ML20070U270
| ML20070U270 | |
| Person / Time | |
|---|---|
| Site: | Shoreham File:Long Island Lighting Company icon.png |
| Issue date: | 03/31/1991 |
| From: | Galyean W, Kelly D EG&G IDAHO, INC. |
| To: | NRC OFFICE OF NUCLEAR REGULATORY RESEARCH (RES) |
| References | |
| CON-FIN-A-6898 EGG-2632, NUREG-CR-5654, NUDOCS 9104080287 | |
| Download: ML20070U270 (208) | |
Text
{{#Wiki_filter:- _ _ -- - NUREG/CR-5654 EGG-2632 =_ - Con':ainmerr: Ven':ing Ana:ysis for the Shoreham Nuclear Power S:ation Prepared by W. J. Galycan, D.1. Kelly Idaho National Engineering I.aboratory EG&G htaho, Inc. Prepared for U.S. Nuclear Regulatory Commission P sP 9 888 M 888&aa F'DR
_~ ~. . . . _ . - - --- - -. .. .. . - - - _ ~ k AVAILADittTY NOTICE Availatxtrty of Refererce Matenals Cited in NRC Pubicatons Most documents cited in NRC publications wS! be avaltable from one of the following sources-
- 1. The NRC Pubhc Document Room,2120 L Street, NW, Lower Level, Washington, DC 20555
- 2. The Superintendent of Documents, U.S. Government Prhting Office. P.O. Don 37082, Washington, DC 20013 7082
- 3. The National Technical Information Service, Sprbgfield. VA 22161 A'though the Esting that fonows represents the majority of documents etted h NRC pubitcations, it is not intended to be exhaustive. ,
Referenced documents avanable for inspection and copying for a fee from the NRC PubHc Docurnent Room hclude NRC correspnndence and intemal NRC memoranda; NRC Office of Inspection and Enforcement butletins, ciredars, Information notices, hspection and investigation notices: Ucensee Event Reports; ven-dor reports and correspondence; Commission papers; and appbcant and imensee documents ar;d corre. spondence, The fortowing documents in the NUREG series are available for purchase from the OPO Sales Program: formal NRC staff and contractor reports, NRC-sponsored conference proceedings, and NRC booklets and brochures. Also avaAable are Regulatory Outdes NRC regu,ations in the Code of Federal Reputations, and Nuclear Regulatory Commission Issuances. Documents available from the National Technical information Service hclude NUREO series reports and technical reports prepared by other federal agencies and reports prepared by the Atomic Energy Commis-skin, forerunner agency to the Nuclear Regulatory Comrnission, Documents ava!!able from pubbe and special technical libraries include aD open Sterature items, such as books, journal and penodical articles, and transactions, Federal Register notices, f ederal and state legista-tion, and congressional reports can usually be obtained from these Ebraries. Documents such as theses, dissertations, foreign reports and translations, and non-NRC conference pro-ceedangs are avahable for purchase from the organ!28tlon sponsoring the pubhcation rMo. Single copies of NRC draft reports are avaRable free, to the extent of supply, upon wittten request to the Office of informat5n Resources Msnagement, Drstribution Section. U.S. huclear Regulatory Comnussion, Washington. DC 20555. Copies of indusuy codes and standards usect in a substanttve manner in the NRC regulatory process are maintained at the NRC Library,7920 Norfolk Avenue, Becesda, Maryland. and are available there for refer. ence use by the pubfic. Codes and standards are usually copyrighted and may be purchased from the originating organitation or, if they are American National Standards, from ti,e American National Standards Institute,1430 Droadway, New York, NY 10018. DISCLAIMER NOTICE This report was prepared as an account of work sponsored by an agency of the United States Govemment. Neither the United States Govemment nor any agency thereof, orany of their employees, 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 en this report, or represents that its use by such third party would not infringe private!y owned rights.
NUREG/CR-5654 EGG-2632 IS, R2, XA Containment Venting Analysis for the Shorenam Nuc. ear Power Station Manuscript Completed: January 1991 Date Published: March 1991 Prepared by W. J. Galycan, D. L Kelly Idaho National Engineering laboratory Managed by the U.S. Department of Energy EG&G Idaho, Inc. Idaho Falls,ID 83415 b Prepared for Division of Systems Research Office of Nuclear Regulatery Research U.S. Nuclear Regulatory Commission Washington, DC 20555 NRC FIN A6898 Under DOE Contract No. DE-AC07-761D01570 1
ABSTRACT An evaluation of the Shoreham Mark Il containment was performed to identify the ef fects of containment venting on core melt frequency, containment failure mode, and offsite consequences. The analysis was based on the Long Island Lighting Company's updated 1988 probabilistic risk assessment of the Shoreham plant with the proposed Supplemental Containment System (SCS). The SCS is a filtered containment vent system based on the Swedish Filtra system installed at the Barseback Nuclear Power Station in southem Sweden.The following three dif-ferent containment venting strategies were examined for their effects on plant risk:
- Venting using the proposed Filtra system
- Venting using the existing equipment at Shoreharn
- Nc venting.
In addition, the consequences of containment venting were examined in con-junction with two sets of assumptions about the effects of a harsh reactor building environment, produced by containment failure or venting through the existing con-tainment and reactor building heating, ventilating, and air conditioning systems, on the equipment located there. Specifically, the analyses studied the consequences when a harsh reactor building environment is assumed to have either no adverse effect on equipment or to fail all equipment. FIN No. A6898-Analysis for Generic issue 29 and NRR Support iii
CONTENTS A B STR A CT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ... iii FOREWORD . . .. . .... .. . ...... ................ .. ... .... .... ix ACRO' 'MS AND ABBREVIATIONS . . . . . . ........... ... ... . .. ... xi EXECUTIVE
SUMMARY
.. ... . . . . .............. .......... . ..... , 1 BACKGROUND... ... ....... . . . .... .. .. ... ... . . .... .. 3 INTRODUCTION . . . . . . . . . . . . . . . . . ... .. . .. .. ....... . .. .. .... .... 4 ANALYSIS OF VENTING STRATEGIES .. . . ..... ................ . . .. 5 Filtra . . ..... .... ........ . .... ...... . , . ..... ...... 5 Existing Vent . . . .. . .. ..... . . . .. .. . .......... ..... . . 7 EFECI'S OF VENTING ON CORE MELT FREQUENCY , . . . . . . . . , , .. ... 9 Simplified Event Tree Approach . . . ... .. .. ............ . ..... .... 9 Assumptions ..., .. . .. ....... . .... . ..... . .. .. ........ 9 Dominant Accident Sequences for Shoreham . ..... ,. . .. . .. ........ 9 Results . .... . .... . . ... ...... ... ........... .... ... ...... 9 C,0NTAINMENT RESPONSE ANALYSIS . . . . . . . . . . . . . . . . . . . . .. ...... .... ... 22 Simplified Containment Event Tree Approach . . . . . . .. . .. ....., ..... 22 Assumptions .. . .. . . ..... ...... ... ........ ... .. ........ 22 l
Comparison of SCET and PRA Results .. ... . .. ..... ... .... ......... . 24 Effects of Venting on the Containment Release Mode . . .. ............. ....... 32 l EFFECTS OF CONTAINMENT VENTING ON RISK . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41 CONCLUSIONS . . . . . . .............. ... .. . ... ... . .. .... .... .. 44 REFERENCES . . . . . . . . ... . .................... ..................... . .. 46 l l APPENDIX A-QUANTIFICATION OFTi(E SilOREllAM SIMPLIFIED ACCIDENT l SEQUENCE EVENT TREES . . . . . , . .... ...... ..... . .. . .... . A-1 l i APPENDIX B-QUANTIFICATION OF Tile SFIOREllAM SIMPLIFIED CONTAINMENT EVENT TREES . . . . . . . . . . . . . . . ... ... .. . ..... B-1 v l ?
APPENDIX C-REACTOR BUILDING ENVIRONMENTS . . . . . . . . . . . . . . . . . . . . . . . . . . C-1 l APPENDIX D-B ACKGROUND AND STRUCTURE OF Tile SHOREHAM PRA . . . . . . . . D-1 l l FIGURES
- 1. Simplified flow diagram of the Filtra vent system ... . . .... ... . ...... . . . 6
- 2. Simplified flow diagram of the Shoreham existing vent system . . . . . . . . . .. .. 7
- 3. Core melt frequency contribution by initiator (Shoreham PRA Filtra case) . . . .. .. . 17
- 4. Core melt frequency contribution by initiator (Shoreham simphfied event tree Filtra case) . . ....... ........ .. .. ... . .. . . .. 18
- 5. Core melt frequency contribution by initiator (simplified event tree-optimistic assumptions) . . . ..... . . . . .. .. . . 19
- 6. Core melt frequency contribution by initiator (simplified event tree-pessimistic assumptions) ... . . ...... ... . . ...... I. '.". . . ......... 19
- 7. Core melt frequency by sequence subclass (simplified event tree-optimistic assumptions) ... ..... . .... .... . . ...... , ... ... . 20
- 8. Core melt frequency by sequence subclass (simplified event tree-pessimistic assumptions) . . . . . . . . . . . . . . . . . . .. . . . .. ............ . 20
- 9. Release mode sensitivity to CV strategy and RB assumptions .. . ....., . .,..... 39
- 10. Dose vs distance (>5 rems-optimistic RB assumptions) . . . ... . . .. . ........ 41
- 11. Dose vs. distance (>200 rems-optimistic RB assumptions) . . ... .... ..... 42
- 12. Dose vs. distance (>5 rems-pessimistic RB assumptions) . .. . . .... ............ 42
- 13. Dose vs. distance (>200 rems-pessimistic RB assumptions) . .. . ...... ........... 43 TABLES
- 1. Accident sequence subclasses . . . . . . . . . ... .... .... .,... ........... 10
- 2. Core damage frequency (per reactor-year) for Filtra venting as calculated by simplified event trees (based on optimistic assumptions) . . . . . . .... I1
- 3. Totals calculated from core damage frequency listed in Table 2 .. . ... . ... . I1
- 4. Core damage frequency (per reactor-year) for existing equipment venting as calculated by simplified event trees (based on optimistic assumptions) . . . .. 12
- 5. Totals calculated from core damage frequency listed in Table 4 . . . .. . ... .. 12 vi
- 6. Core damage frequency (per reactor-year) with no containment venting as calculated by simplified event trees (based on optimistic assumptions) . . . ... 13
- 7. Totals calculated from core damage frequency listed in Table 6 ..................... 13
- 8. Core damage frequency (per reactor-year) for Filtra venting as calculated by simplified event trees (based on pessimistic assumptions) . . . . . . . . . . . . 14
- 9. Totals calculated from core damage frequency listed in Table 8 . . . . . . . . . . . . . . . .. . 14
- 10. Core damage frequency (per reactor-year) for existing equipment venting as calculated by simplified event trees (based on pessimistic assumptions) . ... 15
- 11. Totals calculated from core damage frequency listed in Table 10 . . . . . . . . . . . . . . . , . . . 15
- 12. Core damage frequency (per reactor-year) for no containment venting as calculated by simplified event trees (based on pessimistic assumptions) . . . . . . ....... ......... 16
- 13. Totals calculated from core damage frequency listed in Table 12 . . . . . . . . . . . . . . . . . . . . 16
- 14. Comparison of release mode probabilities for Plant Damage States IA,181 and IB2 . . . . . 25
- 15. Comparison of release mode probabilities for Plant Damage States ICl,IC2 and ID . . , . . . 26
- 16. Comparison of release mode probabilities for Plant Damage States llA, IIB and IIF . . . . . 27 17 Comparison of release mode probabilities for Plant Damage States lilA and 1118 .. .... 28
- 18. Comparison of release mode probabilities for Plant Damage States IllC and lilD ..... . 29
- 19. Cnmparison of release mode probabilities for Plant Damage States IVA,IVF and IVG . 30
- 20. Shoreham simplified event tree results, with Filtra (optimistic RB assumptions) . ....... 33
- 21. Shoreham simplified event tree results, with containment venting through existing equipment (optimistic RB assumptions) . . . . . . . . . . . . , . . . . .... .. 34
- 22. Shoreham simolified event tre ,lts, with no containment venting (optimistic RB assumptions) ........... ... ............... ............. 35
- 23. Shoreham simplified event tre ilts, with Filtra (pessimistic RB assumptions) . . . . .. 36
- 24. Shoreham simplified event tree results, with existing equipment venting (pessimistic RB assumptions) . . . . . . . .. .. ........................ .... ... . 37
