - SEQ-12-3 - - - - - - - -- - - - - - - - SEQ-13 2.2E+04 1.6E+nt 1.0E+011 00E+N 4.7E+04 OE+00 OM+00 OEWO DE+00 OE+00 OE+00 OKWO Dedo 0.0E+00 0.0E40 1AEd)7 - 1.2E+05 :1.0E+03 IE+00 .1.7E-06 13E-06 6.05 05 1ASOS 8.5 & 10 SAG 07 63&o7 8.5E-06 l SEQ-13-1 2.2E+04 1.6E+0* - 1.0E401 UE+00 ' 4.7E+04. 0Edo OE+00 0.0E40 0E+00 ' 0E+00 0 0Ee '*0E+00 00Edo 0.0E40 0.0E400 8 1AE+07 1.2E405 - 1.0E+03 IEd10 135-05 13 & O6 6.0E-05 1/E-05 8.5E-10 8AEC7 635-07 8.5E-06 " SEQ-13-2 - -- - - - - - - - -- - - - - - - SEQ-13-3 - - :- -- - - - - - - - - - - - SEQ-14 2.2E+0* 4.4E+03 1.0E+01 : 33E+04 - 3.5E+0* - 5.6E+03 6.0&ol 3.9E-05 ' 3.9B-04 ' 1.0&O4 4.6&OS 3.6E-10 3.6E-11 3.6&11 4.6E-05
- 3.5E+0' ~ 53E+05 11E+05 4.0E-01 4AE-05 4AE-05 1.2&O3 ' 2AE-O 1.4E-OR 1A&OS 1.0E-05 1 A&o4 SEQ 14-1 . 2.2E+04 4AE+03 1.0E+01 3.3E+0s . 3.5E+04 .16E+03 6.0E-01. 3.95-04 3.9&o4 1.0E-04 4.65-05 3hE-10 3.6E-11 3.6E-11 44&o5 4
3.5E+07 13E+05 11E+05 4.0&01 4A&oS 4.4E-05 1.2403 2A&ot IAE-OR 1A&o5 LOG O5 1A E nt SEQ-14-2. 2.2E+0* 4AE+03 ' 1.0E+01 33E+0S 3.5E+01 5 6E+03 6.0E-01 3.9&Os 3.9&Os 1.0E-04 ~ 4.6&O5 16E-10 3.6E-11 3.6E-11 4 A&o$ 3.5E+07 13E+05 11E+05 4.0&O1 4AE-05 '4A&o5 1.25-03 2.4 & o4 1.4&O8 ' 1.4E-05 1.0E-05 1 AE-Ot ! SEQ-14-3 SEQ-15 22E+04 1.5E+0* IE+01 4.2E+03 4.5E+0* 8.7E+02 8.1502 2.1E-04 2.1E-04 7.9E-05 2A E-05' I 8 & IO 1.8E-11 1.85-11 2AE-05 6AE+08 ' 1.2E+05 8.lE+04 92E-01 1.0&O2 1.0E-02 . 6 8&O3 2.06-03 8.2E-05 4.8&O5 3.6E-05 IJE-43 SEQ-15-1 2.2E+0S 1.5E+0S - 1.0E+01 42E+03 4.5E+04 8.7E402 8.1602 2.1E-04 2.1508 7.9E-05 2AE-45 1.8E-10 1.8B-11 1.8&11 2AE45 6AE+08 .1.2E+05 8.IE+04 9.2E-01 1.0E-02 1.05-02 6 SE-03 22-03 8.2606 4.8E-05 3.6E-05 IJE-03' SEQ-15-2 2.2E+0* 1.5E+ 0! 1.0E+01 4.2E+03 4.5E+04 8.7E+02 8.1E-02 2.IE-04 2.1E-04 7.95-05 2AB-05 1.8E-10 1.8E-11 1.8&Il 2.4E-05 6.4E+08 1.2E+05 8.IE+04 9.2E-01 LOE-02 1.06-02 6.8E-03 2.0 6-03 8.2&oS 4.?B-05 3.6E-05 1.7E n3 SEQ-15 ' _ _ - - -c - - -
&#u.+.. , s_y ,y,, - .h U - ,
7he results of the consequence calculations are however, time and resource constraints did not summarized by source term group in Table 11. permit this. Because the containment failm e mode Table 12 shows the spreadsheet layout used to calculations are in very close agreement with the calculate risk from the consequence and NUREG-Il50 results, the source of the differ-frequency data, ente is most likely within either the source term
- e. smne non<onservatbninnhht have been in-4.3.4 - Comparison with Draft trmlocedn generanng tk M-cliaracter vecim)or NUREG-1150. The r.sk calculated with the c nsequence calculation (newer version of SCETs is compared against the Sequoyah h1ACCS was used).This same problem was en.
NUREG-il50 data in Table 13. Agreement is countered in Reference 15 (Zion CPIP).There the generally gomi considering the uncertainty distr . butions in the NUREG-il50 calculation. Early difference was attributed primarily to the use of tatalities are consistently underestimated with the the central estimates in the SCET approach, as op-SCETs. It is likely that further sensitivity calcula- posed to distributed parameters used in the tions would pinpoint the source of this bias, NUREG-1150 work. Table 11. Consequence data presented by release group and subgroup 50 mi 1000 mi Early Latent Dose Dose Fatalities Fatalities ( Rem) (Rem) Grou L SEQ-01 1 SEQ-01-2 2.07E-415 2.73E+02 3.54E+05 1.60E+06 SEQ-01-3 SEQ-02-1 S EQ-02-2 1.16E-02 3.00E+02 5.47E+05 1.93E+06 SEQ-02-3 SEO--03-1 SEQ-03-2 5.22E-08 4.14E+02 4.34E+05 2.40E+06 SEQ-03-3 SEQ 41-1 SEQ 41-2 6.l l E-4M 8.81 E+ 02 7.04E+05 5.10E+06 SEQ-4M-3 S EQ-05-1 SEQ 4)3-2 7,80E-03 6.21 E+02 7.41 E+05 3.ME+06 S EQ-05-3 SEQ 4b-1 SEQ 4b2 2.63E-04 1.l5E+03 5.55E+05 6.83E+06 S EQ-06-3 43
. .. - . - . , . . - - - - . . = _ - ~ - - - . . ~ . . - ~ . - - . - . . . - . . - . . - . . -
t
-LTable 11,, '(conlinued)'.
-50 mi 1000 mi-- .
Early . Latent -- Dose. Dose Group Fatalities : Fatalities -( RenW (Rem)
' SEQ-07-1 . .
S EQ-07-2 ; 1.95E+00 - 2.01E+03 1.64E406 .1.17E+07 .
- S EQ-07 SEQ-08-1 .
SEQ-08 6.74E44 2.07E+03 . 7.17E+05 1.24E407 -
. SEQ-08-3 -
- SEQ-09-1 r
' SEQ-09-2 -5.52E-03 3.26E+03 9.07E+05 - 1.96E+07 i SEQ-09-3 '
SEQ-10-1 SEQ-10-2 - 1.57E+00 6.08E+03 1.68E+06 3.66E+07 SEQ-10-3
. SEQ-11-1 SEQ-11-2 4.18E+00 3.96E+03 - 1,66E+06' 2.37F407.
SEQ-11-3 -- SEQ-12-1 ~ .0.00E+00 5.40E-03 8.38E+00 - 3.22E+01.
- SEQ-12-2 SEQ-12-3 3 . SEQ-15 1 ' O.00E+00 - 1.94E+00 ' 3.28E+03 1.20E+04 SEQ-13 SEQ-13-3 1 SEQ-14-1 : 0.00E+00 - 3.~2E+01 ~ 3.74E+04 2.16E+05 ,
SEQ-14-2 ' 9.40E-06 < 4.14E+0! 9.38E+04 - 2.39E+05
.. S EQ-14-3 1.86E-06 -5.29E+02 ' 2.41E+05 L SEQ-15-1 ! 3.09E+06 =
- S EQ-15-2 -- 2.63E-05 1.24E+02 2.29E+05 - 7.19E+05
, SEQ-15 44 4_2 ._....a . - _ -
1 Table 12. Sequoyah SCET risk results for five PDSGs 50-mile Cond. Early Cond. Latent SBO-ST LOCA Trans ATWS Cond.
'4ource SBO-LT Early Fatal. Latent Cancer PDSG-3 PDSG-5 PDSG -6 50-mile Pop Dose Terrn PDSG-1 .PDSG-2 Risk Cancers Risk 2.4E-05 2. E-06 Pop Dose Risk Fatal.
Grid 4.6E-06 93E-06 3.5E-05 3.5E+05 1.2E-01 2.1E-05 6.8E-12 2.7E+02 9.0E-05 1-2 1.1E-02 1.9E-02 2.6E-03 1.1E-04 5.6E-03 5.5E405 4.15-02 1.2E-02 8.8E-10 3.0E+02 23E-05 2-2 7.0E-03 4.6E-03 4.1 E+02 1.IE45 1.8E-02 43E+05 1.2E-01 5.2E-08 1.4E-14 3-2 4.8E-03 9.6E-03 33E-03 4.1E-03 7.0E+05 2.4E-02 6.1E44 2.1 E-11 8.8E+02 3.0E45 4- 2 13E-04 1.2E42 3.4E-02 7.8E-03 3.5E-10 6.2E+02 2.8E-05 2.6E-03 1.EE-04 1.5E-03 7.4E+05 5-2 2.6E-03 2.6E45 4.1E-10 1.2E+03 1.8E-03 2.4E-02 5.8E-05 1.9E-02 5.6E+05 8.7E-01 6-2 4.8E-02 5.lE-02 5.85-07 2.0E+03 6.0E-04 6.4E-03 1.3E-01 1.6E+06 4.9E-01 1.9E+00 7-2 1.7E44 2.6E-03 2.9Ec07 53E-04 7.2E+05 8.9E-01 6.7E 04 83E-10 2.1 E+03 8-2 5.5E-02 8.1 E-02 6.5E-03 1.2E-03 6.9FA5 2.0E-03 9.1E+05 3.4E-01 5.5E-03 2.lE-09 3.3E+03 9-2 3,4E-03 2.6E-03 9.5Ec03 6.1E+03 5.0E4* f.7E+06 1.4E-01 1.6E+00 13E-07 10-2 2.7E-03 7.4E-03 4.0E+03 4.9E-G2 1.7E+06 2.1E-01 4.2E+00 5.2E-07 Il-2 1.1E-02 8.0E-03 5.4E-03 1.9E-07 l , 6.6E-01 8.4E+00 3.0E43 0.0E+00 0.0E+00
- 12-1 7.1E-01 5.5E-01 6.8Ec01 93E-01 2.5E-05 33E+03 4.1 E-02 0.6E+00 0.0E+00 1.9E+00 13-1 1.4E-OI 2.4E-01 2.7E-01 3.9E-02 1.5E41 4.8E-03 0.0E+00 0.0E+00 33E+01 4.2E-06 14-1 1.4E-03 13E-02 1.0FcO4 6.8E45 3.7E+04 2.1E-02 9.4E-06 2.l E-12 4.lE+01 9.1 E-06 14-2 3.0E-03 1.5E-03 4.8E-03 5.6E-03 4.4E-03 9.4E+0*
13E-02 1.9E-06 9.8E-14 53E402 2.SE-05 3.0E-03 1.5Er04 2.4E405 15-1 43E-03 2.6E-05 4.8E-13 1.2E+02 23E-06 4.4 E 4 8 4.9E-05 5.9E41 23E+05 4.2E-03 15-2 9.8E-05 8.7E-05 l 335E+00 1.24E-06 7.52E--03 1.00E+00 9.99E-01 9.99E-01 1.00E400 1.00E+00 4.87E-01 234E-O'l 1.16E-03 SBO-LT PDSG-1 1.23E+00 4.22E -07 3.04E-03 SBO-ST PDSG-2 1.06E+00 2.265-09 2.61E-03 LOCA PDSG-3 5.20E-02 3.02E-08 6.23E-05 Trans PDSG-5 5.14E-01 5.49E-07 6.50E4* ATWS PDSG-6 335E+00 1.24E-06 7.52E-03 I . _ - _
1 1 1 Table 13. SCET risk results compared to Draft NUREG-il50 risk results ) (FCMR and MFCR rnethods utilize APETs and are from NUREG/CR-4551, Vol.5, ) Table 5.1-2) ) 1 Core 50-Mile l PDS- - Damage Population Early Im ent i Group Method Frequency Dose Fatalities Cancers SILO-LT FCMR* 4.6&O6 1.3 E+00 1.8E4)6 1.8E-03 PDSG-1 MFCRb 4.5506 9.6E-01 1.7E-06 1.2E-03 SCET N/A 4.9E4)1 2.3 E--07 1.2E4)3 SBO-ST FCMR 9.3 E-06 3.2E+00 4.2E-06 4,0E-03 PDSG-2 MFCR 9.4E-06 2.9E+00 4.7E-06 3.6E-03 SCET N/A 1.2E+00 - 4.2 E-07 3.0E-03 LOCA FCMR 3.5E-05 2.2E+00 4.4E4)7 2.0E-03 PDSG-3 MFCR 3.4E-05 3.4 E+00 3.4 E-06 2.984)3 SCET N/A 1.lE+00 2.3fb09 2.6E-4)3 Event-V FCMR 6.7E-07 1.8E+00 1.8E-05 1 AWO3 PDSG-1 MFCR 8.4E4)7 1.2E+00 1.1E-05 1,4 E-03 Trans FCMR 2.4E-06 6.0502 2.6E-08 7.0E-05 PDSG-5 MFCR 3.2E-06 1.6601 3.4&O7 2.0E-04 SCET N/A 5.2E-02 3.0E-08 6.2E-05 ATWS FCMR 2.1 E-06 4.4 E-01 49E-07 5.3E4)4 PDSG-6 MFCR 2.4E-06 6.4 B-01 1.dE-06 8.0E-04 , 1 SCET N/A 5.1E-01 54&O7 6.5E-04 SGTR FCMR 1.7E-06 3.0E+00 1.4&O6 4.2E-03 PDSG-7 MFCR 2.0E-06 2.7E+00 3.5E-06 3.9E-03 ALL PDSGs - FCMR - 5.6E-05 1.2E+01 2.6E-05 1.4E4)2 MFCR 5.6E-05 1.2E+01 2.6E-05 1.4E-02
- 5-PDSGs FCMR 5.4E-05 7.2E+00 6.9E-06 8.3E-03 MFCR 5.3E-05 8.0E+00 1.2E-05 8.7 E-03 SCET'Ibtals 3.4 E+00 1.2E4)6 7.5E-03 t
l
- a. FCMR-fractional contribution to mean risk.