- 25. Shoreham simplified event tree results, with no venting (pessimistic RB assumptions) . . . . %
26, Summary of conclusions . . .. ... ........ ..,,......... . .. ... . .... 45 l vii
FOREWORD in April 1988, the Office of Nuclear Reactor Regulation requested that the Office of Nuclear Regulatory Research perfonn a review of the proposed use of a filtered vent containment system at the Shoreham Nuclear Power Plant. The review was to consider the benefits of the proposed filtered vent system (similar to that used on the Barsebeck Nuclear Power Plant in Sweden), as well as the existing Shoreham vent systeu, relative to a case where no venting is permitted. The basis for this review was to be the existing probabilistic risk assessment perfonned by the Long Island Lighting Company. The Office of Nuclear Regulatory Research contracted with the Idaho National Engineering Laboratory to perfonn the detailed technical work. The scope of their work was to assess the risk reduction potential of the alternative vent systems, but not to review the probabilistic risk assessment it', elf. This report provides an assessment of the relative risks of the three cases of interest: venting with the filtered vent system, venting with the existing equipment, and no venting. The results of the assessment indicate that the use of the filtered vent system would reduce the estimated risk relative to no venting. However, the cost of this system would be very high. The assessed benefits of the existing vent system are not so definitise, for several reasons. First, the existing vent system permits venting froc, the , containment drywell region, potentially obviating the inherent radionuclide retention capability of the containment's suppression pool. Second, the design of the existing system makes use of low-strength ductwork, if venting were to be initiated, this ductwork could fail, exposing equipment in the reactor building to a high tempe.ature steam environment. Some experimental data suggest that the failure rates of equipment could be high in the estimated environment. Thus, undur some conditions and assumptions, venting with the existing system could result in equipment failures, increasing the possibility of subsequent core damage, and/or the release of radioactise material without the mitigating effects of suppression pool decontamination. From this, it can be concluded that the risk benefit of the existing vent system could be positive or negative, relative to no venting. With the limited scope of this review, and the sensitivity of some results on limited experimental data, the report provided here is, in one sense, inconclusive. However, given the indefinite status of the Shoreham plant, the staff does not now plan to pursue this analysis further. Brian W. Sheron, Director Division of Systems Research Office of Nuclear Regulatory Research ix
ACRONYMS AND ABBREVIATIONS A large loss-of-coolant accident DBA design basis accident AOV air-operated valves DC Failure of the downcomers connect-ing the vessel pedestal region to th, ATWS anticipated tr asient without scram suppression pool B Unavailability of ac power DC Loss of all de power (LOSP only) B- Failure to recover ac power after DF decontamination factor 0.5 hours (LOSP only) DG Failure of onsite emergency diesel generators (LOSP only) B2 Failure to recover ac power after 2 hours (LOSP only) ECCS Emergency Core Cooling System B6 Failure to recover ac power after EDG emergency diesel genenitor 6 hours (LOSP only) BS Containment break size is large ESF Emergency Safety Feature BWR Boiling Water Reactor F1 Early overpressure containment C Failure of the reactor protection failure system F1 Failure of Filtra containment vent C2 Failure to inject boron with the system standby liquid control system (ATWS only) F2 Late containment failure CET containment event trees FWCS Feed Water Control System Cl Failure of containment isolation G Failure of drywell heat removal GT gas turbine generator Cm Failure of the mecham. cal portion of the reactor protection systen) GT Onsite blackstart gas turbine genera-l- tor unavailable (LOSP only) l- CRD control rod drive I 11 Failure to control RPV level causes CS core spray boron to be flushed out of the core CST condensate storage tank til l' drogen burn inside the primary CV Failure to vent containment usinS containment prior to vessel failure e,J. sting equipment 112 flydrogen bum inside the primary D Vapor suppression system failure containment after vessel failure D1 Division i 125V de bus fails IICTL heat capacity temperature limit i xi
.H"P human error probabilities O Failure to detect water Dashing to steam in the reactor level instrument HPCI high-pressure coolant injection reference line E
HPl high-pressure injection OR Operator error causes faHure in op-posite reference leg (Tr sequences HR Failure of long-term containment "'Y' heat removal ORNL Oak Ridge Nstienal Lat. oratory HU No hold up of noble gases by Filtra OT Debris-heating of h drywell atmo-sphere and interna' turkce,. wcurs I Containment failure prior to core (potential overte npme Jro~ f melt (ATWS only) containment) L Failure to initiate RPV Gooding (for P Failure of more than one ratety/ [ RPV rupture, L is the probability relief valve to reclose (ATNS only) that the RPV does not leak before rupturing) PCS Power Conversion System PDS Shoreham plant damage state (e.g., l.ER Licensee Event Reports IA) Limo Long Island Lighting Company PF A driving force for an energetic re-lease exists LPCI low-pressure coolant injection PRA probabilistic risk assessment LPI . low-pressure injection Q Feedwater unavailable LOCA loss-of-coolant accident QU High-pressure injection unavailable LOSP loss of offsite power R Rupture of the reactor pressure ves-M1 . Failure of the operator to maintain sel(for loss of service water RPV water level at the top of active sequences only, R represents failure fuel (ATWS only) to restore reactor building service water after 10 hours) MAAP Modular Accident Analysis Program R1 Failure to recover the core melt se-quence prior to vessel failure MOV motor-operated valve F h m a M M W M is MISV ' bed on the drywell Door outside the main steam isolation valves pedestal region N Coolant makeup to the suppression RB Reactor building retention of fission pool fails following containment product aerosols is ineffective venting (ATWS only) RBCLCW Reactor Building Closed-Loop NPSH - ' net positive suction head Cooling Water System xii 1
RBNVS Reactor Building Normal Ventila. Tf Loss of feedwater events tion System Tmc Main steam isolation valve clo-RBSW reactory building service water sure events RBSYS Reactor Building Standby Ventila- Tc Loss of condenser vacuum tion System events RCIC reactor core isolation cooling Ti Inadvertently opened relief valves RD_ rupture disks Tms Manual shutdowns RHR residual heat removal Te Loss of offsite power RPV reactor pressure vessel Ta ATWS initiating event (ex-RPS Reactor Protection System cludes manual shutdowns) RV relief valve TAF top of active fuel c Td Loss of Division 11125 V de bus S Failure to cross-connect to turbine imtiating event building service water following a loss of reactor building servic Te Loss of offsim pr initising water events St Medium-sized loss-of-coolant Manual shutdown caused by high Th
- accident drywell temperature S2 Small-sized loss-of-coolant Tr Level instrument reference line accident break SBO station blackout Ts Loss of service water initiating SCS Supplemental Containment System SNL Sandia National Laboratories SNPS Shoreham Nuclear Power Station UI High-pressure injection unavailable SRV safety / relief valve (LOSP only)
SP Release products bypass the sup- U2 11igh-pressure injection fails to con-pression pool, either fully or tinue to operate after two hours partially. (LOSP only) - T Anticipated transient initiating UPS uninterruptible power supply
- event, includes V RPV rupture exceeds the capacity of Tt Turbine trips- the vapor suppression system xiil
VI - Low-pressure injection unavailable . VL . Failure of late containment venting V2- ' Low-pressure injection fails follow- VF Primary containment pressure less ing Filtra actuation than the vent setpoira of 60 psig at the end of the early time phase fol-
.VE Failure of early containment venting '
lowing vessel failure through Filtra VF Failure o. h reactor pressure vessel W Failure of evntainment heat removal at low pressure, implying nondisp. ersive exit of core debris from the X Failure to depressurize the reactor vessel pressure vessel s E xiv 0
, r . - ,-
CONTAINMENT VENTING ANALYSIS FOR THE SHOREHAM NUCLEAR POWER STATION EXECUTIVE
SUMMARY
An analysis of containment senting was per- removal. In this situation,25 to 309 of rated, fonned for the Shoreham Nuclear Power Station full-power reactor steam flow may enter the sup-(SNPS) to identify the effects of three different pression pool. This produces a heat load on the containment venting strategies on both core melt suppression pool that far exceeds the heat remov-frequency and offsite release consequences. The al capacity of the resic.aal heat removal system, three veming strategies examined were as follows: which is the nonnal system used for containment heat removal, k
- Filtered venting through a hardened Another characteristic of a desirable sent system such as the Swedish Filtra sys- ~
tem, which has been proposed for in, system is operability during a station blackout wquence. I or example, if ac p(m er is required to stallation at Shoreham actuate the vent valves, a standby or uninterr yt-
- Venting through the existing 6-in. # I" '".urce, such as an invena connected lines from the primary containment
""#U'* N "## # ^ '""F I "'I9nu nta-.
I " "' quired lor proper vent operanon will have wetw ell and dryw ell to be supplied.
- The Filtra system proposed for installation at No ventin8-Shoreham would meet all of these requirements, Venting has been proposed as a means of miti- although at a substantial installation cost. One cal-gating sesere accidents (beyond design basis ac- culated beneficial effect of the Filtra system is a re-cidents in which core damage occurs) by duction in the overall core melt frequency from preventing overpressure containment failure. 4.4E4)S/yr to 16E-05/yr, a decrease of approsi-Venting is beneficial primarily for those accident mately 189 from that of no containment senting.
sequences in which containment integrity is chal- A reduction in the overall offsite consequences is lenged prior to the onset of core melt.The two se- also possible, primarily as a result of Filtra's abil-quences for which venting is expected to have the ity to efficiently remove particulate fission paxi-greatest benefit are an anticipated transient with- ucts such as cesium and iodine from a release and out scram ( ATWS) and a loss-of-containment to provide a hold-up time for the decay of noble heat removal (TWL gas fission products like xenon and kryptoa Be-cause of its hirge size, Filtra is also capable of miti. For venting to be beneficial, it should not ad- gating most ATWS sequences. versely hffect core cooling and containment heat removal equipment. For example, if the contain- In contrast, the existing s ent system was found, ment depressurization associated with venting re- under the most conservative assumptions (re-sulted in inadequate net positive suction head to fened to hereafter as the " pessimistic" assump-the Emergency Core Cooling System pumps tions), to be of no benefit in reducing core melt supplied from the suppression pool, these pumps frequency. This pessimistic case consists of three could fail. in this case, venting would likely in- assumptions. First, there is the assumption in the duce core melt. Another requirement for succes- probabilistic risk assessment (PR A) for sful venting is that the vent flow: path be large Shoreham that venting leads to failure of the enough to mitigate the containment pressure rise equipment located in the reactor building. The pri-expected during a full power ATWS sequence mary effect of this assumption is an increase in the with the main condenser unavailable for heat core melt frequency. In addition, in order to l 1 !