- b. MFCR-mean fractional contribution to risk.
46
- 5. ANALYSIS OF POTENTIAL CONTAINMENT IMPROVEMENTS Based on the risks estimated for Sequoyah re- suming a 40-year plant lifetime, yields a maxi-ported in NUREG-Il50, a number of potential mum modification cost of $480.000.
contairtment modifications have been postulated The procedure for evaluating potential modifi-for improving the performance of the ice con- Mom stuts with a review of the SCETs to iden. denser containment. These modifications lify those events af fected by the modification and I" " to estimate what that effect is. After revision, the T
- 1. Provio.g a hyuropen mitigation sys- SCED are requantified and new source term bin tein biters) that will function in a probabilities are calculated. These conditional station blackout sequence (either a source term ptobabilities are then combined with backup power supply or passive cata, the PDSG frequency and the conditional conse-lytic igniters). quences attributed to each source term bin through matrix multiplication to produce an over.
2, Backup power to the igniters plus all estimate of rbk. backup power to the containment re- 5.1 Backup Power to the + circulation fans. Hydrogen Ignition System
- 3. liigh pressure melt ejection (ilPME) This potential modification addresses the con-mitigation to prevent eontainment fall
- cern of hydrogen accumulation in the ice con-ute caused by the direct impingement denser containment during a postulated severe of core material on the instrumentation accident The threat posed by this situation is that seal table (and subsequently the con- the accumulated hydrogen could ignite and pro, tainment wall). This modification, as duct an explosion of sufficient magnitude to fait envisioned here, entails lining the con- the containment and thereby allow the release of tainment wallin the area of the seal radioactive material to the environment. Current-table with refractory material. ly all ice condenser type containments are fitted with hydrogen mitigation systems that consist of
- 4. Reduce the probability of hydrocen igniter': tglow plugs) dispersed through the con, burns by inerting the containment tainment (see Appendix A),1lowcver, these sys-atmosphere, tems are ac oowered and therefore will not be operable durmg a station blackout (SBO) event.
Each of these modi 0 cations is examined with Because SBO was identified in Draft respect to its risk reduction potential using either NUREG-1150 as a significant contributor to core the SCETs (mods I and 3), the full APET damage frequency and risk, hydrogerwinduced-(mod 2), or by inference (mod 4, which is func. containment failures remain an important issue tionally identical to mod 1). The risk reduction for ice condenser containments, potentials are estimated through idealized evalua-The risk reduction attributable to this potential tions that assume the potential modification oper-m dification was estimated utilizing the SCEh. ates perfectly and is 100% relinble. Because ae power (and therefore the igniters)is al-For comparkon, a bounding analysis is pres- ready available for non-SBO sequences, only the ented here that estimates the cost associated with SBO event trees were modified to model the ef-reducing the total NUREG-1150 risk (50-mile fects of this option. Specifically, the effect of hav, population dose) Ic zero. The total person-tem ing igniters operable during a SBO was reflected risk estimated by the NUREG-il50 analyses of in the SCET by setting the probabilities of early Sequoyah is 12 person-rem per year (from containment failure (event ECF) and late contain-NUREG/CR-4551, Vol.5, Table 5.1-2). At ment failure (event LCF) to zero. These events
$1,000 per person-tem and conservatively as- were utilized to reflect the effects of the improved 47 1
1
hydrogen control because these containment fail- 5.2 Backup Power to the HIS ute events are dominated by hydrogen burns that and Air Recirculation Fans fail containnient. llowever,it is not com pletely ae-This pou ntial containment improsement ad-curate to set these events to zero because there dresses the same issue as the previously discussed should remain a small probability (1E-3) that the nuxlificat.an, namely preventing hydrogen burns conta.mment will fail from slow overpressuriza-from faihng the containment. This potential niod-tion in the cally and late timeframes. In the current f eation takes the more comprehensive approach analysis, this consideration is neglected. The ef- of providing backup power to both the by drogen feet of this simplification will be a negligible over- igniters and the containment air recirculation fan estimation of the potential modificatioris benefit, system (ARFS). The ARFS is designed to circu-The risk reduction resulting from this improve- late the containment atmosphere from the lower ment is summarized in Table 14. compartment, through the ice condenser, to the Table 14. Risk comparison between base case and modification #1 (backup power to igniters), utilizing SCim 50-htile Population Dose (person-rem per Early Fatalities Latent Cancers Plant reactor year) (per reactor year) (per reactor year) Dainage State Base Base Base Group _ Case blod-l Case hiod 1 Case blod-l s SBO-1;r - 0.487 0.255 2.33E-7 2.50E-9 1. I 6E-3 5.41 E -4 PDSG-1 S BO-ST 1.23 0.829 4.22E-7 7.18 E-8 3.04 E-3 ! . 45 E--3 PDSG-2 LOCA 1.07 1.07 2.31 E-9 2.31 E-9 26tE-3 2nlE-3 PDSG-3 Trans 0.052 0.052 3.02E-8 3.02E-8 6.23 E--5 6.23 E-5 PDSG-5 ATWS 0.514 0.514 5.49E-7 5.49 E-7 6.50E-4 6.5E-4 PDSG4 Totals 3.35 2.72 1.24E-6 0.56E-7 7.5 5E-3 5.85B-3 Notes: Totals are for the five PDSG for w hich SCETs have been develciped arxl do not mctude the two containment bypass sequences (Event-V and SGTR). 48
upper compartment, and then back into the lower the steel containment liner and subsequently re-compartment, sult in a breach of containment. A number of pos-sible mitigation strategies have been postulated to Because the fans are not explicitly modeled in prevent containment failure by tlas mechanism, the SCETs, this modification was evaluated uti, among them are depressuri/.ation of the primary lizing the full-sized APETs, The general before vessel failure and lining the containment approach taken in this analysis, which is ex, wallin the seal table area with a refractory materi-plained in more detail in Appendix B, was to re. al. This analysis postulates the latter option and view each question of the APET and to modify assumes it is 100% effective in preventing con-those that address the failure of the igniters or tainment failure by direct impingement of the fans tecause of a lack of ac power. The backup molten core on the containment wall. pow er was assumed to be perfectly reliable. Ilow- The effects of this potential modif cation were ever, other (not power related) failure mecha- modeled in the SCETs by setting the probability ' nisms were retained as modeled in the base case of containment failute by direct impingement APETs. Specifically, the igniters are assigned a (event DI) to zero. Because this modification is failure probability of 0.01, which is attributed to completely passive,its effect was reflected in all the failure of the operators to actuate the system. five SCETs. Further,it is assumed to be 100% The ARFS is assumed to have a failure probabili- reliable. The risk reduction resuting from this im. ty of 0.001 from system hardware faults. Both of provement is summarized in 'Ihble 16. these failure mechanisms were maintained for this sensitivity case. 5.4 Containment inerting The risk reduction attributed to this potential As for the improved igniter performance, con-modification is shown below on Table 15. As can tainment inerting is aimed at preventing contain-be seen this option is estimated to provide mini- ment failure by hydrogen burns. However, mal benefit (less than 2 person-rem per reactor wherena ihe igniter system is designed to burn the year) and would not likely be justified as a hydrogen before an explosive mixture can be backfit. generated, containment inerting precludes the fon. nation of a combustible mixture by maintain-5.3 High Pressure Melt Ejection ing the containment atmosphere in a de-oxygen-Mitigation ated state during normal power operation. Consequently, although the approach is different, Vessel failure at high RCS pressures can result the effect of this modification would be virtually in the rapid ejection of the molten core material, identical to the improved igniter performance The molten core could travel through the instru- modification (mod-1) discussed in Section 5.1. mentation tunnel and impinge on the instrumen- llence, this modification is analyzed identically tation seal table, Once the seal table has failed, the to what was performed for mod-l and those re-ejected material could come in direct contact v/ith suhs are reproduced in Table 17. Table 15. Risk comparison between the base case and modification #2 (backup power to the igniters and fans), utilizing the APETs 50-Mile Population Dose Early Fatalities Latent Cancers (person-rem per year) (per reactor year) (per reactor year) Base Case 10.5 1.89E-5 0.0151 Mod-2 Case 8.7 1.91E-5 0.0117
-19
W _ . . . l Tablo 16. Risk comparison between base case and modification w3 (high pressure melt ejection mitigation, lining the containment wall in the seal table area with refractory material), utilizing SCETs 50-Mile Population Dose (person-rem per Early Fatalities Latent Cancers Plant reactor year) (per reactor gear) (per reactor year) _ Damage State Base Base Base Group Case Mod-3 Case Mod-3 Case Mod-3 0.441 2.33E-7 2.33 E-7 1.16E-3 1,13E-3 SBO-LT 0.487 PDSG-1 SBO-ST l.23 1.10 4.22Fc7 4.19E-7 3.04Eb3 2.93E-3 PDSG-2 LOCA 1.07 1.05 2.31 E-9 2.26E-9 2.64E-3 2.63E-3 PDSG-3 Trans 0.052 0.0516 3.02Fc8 3.02E-8 6.23E-5 6.21E-5 PDSG-5 ATWS 0.514 0.513 5.49E-7 5.49E-7 6.50E-4 6.5 E-4 PDSG-6 Totals 3.35 3.16 1.24E-6 1.23E-6 7.55E-3 7.40E-3 Notes: Totals are for the five PDSG for which SCETs have been developed and do not include the two containment bypass sequences (Event-V and SGTR). 50
l-l Table 17. Risk comparison between base case and modineation #4 (inerting containment atmosphere), utilizing SCirl's 50-Mile Population Dose (persotstem per Early Fatalities Latent Cancers Plant reactor year) (per reactor year) (per reactor year) Damage State Base Base Base Group Case Mod-4 Case Mod-4 Case Mod-4 S BO-LT 0.487 0.255 2.33E-7 2.50E-9 1.16E-3 5.41 E-4 PDSG-1 SBO-ST 1.23 0.829 4.22E-7 7.18E-8 3.(M E-3 1.95 E-3 PDSG-2 LOCA 1.07 1.07 2.31 E-9 2.31 E-9 2.64E-3 2.64 E-3 PDSG-3 Trans 0,052 0.052 3.02E-8 3.02E-8 6.23E-5 6.23E-5 PDSG-5 ATWS 0.514 0.514 5.49E-7 5.49E-7 6.50E-4 6.5 E-4 PDSG-6 Totals 3.35 2.72 1.24E-6 6.56E-7 7.55E-3 5.85E-3 Notes: Totals are for the five PDSG for which SCLTs have been developed and do not include the two containment bypass sequences (Event-V and SGTR). 5l
- 6. RESULTS AND CONCLUSIONS Simplified containment event trees have teen APETs. These modifications include: (a) backup developed for the Sequoyah Nuclear Power power to the hydrogen ignition systein,(b) back.
Plant.These trees are based in the vast store of in- up power to tuth igniters and the air recirculation formation generated by the Draft NUREG-il50 fan sys'em (ARFS), addressed using the APETs, effort and are simplifications of the accident pro- (c) mitigating direct itupingement of core materi-gression event trees developed and used in those al on the contaimner t wall, and (d) preventing hy-analyses, A virtually perfect comparison was drogen burns insid: containment by inerting the achieved for the SCET containment failure mode contaimnent at:.iosphere, results with respect to the APETs, Reasonable ne I the potential contaimnent modifica-agreement was reached between the risk results , uons yequh in signkant rid reduction. Howev-produced by the SCETs and the APETs. Given ey, dus is not surprising because the total plant further refinement of the source term oinner used ated by Drah to generate the 14-character release vector, a
8 I"' b'9" .y a (as ca
)is nly pts n-reins per reac-more precise reproduedon of the APET- I r year. A bounding calculation (see Section 5) predicted risks could no doubt be generated. produces a cost upper knut of $480,000 as being Four potential containment modifications were justifiable on plant backfits, assuming 100% of evaluated utilizing either the SCETs or the full the risk is removed and a 40-year plant life.
52
._ ~. . _ _ _ _ . _ . _ _ . _ _ . _ _ _ _ _ ._ - __ __
__y l i 1 l
- 7. REFERENCES I. Il. V. Noutbakhsh, An Assessment ofIce-Condenser Containment Performance issues NUREG/
CR-5589, July 1990.
- 2. '). J. Gregory et al., An Evaluation ofSevere Accident Risks: Sequoyah Unit I, NUREG/CR-4551, SANL)86-1309, Volume 5, Part 1, Revision 1. December 1990.
- 3. K. C. Wagner, R. J._ Dallman. W. J. Galyenn, An Overview of Boiling Water Reactor Afark-l Con.
tainment Venting Risk Implications, NUREG/CR-5225, EGG-2548,0ctobet 1988.
' 4. ' ETA-II, Version 1.2, Los Altos, California: Selence Applications International Corporation, November 4,1987,
- 5. Lotus 1-2-3, Release 2.01, Cambridge, Massachusetts: Lotus development Corporation,1987.
'6. 3. Michael Griesmeyer, L. N. Smith, A Reference Afanualfor the Event Progression Analysis Code (EVNTRE), NUREG/CR-5174, S AND88-1607, September 1989.
r
.7l 5arah 3. Iliggins, A _ User's A1anualfor the Postprocessing Program PSTEVNT, NUREG/CR-5380, S AND88-2988, November 1989.
- 8. 11. N. Jow et al.,XSOR Codes User's Afanual, (Draft), NUREG/CR-5360, S AND89-0943, (avail.
able from Public Document Room, Washington, D.C.).
- 9. R. L Iman, J. C. Iiciton,1. D. Johnson, A User's Guidefor PARTITION: A Programfor Defining the Source Term / Consequence Analysis Interface in the NUREG-Il50 Probabilistic Risk Asses-sments, NUREG/CR-5253, SAND 88-2940, November 1989.
' 10. D, l. Chanin el al., AIELCOR Accident Consequence Code System, NUREG/CR-4691, September i1988.
I: . p --11.1 ETLOAD, Version 1.0, Aiken, South Carolina: Westinghouse Savannah River Company (Contact:
.N. Douglas Woody, Reactor Safety Research Section),1990.-
- il21-U.S. Nuclear Regulatoiy Commission, Office of Nuclear Regulatory Research, Severe Accident
- Risks: An Assessmentfor Five U.S. Nuclear Power Plants. NUREG-l !50, Vol.1, Second Draft for i Peer Review, June 1989.
' 13. R. C. Bertucio and S. R. Brown, Analysis of Core Damage Frequency: Sequoyah, Unit I internal _
Events, NUREG/CR-4550,-Vol.5, Rev.1, Parts 1 & 2, April 1990.
- 14. R. L lman, M. J. Shortencarier. A User's Guldefor the Top Event hfatrh Analysis Code (TEAIAC),
i NUREG/CR-4598, SAND 86-0960, August 1986. I5. ' D. L. Kelly et al., Guantitative Analysis ofPotential Performance improvementsfor the Dry Pres. surl:ed lHiter Reactor Containment, NUREG/CR-5575, EGO-2602, August 1990. 53
APPENDIX A ICE CONDENSER DESIGN FEATURES A-1
1 APPENDIX A ICE CONDENSER DESIGN FEATURES Design Feature Comparison Cook 1 + 2 Scouoyah 1+2 McGuire 1 +2 Catawba 1+2 Watts Bar 1+2 Ind./Mich. Utility Power Co. TVA Duke Duke TVA Comm. Op. 895, 7R8 7/81, 6/82 12/81, 3/84 6/85, 8/86 N2, indef. Power 100% 1020/1060 1148 1129 1129 1177 MWe (net) MWt 3250/3411 3411 3411 3411 3411 3 1.19M 1.196M 1.18M 1.19M Cont. Vol. (ft ) 1.2M Design pr. 12 psig 12 psig 15 psig 15 psig 15 psig Est. Fall. 60 psig 84 psig 120psig Annulus 3 375,000 426,760 484,090 375,000 Vol.(ft ) N/A RWST Vol. (gaI.) 350,000 370,000 372,100 350,00() 350,(XX) Cont. Sp.
# pumps 2 2 2 2 2 Cap. @ 3,200 gpm 4,750 gpm 3,400 gpm ' K) gpm 4,000 gpm RilR Spray
# pumps 2 2 2 2 Cap. @ 2,(XX) gpm 2,000 gpm _ 1,600 gpm 2,000 gpm Service Water Cooling Towers, 1.8E+8 gal 2.7E+8 gal Source river NSW pond NSW pond Fire Prot.
System
# pumps 4 3 3 Cap. @ 1,500 gpm 2,5(X) gpm 2,500 gpm A.3
i,
., %. s C, ' Nr -Ae
<,/ 7 s .perratw s:nru,* 4
-. _. r.r.WNN '."
.iis ,tawraeggy . . . .
6 .a e N' % s. _./ , , _- -a ,g . . . . . . . ,
. p m w s m _~
c
/
/
=
/.sw
. u_.
Aa ews estsautumw 4 m.mm-d . , . ,
< 3..* e. e
/ g,g t *$+rer*;.r;,; marse-e.stes-rem n., \
{ ..
\ se '
+
\ ir ; **** V h
\ 3 g ; -* - ,, /,
\
\. i 1 .J 3$ y.t --
/g _
T I N x
. }
tw,'% .._ ((JN_ _h_ rgG s .__,4,t , w J ,,,,, ,,,,
.y !.yj y 7 p;.,-_?.w 2
[
,on:r ,\ .,
Q.k' .
,,1 . -
m I. Q. g i
;n~ic-'
g,.. . . .m, ...,,,, , pM ,-J (J. .- s..q ng .. j, g ,. ,, x
.r r m
2 ;~. ., a...- ,m,, t7.a *i i [ u.
,,',"S, a .
L. :..a N
/- p "'~ , " , ,
,,,,,,,,,,,,,p.
q "y. . 74 ,, . g ggk M9h,.
~~
7 g m
" ~T. ~.TQ . . , . , , , ,
"A _ _ . . _
.h j "e
@!' P.*D..7 MtW'g i i
l' g.. j 7m W. m W{ f:<t ll' i l. n.y -: / s q pi k $ s ,e
. .w. _.x.5 m_ s i.
~6a' *** [1,;
l dJ. . N*] r rn n =g, wraj g,. ,
' if.*L".'"
r: A. N=.rt.;t.@ tb. 4 5.f{ 3' P "
- J. {.,.j -
..,g'{t"~'t
,j ,, 7, . ., ;
l- \1 (3dj
~
i c,' il g Gl.:)![.n,,g j gi.-
- 0 I : R t .
a .: j' q h -1,,
- n. .A w .q,, f=)
q- n 1
?
'c t E, dr,4*aN.
'd-
',(g, S g - .;
ki '+ 7 ,\ .,_.
' . M.'.%
..u,,,,,
4
- s. a a
~
. e.
- m. ,
h k p"'*'.Lj ,, 1 i j,.
.b 1
7h n- i" A
)V.
idnudnda -.K. .
} ppppg h '~
l3 , L' ' , ' ' " c .- .
""u _- .uz,,, g I 3 '
_ M. . Q, ,,a
, _ _'.;h p a
,'N'-
c.- , ra ,s - a - P --) .'
~
.,{c *
- , :.k.{,l[
i
",% .:. l .
c.--.;5 s b ,
, 4..,4-g_..k.- g. '.
>..E*
& Li i
. l i
'. . - : ;,7- a.b+.% .g,g .g.w
'1 pC; },'htb,.. r 7,7 ..
3.. .m. ..g._.. J ! ',9, LL '
- t. . :=. o . -/q 1,4,g.,
[1
,) kil$ .. . /'~
l v, lw %e-
. o.3 - .. t l
+wy -- V '.4 g 3: a9 h I N
WL-(&% p} jul II I i Figure A-1. Ocneral Arrangement of Containment - Sideview (frora Catawba FSAR figure 1.2.2-15). A4
O (O A ' 1 Il "Ibly E(
/; " ,
/ k '1 ~
Pj%gn f~.J y'- 7 N. ,,,
// -4p '
/ G h f.,tM ' -\
p, u,,4 # 1. i &,...u~!.h. p, r aiph.p,iMr.h,A p.#
's. 7 f g yg.y([
,. .m ..
g'w'st s.l v%. i . $n.jj,9MfffghfNhI f, /h-/,hf'.h%k[7$)E i f 3
%t .e. i b
^ * %.
rAu c.ypg y b 8 $s
. k.')[i a,(5 \s >
I 3
,4 *
,y B a
^,
74 (p.- g(W. .. ,\ 'gh
- E.) y I"L N '
7s T ,7
^
. [1 y % q ,. I --
fll) B ,
. - ( .,y yif g
.*g'h'('s'.
po O x #, gs \\.
,, /g , J.x:
(j cf:
- [f .. ,//'s, 7'
( ' (i/4,$3], j, [
/
f 4g " _ Q . }a.C '; 'a .,u T!?p46 .
.\ wa vsqfg~
l I. I n i
- i. 1,..,}Pj, l
; i ,
pp l \ b C ~.-- 3. .sc' c \1, g q[ ;c ll
'l I f 4 b t IL
,i, q bl{b)y---(.ttd p ) - e -
ll j P 5; :? E J!L M ' A '$ 1 Nf h/ .. 14 ji UQY I' [3'/,-! . k. jl \ k(\d \;,tN , hgg4,. ) e .()) 4, # . (
) q,,
$\s ll \ s' -
! .yt.'(.- g,g.< 3,/
f */ " J l '* l
'\ /
l *\l 1 '~ I; ~. M,
, j.- # !! l $
({'!i I'igure A.2. General Arrangement of Containment . To[r.iew (from Catav tia FSAR ligure 1.2.210). A5
,a 1 Hydrocen MitJ.gation Sntem hicGuire - (from NUREG-0422, McGuire SER supplement 7 May 1983, also see 1984 update of FSAR section 6.2.7).
- 1) Model 70 glow plug manufactured by Genera! Motors AC Division (1700 F).
- 2) Powered directly from a 120/14 We transformer, cach plug has its own transformer.
- 3) Two redundant group, with five separate circuits und circuit breakers per group, the number of igniters on each circuit ranges from 1 to 10,
- 4) System is manually actuated from the auxiliary building by switching a total of 14 breakers at six locations in the aux, building (estimated to take 10 min.). System can also be actuated from the main control room and has the means for verifying the system status from there.
- 5) McGuire site contains two diesel generators per unit with manual cross. tics between units. Also, McGuire has a safe shutdown sptem with a fifth dedicated diesel generator.
- 6) HMS consists of 66 igniter assemblics (staff requested two additional igniters in the lower compartment and four additional igniters in the upper compartment).
. Lower compartment 20 (10 per traln)
. Ice condenser upper plenum - 12
. Upper compartment 8
. Dead.cnded region 2 in each of eight rooms and 5 pairs in instrument areas.
- 7) System is surveillance tested quarterly, fatawba - (Catawba SER, NUREG.0954 Supplement No. 2, June 1984)
The hydrogen mitigation system (HMS)lnstalled at Catawba is identical to that installed at McGuire, except for minor differences in terminal box designation and igniter location. Sequoyah Sequoyah SER, NURE04 Kill supplement 6, December 1982 and FSAR Section 6.2.5A.2 Rev.2)
- 1) Sequoyah uses 120 Vac hermetically scaled thermal igniters manufactured by Tayco Engineering.
- 2) Tayco igniters exhibit a tendency to cool (to 1500 F) significantly in a spray environment.
- 3) Igniters are equally divided into two redundant groups, with 16 separate circuits per group, cach with an independent circuit breaker and two igniters per circuit. Manual actuation capability for each group is provided in the main control room (one switch per group), along with the status of each group (on.off).
- 4) The splem consists of 68 igniters:
A.6
. imwer compartment 22
. Ice condenser upper plenum 16
. Upper compartment 14 (4 around the dome,4 at intermediate clevations around the S/G enclosures,4 around the top inside of the crane wall, and one above cach of the two air return fans).
. Dead.end regions 16 (2 in each of eight room). *
- 5) Testing and surveillance will be donc to verify that the igniter temperature will be at least 1700 F-Watts Bar FSAR Section 6.2.5A, Amendment 55.
1) Watts liar uses 120 Vae Tayco igniters
- 2) A total of 68 igniters equally divided into two redundant groups:
. 22 in the inwer compartment inside the crane wall
. 16 in the upper plenum
. 16 in dead.cnded compartments
. 4 atound the upper compartment uome
. 4 at intermediate clevations on the outside of the steam generator enclosures
. 4 around the top inside of the crane wall
. One above cach of the two air return fans.
The llMS will be energized manually from the main control room follovring any accident where 3) conditions indicate inadequate core cooling. Containment Spray System hwoyg i? rom NUREG/CR-4550, Rev.1, October 1988, Section 4.6.8. Both CSS pumps start automatically upon receipt of a Phase B containment isolation signal from the Enginected Safety Feature Actuation System (1.sFAS). The Phase B isolation signal is initiated by a containment pressure differential of 2.81 psi between the lower compartment and the annulus. A 30 second time delay is included in the CSS pump start circuit.
- gluovah From Sequoyah FSAR Section 6.23.1.1, page 6.2-97.
"There are no formal design bases established for air cleanup by the Containment Spray System.
This was done with the knowicdge that water from the Containment Spray System will remove halogens and particulates from the containment atmosphere following a LOCA. No credit, however, was taken for A7
. ._. . ... . ._.~ .-. . . .-. - .- . . - - . .- - .
. this removal process in accident analyses presented in subsections 15.4.1. In such circumstances, no design bases are needed for this air purification action."