. - . - _ - - _ . . _ . - . _ . - . - . - . _ . - . . - _ - . . - - - . . ~ . - _ . - . - . . - . - . .
1 mitigate the containment pressure rise expected to consequences approximately five tir,es inore se-occur during a TW sequence, both existing vent vere than those for the no-venting cass , lines (a (riri, wetwell and a (win. drywell) would
. have to be opened, which would result in a fission The small size of the existing vent lines also ,
pnxtuct release that bypasses the suppression pool makes the system ineffective for mitigating a se-(as indicated by Modular Accident Analysis Pro. vere ATWS sequence. For this cwe, there is no gram calculations performed for the Shoreham significant difference in terms of either core melt PRA). This contrasts with the overpressure con- frequency or resulting consequences between not tainment failure that would occur if the vents were venting and venting through the existing lines be-not actuated. In the case o' overpressure failure, a cause steaming of the suppression pool would number of possible containment failure modes challenge containment integrity prior to core melt would produ a scrubbed release through the in either instance. suppression p;
- resulting in less severe conse-quences in terms of the potentialindividual For station blackout sequences, the existing whole-body dose at a given distance from the vent system is again ineffective. This is because plant. The third assumption is based on an analysis the 120 V ac control power required to operate
- of Shoreham by Oak Ridge National Laboratory the vent valves is supplied from the emergency ac that indicates the post-venting reactor building buses, with no backup from an uninterruptible decontamination factor for particulate fission power supply. Thus, the valves will not be able to products is likely to be unity. This is based on the be opened remotely and local operation will be combined effect of the low exhaust capacity of the prohibited by the adverse, post-venting, reactor reactor building fission pnx!uct control systems building environment. This environment would and the predicted high likelihood of global hydro- similarly prevent entry into the reactor building to gen bums in the reactor building following prima- repair failed equipment or align alternate vessel ry containment failure or venting. The effect of injection systems, such as condensate transfer these second two assumptions is to make offsite pumps or fire-water pumps.
2
BACKGROUND in 1987 the Long Island Lighting Company cations, it is considered to be an alternate source (Lilco) proposed the installation of a Supplemen- of ac power in the probabalistic risk assessment tal Containment System (SCS) at the Shoreham (PRA). As a consequence of these alternate Nuclear Power Station (SNPS) to the Nuclear sources of ac power, the potential for core melt Regulatory Commission. SNPS is a boiling water due to station blackout events is much lower than reactor (BWR) of the General Electric BWR/4 that for other BWRs. product line with a Mark 11 primary containment. The SCS would provide filtered venting of the primary containment in the event of a severe acci- Shoreham's reactor building design also ap-dent. Venting has been postulated as a last-<litch pears to be unique among U.S. Mark 11 BWRs. effort to prevent the possible overpressure con- First of all,its intemal geometry is that of an open tainment failure that could result from a severe cylinder, with very few compartments outside of accident (an accident that exceeds the design the primary containment boundary. For example, basis). In some accident sequences, venting all of the Emergency Core Cooling System might provide a means to prevent core damage, (ECCS) pumps are located in an open annular while in others it could be used as a possible way area outside of the suppression pool in the base-to mitigate accident consequences, ment of the reactor building; at other plants, such as La Salle and Susquehanna, the ECCS pumps Shoreham is rather unique in certain respects are individually located in watertight compart-from other U.S. BWRs. Perhaps the most signifi- ments. There is also a high degree of connectivity cant contrasting feature is the plant's emergency among the different elevations in the Shoreham ac distribution system, which uses six emergency reactor building. Should hydrogen be released diesel generators supplied by two different ven- into the reactor building as a result of primary dors [three each from Transamerica DeLaval in- containment failure or venting, this high :onnec-dustries (TDI) and Colt]. One TDI generator and tivity would tend to promote global hydrogen one Colt generator supply power to each of the burns over less severe localized burns. Finally, three emergency ac electrical divisions. However, the low exhaust capacity of the fission-product only Divisions I and 11 are independently capable control systems for the reactor building makes of allowing the plant to achieve a safe shutdown overpressurization of the refueling-floor walls condition; Division til does not supply power to likely in the event of primary containment failure any primary system depressurization equipment or venting through the existing heating, ventilat-or sources of high-pressure injection. Shoreham ing, and air conditioning (IIVAC) ductwork. This also has a gas turbine generator with blackstart has an important effect on offsite consequences capability located at the plant site. Although it is because it resulty in a reactor building decontami-not covered by the Shoreham Technical Specifi- nation factor (DF) that is essentially unity. l 1 l l 3
INTRODUCTION o The accident sequence analysis described in specific whale-body dose at a given distance this report uses simplified systemic event trees from the plant is calculated using dose-versus-(see Appendix A) to identify the dominant core- distance consequences. Ilecause Shoreham did melt accident sequences and to calculate their as- not have an approved offsite emergency response sociated frequencies of occurrence. These esent plan at the time of this analysis,it was deemed in-trees are based on those presented in the Proba- appropriate to postulate an evacuation strategy, bilistic Ris A Assessment, Shorcham Nuclear Pow-lier , only potential w hale-body doses to an in-er Station.1 dividual at a given distance from the plant were calculated. No population doses or demographic-The containment analys.is also uses simplified based consequences were calculated, if evacua-containment event trees (SCETs) to model the tion had been considered, few, if imy, early health containment s response to each plant damage - state (see Appendix 11).
# .'"*" "Y " "I'"4"#"C#*i"""i*'
ing a loss of long-term containment heat removal In calculating severe accident risk, the output (TW), because of the long delay period asso-from the front-end analysis (the accident se. ciat,;d with the release. Ilowever, latent health ef-quence frequency) is combined with the condi. fects are typically not significantly affected by tional release mode probabilities determined evacuation because they are largely a result of from the SCETs, which are then binned into re- long-term chronic esposure to slightly contami-lease categories to detennine the probability per nated land (i e., land that is contaminated at levels reactor-year of a radioactive release of a specific below the threshold where foodstuffs are con- - severity. Then the probability of exceeding a demned or land use is restricted). 5
ANALYSIS OF VENTING STRATEGIES The proposed SCS would be modeled on the advertent Filtra actuation (e.g., during a design Swedish Filtra design that uses a large gravel bed basis accident when containment isolation is to filter the primary containment effluent during required) is prevented because the two isolation severe accident sequences that threaten contain- MOVs automatically close if flow is detected ment integrity. Presently, the capability to vent down(tream of the weelt airspace penetration the SNPS primary containment is provided by and containment pim.are is less than 48 psig. two 6-in, vent lines, one connected to the w etw ell This automatic closure signal is scaled out if con-airspace and the other to the drywell. Each of tainment pressure is greater than 55 psig. these venting mechanisms is described in more detail below. The Filtra system is aho designed to operate during a station blackout. The isolation MOVs Filtra ""' P " e" by "" ""i"'c""Ptible Po" c' s"PP ly (UPS) from a dedicated battery. A dedicated power supply is also provided for the Filtra sys-The Swedish Filtra contamment sent system tem instrumentation and controls. consists of a large gravel-bed filter housed in a reinforced concrete structure separate from the Pressure control for the Filtra structure is pro-reactor building. This system is connected to the v ded by a pressure control station, which con-primary containment through one or more hard- a sts of a 12-in. globe valve and two 24-in. relief pipe penetrations. A simplined now diagram of vahes with a 2-psig rupture disk downstream of the Filtra system is shown in Figure 1. The pro- each valve. The actuation setpoints of these three posed installation at SNPS would use the existing valves are staggered so that pressure relief can be persomel access hatch as the wetwell airspace provided while limiting the amount of mass re-penetration, and would use a 24-in. diameter pipe leased to the environment. passing through the wall of the reactor building, which expands to 30 in, in diameter before enter- The Filtra system is inerted with nitrogen to ing the gravel bed. This flow path would include minimize the potential for hydrogen detonation. two motor-operated valves (MOVs) that are not in addition, the Filtra system piping is designed to mally open and can be operated remotely, and w thstand the forces generated by a local two 60-psig rupture disks to relieve the contain' hydrogen detonation. ment wetwell airspace pressure to the Filtra system. The Filtra system is designed to reliese up to 8% of rated steam now from the primary contain-There is also a 6-in. line from the drywell at- ment for up to two hours. Furthermore, the dis-mosphere that would join the 24-in. wetwell vent cussion in Section 4 of :he Shorcham Nucicar downstream from the rupture disks. The MOVs Power Station Full Power PRA, PRA Update: in this line would normally be closed, but could Supplemental Containment System Implementa-be remotely opened by the operator to establish a tion indicates that, through the operation of the vent path for long-term heat removal or for con- pressure control station, Filtra has the capability trolling any hydrogen accumulation inside the to relieve up to 18% of rated steam flow from drywell. containment.2 The Filtra system is passive, so no operator ac. The benefits of venting through the Filtra sys-tion is required to initiate venting. When the pres. tem include the following: sure inside the wetwell airspace reaches 60 psig, the rupture disks in the vent line actuate and the
- Prevention of containment overpres-containment is vented to the gravel filter bed. in- sure failure.
5 l l _ - - = _ _ _ _ _ _ _ _ _ _ _ _ . _ _ _ _ _ . _ _ _
Koy e Hard piping
--- Ductwork llll Flexible connectors k Alr operated valvos a To atmosphoto
"~ "
k Motor. operated valvos Filtra N buildhg G g] 2 psig RD Rupture disks RV 2 { l'h on}U RV 1 RV Reliof valves 1 Filtra roliof station sotpoin',s -n RV 1 50 psia RV 2 54 psia . AOV 49 psia o h
> :- ) r Drywell MOV 003 MOV 004 Gravel filter bod 10 in.
l o 60 psig 60 psig Wotwoll MOV 001 MOV 002 ( -** f j c
= =
RD 001 RD 0021 r MK il containment Ol[ 01: c ,f From N2 Inerting system 0 0635 Figure 1. Simplified flow diagram of the Filtra vent system. l l l i l
- Removal of particulates from the outside the containment; for the drywell line, one effluent gas stream by the gravel bed. AOV is inside the containment while the other is outside. The two lines are connected by a flexible ne prene sock to a common ventilation duct up-e lloid up of noble gases to allow time stream of the primary containment vent / purge fil-for decay prior to release. Hold up is ter. D wnstream of this filter, the ducting ties into provided only as long as steam flow the Reactor Building Normal Ventilation System does not saturate the gravel bed.