McGuire . From McGuire FSAR Section 6.5.5, page 6.5 9 (1985 Update to FSAR). Under accident conditions, the Air Return System and the Containment Spray System are activated when the internal pressure in the containment reaches 3 psig. hu out of three hi pressure signals (1 psig) produce an Ss signal or safety injection signal. Two out of four hi-hi pressure signals (3 psig) produce an Sp or containment spray signal. l l i. l A8 1 r
D. C. Ook D. C. Cook FSAR, July 1982, page 6.3 2.
'The secondary purpose of the Containment Spray System is the removal of fission products (radioactive kidine isotopes) from the containment atmosphere. The Containment Spray System is designed to deliver sufficient sodium hydroxide solution which, when mixed with water from the Refueling Water Storage Tank which contains approximately 1.5% by weight boric acid (2(XX) ppm Boron), reactor 4
coolant system water and the melted ice, gives a final spray water pH of approximately 9.3 after the spray additive (Naoll) tank is emptied. The performance of the Containment Spray System for h> dine removal with a single Containment Spray Pump operating adequately fulfills the requirements of 10 CFR 100 as described in Chapter 14.' NOTE: Cook has sprays in both the upper and lower compartments, the other ice condenser plants have sprays only in the upper compartment. D. C. Ook D. C Cook FSAR, July 1982, page 6.3 6. The CSS is automatically initiated on receipt of a hi hi containment spray signal (3.0 psig,2/4 logic). Similarly, a hi containment pressure signal (1.2 psig,2/3 logic) initiates a Safety Injection Actuation signal. Catawba Catawba FSAR Section 6.2.4.1 page 6.2 50. Phase A containment isolation is initiated by a hi containment pressure signal (ST). An ST occurs i when a containment pressure of 1 psig is sensed by 2/4 containment pressure sensors or upon receipt of a safety injection signal (SS). A phase B containment isolation is initiated by hl.hl containment pressure signal (SP), which occurs when a containment pressure of 3 psig is sensed by the same pressure sensors. A9
l Catawba "The Containment Spray System is assumed to remove no fission products following a design basis accident and no credit is taken in offsite dose calculations." (Catawba FSAR Section 6.5.2 page 6.5 3.) The ice condenser design basis includes the removal of iodine from the post LOCA containment atmosphere and thereby reducing the offsite doses following a LOCA. This is accomplished by chemically controlling the ice pil to an alkaline range of 9.0 to 9.5. (Catawba FSAR Section 6.5.4. page 6.5 4, also see the Westinghouse nonproprietary topical report WCAP 7426.) Watts flar . Watts Bar FSAR Section 6.2.21 Section 6.2.21, describes the design bases for the Containment Spray System as primarily containment presure suppression to " . ensure that the containment pressure cannot exceed the containment shell design pressure of 15.0 psig at 250 F." A secondary design basis is the removal of energy directly frorn containment after the ice has melted in the ice chest. There is no mention of a radioactivity removal function in the design basis description for the CSS. i t A 10 l
- ..-. - . - - -. . . _ _ . . ~ . - - . - - . . . . - - . - - . - - .
U l l APPENDIX B APET BASED RISK ANALYSIS OF HIS AND ARFS IMPROVEMENT B-1 I
APPENDIX B APET BASED RISK ANALYSIS OF HIS AND ARFS IMPROVEMENT The potential containment improvement of backup power to the Air Return Fan System (ARFS) and llydrogen Ignition System (HIS) has been evaluated. Unlike the other potential improvements discussed in this report, this fix was examined utilizing the complete APET, as developed by the draft NUREO.1150 effort. The modification was modeled, such that both systems could continue to operate regardless of a loss of all plant power by virtue of a dedicated backup power suppiy that would be independent of the existing ac power system. With this improvement both fans and igniters would be operable under SBO conditions. The boundary conditions for this analysis include the assumption that the backup ac power supply would always available. The probability that the operators fall to actuate the lilS when required was retained from the base case APET value of 0.01. The hardware unavailability of the fan sptem, was also retained from the base case value (0.001). Only the two station blackout PDSs were included in this evaluation because for the other PDS ac power is already available. B.I. APEP Modifications Only the APETs for PDSs 1 and 2 (SBO.LT and SBO ST) were modiried. The hardware availabilities of the improved ARFS and HIS are assumed to be unchanged from the base APET values of OW) and 1.0, respectively. The first APET question that is modified is Question 13, which asks whether the operators actuate the HIS. In the base case,if there is no ac power at the time of uncovering the top of the fuel (UTAF), the igniters will not be actuated. For the improvement analysis, it was assumed that a backup 30urce of ac power was always available and that the unavailability of the igniters, caused by operators falling to actuate them, remained at the base case value of 0.01. The next question to be modified is Questien 14, which addresses the status of the ARFS at UTAF. In the base case, because ac power is unavailability for the SBO PDSO, the air return fans are not opera ing at UTAF. The probability that the fans can operate if power is recovered during core de;.udation is OW). The probability that the fans have failed and cannot operate upon demand is 0.031 s nardware unavailability of the fans system). For the improvement analysis,it was assumed that a backup source of ac power was always available and that the unavailability of the fans remained at the base case B.3
l value. Therefore,in the improved case, with ac power available to the ARF system, the availability of the fans at UTAF is 0.999. 1 - The next APET question to be modified was Question 28, which asks whether the fans are operational in the early time period (between core uncovering and VB). In the base case, there are four possible cases. First, if the fans were operating at UTAF, then they continue to operate. (This case was not applicable for SDO.) Next,if the fans were failed at the start of the accident then they remained failed. Third, if the fans were available to operate at the start of the accident, and ac power was recovered, then the fans now operate. Finally,if the fans were available to operate at the start of the accident, ard at power was not recovered, then the fans remain availabic to operate when power is restored. In the improvement analysis, with backup ac power available to supply the ARF. system, the last two cases described above are no longer applicable and were removed from the APET. t Question 31 asks whether the air return fans are effective in mixing the containment atmosphere before hydrogen ignition occurs. For the base case, three cases are possible. First,if the lans are inillally operating, then the fans are effective and the hydrogen is uniformly distributed throughout the containment. Second if the fans are available to operate and ac power is recovered in the early period of a station blackout, the probability of the ARF" to mix the containment atmosphere prior to hydrogen ignition was 0S30. Finally, for cases without ac power recovery or in which the fans have failed, the fans will be ineffective. For the improvement analysis, the second case is no longer applicable and was removed. Also, the sampling file was modified to climinate the sampling of Case 2. Question 47 asks whether the hydrogen igniters are operating during core degradation and addresses accidents involved with loss of offsite power und subsequent power recovery. llowever, in the improvement analysis, an ac power supply to the igniters is always available. Therefore, this question was modified to climinate those cases involving ac power recovery which are no longer applicable. Question 49 addresses whether hydrogen ignition occurs in the ice condenser during core degradation (CD). Case 2 of this question was modified to eliminate those sequences involving fan
-recovery that were no longer valid for the improved case.
Question 50 and $1 address whether hydrogen ignition occurs during CD in the upper plenum and upper compartment, respectively. Case 3 of both questions was modified to climinate those sequences involving fan recovery that were no longer valid for the improved case. B.4
l Question 92 asks if the fans are operating late. Case 3 was climinated since it involved only sequences in which the fans operate after ac power is recovered, in the improved case backup power to the fans is always available. The final question modified was Question 100. This question asked if the igniters were operating late. Cases 2 and 3 were climinated. Both of these cases involved station blackout sequences with ac power recovery after VB. For the improved case with backup power supply to the igniters, these two case are no longer valid. 11.2. Accident Progresskm Finding Table B 1 show? the effects of the backup power supply to the fans and igniters on the conditional probabidtics of the accident progression bins for the two SBO PDSs. B5
Tubic 11 1, Conditional probability of accident progression bins at Sequoyah, with backup power to fans and Igniters. Conditional Probabilities PDS1 PDS 2 Accident Progression Bin fing (SDO LT) (SHO CD Cl;a before VD,b carly CF Sensitivity: 1.16E-03 2.30E-03 Basecase: 1.25E 02 1.50E-02 VD, alpha, cally CI' Sensitivity: 7.09E4M 2.55E 03 Basecase: 6.87E4M 2.51E-03 VII,itCSc pressurt > 260 psi, early CF Sensitivity: 4.60E 02 4.64E-02 Basecase: 5.55E 02 6.57E 02 Vil, itCS pressure e 2(O psi, early CF Sen itivity: 1.93E 02 2.98E 02 Base case: 3.24E 02 6.53E-02 Vil,11 2burn, late CF Sensitivity: 5.84E44 1.66E 03 Base em.c: 9.60E 02 1.76E-01 VD, BM'Idor scry late ope Sensitivity: 8.16E 02 1.42E 01 Basecase: 4.53E-02 7.69E 02 Bypass Sensitivity: 1.43E4M 1.86E 03 Basecase: 1.32E 04 1.74E 03 VD, no CF Sensitivity: 2.53E 01 4.05E 01 Base case: 1.58E 01 2.27E-01 No VB, early or no CF Sensitivity: i.73E-01 3.48E 01 Basecase: 5.71E 01 3.46E 01
- a. Containment Vessel.
- b. Vessel Dreach,
- c. Reactor Coolant Sp, tem.
- d. Ilassmat mell through
- c. Ove. pressurization.
B6
'Ihe clicets of this potential modification nri the accident progression bins are as follows. For both PDS 1 und PDS.2, given vessel breach, the probability of estly containment failure (ECF) decreased 34% and <t$'li, respectively, from the base case values. For PDS 1, the percentage of sequences resulting in ECF is 6.7% 'ersus 101% for the bne case. For PDS 2,8.l'1 of the sequences resulted in LCF versus 14.9'1 for the base case. The conditional probabilities were decreased for each of the four APBs that contribute to ECF with the cucption of the alpha mode failure bin which increased 3.2'fi and lu4, for PDS 1 and PDS.2, tespectively. These small increases are due to the increased number of seque,nces that do not fait containment prior to vessel breach.
The probability of late containment failure from late hydrogen burns was decreased more than two orders of magnitude. For PDSs 1 and 2, the conditional probabilities were reduced by factors of tr 4 and 106, respectively. These substantial decreases are the result of the combined effects of the fans and igniters operating throughout the accident. The conditional probabilities of basemat meltthrough (BMT) or very late overpressur9ation (OP) from a build up of steam and noncondensibles increased 80'1 and 85% from base case values of 4.53E 02 and 7.69E 02, for PDSs I and 2, respectively. Of the two very late failure modes (BMT and OP), BMT is the more prevalent failure mode by at least an order of magnitude. For example,in the base case analysis, the conditional probabilities of BMT and OP, for PDSs I and 2, are 4 4E-02 end 4.0E 03, and 7.87E 02 and 1.7E 03, respectively, in the sensitivity analysis, the conditional probability of both fallute modes increased because the early failures are averted and more scenarios reach the point where late failures are poulble. The conditional probabilities of bypass reported in Table B 1 show increases of 8% and 7'1 above the base case values of 1.32Eol and 1.74E 03, for PDSs I and 2, respectively. Ilowever, these inercai.es are artifacts of truncating paths through the accident progression event tree when the path Ircquency is less than 1.0E 05 in fact, the only failure mode contributing to the bypass bin for the station blackout PDSs is temperature induced steam generator tube ruptures (SGTR) which are not affected by the addition of backup power supply to the fans and igniters. The conditional probability of a temperature induced SGTR reported in the ' Branch and Case Frequency' table listed in the EVNTRE output is 1.78E-
')4 and 2.14E 03, for PDSs 1 and 2, respectively. These probabilities remained unchanged in the sensitivity analpis. However, due to the trec modifications in the sensitivity analpis, different paths frequencies developed, so that different paths were truncated than for the base case tree These differences in the truncated paths resulted in different probabilities of bypass for the base case and sensitivity case.
The probability of vessel breach with no containment failure increased Wi and 78'll above the base case values of 1.58E-01 and 2.27E 01, for PDSs 1 end 2, respectively. These increases in no B7
containment failures are due to the fans and igniterri climinating failures from early and late hydrogen burns. The piabability of no vessel breach remains unchanged within the uncertainty of the anrJysis. Table 112 prescrits the accident progression bin probabilitem weighted over the two station blackout PDSs, for the base and sensitivity cases. The results are similar to those discussed above for the individual PDSs. With the addition of backup power to the fans and igniters, early containment failures are seen to decrease by 42% from a base cr.e value of 1,3E.01. Alpha mode failures increase by INii. Late failures caused by hydrogen burns decreased by a fact; r of 115 and very late failures from IIMT or late overpressurintions increased N3%. Dypass failures increased 6F;i. Again this increase in bypass failures is an artifice of the event tree truncations, The probability of no containtnent failure increa<.ed 74% and the probability of no vessel breach remained unchanced. n kh'b
mr - - _ _ _ h o .- Table 112. Comparison of station blackout weighted averages of the accident progression bin probabilities for Sequoyah, with backup power for fans and igniters.
$150 Weighted Ascrage Conditional Probabilities 6sEjfpnt Progtmi<23,1 tin lids Sen'ithily Cl.a tactore YB,D carly CF 1.42E 02 1.92 B413 1.91 E-03 1.94E413 VB, alpha, early CF VB, RCSC > 2(0 psi, early CF 6 23E 02 4.62E 02 VB, RCS < 2(U psi, curly CF 5.44E 02 2.64E 02 YB,112burn, late CF 1.49E 01 130E-03 VB, BMTdor very late OP' 6.65E 02 1.22H 01 1.21E 03 1.29E 0T Bypass VB, no CF 2.04E 01 3.55E 01 No VB, early or no CF 4.21E 01 4.22E 01
- a. Containment Failure.
- b. Vessel Breach,
- c. Reactor Coolant System.
d Basemat mell through.
- c. Overpressurl/ation.
Table B 3 presents the accident progression bin probabilities weighted over all PDSs for the base and sensitivity cases. Because, the backup power supply to the fans and igniters only affeettd the two station blackout PDSs, the effect of the sensitivity is diminished when weighted over all PDSs. Ilowever, the overall trends remain: early coritainment failures are deercased by 219, late failures are decreased by a factor of 35, very late failures from BMT and overpressuritation are increased 7.5'li, and no containment failures are inercased 1491. B9
Table 113. Comparison of weighted average n'cident progression bin probabilities for lequoyah with backup power for fans and Igniteis. Weighted Average Conditional Probabilities liase: Semitivity: Accident Progreulon Din All PDSOs All PDC.Os ) CFs before VB,b carly CF 5.llE 03 2.0SE-03 VD, alpha, early CF 1.69E 03 1.69E 03 Vil, RCSc > 200 psi, early CF 3.61E 02 3.21E.02 VD, RCS < 200 psi, early CF 2.29E 02 1.59E 02 VD,112burn, late CF 3.78E.02 1.09E 03 Vil, DM1dor very late OP' l.87E-01 2.01E 01 Ilypau 4.82E 02 4.82E 02 e Vil, no CF 2.63E-01 3.0lE.01 No Vil, carly or no CF 3.76E-01 3.77E 01
- n. Containment Vessel.