(RilNVS). Dunng an accident, when the RUNVS is isolated and the RBSYS is operating, the flow Existing Vent path will be through the RBSYS filters (as long as the ventilation ducting remains intact). Shoreham presently has the ability to manually vent the wetwell and the drywell to the Reactor The Shoreham Emergency Operating Proce-Building Standby Ventilation System (RBSYSh dure (EOP) for primary containment control di-Figure 2 shows a simplified flow diagram of the rects the operator to vent the wetwell before R existing vent lines. Both vent lines are 6 in. in di- suppression chamber pressure reaches 60 psig, ameter and contain air-operated vahes (AOVs) and to vent the drywell if the wetwell water level to automatically isolate the containment during is above the vent line penetration, or if the wet-an accident. Both AOVs in the wetwell line are well cannot be vented. Because an accident will p_________________________________ i Reactor building $ l e Hard piping l
--- Ductwork i
I To reactor building no, mal venting system h Flexible connectors l Dig Air operated valves i l i k Motor operated valves l I
'r.-_.-ll_
l 1 b I I l b_... 610. Drywell i 2 o h a i l h "_- _ 8e .. A r,- - r, I 6 in. AOV 078B AOV 078A o I l l M' **- Wetwell i AOV 0798 AOV 079A i l 1 I I MK il containment 0-0636 Figure 2. Simplified flow diagram of the Shoreham existing vent system. 7
result in a closure signal being sent to the isola- pumps) located in the low er elevations. Ilowever, tion AOVs in the vent lines, the operator will be in the containment analysis in Reference 3, vent-required to install electrical jumpers in the main ing was assumed to resuh in the failure of reactor control room to allow these valves to be opened, building equipment because of the adverse envi-Also, these valves cannot be opened during a sta- ronment created by the rupture of the ventilation tion blackout (SBO) because the 120V ac control ducting. For this report, two bounding sensitivity power to the valve solenoids is not supplied from cases were analyzed. The first case assumes that a UPS source. venting through the existing lines does not signif-icantly affect equipment located in the reactor The ventilation ducting connected to the vent building, his is consistent with the (Draft) NU. lines is not designed to withstand the pressure REG-1150 assumptions for Peach Bottom. In ad. loading that could occur during venting. Conse- dition, this case assumes that venting does not quently, venting through the RBSYS would most adversely affect the reactor building aerosol DF. likely cause the ducting to fail and release steam The second case assumes that venting leads to the and noncondensible gases into the reactor build- failure of coolant injection which is consistent ing. The atmosphere in the reactor building fol- with the Shoreham back-end analysis.3 Also, as lowing the rupture of the ventilation ducting and outlined in the Oak Ridge National Laboratory the effects on the equipment located there are (ORNL) analysis of the response of the Shoreham both uncertain. The front-end accident sequence secondary containment to a severe accident,4 analysis in Reference 2 assumed, based on engi- venting is assumed to lead to reactor building by-neering judgement, that the rupture would occur pass. In the discussion that follows, these two in the upper elevations of the reactor building and cases are referred to as the optimistic and pessi-would not affect equipment (such as ECCS mistie reactor building assumptions, respectively. 4 8 l
EFFECTS OF VENTING ON CORE MELT FREQUENCY The initial phase of the analysis ccnsisted of cation of the accident sequences was not possiti quantifying the change in core melt frequency af- Such a course would require obtaining, reviewing fected by each containment venting strategy, in detail, and rerunning the system fault trees, From the perspective of being able to prevent which is beyond the scope and resources of this core damage from occurring, containment vent. study, ing affects the TW and possibly some ATWS se-quences (i.e., those accident sequences for which Dominant Accident Sequences containment failure precedes core damage). The for Shoreham approach used to quantify the effect of venting on core melt frequency is described below. Each path of the event trees represents a possi-Ne sequ ne that rnay or inay not asuh in me Simplified Event Tree Approach damage. Those that do are referred to as accident sequences, which are then grouped or combined The first step in determining the effect of vent-into accident sequence classes (sometimes re-ing on core melt frequency was to develop a ferred to as plant damage states). These classes working model of the front-end core melt acci-are characterized as follows: 3 dent sequence analysis. This involved the cre-ation of simplified accident sequence event trees
- I-Loss--of-core -cooling accidents to approximate the dominant core melt se-quences. These event trees then provide the mechanism for detennining the effects of the dif-
- II-Loss-of-containment-heat-removal accidents ferent venting strategies on core melt frequency.
A simplified event tree was developed for each
- III-Loss-of-coolant accidents (LOCAs) of the initiating events identified as being impor-tant to risk in Reference 2. Furthermore, these * 'V-Anticipated transients without scram simplified event trees were constructed such that (ATWS) each sequcnce contributing a significant percent-age to any of the 14 accident sequence subclasses These sequence classes are further divided into was represented (albeit in simplified fonn). subclasses in the Shoreham PRA, as shown in Table 1.
The simplified event trees, along with descrip-tions of how each event was quantified, are con- Results tained in Appendix A. Wherever possible, insignificant or irrelevant branches were omitted. Once the simplified event trees were devel-This resulted in a significant reduction in the size oped, they were used to determine the frequency and complexity of the event trees, of the accident sequences and accident sequence subclasses. The event trees were quantified for Assumptions each of the three containment venting strategies using both the optimistic and pessimistic reactor Although modifications to the logic and quan- building assumptions on the effects of venting tification of the PRA event trees were made as through the existing lines on equipment located in necessary, there was no attempt to verify or vali- the reactor building. Tables 2 through 13 show date the system unavailabilities and failure proba- the core melt frequency contribution for each ac-bilities that were presented in the PRA. Given the cident sequence subclass, listed by initiating available documentation, a complete requantifi- event. This same information is also shown i 9
h y. Table 1. Accident sequence subclasses Subclass Desenption IA-TQUX Subclass I A accident sequences involve a loss of inventnry makeup in which the reac-tor pressure vessel remains at high pressure during core damage. l IB - TeQUV(X) Subclass IB accident sequences represent both the short-tenn (1B1) and long-term (IB2) (station blackout sequences and comprise a loss of offsite pow er (Te) and a loss of coc. ant inventory makeup (injection failure at battery depletion for the long-term and immediate failure of all injection for the short-term sequences). These sequences ' include both the high and low RPV pressure sequences (TeQUX and TcQUV, respectively )). IC - TaCmQU These are anticipated transient without scram ( ATWS) sequences that induce a loss-of-coolant inventory makeup and result in core melt before containment failure. The IC plant damage state is subdivided into classes ICI (SCS failed during the accident sequence) md IC2 (SCS not failed during the accident sequence and pmbably avail-able to mitigate the post-core melt phase of the accident). ID - TQUV Subclass ID is comprised of the loss-of-coolant imentory makeup sequences in which the RPV has been depressurized to 200 psi or less prior to the onset of severe core damage. IIA - TW-CF This subclass includes those sequences that involve a loss of long-term containment heat removal with the RPV intact at the time of containment tailure Containment fail-ute induces a loss of RPV injection and subsequent core damage. IIB - AW-CF These are also loss of long-term containment heat removal sequences but they in-clude a prior rupture of the primary coolant system (LOCAL llowever, RPV injection is initially successful and core damage does not begin until after containment f ailuie. IlF -TW-F Subclass lif* sequences are similar to the previous two subclasses with the exception that the containment is vented. RPV injection failure after containment venting leads to core damage. Illa - R Large RPV ruptures (larger than a large LOCA) that exceed the capability of the ECCS. IIIB - SlQUX nis subclass includes small and medium LOCA accident sequences for which the RPV remaim pressurized. lilC - SlQUV This subclass includes medium and lari e LOCA sequences for which the RPV is depressurized. IllD - AD This subclass is comprised of large LOCAs or RPV ruptures that exceed the capabili-ty of the vapor suppression system, thereby challenging containment integrity. IVA -TaCmC21 Subclass IVA is comprised of ATWS sequences with reactor power greater than the SCS capacity and/or with the SCS not available. Containment failure occurs prior to core melt. IVF-TaCmC2V2 Subclass IVF is comprised of ATWS sequences with reactor power within SCS ca-pacity and with the SCS available injection systems are assumed to fail following SCS actuation.Thus, steam flow to the SCS is not maintained and the gravel bed does not become saturated. Subclass IVF is not defined for sequer,ces w here venting occurs through the existing 6-in lines. IVG -TaCmC2N Subclas3 IVG sequences nre similar to those of subclass IVF except that coolant injec-tion continues for some time after SCS actuation, resulting in continuing steam flow to the SCS and saturation of the gravel bed. As in subclass IVF subclass IVO se-quences are not defined for venting through the existing (>-in. lines. 10
l Table 2. Core damage frequency (per reactor-year) for Filtra venting as calculated by simplified event trees (based on optimistic assumptions) Imtutmgvent __ Ty pctJ Subclau Trans Tde NL Tsw %)* t 3p_ R $2__ St A ATWS Sequenet IA 2. lE-05 3.lE4b - 1.10-06 - 2.3E46 - - -- - 7.6E4m TQUX IB - - 4.0E-07 - - - - - - - - SBO IC - - - - - - - - - - IEE-OR TCQU ID 2.7E-07 6.9E-10 - 1.0E-Oh 4 7E.07 4.0E4D - 4.00-07 - - SEE-09 TQUV llA 5 IE4)8 23E4N - 9.2E4N - 1.lE-10 - -- - - TW-CF llB - - - - - - - 2 4t-10 2 3E-!O 640-10 - AW-CF llF 3.5 E-07 1.4E4)7 - 1.3E-06 - 7 KE-.10 - 6 5E-10 63E419 1.NE4M - EF) tilA - - - - - - 2 9E4D - -- - - R lllB - - - - - -- - 4 2E4m 6 9E-41 - - SIOUX