- b. Vessel Breach.
- c. Reactor Coolant System.
- d. Basemat melt.through.
- c. Overpressurization.
B 10
113. Ithk Raults As was done in the base case benchmatting analysh, twu methods were used to estimate the affects of the potential modification on risk. 'Ihe first method used the PARTITION code to estimate the mean risk potentiah in terms of early and latent fatalities. Tlw second h the more detailed and preciw method of using the conditional consequences calculated by the MACCS code. The mean thL potential estimates for both this sensitivity case and the base case are phen in Table 11-4. The rnean risk measures uileulated using the MACCS consequences are shown in Table D.5. The effect of thh potential improvement on the mean risk potentlah, presented in Table B 4, h an 18'l; and 19'?; reduction in the incan early and latent fatality potentiah, respeethcly. The effect on the mean rbk estimates (shown on Table B 5) b a 17'T to 22't reduction in all of the rkk estimates, with the exception of early fatalities, which increased slightly by l'!. This latter result is somewhat surptking since the mean risk potential for early fatalities decreased nmtly 18':i. B 11
Table b 4. Sequoyah mean risk potentials: comparison of base case with the backup power supply to fans and igniters scnsitivity case. Mean Early Mean 1.alent Fatalities Fatalities (per year) {per scar) Base Case 8.25E-05 1.13E-01 Sensitivity 6.77E 05 9.19E-02 Reduction 17.9 % 18.7 % Tabic 115. Sequoyah mean risk measures: comparison of base case with the backup power supply to fans and igniters sensitivity case. Mean Dose Mean Dose Mean Early Mean Latent 50-Mile 1000-Mile Fatalitics Fatalitics (Person. Rem (Person. Rem (neryear) (per year) .per year) per year) Dase Case 1.89E 05 1.51E 02 1.0$E+ 01 8.93E+01 Sensitivity 1.91E-05 1.17E-02 8.69E +00 6.90E+ 01 Reduction s 1.1% 22.5 % 17.2 % 22.7 %
- a. & negative value indicates an increase in the risk measure.
B 12
l l APPENDIX C DATA FILE LISTINGS FOR SEQUOYAH SCET DEVELOPMENT C-l i
APPENDIX C DATA FILE LISTINGS FOR SEQUOYAH SCET DEVELOPMENT
he listings provided in this appendix are of two types. The first $ct (four listings) cotuists of the EVNTRE binning data used to extract simplified containment event trees (SCETS) from the accident progrew.lon event trees (APETs) used in the NUREO 1150 analysis of Sequoyah. The question references in this set refer to APET questions documented in the NUREO/CR 4551 volume for Sequoyah. The r,econd set (also four ilstings) consists of the PSTEVNT input used to reduce the SCET endstates to an approximation of the NUREO 1150 accident progression bins. The question references in this set refer to SCET events, or quest.ons, as defined in the comments of the first four listings.
LISTlHG 1 Sequoyah SCE1 Binning -- PDSG 1, Slow SB 16 l'
', 6 '
'2' '3' '4' '5'
'7' '8' '9' '10'
'll' '12' '13' '14' '15'
'16' 2 2 nob L B-L $ 1. Initial containment leak or 1 1 12 $ isolation f ailure.
2 nob Leak 1 2 12 1 B leak 2 2 nob 1 B1 $ 2. Failure to restore AC power early. I 1 22 1 E-ACP 2 2 22 22 2 +3 EaACP EfACP 2 2 NoCfl Cfl $ 3. Cont, fails pre-VB (H2 burn). 1 1 58 1 EnCF 1 2 58
/1 EnCF 2 2 NoE-1BP E-IBP $ 4. Large ice bypass prior to VB.
2 1 59 59 3 +2 E2nlBP E2-1BP2 ' ( C-3
l 1 2 59 1 E2-IBPI 2 2 NoDP DP $ 5. Failure to depressurize the reactor. ; 1 1 25 i 4 l 1-loPr 1 2 25
/4 1-LoPr 2 2 NoVB VB $ 6. RPV fails (vessel breach).
I 1 26 1 noVD 1 2 26 2 VB 2 2 NolVSE IVSE $ 7. In vessel steam explosion 1 1 64 $ fails the vessel and containment. 2 NoAlpha 1 2 64 1 Alpha 2 2 NoEVSE EVSE $ 8. Ex-vessel steam explosion at VB 2 1 71 71 1 +3 EVSE noEVSE 1 2 71 2 EVSE-CF 2 2 No0PVB OPVB $ 9. OP fails cont, at VB 4 1 64 70 71 B2
/2 + /3 + 2 + 1 NoAlpha NORkt EVSE CF InCF l 4 2 64 70 71 82 2 *3 * /2 * /1 NOAlpha NoRkt EVSE-CF InCF 2 2 NODI Di $ 10. Direct impingement on the 1 1 78 $ seal table wall (and hence 2 $ containment failure)
InCDFlmp 1 2 78 1 I-CFD!mp 2 2 Nol2-!BP 12-1BP $ 11. Large ice bypass after VB. 2 1 83 83 3 +2 12niBP 12-1BP2 ( l 2 83 1 12-IBP1 2 2 NoCCI CCI $ 12. Prompt CCI occurs. 2 1 89 89 4 +5 C4 l
s S0lyCCI nol2CCI % 3 2 89 89 89 1 +2 +3 DryCCI SScrCCl OScrCCI 2 2 nob 2 B2 $ 13. Failure to restore AC power late 'n 1 1 90 y 1 L-ACP 1 2 90
/1 L-ACP 2 2 No0PL OPL $ 14. Late over-pressure (H2 or slow OP) 1 1 103 1
LnCF 1 2 103
/1 LnCF 2 2 nob 3 B3 $ 15. Failure to restore AC power very late 1 1 105 1
L2 ACP 1 2 105
/1 L2-ACP 2 2 NoVLCF VLCF $ 16. Very late failure (OP or BMT) 2 1 107 109 2 *1 noMT L2nCF 2 2 107 109 1 4 /1 BMT L2nCF 1
16 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 SORT ALL BINS --> C5
LISTING 2 Sequoyah SCET Binni g PDSG 3, LOCA 13 'l' '2' '3' '4' '5' oy, ogo ogi ogno
>$o
'll' '12' '13' 2 2 Cl NCI $ 1. Containment isolation failure 1 1 12 2
nob Leak 1 2 12 1 B Leak 2 2 NoECF ECF $ 2. Cont f ails pre-VB (H2 burn). 48 49 50 51 58 5 1 (/l * /1 * /1 * /l) + 1 EnCF 5 2 48 49 50 51 58 (1 +1 +1 + 1) /1 EnCF 2 2 LP HP $ 3. Failure to depressurize the reactor. I 1 25 4 1 LoPr 1 2 25
/4 1-LoPr 2 2 NoVB VB $ 4. RPV fails (vessel breach).
1 1 26 1 noVB 1 2 26 2 VB 2 2 NoE-HR E-HR $ 5. Early containment heat 4 1 27 28 29 30 $ removal failure. 1
- 1 * /1
- 3 E-Sp E-Fan E-Mitl EnlBP 4 2 27 28 29 30
/1 + /1 + 1 + /3 E-SP E-Fan E-Mitl EnlBP 2 2 NolVSE IVSE $ 6.'In-vessel steam explosion 2 1 64 70 $ fails the vessel and containment.
2 * /1 NoAlpha Rkt-CF 2 2 64 70 1 +1 Alpha Rkt-CF 2 2 NoEVSE EVSE $ 7. Ex vessel steam explosion at VB 2 1 71 71 1 +3 EVSE noEVSE 1 2 71 2 C-6
IVSE CF 2 2 NoCf/B CfvB $ 8. OP fails cont. at VB 4 1 58 64 71 82
/1 + /2 + 2 + 1 Entf NoAlpha EVSE-Cf InCF 4 2 58 64 71 82 1 *2 * /2 * /1 Entf NoAlpha EVSE Cf InCF 2 2 NODI DI $ 9. Direct impingement on the 1 1 78 $ seal table wall (and hence 2 $ containment failure)
InCDfimp 1 2 78 1 1 CfDimp 2 2 NoDNC DNC $ 10. Coolable debris bed not formed 1 1 88 1 1-CDB 1 2 88 2 InCDB 2 2 NoL HR L-HR $ 11. Late and V late cont, heat removal f ailure 3 1 91 92 106 1 *1 *1 L-Sp L fan L2 Sn 3 2 91 92 106
/1 4 /1 4 /1 L-Sp L fan L2-Sp 2 2 NoLCf LCf $ 12. Late over pressure (H2 or slow OP) 1 1 103 1
l.nt f 1 2 103
/1 LnCF 2 2 NoVLCf VLCf $ 13. Very late failure (OP or BMT) 2 1 107 109 2 *1 noMT L2nCF 2 2 107 109 1 + /1 BMT L2nCf 1
13 1 2 3 4 5 6 7 8 9 10 11 12 13 SORT ALL BINS ==> C-7
LISTING 3 Sequoyah SCET Binning PDSG 5. Transient , , 10 'l' '2' '3' '4' '5'
'6' '7' '8' '9' '10' 2 2 Cl NCI $ 1. No containment isolation.
I 1 12 2 nob Leak 1 2 12 1 B Leak 2 2 NoSGTR SGTR $ 2. Temperature induced SGTR. I 1 20 2 noE SGTR 1 2 20 1 E SGTR 2 2 HP LP $ 3. PCS fails before VB. I 1 25 $ (low pressure)
/4 IntoPr 1 2 25 4
l-LoPr 2 2 NOVB VB $ 4. RPV fails (vessel breach). I 1 26 1 noVB 1 2 26 2 VB 2 2 NoEVSE EVSE $ 5. Ex vessel steam explosion at VB 2 1 71 71 1 +3 EVSE noEVSE 1 2 71 2 EVSE CF 2 2 No0PVB OPVB $ 6. OP fails cont, at VB 3 1 58 71 82
/1 + 2 + 1 EnCF EVSE CF InCF 3 2 58 71 82 1 * /2 * /1 EnCF EVSE CF InCF 2 2 nod 1 DI $ 7. Direct impingement on the 1 1 78 $ seal table wall (and hence 2 $ containment failure)
InCDFimp 1 2 78 1 1-CFDimp C-8
2 2 NoDNC DNC $ 8. Coolable debris bed not formed 1 1 88 1 1 008 1 2 88 2 InCOB 2 2 Nol llR L HR $ 9. Late and V late cont, heat removal failure 3 1 91 92 106 $ (4 to 6 hours) 1 *1 *1 l Sp L fan L2 Sp 3 2 91 92 106
/1 + /1 + /1 .
L-Sp L-fan L2 Sp 2 2 NoVLCf VLCf $ 10. Very late failure (OP or BMT) 2 1 107 109 2 *1 noMT L2ntf 2 2 107 109 1 4 /1 BMT L2nCf 1 10 1 2 3 8 5 6 7 8 9 10 SORT ALL BINS ==> C9
LISTING 4 Sequoyah SCET Binning -- POSG-6, ATWS 13 'l' '2' '3' '4' '5'
'6' '7' '8' '9' '10'
'll' '12' '13' 2 2 NoSGTR SGTR $ 1. SGTR occurs before VB 1 1 1
/5 B-SGTR 1 2 1 5
B SGTR 2 2 NoCl Cl 5 2. Containment isolation failure 1 1 12 2 nob Leak 1 2 12 1 B-Leak 2 2 Notf1 CFl $ 3. Cont, f alls pre VB (H2 burn). 48 49 50 51 58 5 1 (/l * /1 * /1 * /l) + 1 EnCF 48 49 50 51 58 5 2 * (1 +1 41 + 1) /1 EnCF 2 2 LP HP $ 4. Failure to depressurize the reactor. I 1 25 4 ' 1-LoPr 1 2 25
/4 1-LoPr 2 2 NoVB VB $ 5. RPV fails (vessel breach).
I 1 26 1 noVB 1 2 26 2 VB 2 2 HolVSE IVSE $ 6. In vessel steam explosion 2 1 64 70- $ fails the vessel and containment. 2 * /1 NoAlpha Rkt-Cf 2 2 64 70 1 +1 Alpha Rkt-Cf 2 2 NoEVSE EVSE $ 7. Ex-vessel steam explosion at VB 2 1 71 71 1 +3 EVSE noEVSE 1 2 71 2 C-10
i EVSE-CF l 2 2 No0PVB OPVB $ 8. OP fails cont, at VB ! 4 1 58 64 71 82
/1 + /2 + 2 + 1 EnCF NoAlpha EVSE CF :nCF 4 2 58 64 71 82 1 *2 * /2 * /1 EnCF NoAlpha EVSE CF InCF 2 2 nod 1 D1 $ 9. Direct impingement on the 1 1 78 $ seal table wall (and hence 2 $ containment failure)
InCDFimp 1 2 78 1 1-CFDimp 2 2 NoCD CD $ 10. Coolable debris bed not formed 1 1 88 1 1-CDB 1 2 88 2 InCDB 2 2 Not-HR L-HR $ 11. Late and V late cont, heat removal failure 3 1 91 92 106 $ (4 to 6 hours) 1 *1 *1 L-Sp L-Fan L2-Sp 3 2 91 92 106
/1 + /1 + /1 L-Sp L-Fan L2-Sp 2 2 No0PL OPL $ 12. Late over-pressure (H2 or slow OP) 1 1 103 1
LnCF 1 2 103
/1 LnCF 2 2 NoVLCF VLCF $ 13. Very late failure (OP or BMT) 2 1 107 109 2 *1 noMT L2nCF 2 2 107 109 1 + /1 BMT L2nCF 1
13 1 2 3 4 5 6 7 8 9 10 11 12 13 SORT ALL BINS ==> C-ll
LISTING 5 Sequoyah Source Term Rebinning - PDSG 1 and 2, SB0 14 Cf-Time Sprays CCI RCS Pres VB Mode SGTR Amt CCI Zr-0x HPME CF Size RCS Hole E2-IC 12-1C ARfans 7 7 V-Dry V Wet CF Early CF atVB CF-Late CF-Vlate NoCf 2 1 1 1 $ A. Event V, not scrubbed 1 * /1 V-Dry 2 2 1 1 $ B. Event V, scrubbed 1 * /1 V-Wet 2 3 1 3 $ C. CF during core degradation 2+2 CF Early 4 4 7 8 9 10 $ D. Cf at vessel breach 2+2+2+2 CF atVB 1 5 14 5 E. Late CF 2 CF-Late 1 6 16 $ f. Very late CF 2 CF-Vlate 8 7 1 3 7 8 9 10 14 16 $ G. No containment failure 1*1*1*1*1*1*1*1 NoCf 8 8 Sp-Early Sp-E+1 Sp-E+1+L SpAlways Sp-Late Sp L+VL Sp VL Sp-Non0p 3 1 2 13 15 $ A. Sprays operate only early 1*2*2 Sp-Early 2 2 1 1 $ B. Early and intermediate 1 * /1 Sp E+1 3 3 2 13 15 $ C. Early and late, not very late 1*1*2 Sp-E+1+L 3 4 2 13 15 $ D. At all times 1*1*1 SpAlways 3 5 2 13 15 $ E. Late only 2*1*2 Sp Late 3 6 2 13 15 $ f. Late and very late 2*1*1 Sp-L+VL 3 7 2 13 15 $ G. Very late only 2*2*1 Sp-VL 3 8 2 13 15 $ H. Never 2*2*2 C-12 l
-_ . . . ~ _ _ . _ _
Sp-Never or Sp final ' I 4 6 6 Prmt Dry Prmt-Sh1 No CCI Prmt-Dp SD1y Dry LD1y Dry 4 1 2 5 6 12 $ A. CCI is dry and starts inmediately 2*2*2*2 l 4 Prmt Dry 4 2 2 5 6 12 $ B. CCI occurs under 5 ft of water 2* 1*2*2 Prmt-Sh1 3 3 6 6 12 $ C. CCI does not occur 1 +(2
- 1)
No CCI 2 4 6 12 $ D. CCI occurs under 10 ft of water 2*2 Prmt-Dp 2 5 1 1 $ E. CCI occurs after a delay,
- 1 * /1 $ cooling water not replenished SDly Dry 2 6 1 1 5 F. CCI occurs after a long delay 1 * /1 LD1y Dry 4 4 SSPr HiPr ImPr LoPr 2 1 1 1 $ A. System setpoint pressure (2500) 1 * /1 i SSPr 1 2 5 $ B. 1000 2000 psia 2
HiPr 2 3 1 1 $ C. 200 1000 psia 1 * /1 ImPr 1 4 5 $ D. < 200 psia ,
~
1 LoPr 6 6 VB HPME VB-Pour VB-BtmHd Alpha Rocket No-VB 3 1 5 6 7 $ A. HPME and DCH 2*2*1 V8 HPME 3 2 5 6 7 $ B, Molten pour at low pressure 1*2*1 V8-Pour 2 3 1 1 $ C. Gross failure of bottom head 1 * /1-VB BtmHd 2 4 6 7 $ D. Alpha mode failure 2*2 Alpha 2 5 1 1 $ E. Upward acceleration of vessel 1 * /1 Rocket 1 6 6 $ F. No vessel breach 1 No VB 3 -3 SGTR SG SRVO No SGTR 2 1 1 1 $ A. SGTR occurs, secondary RVs reclose 1 * /1 C-13
SG1R 2 2 1 1 $ B. SCTR, secondary RVs stuck open 1 * /1 SG-SRVO 2 3 1 1 $ C. SGTR does not occur 1 + /1 No SGTR 4 4 Hi CCI Med CCI Lo CCI No CCI 1 1 12 $ A. 001 involves 70 100% 2 Hi CCI 2 2 1 1 $ B. CCI involves 30-70% 1 * /1 Med CCl 2 3 1 1 $ C. 001 involves 0 30% 1 * /1 Lo CCI 1 4 12 $ 0. No CCI occurs 1 No CCI 2 2 Lo Zr0x Hi Zr0x 1 1 5 $ A. 0 40% of core Zr was oxidized I lo Zr0x 1 2 5 $ B. > 40% of core Zr was oxidized 2 Hi Zr0x 4 4 Hi HPME Md HPME Lo HPME No HPME 3 1 5 6 7 $ A. Pct core ejected > 40% 2*2*1 Hi HPME 2 2 1 1 $ B. Pct core ejected 20 40% 1 * /1 Md HPME 2 3 1 1 $ C. Pct core ejected < 20% 1 * /1 Lc HPMC 3 4 5 6 7 $ 0. No HPME 1+1+2 No HPME 4 4 Cat-Rpt Rupture Leak No-Cf 3 1 3 9 14 5 A. Gross structural failure 2+ 2 + 2 Cat Rpt 2 2 7 8 $ B. Hole > 7 ft2 2+2 Rupture 3 3 10 1 16 $ C. Hole is about 0.1 f t2 2+2+2 Leak or BMT 8 4 1 3 7 8 9 10 14 16 $ D. No containment failure 1*1*1*1*1*1*1*1 Bypass or No Cf 2 2 1-Hole 2 Holes 1 1 7 $ A. One hole - no natural cire. C-14
1 1 Hole 1 2 7 $ B. Two holes - natural circulation 2 l 2 Holes 3 3 E2 inByP E2-lpByP E2 1ByP 1 1 4 $ A. No ice condenser bypass 1 E2 inByP 2 2 1 1 $ B. 10% ice bypass 1 * /1 E2 lpByP 1 3 4 $ C. Total ice bypass 2 E2-1ByP 3 3 12 inByP 12 lpByP 12-lByP 1 1 11 $ A. No ice condenser bypass 1 12 InByP 2 2 1 1 $ B. 10% ice bypass 1 * /1 12-IpByP 1 3 11 $ C. Total ice bypass (or melted) 2 12 1By? 4 4 ARF-Erly ARF-E+L ARF Late lio- ARF 2 1 2 13 $ A. Air return fans early only 1* 2 ARF-Erly 2 2 2 13 $ B. early and late 1* 1 ARF E+L 2 3 2 13 $ C. late only , 2* 1 ARF-Late 2 4 2 13 $ D. No air return fans t*2 No-ARF C-15
LISTING 6 3 Sequoyah Source Term Binning - POSG-3, LOCA 14 CF-Time Sprays CCI RCS Pres VB Mode SGTR Amt-CCI Zr-0x HPME CF Size RCS Hole E2-IC ' 12-10 ARFans 7 7 V-Dry V Wet CF Early CF-atVB CF-Late CF-Vlate NoCF 2 1 1 1 $ A. Event V, not scrubbed 1 * /1 V Dry 2 2 1 1 $ B. Event V, scrubbed 1 * /1 V-Wet 2 3 1 2 $ C. CF during core degradation 2+ 2 CF Early 4 4 6 7 8 9 $ D. CF at vessel breach 2+2+2+2 CF-atVB 1 5 12 $ E. Late CF 2 CF Late 1 6 13 $ F. Very late CF 2 CF-Vl ate 8 7 1 2 6 7 8 9 12 13 $ G. No containment failure 1*1*1*1*1*1*1*1 NoCF 8 8 Sp Early Sp E+1 Sp E+1+L SpAlways Sp Late Sp LtVL Sp-VL Sp-Non0p 2 1 5 11 $ A. Sprays operate only early 1* 2 Sp Early 2 2 1 1 $ B. Early and intermediate 1 * /1 , Sp E+1 2 3 1 1 $ C. Early and late, not very late 1 * /1 Sp-E+I+L 2 4 5 11 $ D. At all times 1* 1 SpAlways 2 5 1 1 $ E. Late only 1 * /1 Sp-late 2 6 5 11 $ F. Late and very late 2* 1 Sp t+VL 2 7 1 1 $ G. Very late only 1 * /1 Sp-VL 2 8 5 11 $ H. Never 2* 2 C-16
Sp Never or Sp-final 6 6 Prmt-Dry Prmt-Shi No CCI Prmt-Dp SDly Dry LDly Dry 2 1 1 1 $ A. CCI is dry and starts immediately 1 * /1 Prmt Dry 2 2 1 1 $ B. CCI occurs under 5 ft of water 1 * /1 Prmt-Shi 1 3 10 $ C. CCI does not occur 1 No CCI 1 4 10 $ D. CCI occurs under 10 ft of water ' 2 Frmt Dp 2 5 1 1 $ E. CCI occurs af ter a delay, 1 * /1 $ cooling water not replenished SD1y Dry 2 6 1 1 $ F. CCI occurs af ter a long delay 1 * /1 LDly Dry 4 4 SSPr HiPr ImPr LoPr 2 1 1 1 $ A. System setpoint pressure (2500) 1 * /1 SSPr 1 2 3 $ B. 1000-2000 psia 2 HiPr 2 3 1 1 $ C. 200-1000 psia 1 * /1 ImPr 1 4 3 $ D. < 200 psia 1 LoPr 6 6 VB-HPME VB-Pour VB BtmHd Alpha Rocket No VB 3 1 3 4 6 $ A. HPME and DCH 2*2*1 VB HPME 3 2 3 4 6 $ B. Molten pour at low pressure 1*2*1 VB-Pour 2 3 1 1 $ C. Gross failure of bottom head 1 * /1 VB 8trnHd 2 4 4 6 $ D. Alpha mode failure 2*2 Alpha 2 5 1 1 $ E. Upward acceleration of vessel 1 * /1 Rocket 1 6 4 $ F. No vessel breach 1 No-VB 3 3 SCTR SG-SRVO No-SGTR 2 1 1 1 $ A. SGTR occurs, secondary RVs reclose 1 * /1 C-17
SGTR 2 2 1 1 $ B. SGTR, secondary RVs stuck open 1 * /1 SG SRVO 2 3 1 1 $ C. SGTR does not occur i- 1 + /1 No-$GTR 4 4 Hi-CCI Med CCI Lo CCI No CCI 1 1 10 $ A. CCI involves 70-100% 2 Hi-CCI 2 2 1 1 $ B. CCI involves 30-70% 1 * /1 Med CCI 2 3 1 1 $ C. CCI involves 0-30% 1 * /1 Lo-CCI 1 4 10 $ 0. No CCI occurs 1 No CCI 2 2 Lo-Zr0x Hi-Zr0x 2 1 1 1 $ A. 0 40% of core Zr was oxidized 1 + /i Lo Zr0x 2 2 1 1 $ 8. > 40% of core Zr was oxidized 1 * /1 Hi-Zr0x 4 4 Hi HPME Md-HPME Lo HPME No HPME 2 1 3 4 $ A. Pct core ejected > 40% 2*2 Hi HPME 2 2 1 l $ 8. Pct core ejected 20 40% 1 * /1 Md HPME 2 3 1 1 $ C. Pct core ejected < 20% 1 * /1 l Lo-HPME 2 4 3 4 $ D. No HPME 1+1 No-HPME 4 4 Cat Rpt Rupture Leak No CF 5 1 2 6 7 8 12 $ A. Gross structural failure 2+2+2+2+2 Cat Rpt 1 2 13 $ B. Hole > 7 f t2 2 Rupture 2 3. 1 9 $ C. Hole is about 0.1 f t2 2+2 Leak or BMT 8 4 1 2 6 7 8 9 12 13 $ D. No containment failure 1*1*1*1*1*1*1*1 Bypass or No-CF 2 2 1-Hole 2-Holes 1 1 6 $ A. One hole - no natural cire. C 18
l 1 l l-Hole 1 2 6 $ B. Two holes natural circulation 2 2-Holes 3 3 E2-InByP E2 lpByP E2-1ByP 2 1 1 1 $ A. No ice condenser bypass 1 + /1 E2 InByP 2 2 1 1 $ B. 10% ice bypass 1 * /1 E2-lpByP 2 3 1 1 $ C. Total ice bypass 1 * /1 E2-1ByP 3 3 12 InByP 12 lpByP 12 IByP 2 1 1 1 $ A. No ice condenser bypass 1 + /1 12-InByP 2 2 1 1 $ B. 10% ice bypass 1 * /1 12-lpByP 2 3 1 1 $ C. Total ice bypass (or melted) 1 * /1 12-1ByP 4 4 ARF Erly ARF-E+L ARF Late No ARF 2 1 2 4 $ A. Air return fans early only 1*2 ARF Erly 2 2 2 4 5 B. early and late 1*1 ARF E+L 2 3 1 1 $ C. late only 1 * /1 ARF Late 1 4 2 $ 0. No air return fans 2 No ARF C-19
LISTING 7 i Sequoyah Source Term Binning -- PDSG 5, Transient 14 CF Time Sprays CCl RCS Pres VB Mode SG1R Amt CCI Zr Ox HPME CF-Size RCS-Hole E2-IC 12 10 ARfans 7 7 V Dry V Wet CF Early CF-atVB CF-Late CF Vlate NoCf 2 1 1 1 $ A. Event V, not scrubbed 1 * /1 V Dry 2 2 1 1 $ B. Event V, scrubbed 1 * /1 V Wet 1 3 1 $ C. CF during core degradation 2 CF Early 3 4 5 6 7 5 O. CF at vessel breach 2+2+2 CF atVB 2 5 1 1 $ E. Late CF 1 * /1 CF Late 1 6 10 $ f. Very late CF 2 CF-Vlate 5 7 1 5 6 7 10 $ G. No containment failure 1*1*1*1*1 NoCf 8 8 Sp-Early Sp E+1 Sp-E+1+L SpAlways Sp-Late Sp L+VL Sp VL Sp Non0p 1 1 9 $ A. Sprays operate only early 2 Sp Early 2 2 1 1 $ B. Early and intermediate 1 * /1 Sp-E+1 2 3 1 1 $ C. Early and late, not very late 1 * /1 Sp E+1+L 1 4 9 $ D. At all tiri,es 1 SpAlways 2 5 1 1 $ E. Only late 1 * /1 Sp Late 2 6 1 1 $ f. Only late and very late 1 * /1 Sp L+VL 2 7 1 1 $ G. Only very late 1 * /1 Sp-VL 2 8 1 1 $ H. Never 1 * /1 C 20
Sp Never or Sp final 6 6 Prmt Dry Prmt Shi Ho CCI Prmt Dp SD1y Dry LD1y Dry 1 2 1 1 1 $ A. CCI is dry and starts immediately 1 * /1 Prmt-Dry 2 2 1 1 $ B. CCI occurs under 5 ft of water 1 * /1 Prmt Shl 1 3 8 $ C. CCI does not occur 1 No-CCI 1 4 8 $ D. CCI occurs under 10 ft of water 2 Prmt Op 2 5 1 1 $ E. CCI occurs after a delay. 1 * /1 $ cooling water not replenished SD1y Dry 2 6 1 1 $ f. CCl occurs after a long delay 1 * /1 LD1y D.*y 4 4 SSPr HiPr ImPr LoPr