!!1C - - - - - - - -
1.5E-07 1.7E-08 - 51QUV lilD - - - -- - - 9.0E4N - ~ 7. l E-08 - AD IVA - - - - - - - - - - 15E4)7 TC-CF IVF - - - - - - - - - - T.2E-07 TC-Fl IVO - - - - - - - - - - 1.4 E 4M rC' Sat Table 3. Totals calculated from core damage frequency lis'-d in Table 2 PRA initiating PRA SubcIns Comparison Imtiating Es ent Comp.!nson Subclaw Subtotal Subtotal Es ent Subtotal Subtotal _ IA 2,7E-05 2.6E-05 Trans 2.lE-05 2. l E-05 IB 4.0E-07 4.lE-07 Tde 3.3E-4M 3.3 E-06 1C 1.8E-08 1. l E48 Tlosp 4.OEJ7 4. l E-07 ID 2.6E46 2.0E4M Tsw 3.4E-on 4.50-07 IIA 6.4E-08 2.7 E-08 Thdwt 4.7E-07 .* 6E47 11 0 i . l E--09 5.7E-09 Trup 2.7E4h 3.5E46
!!F l .8E4M l .6E-06 R 3 OE-07 3.0E-07 tilA 2.9E-07 2.9E-07 S2 4.4 E-07 4.8E-08 IllB 7.3 E-07 7.4E-07 S1 8.5 E-07 1.3E= 06 lilC 1.7E47 3.6E-07 A 1.lE-07 2.RE-07
!!!D 8 OE48 8. lE48 ATWS 2,5 E46 1.5E-06 IVA 3.5E-07 4.4 E-07 .- - -
IVF 7.2E47 3.6E-07 - - - IVG 1.4 E4M 6 4E-07 - - -- Total 3.6E-05 3.3 E-05 3 AE-05 3.3E45 i I1 I
i Table 4. Core damage frequency (per reactor -year) for existing equipment venting as cakulated by i simplified event trees (based on ry imistic asstimptions) initiating Evert Sukiass T1osp Tsw ht Trop _ R S2 St A ATWS Trans _Tde l 7.4E4)F a IA 2.10415 3.lE4)6 - 1.llMM -- OJE-06 - - - - til - - 4.0E-07 - - - - -- - - - IC - - - - - - -- - - - 2.5 E-06 ID 2.7E-07 6.9E-10 - 1.0E46 4.7E-07 4.0E417 -- 4.0E417 - - 5 8E4N llA 7.lE40 3.9E40 - 1.3E4m - I hE-M - - - - - IIB - - -- - - - - 3]E46 3.2E-09 8.9E49 - t.9E-10 5.7E4W 1,bE-08 7 IIF 3.2E4)7 0.0E40 - 4.2E-11 -
- 7. l E-10 - - - -
I LilA - -- - - - - 2.9E417 - -- - - 3 IIIB --. - - - - - - 4.21i-08 69640 - - IllC - - - - - - - - 1.5E-07 1.7E4)fi - IMD - - - - - - 9.0E 4)9 - - 7.lE45 - IVA -- - ~, - - - - - - - 1.l E-06 IVF -- - - - -- - - - - - - - 0.0E+00 iVG - -- - - - --- - -- - - 0.( L (1) Table 5. Totals calculated from core damage frequency listed in Table 4 , ( Intiadng Subclass Initiatmg Ewnt Suklass Sub:otal Event Subtotal IA 2.7E-05 Trans 2.2E45
!!! 4.0E4D Tde 3.5 E-(E IC 2.5E4M Tlosp 4.0E47 ID 2.6EJ6 Tsw 3 4E-06 IIA 2.4 E-06 Thdwi 4.7E-07 IIB 1.5E-08 Trup 2.7E4%
IIP 3 4E-07 R 3.0E-07 tilA 2.9E-07 S2 4.5E40 IllB 7.3E-07 SI 8SE-07 lilC 1.7E-07 A 1.lE4D lilD 8.0E-08 ATWS 3.7E-06 IVA til E-44 - - IVF 0.0E4X) -- - IVG 0.OE4X) - - Total 3.8E-05 J.8 E-05 l 12 l
I Table 6. Core damage frequency (per reactor-year) with no cotl# ment venting as calculated by simplified event trees (based on optinustic assumptions)
- ~ ~
I i, _ , , , _ , Initiati4 L ent Subcla9 fm , TL g) Tsw Thdwt Trup R .N ' Si A ATWS lA 1.JE-D) 3.tE 06 --
. llM6 - 2,'E '6 - - - - 7.6E-08 IB --
- 4.0E417 - - - - - - - -
IC - - - . . - - - - - - - 2.5E-06 ID 2 7E-07 6.9E-10 - 3 A-On 4.7E-.07 4.0E47 - 4.0FA)7 - - 5.8E-09 IIA 7.lE4e 3.9E-07 - 1.3 E-06 - t hE4)8 - - .- - - IIB - - - -- - - - 13E-08 3.(E '18 8.9E4)8 - IIF 0.0E40 00E40 - 0.0E+00 - 0.0E+00 - 0.0 EMU 0.0E+00 0 0E+00 - Illa - - - - - - 2 M 07 - - - - lilB - - - - - - - 4.2E-08 6,9E47 - - IllC - - - -- - - - - 1.5E-07 1.7E-08 - 111 0 - - -- - - - 9.0E-09 ~- - 7.lE4m - IVA - - - - - - - - - -
- 1. lE4h IVF - - - .. - - - _ _ _
o m oo ) Ivo - - - ~. -- - - - -_ _ 0.0e w 4 Table 7. Totals calculated frora ec e damage frequency listed in Table 6 . Ini tianng ( Sutdus initiating Event Subclass Subtotal Event Subtotal I i IA 25E-05 Trans 1.RE4)5 1 IB 4.0E-07 T de 3.5E46 - i r IC 2.5E-06 'losp 4.0E-07 h ID 2.6E46 T.v 3.4E46
!!A 8.8E-06 Thdwt 4.7E-07 11B 1.5E4)? Trup 2.8E-06 I I!F 0.0E+00 R 3.0E-07 L
I!!A 2.9E-07 S2 4.7E-07 lllB 7.3E-07 51 8.8E-07 IIIC 1.7E-07 A 1.8E-07 tilD 8.0E 08 ATW5 3.7E-06 IVA 't .1 E4h - - IVF 0.0E+00 - - WG 0.0E+00 - -
otal, A4E-05 4.4E-05 g . _--.- ._. -_
G K ___ ._ _ _ _ _ _ . - _ _ ._.___. . _ _ _ _ _ _ . _ _ _ . _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ . _ _ . _ . _ . _ . _ _ _ _ _ _ _ _ _ _ _ _ _ _ . _ _ . _ _ _ . . _ _ _ _ _ _ . _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ . _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
Table 8. Core darnare frequency (ger tractor-year) for lititia s enting c cnicutr.ttd tiy siingilined es 1.5 trees (t.ased on gessitnistic nwurnjitionu
... :. -~
I litibeltlij i tfitt 1)pis al hoh tau 1rans Ttk ll$p_ __,'r sw_ ._1kle t_ _1 tyL __ k_. , h2_ hl A. . ., AIM S P'T r'M l 16t a* IA 2 nais .i a tv- -. u l Aw, - 2 r4 v, - - - Wl'x in _ ._. 4 olai7 - - - - - -- Sito Ic - . . -, ._ . - - - - I kl. 4m 1CQ ID 2.71417 6 Vl;- 10 -- I 01. 4Wi 4 71 -437 4 01 -417 - 4 UI 4'7 - S nllH IVl'Y llA $ !!' 4* 2 tilAN 9 214N - 1!!'-10 - - - - -- - TW-Cl ini _ - -. - . - -- 2 4tcl0 211-10 6 41.--10 - AW 41 11): 331 417 1.vl A17 .- 131 4 w, - 7 ki'.- 10 - 6%i-10 6 TilN . kl (* - l W II l lilA - -- . -- - - 2 91 -07 - - - R 6 11111 - -- - -- .- - - 4 2tA* 6 91. 07 - -- SlQt!A lllC - - - - - - - -- 131'07 17t M* - S IV'N Inn _ _ . - - 901.IN - - 7,Il -4* -- AD IVA - - - - - - - - - 1 SI '07 10 -CI gyr _., _ . _ .-. _. _ __ -. - 721 Jt? Ic.fi gyg _ _ _ _ - . . - . - - 1 41, . 3 104.t l Table 9. Totals calculated f roin core darnage frequency listed in Tat le 8 IHA Imtiannt IHA hulwlau Companwn Imtiatmp in ent Compariwn Sutwlaw Subtotal Sobtotal I Aent Subiotal hubtotal IA 2.7tMi$ 2 6E415 Tram 2.llM)$ 2.llWi$ 1 11 4 016 07 4.1E417 1 de 3.31i 416 3.3tM W> IC l.804m 1.llMM Tlosp 4 01:-07 4.llMil ID 2 blMW, 2.0E.4 r> l u. 3 4f Mki 4.SIMO i IIA 6.41Mm 2.7tMm 'thdw t 470-07 $60-07 1111 1. IIMN 5,7049 1rup 2.71M W> 3.51Mki llF 111Mxi 1.61M r., it 3 00-07 300-07 lilA 2.90.-07 2VE4D S2 440-07 4 klMM lilli 7.3tMD 7.4 tLU7 S1 8.$E417 1.31M6 IllC 1.7E-40 3 6tMD A l.1 F-40 2.NE-40 tilD 8.004m H ;tMl8 ATWS 2.$1Mki 1.51M r> IVA 3.5E-07 4 4fMD - - - IVF 7.2iMD 3 blM0 - - - IVO 1.4IMi6 6.40-07 - - - Total 3 61M)5 3.311-05 36E-05 3.31605 14
Tcbb 10, core darnage frequency (m reacle.-year) for existing equipnient venting as calculated by 6implified event trees (based e russlinistic assurnptions) Initiating Esent Sut(lass Trans 1de Tio t 1sw Mgiwi Trug R S2 51 A ATWS lA 2.10-05 311M6 - 1.10416 - 2.3E4m - - - - 7.41Mik lit - - 4 0E417 - - - - - - - - IC ~ - - - - - - - - - 2.5E40 ID 2.7E417 tt9E-10 - 1.0!M16 4.71M)7 4.0E40 - 4.00-07 - - 5 8tMN llA 7.lE417 3.91MO ft40419 1.3E-06 - 1.6fMN - - - - -
!!b - - - - - - 3,7E-11 3.31MN 3 2E4N 8.9E4N -
IIF 6 4E46 0.nE+fK) 3.704)M N4E-10 - 1.41M18 3.41A0 3 OL-Oh 2.81MW N.0E418 - Illa - - - - - - 2.4E417 - - - - , Illil - - - - - - - 4.2E4th 6.91MG - - t IllC - - - - - - - - 1.51M17 1.7tMM - lilD - - - - - -- 9.004N - - 7.1E4* -- IVA - - - - - - - - - - 1.t!M6 IVF - - - - -- - - - - - - IVO - - - - - - - - - - - Table 11. Totals calculated frorn core darnage frequency listed in Table 10 Initiating Sutclass initiating i vent Sutriass Subtotal Event Suhtotal IA 2.7E4)5 1rann 2.EE415 lli 4OE40 TA 3.5E4W> IC 2.5E-06 Tiosp 4 4EMD ID 2.6fMr> Thw 34E46 IIA 2.41M16 1hdwt 4.7E40 1111 1.5E4m Trup 2.81M16 IlF 6.6E416 R 3.0E40 tilA 2.9040 S2 4 7E-07
!!!!! 7.3E40 5i H.8E-07 IllC 1.7E40 A 1.NE-07 tilD h.OlMJ ATWS .7E46 IVA 1. llMA - -
IVF - -- - IVO - -- - Total 4.4 E-05 4.41Mp3
+
15
Table 12. Core damage frequency (per reactor-year) for no containment venting as calculated by sirnplified event trees (based on pessimistic assumptions) Initiating Event , TMat A AT% S Sutwlau Trans Tde Tlosp. Tsw Trup _ R $2 _ _ St LA 210-05 3 llM6 - 1.lFM6 - 2.30416 - - - - 7 60-08 1 11 - - 4.0E47 - - - - - - - - IC - - - - - - - - - - 2.51M6 ID 2.70-07 690-10 - 1.0E40 4.7tM)7 4 OIM)7 - 4 OIM17 - - 5.8tMW
!!A 7.lE-06 3 9fM17 - 1 JE4R. - 161'Am - - - - -
1111 - - - - - - - 3 3E4m 3.llMm 8.9E4m - 0.0E+00 0.0F+00 - 0 0E+00 - 0.0E+(0 - 0.01;+(0 0=0E+00 0.00+00 - Ilf tilA - - - - - - 2.9tM17 - -- - - 11111 4.20-08 6 9E-07 - - IllC - - - - - - - - 1.5b07 1.7tx08 - IllD - - - - - - 9.0!MN - - 7.1[X* - IVA - - - - - - - -- - - I1[-06 IVF - - - - - - IVO - - - - - - Table 13. Totals calculated fram core damage frequency listed in L.ble 12 c imtioung Subcim initiating Enent Subclau Subiotal Es ent Subtotal IA 2.7E-05 Trani 2.8FM15 Ill 4.0lM17 1de 3.50MW IC 2.5tMb Tiosp 4 OE417 ID 2.61Mb 'Is w 3 4E46 11A 8.8E46 Thdwt 4.71M)7 till 1.5E-07 Trup 2.80416 IIF 0.0E+00 R 3.0lM17 111A 2.9E-07 S2 4.7E-07 11111 7.31M17 SI 88E-07 "ilC i.7tMW A 1.8E-07 tilD 8.0E4* ATWS 3.704ti IVA l.lE46 - - IVF - IVO - - Total 4.4 E413 4.4EM)$ 16
. - -- . .- - . . - . _-_ - . - - - . - - . _ _=_ __ _ - - - -
l l graphically in Figures 3 through M. The core melt the esisting equipment senting (which is inade-frequency is shown for each of the containment quate for ATWS sequences) and the nonenting venting strategies previously described. As cases, shown from the results, the yearly core melt fre-quency ranges from 3.6EM15 with the Filtra es- As discussed earlier, the design or the emer-tem to 4.41?-05 with no venting. Venting pency ac electrical power system at Shoreham is primarily affects the core melt frequency for more redundant than at other commercial nuclear Class 11 and IV sequences, and Class I to a lesser power plants because of the relatively large num-extent. Classes 11 and lY are the loss of long-term ber of on:,ite emergency diesel generators. There-heat removal (TW) and ATWS sequences, re- fore, the core melt frequency contribution from spectively, for w hich containment failure induces SliO is much smaller at Snoreham than has tren core melt. For ATWS sequences with ve *ing found at other plants. Also, compared to the re-failure or no seating, containment failure tiefore suits from the (Drafo NURl!G-il50 analysis of core melt is calculated probabilistically; there- the Peach flottom Atomic Power Station (both fore, a percentage of these sequences will result Shoreham and Peach Ilottom ate llWR/4 de-in containment failure succeeding core snelt and signs), Shoreham has a rnuch highc probability thus are binned into Class 1. llence, there is an in- of failure to depressurite the RPV than does crease in the frequency of Class I sequences for Peach llottom (3f.-03/ demand at Shoreham ATWS (1.6%) A (0.9%) 31 (4.0%) m S2 (0.1 %) Trup. (10.7%) - g R (0.9%) m Thdwt (1.7 %) N Tsw (1.4 %) ~ =N Tlosp (1.3%) Tde (10.1 %) Trans (64.3%) 00643 Figure 3. Core melt frequency contribution by initiator (Shorch2m PRA Filtra case). 17
, ATWS (7.0%)
A (0.3 %) S1 (2.4%) S2 (1.2 %) Trup. (7.6%) R (0.8%) Thdwt (1.3%) I Tsw (9.5%) Tiosp (1,1%) Tde (9.1%) L Trans(59.6%)
- 0-0644 Figure 4. Core melt frequency contribution by initiator (Shoreham simplined event tree Filtra case).