) 1 3 $ A. System setpoint pressure (2500) 1 SSPr 2 2 1 1 $ B. 1000 2000 psia 1 * /1 HiPr 2 3 1 1 $ C. 200 1000 psia 1 * /1 ImPr 1 4 3 $ D. < 200 psia 2
LoPr 6 6 V8 HPME VB Pour VB-BtmHd Alpha Rocket No VB r 2 1 3 4 $ A. HPME and DCH 1*2 i VB-HPME 2 2 3 4 $ B. Molten pour at low pressure 2*2 VB Pour 2 3 1 1 $ C. Gross failure of bottom head 1 * /1 VB BtmHd 2 4- 1 1 $ D. Alpha mode failure 1 * /1 Alpha 2 5 1 1 $ E. Upward acceleration of vessel 1 * /1 Rocket 1 6 4 $ f. No vessel breach 1 No VB 3 3 SGTR SG-SRVO No SGTR 1 1. 2 $ A. SG1R occurs, secondary RVs reclose 2 C-21
SGTR 2 2 1 1 $ B. SGTR, secondary RVs stuck open 1 * /1 SG-SRVO 1 3 2 $ C. SGTR does not occur 1 No SGTR 4 4 Hi-CCI Med CCI Lo CCI No CCI 2 1 1 1 $ A. CC1 involves 70-100% 1 * /1 Hi-CCI 2 2 1 1 $ B. CCI involecs 30 70% 1 * /1 Med CCI 1 3 8 $ C. CCI involves 0-30% 2 Lo CCI 1 4 8 $ 0. No CCI occurs 1 No CCI 2 2 Lo Zr0x Hi-Zr0x 2 1 1 1 $ A. 0-40% of core Zr was oxidized 1 * /1 Lo Zr0x 2 2 1 1 $ B. > 40% of core Zr was oxidized 1 + /1 Hi-Zr0x 4 4 Hi HPHE Kd-HPME Lo HPME No HPME 2 1 3 4 5 A. Pct core ejected > 40% 1*2 Hi HPME 2 2 1 1 $ B. Pct core ejected 20 40% 1 * /1 Md-HPME 2 3 1 1 $ C. Pct core ejected < 20% 1 * /1 Lo-HPME 2 4 3 4 $ 0. No HPME 2+1 No-HPME 4 4 Cat-Rpt Rupture Leak No CF 2 1 5 6 $ A. Gr e structural failure 2+2 Cat-Rpt 1 2 10 $ B. Hole > 7 ft2 2 Rupture 2 3 1 7 $ C. Hole is about 0.1 ft2 2+2 Leak or BMT 5 4 1 5 6 7 10 $ D. No containment failure 1*1*1*1*1 Bypass or No CF 2 2 1-Hole 2 Holes 2 1 1 1 $ A. One hole - no natural cire. C 22 1 l l
l 1-+ /1-l-Hole 2 ? 1 1 $ B. Two holes - natural circulation 1 * /1 2 Holes 3 3 E2-InByP E2-lpEyP E2 1ByP 2- 1 1 1 $ A. No ice condenser bypass I + /1 2 2, E2-In9yP' 1 $ B. 10% ice bypass 1 * /1 E2-IpByP 2- 3 1 1 $ C. Total ice bypass (or melted) 1 * /1. E2 IByP 3 3 12-InByP 12-IpByP 12-IByP 1 1 5 $ A. No ice condanser bypass 1 12-InByP 2 2 1 1 $ B. 10% ice bypass
-l * /1 12-lpByP 1 3 5 $ C. Total ice bypass (or molted) 2 12-IByP 4 4 ARF-Erly ARF-F+L ARF Late No-ARF 6 $ A. Air return fans early only 2- 1 5 2+2 ARF-Erly-2 2 5 6 $ B, early and late 1*1 ARF-E+L late only t
2 3 1 1 $ C. 1 * /1 ARF-Late 2 4 1 1 $ D. No air return fans 1 * /1 No-ARF C-23
1 LISTING 8 Sequoyah Source Term Binning -- PDSG-6, ATWS la CF-Time Sprays CCI RCS Pres VB-Mode SGTR Amt-CCI Zr-0x HPME CF-Size RCS-Hole E2-IC 12 .'- ARFans 7 7 V-Dry V-Wet CF-Early CF-atVB CF-Late CF-Vlate NoCF 2 1 1 1 $ A. Event V, not scrubbed 1 * /1 ' V-Dry 2 2 1 1 $ B. Event V, scrubbed 1 * /1 V-Wet 3 2 3 $ C. CF during core degradation 2+2 CF-Early 4 4 6 7 8 9 $ D. CF at vessel breach ks 2+2+2 CF-atVB 1 5 12 $ E. Late CF 2 CF-itte 1 5 13 $ F. Very late CF 2 CF-Vlate 8 7 2 3 6 7 8 9 12 13 $ G. No containment failure 1*1*1*1*1*1*1*1 NoCF 8' 8 Sp-Early Sp-E+1 Sp-E+I+L SpAlways Sp Late Sp-L+VL Sp-VL Sp-Non0p 1 1 11 $ A. Sprays operate only early 2 Sp-Early 1 2 2 1 1- $ B. Early and intermediate 1 * /1 Sp-E+1 2 3 1 1 $ C. Early and late, not very late 1 * /1 Sp E+I+L 1 4 11- $ D. At all times 1 SpAlways 2 5 1 1 $ E. Only late 1 * /1 Sp late 2 6 1 1 $ F. Only late and very late 1 * /1 Sp L+VL 2 7 1 1 $ G. Only very late 1 * /1 Sp-VL 2' 8 1 1 $ H. Never 1 * /1 C-24 i
Sp-Never or Sp Final l 6 6 Prmt Dry Prmt-Shi No-CCI Prmt Dp SD1y Dry LDly Dry 2 1 1 1 $ A. CCI is dry and starts immediately 1 * /1 Prmt-Dry 2 2 1 1 $ B. CCI occurs under 5 ft of water 1 * /1 Prmt-Shl 1 3 10 $ C. CCI does not occur 1 No-CCI 1 4 10 $ D. CCI occurs under 10 ft of water 2 Prmt-Dp 2 5 1 1 $ E. CCI occurs after a delay, 1 * /1 $ cooling water not replenished SDly Dry 2 6 1 1 $ F. CCI occurs after a long delay 1 * /1 LDly-Dry 4 4 SSPr HiPr ImPr LoPr 2 1 1 1 $ A. System setpoint pressure (2500) 1 * /1 SSPr 1 2 4 $ B. 1000 2000 psia 2 HiPr 2 3 1 1 $ C. 200 1000 psia 1 * /1 ImPr 1 4 4 $ D. < 200 psia 1 LoPr 6 6 VB-HPME VB-Pour VB BtmHd Alpha Rocket No VB 2 1 4 5 $ A. HPME and DCH 2*2 VB-HPME 2 2- 4 5 $ B. Molten pour at low pressure 1*2 VB-Pour 2 3 1 1 $ C. Gross failure of bottom head 1 * /1 VB BtmHd 2 4 1 1 $ D. Alpha mode failure 1 * /1 Alpha 2 5 1 1 $ E. Upward acceleration of vessel 1 * /1 Rocket 1 6 5 $ F. No vessel breach 1 No-VB 3 3 SGTR SG-SRVO No-SGTR 1 1 1 $ A. SGTR occurs, secondary RVs reclose 2 C-25
SGTR-2 2_ 1: 1 $ B. SGTR, secondary RVs stuck open l * /1 SG SRVO 1 3 1 $ C. SGTR does not occur 1 No SGTR 1- 4 Hi-CCI Med-CCI Lo-CCI No CCI 1 1 10 $ A. CCI involves 70100% 2 Hi-CCI 2 2 $ B. 1 1 CCI involves 30 70% 1 * /1 Med-CCI 2 3 1 1 $ C. CCI involves 0-30% 1 * /1 Lo-CCI 1 4 10 $ D. No CCI occurs ; I No- I 2 2 Lo >x Hi-Zr0x ' 2 1 $ A. 1 1 0 f0% of core Zr was oxidized o 1 * /1 Lo Zr0x 2 2 1 1 5 8. > 40% of core Zr was oxidized 1 + /1 Hi Zr0x 4 4 Hi HPME Md-HPME Lo-HPME No HPME 2- 1 4. 5 $ A. Pct core ejected > 40% 2*2 Hi-HPME 2 2 1 1 $ B. Pct core ejected 20-40% 1 * /1 Md-HPME i
'2 3 1 1 $ C. Pct core ejected < 20%
1 * /1 Lo HPME 2 4 4 5_ $ D. No HPME 1+1 No HPME 4 4 Cat-Rpt- Rupture Leak No-CF 3 1 3 8 12 $ A. Gross structural failure 2 +.2 + 2 Cat-Rpt 3 2 6 7 13 $ B. Hole > 7 f t2 2+2+2 Rupture
-2 3 2 9 $ C. Hole is about 0.1 f t2 2 + 2-
- l. -
Leak or BMT
- i. 8 4 2 3 6 7 8 9 12 13 $ D. No containment failure 1*1*1*1*1*1*1*1 Bypass or No-CF 2- 2 1-Hole 2-Holes
-1 1 -6 $ A. One hole - no natural circ.
C-26 l l:
1 1 Hole natural circulation i 2 6 $ B. Two holes 1 2 2 Holes 3 3 E2-InByP E2 lpByP E2-IByP 1 $ A. No ice condenser bypass 2 1 1 1 + /1 E2-inByP 1 $ B. 10% ice bypass 2 2 1 1 * /1 E2-lpByP 1 1 $ C. Total ice bypass (or melted) 2 3 1 * /1 E2-1ByP 3 3 12 InByP 12 lpByP 12 IByP 1 $ A. No ice condenser bypass ' 1 2 1 1 1 + /1 12-inByP 1 $ B. 10% ice bypass 2 2 1 1 * /1 12-lpByP 1 $ C. Total ice bypass (or melted) 2 3 1 1 * /1 12-1ByP 4 ARF Erly ARF-E+L ARF-L.'e No ARF 4 3 i A. Air return fans early only 1 1 2 ARF-Erly carly and late 1 2 3 $ B. 1 ARF E+L late only 2 3 1 1 $ C. 1 * /1 ARF-Late 2 4 1 1 $ D. No air return fans 1 * /1 No-ARF C-27
?$
/
APPENDIX D CATALYTIC HYDROGEN IGNITERS 4 D-1
APPENDIX D CATALYTIC HYDROGEN IGNITERS Development work on catalytic igniters has progressed in both the United Stv%s m.d the Federal Republic of Germany over the last couple of years.N,42 In the U.S., the prime developer has been Sandia National Laboratories in Livermore, CA. They have dewloped a prototype catalytic igniter that lequires no electric power and can Ignite mixtures as lean as 5.5% hydrogen. The igniter is susceptible to failure from high gas velocitics, water spray, sicam and iodine-containing compounds, however shiciding and semi permeable coatings could overcome these problems. (Recent, yet to be published work, has succeeded in developing a fin type, wet proofed ljniter.) Catalytic igniter development has also been carried out by ) Siemens/Kraftwerk Union (KWU). However, the KWU igniter is a fully engineered device with an enclosure for the igniter element, which has been tested and qualified for use in German reactors. KWU has also developed a battery powered spark igniter for use in hydrogen control applications. This spark igniter has an adjustat.3 actuation temperature and pressure and a discharge life for the batteries of at least five dap at 130 C, l [ D3
REFERENCES D 1. L R. Thorne, J. v. Volponi, and W. J. McLean, Platiistin Catalytic /gniters for Lean flydrogen Air hiittures, NUREGICR 50%, September 19M. D.2. R. Hccht,'Cata\ytic ignitet. Spark igniter l ptnenteil at the Catalytic Igniter Evaluation hiteting sp<ntsored by the Electric Power Research hustitute, Palo Alto, Califontia, Novensber 21,1988. i I D4
- w. - - - . - . - - . - , - , - - - - - - - - - - - _ - - _ _ - - , - - _ _ - - , - - _ - . _ . , - - - - - . _ - - - - _ , - - -
APPEND,X E FRONT END ISSUES ANALYSIS E-l
APPENDIX E FRONT END ISSUES ANALYSIS ECR Risk Sensitivity Sensitivity analyses were peiformed to estiniate the effect on both core melt frequency and risk of two approaches for improving the reliability of emergency coolant recirculation (ECR). Pending availability of the updated Sequoyah containment analysis,E4 interim sensitivity calculations were made by combining the updated version of the cure damage analysisE .2 with the original version of the containment and consequence analyses.E-3 Two possible modifications were evaluated as Case Studies and these are discussed as follows. Case 1 1 examined the benefits of improving the performance of the reactor operators (i.e. reducing the human error probhbilities or HEPs) in' emergency core coolant system (ECCS) realignment from hic,h pressure injection (HPI) to high pressure recirculation (HPR). The human errors that were identified in Reference E.3, as important to the core damage frequency are ilsted in Table E.1 Case.2 analyzes the benefits of the capability of prolonging HPl during small LOCA1 by refilling the refueling water storage tank (RWST). This would eliminate the need to switch to HPR (Note this was not considered effective for the interfacing system LOCA sequences, see following section). Tables E 2 and E 3 display the results of these two cases respectively, for different levels of improvement. Possible ECR Imnrovements It has been proposed that refilling the RWST would improve the severe a;cident performance of Ice Condenser type of plants by extending the amount of time available to the plant operators for carrying out required emergency proceduas. Although the basic idea has merit, a number of influencing factors must be accounted for to actually produce a algnifictml benefit in terms of risk reduction. This proposed improvement would benefit those accident sequencerthat involve a loss of. coolant accident (LOCA), with successful emergency coolant injection (ECl) but with a subsequent failure of emergency coolant recirculation (ECR).