versus approximately 3E-04/ demand at Peach reactor building continues to operate such that Bottom). This results in a significantly larger RPV injection continues. contribution to the Shoreham core melt frequency from the TQUX type of sequence, the frequencies Any calculated benefit from containment vent-of which are not affected by venting. Because the ing is very dependent on the treatment of post-TQUX sequences are such dominant contributors venting RPV injection. Two questions must be to the total core melt frequency (approximately considered w hen calculating these benefits. First, 60%), they minimize the relative benefit how do coolant injection systems that are sue. attributable to containment venting in reducing cessfully operating react to the effects of the rapid core melt frequency. containment depressurization that results when the containment is vented? And second, when ex-isting equipment flow paths are used that result in With respect to core melt frequency, the bene fit a release to the reactor building, how well does derived from containment venting depends on the any vital equipment survive the adverse condi-system used and the subsequent effects on reactor tions created by venting? The failure of RPV in-building equipment. Obviously, the most desir- jection after venting results in a sequence similar able strategy is to have a hardened pipe, filtered to an unmitigated TW sequence except that the vent system such as Filtra. This provides opti- release (at least initially) may be scrubbed mum operability along with a filtered release. Us- through the suppression pool and it occurs earlier ing existing equipment results in a core melt (the venting setpoint is lower than the ultimate frequency reduction only if vital equipment in the containment failure pressure). 18
3 T' ' ' ~' ' ' ' ' i ' ' 2.8 - 7 - 2.6 - - 2.4 i G PRA - 2.2 i M Filtra - h- 2 e I i Ex Eq CV 3$ 1.8 ' - g8 - ' A NoCV
$ 1.6 < -
=" / <
g 1.4 f 4 - g 1.2 < -
/
1 f 0.8 , 4 - 0.6 , l - 0.4
, f , ,
0.2 / b ? i 0', '
, T T r '", "" - - "
Trans Tde Tiosp Tsw Thdwt R Trup. S2 S1 A ATWS Accident sequence Initiators oma Figure 5. Core melt frequency contribution by initiator (simplified event tree-optimistic assumptions). 3 i , i i , , , i i i i 2.8 , 2.6 -
.j. -
2.4 - j PRA .- 2.2 3 Filtra C 2 / - - g 1.8 ' ' 2d 1.6 l
'~ No CV _
tS f Tu a 1.4 < 4 - Eg ( eg e.2 ,, - b 1 '; j - 0.8 < ; - 0.6 [ - 0.4
' I -
0.2 h i 0 ', """
,'^ Y ~7 ', ? r ,'
Trans Tde Tiosp Tsw Thdwt R Trup. S2 S1 A ATWS 0.osn Figure 6. Core melt frequency contribution by initiator (simplified event tree-pessimistic assumptions). 19
r----___________ _ _ _ __ _ _ _ _ _ _ _ - . . _ - - - - - - - - . - . - - -- 20 t, i i i i i i i i i i i i i 2.6 - 2.4 , PRA 2.2 _ u 2 M Filtra - 1B {$ 16 R Ex Eq CV - J$ No Cv
=- 1.4 - -
$E g,E 1.2
) -
O~ 1 - 0.8 - 0.6 - 0.4 - 0.2 {" O V '9 ,- -- P , , p ,q th 1A 18 1C ID 11A 110 11F 111 A 1118 iiiC 1110 1VA 1VF 1VG Accident segaence subclass 40888 Figure 7. Core melt frequency by sequence subclass (simplified event tree-optimistic assumptions). 2.8 ,, , , , , , , , , , , , , , 2.6 - 2.4 . 2.2 , Q PRA _ 2 M Filtra - gg 1.8 I IEtEqCV - S8 1.6 ~ i
!!)W 'h NoCV
=- we 1.4 Eg 1.2 .
ga. o 1 , 0.8 . 0.6
} .
I 0.4 0.2 "r - ! J . 0 r
, ,- y ? ', dq rk 1A 1B 1C 10 11A 118 11F 111 A 111B 1110111D 1VA 1VF 1VG Accident sequence subclass 40645 Figure 8. Core . - frequency by sequence subclass (simplified event tree-pessimistic assumptions).
20 I l l _ _ _ _ _ _ _ _ _ - _ _ _ _ _ _ _ _ _ _ _ _ _ . _ _ _ l
i The rapid depressurization of the primary con- Other potential 60urces of injection include ser- l tainment that accompanies venting (or contain- vice wster, condensate transfer pumps, and the ment failure) will likely result in saturated plant fire-water system. However, the first of conditions in the suppression ;ool. This can pro- these requires the opening of two MOVs located duce inadequate net postuve suction head in the reactor building. The latter two would re-(NPSH) or vonex problems for the low pressure quire entry into the reactor building to open man-pumps taking suction frem the suppression pool, ual valves or connect a hose in place of a blind resultinF in probable pump failure. This problem Dange. is ad:lressed in the Shoreham EOPS by instmeting the operators to realign the core spray pumi sue- Ensuring continued RPY injection after vem tion back to the condensate storage tank (CST). ing is further compheated in the existing equip-NOTE: At Shoreham, the core spray but not nwntpow path h the equ5 ment suWaMty qmuM gnen the post-venting environment in the LPCI pumps are capable of taking suc-tion from the CST. the reactor build,mg. l'he esist,mg equ pment now; path uses a 6-in. pipe that connects to llVAC ductwork inside the reactor building through a The perfonnance of this task requires the oper-neoprene sock If the primary conta,nment i were utors to bypass some interlocks with electrical vented using the Mn. line, this neoprene sock jumpers. For this reason, the Shoreham PRA gave andkr the llVAC ductwork w ould likely fail, pro-4 reto credit for realigning core spray pump suction ducing a sewn envimmnent in the reactor bum l to the CST before venting. Given failure of the ing. In addition to making the rmetor building low-pressure pumps caused by s enting, it may be uninhabitable for plant personnel, the effects of possible to initiate an alternative source of RPV this severe environment on the pumps. valves, injection. One likely source is the control rod and motor control centers hicated in ti c reactor drive (CRD) pumos. Ilowever, the CRD pumps buihhng are difficult to detenmne renl .tically. require pump cooling by the Reactor flullding Closed-Loop Cooling Water System (R11CLCW), which will has e previously isolated Work perfonned at ORNL (Reference 4) and all nonsafety loads (such as the CRD pumps) in (Sandia National Laboratory (SNL) Reference 51 response to the emergency safeguards feature indicates that survival of reactor building equip. (ESP) actuation signal. Operating the CRD ment may not be likely following s enting through pumps would require bypassing these isolation the existing lines. The details of this work are dis-interkicks with electrical jumpers; this is not cov- cussed in Appendix C. Under the pessimistic as-ered by procedures. Another possible source of sumptions concerning the effects of venting on RPV injection is the condensate and condensate the reactor building, venting using existing
. bomter pumps, but this depends on maintaining equipment has no benefit in preventing core inventory and vacuum in the condenser hotwell. damage.
21
CONTAINMENT RESPONSE ANALYSIS in additton to core melt frequency, contain- ability of early overpressure or containment iso-ment s enting also affects containment release / lation fullure, gis en that early venting through the failure modes. Examining the effects of venting SCS had failed. Failure of early venting was on the containment respmse to a severe accident quantified using the early time-phased tree by provides usefulinsights on what drives any summing the probabilities of the vented end-changes in possible consequences from a core states and subtracting the total from one. 't he ~ melt accident. For example, venting might delay probability of having a diiving force for a pulf re-or speed up the occurrence of a telease or contain- lease given early containment failure was gener-ment failme. Venting might also affect the path- ally taken to be the probability of RPV failure at way a release would take, either passing through high pressure, with this probability taken from or bypawing the suppression pool and the reactor the early time-phased tree. building. The discussion below describes the Shoreham containment response to a severe acci. This method of modeling containment re. dent both with and without containment venting. sponse to severe accident sequence. is valid and rnathematically correct, liow ever, it tendt to ob-scure the explicit sequence of events leading to a Simpilfled Containment Event particular end state. In addition, for some plani Tree Approach damage states such as lilA and lilD, the transfer from the early time-phased trees to the main event trees is not consistent with the plant damage Because of sire-limiting constraints in the , event tree analysis software used in the develop- state definition (this is discussed further in the ment of the Shoreham PRA, two basic types of I"II"*I"E '#Cli"""h containment event trees (CETs) were developed To make the sequence of events more readily
- m Reference 3. The first type is the early time-appm nt ud to avoid inconsistencies in the inod-phased phenomenological event tree that is only cling, the two types of event tree were combined quantified for plant damage states (PDSs) that in-for the present analysis. This makes sensitivity volve core melt prior to contamment failure studies easier to perform because the need to (plant damage states I and !!!). Events on this tree carry out intermediate quantification is elimi-span the time period f rom the onset of core degra' nated. Also, some events were escluded, climi-dation until just after vessel failure. The second nating irrelevant or insignificant sequences and type is the main CET, which overlaps the early reducing (simplifying) the size of the CETs, time-phased tree and models the containment re-which avoids limitations on the event tree size in sponse up to the point at which the sequence is re-covered or a release to the envimnment occurs, it theso h n is the only CET developed for PDSs 11 and IV be-Assumptions cause the inclusion of the potential for early con-tainment f ailure, by definition, obviates the need The source of these CETs (Reference 3) often i for the early time-phased trees
- provided no basis for the failure probabilities used in the early time-phased phenomenological For PDSs I and til in the Shoreham PRA, the event trees. In particular, the quantification of early events on the main CET (early venting containment isolation failure is problematic in through SCS, early containment failure, and puff that there are often inconsistencici, among the release driving force) were quantified on a condi- values used for different venting strategies within tional basis, using the early time-phased pheno- the same plant damage state. For example, in the menological event trees. For example, the SBO sequences (plant damage states 1B1 and probability of early containment failure on the 182), a containment isolation failure probability main CET was calculated as the conditional prob- of 5.1E-03 was used if the SCS was available, 22
I while 2.llMO was used otherwise. For other se- The second assumption is related to the first quences, the containment isolation failure proba- and intohes the likelihood of ex-vessel steam bihty was independent of the particular venting explosions occurring w hen the molten core debris strategy. For this report, the values given by Ref- makes contact with water, either on the pedestal crence 3 for the early time-phased events were noor or in the suppression pool. Appendix 1 of used unless there was an obvious question atout Reference i probabilistically analyzed the sikeli-the validity of the value. No attempt was made to hood of ex-vessel steam explosions and con-rigorously requantify these early esents because cluded that a steam esplosion of a magni'.ude that many of them, such as hydropn bums, pedestal could fail the containment was not post ible. The downcomer failure, and containment pressure reasons for this conclusion are as follows: follow ing vessel failure, invoh e a large degree of uncertainty; at this time, the probabilities used for
- There is insufficient ptemixing of these events are based primarily on engineering rnolten debris and water.