' These types of sequences were found to be significant for both the Sequoyah core melt frequency and its risk.
Also, it is important to note that this class of sequences might include the interfacing system LOCA (ISLOCA). In the ISLOCA sequence, the lost coolant bypasses containment and therefore results in a failure to fill the containment sump, which is-the source of water for ECR However, for this option to be effective E3
for the ISLOCA sequences, a virtual inexhaustible supply of water must be provided, and the consequences of the ISLOCA Gooding areas of the plant must be accounted for. One aspect of refilling the RWST centers on the tale at which makeup can be provided. The major motivation behind this actions originates from the relatively rapid rate at which the RWST drains during a postulated small LOCA. The specific accident scenario of interest involves a LOCA of such a size that the RCS does not depressuri/e, but does result in the containment pressure setpoint (typically about 3 psig) being reached that actuates the containment spray splem (CSS) The large now rate of the CSS pumps (about 4000 ppm for each of two pumps) results in the rapid depletion of the RWST to the low low level at which time the operators are required to realign the safety injection system for high pressure recirculation (this time has been estimated in support of the NUREG ll50 effort at about 30 minutes, see section 4.8.4.3 of NUREG/CR-4550, Rev.1, Vol.5). The low-pressure portions of the safety injection system realign to recirculation automatically, however manual actions are needed to align the high pressure pump suction to the low pressure system discharge (i.e., the normal alignment for HPR). Normal makeup for the RWST, is provided via the CVCS and is capable of supplying approximately 140 ppm of borated water Therefore, for this strategy to be effective, either the CSS flow needs to be severely limited or an alternate means of makeup is required. I in the FSAR LOCA analyses, four of the five ice condenser plants (D. C Cook being the exception), rely upon the CSS for containment pressure suppression only, v.hile the ice condenser system provides the containment atmosphere radionuclide removal function, it therefore annears feasible that for these four plants, the CSS operation could be delayed without major complications to the DBA analpis. However, detailed containment calculations would be needed to ensure that the DBA containment pressure remains below the design pressure (12 to 15 psig). These containment calculations would provide a re-estimate of the peak containment pressure reached during a severe accident, which would likely exceed the original design basis accident estimated peak pressure of 10.8 psig (note that the actual failure pressures for three of the five ice condenser containments have already been estimated at from 60 to 120 psig). Another alternative is to lower the containment pressure setpoint for actuation of the containment air recirculation (CAR) fans, which circulate the containment atmosphere from the lower compartment through the ice chest to the upper compartment and then back to the lower compartment. Currently the fans actuate at the same time the sprays do, at a bl.hi c(mtainment pressure (typically 3 psig). If instead, the CAR fans were started on the Safety injection (SI) signal, at a hi containment pressure of 1 psig,it might extend the time it takes for the containment pressure to reach the 3 psig setpoint and CS actuation. This would likely provide additional time (hence improve their reliability) for the operators in responding to the accident, including performing the realignment to recirculation. However, detailed containment modeling calculations would need to be done to accurately estimate the effect of this modification. E-4
If none of the above options are practical, the last alternative is to install a new capability for refilling l the RWST at very high flow rates (at least 4tKX) gpm). Further, the makeup should be borated water to ensure reactor suberiticality. One possible source for providing RWST makeup could be the fire protection system, modified to filter and borate the wt,;er before being added to the RWST. This option has the advantage of: (a) large flow rates,(b) a virtually unlimited supply of water (although a means of borating the water would need to be added), and (c) the pumps are already in place, are (for Sequoyah at least) powered from a emergency sources, and 5:ismically qualified. This modification also has the benefit of potentially being effective during ISLOCA sequences, which for Sequoyah are the dominant contributors to plant risk. llowever, before estimating the benefit with regards to the ISLOCA, the effects of possib!c flooding, as a result of the ISLOCA, on required equipment would need to be analyzed. Table E-1. Iluman Errors Dominating the Failure to Perform the Realignment from IIPI to llPR for Sequoyah. Event ID Description {!ase Case ilEP llPR XilE FO SIMIN Operator fails to close Si mininow lines to RWST 2.85E 3 during S2 LOCA. IIPR-XHE FO-SIMN2 Operator fails to close si mininow lines to RWST 2.51 E-3 during S3O dsequences. HPR XilE FO V8Vil Operator fails to open ilPR flow control valves 63 8 and 6311. 2.05 E-3 E-5
n i
- l Sequoyah Risk Sensitivity Analysis to Improving Operator Perfortnance in Realigning from j Table B-2.
HPl to HPR. (Risk per reactor year.) i I Base Case _ Sensitivity Cases Factor Reduction - > _qo_. 0.1 0.5 0.9 ,1.0 in HEPs Core Melt Freq: 5.90E 0$ 5.62E-05 4.53E 05 3.44E 05 3.17E 05 Early Fatalitics 3.7E 05 3.7E-05 3.6E-05 3.6E 05 3.6E 05 ; Early injuries 8.8E 05 8.8E 05 8.6E 05 8.4E.05 8.3E-05 Latent Cancers 4.4E 02 4.3E 02 4.lE 02 4.0E @ 3.9E 02 Ind Risk of Fat. 1.0E 07 1.0E-07 9.9E 08 - 9.8E 08 9.7E 08 Offsite Costs 1.3E+44 1.2 E +04 1.2E + 04 1.lE + 04 1 1E+04 ($) Pop. Dose 8.7E+0! 8.6E+01 8.l E+ 01 7.6E +01 7.5 E+ 01 - (man rem)
- a. These Sensitidly caluclations represent unspecified improvements in the performance of the human operator accomplishing the realignment from high pressure injection to high pressure recirculation. These improvements could be accomplished through such things as better training or better procedures (although it is unrealistic to expect a reduction factor of 1.0,i.e., reducing the HEP to 0.0,
- b. The factor redu; tion in HEP is utilized in the following manner: new HEP = (X) HEP, wherc X represents 'I the reduction fac'.or listed in the table.
l-l- l. l l E-6 L
Table E-3. Sequoyah Risk Sensitivity Analysis to Refilling the RWST for Continued IIPI Operation (l.c., obviating ilPR failures) (Risk per reactor year.) Ilase Cne Sensitivity Cases Probebility of Successfully > 0.0 01 0.5 09 1.0 Refilling RWST Core Melt Freq 5.90E 05 5.50E 05 3.93E 05 2.36E 05 1.% E 05 Early Fatalities 3.7E 05 3.7E 05 3.5E 05 3.4E 05 3.4 E.05 Early injuries 8.8E-05 8.7 E-05 8.2E 05 7.7E 05 7.6E 05 1.atent Cancers 4.4E 02 4.2E 02 3.8E-02 3.3E-02 3.2E.02 Ind. Risk of Fal. 1.0E 07 1.0E 07 9.6h 08 R2E 08 9.lE 08 Offsite Costs 1.3 E + 04 1.2 E + 04 1. l E + 04 9. l E + 03 8.7F + 03 ($) Pop. Dose 8.7E4 01 8.4 E + 01 7. l E + 01 5.8E + 01 5.5 E+ 01 (manrem)
- a. This sensitivity exarnines the idealistic situation where the RWST can be refilled and a source of injection water maintained indefinitely. The different cases estimate the effect of different levels of reliability for this process. That is,in the base case, no credit is assumed for refilling the RWST, therefore, the probability is shown as 0.0. the sensitivity case represented as 1.0 estimates the effect on risk of a perfectly reliable system for refilling the RWST, which translates into the complete removal of all HPR failures from the sequence cutsets.
\.
E-7
REFERENCES E 1. 3. 3. Gregory et al., Evaluation of Severe Accident Risks: Sequoyah Unit 1, Sandia National Laboratories, NUREO/CR 4551, Volume 5, Revision 1, SAND 861309, Decernber 1990. U.2. R. C. Dertucio and S. R. Brown, Analysis of Core Damage Frequency: Sequoyah Unit 1, Sandia National Laboratories, NUREG/CR 4550, Volume 5, Revision 1, Parts I and 2, SAND 8(32081, June 1989 E 3. A. S. nen} amin et al., Evaluation of Severe Accident Risks and the Potentialfor Rhk Reduction: Sequoyah Power Station, Unit 1, Sandia National Laboratories, NUREG/CR 4551, Volume 2, Draft Report for Comment, SAND 8tr1309, February 1987. E8 l l
gayoav ns u s Nuctran accutaroav couwssioN t ga, ryt,a_ EE BIBLIOGRAPHIC DATA SHEET
~^ ~
is., ,,,,, w, , e, r, ,,,,s,i NUREORR-5602 EOO-2606 L ht Li A40 5V0 fif Lt Simplified Containment Event Tree Analysis for the Sequoyah ice Condenser Containment 3 oati ataoa t ausus,.s o m ., .- December 1990 4 FIN OR GR ANT NUYe(R A6900
% AV1HORis' 6 T v PE of MtPORT I
W. J. Galycan Technical J. A.Schroeder tceecocovssio~,.,. om. D. J. Pafford aSej,o,ngNgeaaNa , A ooN - Nave aNo aoouss au s.,c - .e o . . o,,.a - a - as o.- a e.,, cn... , .- ,,,a u,na,e e.. Idaho National Engineering Laboratory EO&OIdaho,Inc. P.O. Box 1625 Idaho Falls, Idaho 83415 o gogoag04N a A T ion - Nav s ANo acoa t ss <,, oc ,, s- .~ < n-. -,- ~.e o .... o~. - ~, c 4 . ., um, , c -.. Division of Safetyissue Resolution Office of Nuclear Regulatory Research U, S. Nuclear Regulatory Commission Washington, D.C.20555 le sUPPLEVt NT ARY NoTis il, ASSrRACT tw w a e aner An evaluation of a Pressurized Water Reactor (PWR) ice condenser containment was performed. In this evaluation, simplified con-tainment e- at trees (SCETs) were developed that utilized the vast storehouse of information generated by the NRC's Draft NUREO- ..i0 effort. Specifically, the computer programs and data files produced by the NUREG-1150 analysis of Sequoyah were used to electronically generate S CETs,as opposed to the NUREO-i l50 accident progression cvent trecs (APETs).Thb sim plification was performed to allow graphic depiction of the SCETs in typical event tree format, which facilitates their understanding and use. SCETs were developed for five of the seven plant damage state groups (PDSOs) identified by the NUREO-Il50 analyses, which includes: both short- and long-term station blachut sequences (S BOs), transients, loss-of-coolant accidents (LOCAs), and antici. patcJtransientwithoutscram(ATWS),Steamgeneratortuberupture(SOTR)andevent-VPDSOswerenotanalyzedbecauseof their. containment bypass nature. After being benchmarked with the APETs,in terms of containment failure mode and risk, the SCETs were
- used to evaluate a number of potential containment modifications. The modifications were examined for their potential to mitigate or prevent containment failuro from hydrogen burns or direct impingement on the centainment by tne core, (both factors identified as significant contributors to risk in the NUREO-1150 Sequoyah analysis). However, kcause of the relatively low baseline risk pos-tulated for Sequoyah (i.e.,12 person-rems per reactor year), none of the potential modifications appear to be cost effective, u x t v ovnussot sc8 m ua s mn ,. . ,m, u., u m.u , ~ . w-, u.. ,..~, , u ava.sw r , ummi e imi:_nmt s4 $(GTT V 7.'A M i ca l .v,,
SCET ,,,,,,,,,,, NUREO-il50 PDSO ,,,,ygj^d Sequoyah Iinchesified l*> NUV8kNOFPAuts
' 6 P R h.,6 MC PORY 3J's 12 49:.
l l i l t l THIS DOCUMENT WAS PRINTED USING RECYCLED PAPER.
UNITED STATES 5"jpl4?,"l"ll3^5MA" NUCLEAR REGULATORY COMMISSION uwe WASHINGTON, D.C. 20555 n.uit % o o 0FFICIAL BUS! NESS i PENALTY FOR PRIVAYE USE. 4300 05 k_ h [n'dl I 14NIRCl411\Al
?H ?
v m !! ggguc Ancys sus WASHINcing DC 20555
.}}