Judgement.
- There is insufficient energy to rapidly mix the molten debris and water on an For the post-vessel failure events, Referenec 3 C*PI "' IVC II* C "C"IC-provided more details of the quantiGcation pro-cess. Ilowever, inconsistencies were still found.
l '" I' "" I"'"E'i'"I 'fi EC' E I"' *. ." tiating an explosion. end the actual steps undertaken in quantifying each event in each sequence could nN always be
- a h a lack d a strong coupung reproduced. Appendix B gives the rationale used rnechanism, such as slog impact, that in quantifying the CETs. Some of these values are could transfer enough kinetic energy especially sensitive to the key assumptions out-lined below.
to pnemte a nunk ccpaNe of pene trating the containment wall. Therefore, for sequences where venel failure The first assumption is based on the unique occurs at low pressure, all debris that nows out of configuration of the Shoreham pedestal region, the vessel is auumed to be quenched in the sup-which has a corium ring installed around the pe- pression pool with no deleterious effects on the rimeter and four pedestal downcomers connect- containment. ing the area beneath the venel to the suppression pool. if the RPV fails at low pressure, the assump. The third assumption is that venting through tion is that no significant amount of core debris is the existing lines automatically leads to bypan of dispersed outside the pedestd region. Instead, es- the suppression pool. This is based on the con-sentially all of the debris is assumed to flow out of tainment-vent analysis in Appendix K of the vessel onto the floor of the pedestal cavity. Reference 2. The quantificatior, of containment The debris is then directed into the suppression venting through the existing vent lines is based on pool by the corium ring through the pedestal the auumptien that both the wetwell and drywell downcomers. This assumption was stated in Ref. vent lines would have to be opened to relieve the crence 3 but was not followed consistently, as dis- pressure cxpected during a TW sequence. This cussed in Appendix 13. The net result is that, for assumption is important because bypass of the sequences where the vessel fails at low preuure, suppression pool would occur every time venting debris-heating of the drywell atmosphere and in- was carried out using existing equipment because ternals is auumed not to occur. All the core debris the now path through the drywell vent would pro-is assumed to be quenched in the suppression vide very little flow resistance compared to the pool where ex-vessel recovery, which entails the path through the suppression pool. Ilowever, this establishment of a coolable debris bed,is be- assumption was not followed in the CET evalua. lieved to be assured. tion in Reference 3, wheie suppression pool 1 23
f ; I
; bypass was assigned a probability of 3.0E-01 for sure rise expecOd to occur during an ATWS no- ; sequences with successful venting through the quence. This is based on the small sl e (6 inches) existing lines, of the Shoreham vent lines ne result of this as-sumption is that not venting and venting through The founh assumption is that venting through the existing lines are equivalent in terms of the ! the existing lines results in a release of steam and challenge presented to containment integrity by I noncondensible gases to the reactor building, ne ATWS sequences.
release to the reactor building is caused by the rupture of the ventilation ducting used in the COmparlton of SCET and PRA venting flow path. As discussed earlier, the ef- gg, fects of an adverse reactor building atmosphere on vital equipment are uncertain. Reference 3 j states that use of the existing vent lines to relieve Tables 14 thro.igh 19 show a comparison of the containment pressure is assumed to fall all equip- release mode conditional probbilities as calcu. ment located in the reactor building. However, lated in Reference 3 and in this relort using the this assumption was not followed consistently in SCETs with the optimistic reactor building as-Reference 3, as can be seen by examining the sumptions. Overall, the results from the SCETs main CET for PDS lA with venting through the and the PRA cornpare well with one another. Dif-e existing lines (Figure D-1 of Reference 3), For ferences that exist result in large part from the in. the sequence with successful in-vessel recovery,- clusion of external events (such as carthquakes but subuquent loss of containment heat removal, and fires)in the PRA event tree development, t,uccessful venting through the existing lines whereas these were excluded in the present leads to an "OK" end state, llowever, if venting analysis. failed all equipment in the reactor building, cool-ant injection to the vessel would be lost and core The overall agreement between the values tak-melt would ensue, en from Reference 3 and the results calculated from the SCETs is quite good for PDS IA. The in addition, the reactor building atmcsphere frequency of several of the 11 release modes is less following venting would be severe enough to pre- than what was calculated in Reference 3 because - i vent personnel entry into the building. This of the exclusion from the SCETs of vapor sup-would preclude the use of such attemative means pression failure and overpressure containment of injection as condensate transfer pumps through fa!!.we prior to vessel breach. The exclusion of the RHR and core spray injection lines, and the ths .e two events is based on their low frequency . fire pumps through a hose connection to the RilR of occurrence, llowever, this assumption is eva- , system because operator entry into the reactor luated further in the consequence analysis be-building is required to establish these flow paths, cause the B release modes, due to their relative ne only available means of attemative injection severity, are significant co-Hbutors to the overall would be service water through the ultimate heat risk. sink connection to RilR (assuming the two MOVs located in the reactor building were not here are several notable differences between disabled by the adverse environmenti ad the . the Reference 3 results for PDS 181 and those condensate pumps.1\vo restrictions would limit calculated using the SCETs. First of all, early (in- - the use of the latter, namely a maximum makeup vessel) and late (ex-vessel) recovery have higher flow rate of 1000 gpm to the condenser hotwell probabilities in the SCETs than in Reference 3-L from the CST, and the requirement to maintain a because of the exclusion of extemally initiated vacuum in the main condenser. sequences. This leads to a relative increase in the ! frequency Ithe less severe A and D release The fifth assumption is that venting through modes. Second, the frequency of release mode C2 the existing lines is inadequate to relieve the pres- is calculated to be 0.0 because of the assumption, D 24
I Table 14. Comparison of release male probabilities for Plant Damare States I A.1B 1 and 1112 l 3 Plant Damge States __ IA 181 lil2 Release SCET PRA SCET PRA SCET PRA Mode' Probabihty 1lrobability Probability Probability Probability Probability AI 7.89E4)1 7.89E4)1 4.90! Mil 2.00E-01 1.10lMil 2.00E412 A2 7.84 E44 7.84E44 4.86E44 1.991M4 1.09E4L1 1.99E-05 A3 5.691M16 5.69E456 3.53E-06 1.43fM16 7.92iM17 1.441607 B1 5.291Mb 5.33E-(Ki 5.72iM4 3.93E44 8.211M4 1.2iMO 11 2 6.28fM)6 6.32E4)6 6.79tM4 7.88E44 9.811M4 2.29E413 4 B3 1.671M)6 1.68E-06 1.80lM4 1.731M4 2.57IM4 5.02E44 11 4 9.50f!-09 9.66E-08 5.77E-07 1.0$fM)$ 9.50E-06 3.78fM15 11 5 5.95E-09 1.791M18 3.62E-07 3.751606 6.86E46 1.201605 11 6 8.08E- 10 6.281M)9 4.90E-08 7.68tWO7 8.55E417 2.64E4)6 B7 5.24fM4 5.28FM4 4.68E.44 7.971M4 1.00lMO 3.37E44
!!8 4.57E-44 4.59E44 4.081M4 1.27E40 8.85IMn4 5.1 ilM4 11 9 3.30E44 3.331M4 2.951M4 - 6.83EM4 6.30fM4 2.75tM4 B10 9.411M17 9.56E-06 4.73fM)7 2.14E415 1.16E-05 1.071b05 Bil 5.09E-07 1.12E4)6 2.$6E-07 5.96fi-06 7.34 E-06 2.65E466 B12 1.60E417 1.271M)6 8.03E-08 3.21fb06 2.09E46 1.49E416 Cl 8.90E416 9.00E4)6 3.74E40 1.841!-01 4.20lM12 8.211M12 C2 2.72E44 2.75fM4 0 3.50fMil 4.14FM12 4.42iM)3 C3 2.18EM4 2.20E44 2.00h40 4.38E412 5.901603 1.59fM12 C4 1.18E44 1.19E44 1.07E40 2.36fM12 3.18IMO 8.581MO C5 4.29E4N 3.5011-09 2.61 E-06 4.44E415 1.141M15 1.27E415 C6 4.34I M N 5.241M)9 2blE-06 1.081M4 1.231M15 3 00E-05 C7 4.79E-10 4.378-10 2.91 E-07 4.941Ml6 1.29fM16 1.90!h06 C8 4.81 E-10 5.34E-10 2.92E417 1.20E415 1.34LM)6 2.85fM16 C9 2.05E4)6 2.08E-06 1iMIMIS 1.60E-03 2.081M)$ 9 22E415 C10 1.llE4)6 1.12 E-06 5.56E-06 8.63fM4 1.12iM)5 4.97E415 D1 5.261M12 5.28tM12 5.69E412 3.611M12 3.241 Mil 1.57E-01 D2 7.33E-03 7.35 E-03 1.07E412 l.90lMO 4.03E412 2.431602 D3 1.48E4)I 1.481!-01 4.13E411 1.54E411 4.28!b01 6.87E-01 D4 1.08E-05 1.08 E-05 3.16E-05 1.10fM)5 3.12tMIS 4.991M)5
- a. Release mode classes are as follows: .
- A-Recovered irt-vessel 1
e Il-Early release of radioactise material i e C-Late release of indioactive material e D-Recos cred es-v essel, t 25 I
Table 15. Comparison of release mode probabilities for Plant Damage States Icl. IC2 and ID Plant Damage States ICI IC2 ID Release SCET PRA SCET PRA SCET PRA Mode' Probability Probability Proba_bility Probability Probability Probability Al 0 0 0 0 - - A2 0 0 0 0 - - A3 0 0 0 0 - - til 3.08E-(M 1.52E-03 1.06E-05 2.63E-05 8.32M4 5.5tE44 11 2 9.02E4M 7.19E44 1.64E4)5 1.66E415 9.85MM 1.32E-03 11 3 3.32E4)$ 6.26E4)5 2.87E46 3.17E46 2.62E4M 2.51 E44 B4 7.00E44 1.79E4)3 5.(ME-06 2.07E4)$ 4.541M18 3.39E415 B5 4.62E44 3.90E44 3.33606 3.81E46 2.851M18 1.36E415 B6 6.07tMi$ 1.16E44 4.371M17 1,351M)6 3.861M)9 2.39E-06 11 7 3.(ME4)2 1.50E-01 1.0$E-03 2.60E-03 8.40lM)6 5.57E4)6 11 8 8.61 E-02 6.58E4)2 1.34E-03 1.31 E-03 7.31 E-M 1.06E4)5 B9 6.57E-03 1.16E412 5.671MM 6.47E44 5.29E4)6 5.23E4)6 1110 6.93E412 1.78E-01 4.99E44 2.05E4)3 4.59E-10 3.43E4)7 m Bil 3.98E412 2.71 B-02 2.86E4M 2.46E44 2.49E-10 1.ll E417 B12 1.20E-02 2.30E-02 8.65E4)$ 2.67E4M 7.79E-I l 5.00E-08 Cl 0 0 4.28&OS 2.24MM 8.85E-05 2.36E4)1 C2 0 0 1.31 E-03 1.75E4)3 3.98E-07 1.82E413 C3 0 0 1.05E4)3 6.52E4M 3.191M)7 2.32E4M C4 0 0 5.64E4M 3.51E4M 1.72E417 1.25E44 C5 4.89E-06 6.21E4M 2.53E-10 4.96E-10 6.16E418 1.311MM s C6 5.15E-06 1.06E-03 2.67E-10 7.48E-10 6.22E-08 3.90E4M C7 5.50E-07 8.12E-05 2.85E-Il 6.21 E-11 6.86E-09 2.17E-05 C8 5.65E4)7 1.06E4)4 2.93E-11 7.62E-11 6.90E-09 3.61IM)$ C9 1.44E-03 4.91 E4)$ 7.44E-08 1.78E-07 3.01 E-09 1.12E4)5 C10 7.73E4M 2.65E-05 4.01 E -08 9.58E-08 1.62E-09 6,01E416 1,1 0 0 8.80lM)1 8,79E411 3.40lM)3 2.74E4)3 D2 0 0 9.79E-02 9.78E-02 1.02E4)2 7.80E413 D3 7.441M)1 5.33E-01 1.52E-02 1.26E4)2 9.84tM)' 7.49E-01 D4 7.51E-03 5.32E-03 1.10E-06 9.19E-07 7.16E4); 1.85E4)5
- a. Release mode classes are as follows:
- A-Recovered in-vessel
- Il-Early release of radioactive material
- C-Late release of radioactive material
- D-Recovered ex-vessel.
b 26 t
Table 16. Comparison of release mode probabilities for Plant Damage States IIA,llB and !!F Plant Damage States IIA IIB llF Release SCET PRA SCET PRA SCET PRA hkxie' Probability Probability Probability Probability Probability Probability B1 4.45 E-01 2.02E-01 4.45E-01 2.02E-01 - - B2 3.37E4)3 2.36E-01 3.37E413 2 36E-01 - -
- 83 1,12E-03 1.13E-02 1.12E-03 1.13E412 - -
B4 4.45E-02 3.38E-02 4.45E412 3.38E-02 - - B5 2.75E4)3 1.37E412 2.75 E-03 1.37E4)2 - - B6 2.75E-03 2.50E4)3 2.75E4)3 2.50E-03 - - B7 4.45E4)1 2.02E-01 4.45E411 2.02E-01 -- - B8 2.25E-03 2.25E4)1 2.25E4)3 2.25E-01 - - B9 2.25E413 2.25E-02 2.25E413 2.25E-02 - - B10 4.45E-02 3.38E-02 4.45E412 3.38E-02 - - Bil 0 1,12E-02 0 1.12E4)2 - - B12 5.50E-03 5.00E-03 5.50E4)3 5.00E-03 - - C1 - - - - 0 0 C2 - - - - 9.96E-01 9.95E4)1 C3 - - - - 2.34E-03 3.30E-03 C4 - - - - 1.22E4)3 1,75E4)3 D1 - - - - 0 0
- a. Release mode classes are as follows:
- A-Recovered in-vessel e B-Early releaw of radioactive material e C-Late release of radioactise material e D-Recovered ex-vessel.
27
Table 17. Comparison of release mode probabilities for Plant Damage 3tates !!' A and 11111 Plant Damage States Illa 11111 Release SCET PRA SCET PRA Mode' Probability Probability Probability Probability Al - - 7.89E-01 7.891 Mil A2 - - 7.841MM 7.841MM A3 - - 5.69E4b 5.691 M ki 11 1 5.41E4M 5.51 E44 1.76E46 1.87E4)6 11 2 1.29E4)3 1.32E4)3 2.10E-06 2.131Mb 11 3 2.46E4)4 2.5 iE4M 5.56E- 07 5.66E4)7 11 4 0 3.391M15 9.54 E-09 9.751 418 11 5 0 1.361 MIS 5.981M)9 1.80lM18 11 6 0 2.39E46 8.12E-10 6.341M)9 11 7 5.461M)6 5.57E4)6 1.75E41 1.85E4M 11 8 1 (ME4)5 1.06E415 1.531MM 1.57E41 11 9 5.13E4ki 5.23&O6 1.10E41 1.091MM 1110 0 3.43E417 9.451M17 9.65tMb 1111 0 1.llE-07 5.12E417 1.161Mb 1112 0 5.00E418 1.61 E417 1.251M)6 C1 0 2.94E413 8.95E4)6 9.04IM)6 C2 0 2.27E-05 2.73E41 2.761MM C3 0 2.90E4b 2.19E4M 2.211MM C4 0 1.56E416 1.181MM 1.19E44 C5 0 1.63606 4.31 E4)9 3.51 E-09 C6 0 4.87E4)6 4.36E-09 5.28 & O9 C7 0 2.72E-07 4.81 E-10 4.39E-10 C8 0 4.51 E-07 4.83E-10 5.38Fcl0 C9 0 1.40E-07 2.06E4)6 2.10E-06 CID 0 7.51E4)8 1.12b06 1.13E4ki DI 8.83E44 3.59E-03 5.281M12 5.281M12 D2 1.00E-02 1.02E-02 7.35E-03 7.35FA13 D3 9.87E4)I 9.811M)1 1.491Mll 1.49fM)1 D4 7.18E-05 2.43E-05 1.08E4)$ 1.08E415
- a. Release mode classes are as follows:
- A-Recm cred in-vessel
+ B-Early release of radioactive material
- C-1. ate rel.ase of radioactive material
+ D-Recovered ex-vessel.
28
Table 18. Comparison of reletse mode pmbabilities for Plant Damage States IllC e.nd tilD Plant Damage States tilC _ lilD Relcase SCET PRA SCET PRA Mode
- Probability Probability - Probability Probability B1 8.32E44 8.05E-04 0 0 11 2 9.85E-(M 1.05E413 3.47E413 0 B3 2.62 E-(M 2.63EAM 8.91 E-05 0 B4 4.54E-08 3.89E-05 1.60E-03 0 B5 2.85E-08 8.45E-06 1.78E4)3 6.09E-01 B6 3.86E-09 2.52E-06 1.76E-04 2.03E-01 B7 8.40E-06 8.14E-46 0 0 B8 7.31 E4M 8.10E4m J.42E-05 0 89 5.29E4m 5.18E4M l .80E-06 0 BIO 4.59E-10 3.93E-07 1.62E415 0 Bil 2.49E-10 5.99E418 1.62 E-05 4.10E-03 B12 7.79E-l l 5.10E418 3.56E-06 4.10E413 Cl 8.85 E-05 2.94E-02 8.85E-03 0 C2 3,98E-07 2.27E4M 3.98E415 0 C3 3.19E-07 2.90E4)5 3.19F415 0 C4 1.72E-07 1.56E-05 1.72E-05 0 C5 6.16E-08 1.63E4)5 -- -
C6 6.228-08 4.87E-05 - - C7 6.86E-09 2.72E-06 - - C8 6.90E-09 4.51 E-06 -- - C9 3.01E4)9 1.40E-06 - - CIO 1.62E-09 7.51E-07 - - D1 3.40E-03 3.49E4)3 8.85 E-01 0 D2 1.02E-02 9.96E4)3 9.84E412 0 D3 9.84E411 9.55E4)I - - D4 7.16E-05 2.36E4)5 - -
- a. Release mode classes are as follow s:
- A-Recovered in-vessel
- B-Early release of radioactive material
- C-Late release of radioactive material
- D-Recovered ex-vessel.
29
Table 19. Comparison of release male probabilities for Plant Damage States IVA. IVF and IVO Plant Damage States _ IVA IVF IVO Release SCET PRA SCET PRA SCET PRA Mode' Probability Probability Probability Probability Probability Probability B1 3.60E-03 7.20E44 - - - - B2 1.27E-.03 2.30E-(M - - - - 11 3 1.26E44 5.(M E-05 - - - - B4 4.00E-03 7.20E-03 - - - - B5 7.20E44 1.35E4)3 - - - - B6 2.80E44 4.50E44 - - - - B7 3.56E-01 7.13E4)2 - - - - B8 1.14E-O f 2.27E4)2 - - -- - 13 9 2.49E-.02 4.99E-03 - - - - B10 3.96E411 7.13E4)I - - - - B11 4.45E412 8.91E4)2 - - - - B12 5.45E-02 8.91 E-02 - - - - C1 - - 1.51E-02 9.91 E-01 - - C2 - - 1.02E412 8.91 E4)3 0 9.91E4)I C3 - - 3.02E-03 6.50E-05 0 8.98E-03 C4 - - 1.63 E-03 3.50E-05 0 3.50E-05 DI - - 8.73E-01 0 9.00E-01 0 D2 - - 9.70E-02 0 1.0GE411 0
- a. Release mode classes are as follows:
- A-Recovered in-vessel
*- B-Early release of radioactive material e C-Late release of radioactive material e D-Recovered ex-vessei.
in the SCET, that late containment failure is crease in the frequency of the C release modes certain given failure of event OT (overtempera. and an increase in the frequency of the D modes, ture in the drywell, see Appendix B for further which represent ex-vessel recovery, discussion). Third, the assumption that core de-bris is quenched in the suppression pool if the Similarly, the calculated failure probabilities vessel fails at low pressure leads to a relative de- for early and late recovery in PDS IB2 dif fer from 30
_ _ _ _ _ _ _ _ _. __ m._ _ _ _ _ _ _ . _ l those in Reference 3 because of the exclusion of later w hen estimating the probability that the con-1 externally initiated sequences. The higher proba- tainment break site is small. Thus, because the i oility of successful in-vessel recovery leads to a above release males are characterized by a large relatise increase in the frequency of the A release break, their frequencies will be increased if fail-modes. Again, there is a reduction in the frequen- ute of containment isolation is included in the cy of the C release modes because of the assump- quantification. This is mitigated somewhat by the j tion that core debris is quenched in the inclusion of containment isolation failure etplic-suppression pool if the vessel fails at low illy in *he break size quantification. Ilowever, be-pressure. cause containment isolation failure is also 3 included in the quantincation of suppression pool I Reference 3 subdivides PDS IC based on Filtra bypass and reactor building retention, the prob-availability by defining PDS ICl for Class IC se- lem still remains. As for PDS lA, the effects of
, quences for which the SCS has failed and PDS cxcluding vapor suppression failure and very ' early cornainment failure from the SCETs are IC2 for Class IC sequences in which operation of the SCS has not yet been demanded. Reference 3 evaluated in the consequence analysis, uses a probability of 1.0E-02 for event OT if the SCS is fr.ded and 9.9E-01 if SCS is available. No There is generally goal agreement between the reason ror this difference is given, nor could one SCET results for plant damage states ll A and Ill! .
be inferred. For this report, a value of 9.9E-01 and those presented in Reference 3. Any marked was used for both plant damage states because the differences can be attributed to changes in the RPV is likely to fail at high pressure regardless of quantification of suppression pool bypass and whether or not venting to the SCS is successful, reactor building fistion product retention, as dis-This results in a marked increase in the frequen. cussed in Appendix 11. cies of release modes C9 and C10 for PDS ICl. Again, the assumption of debris quenching in the A s explained in Appendix B, the quantification suppression pool Iollowing vessel failure at low of the SCET for PDS lil A (large RPV ruptures)is pressure causes a relative decrease in the frequem a major departure from that in Reference 3.The as-cy of the C release modes. Other differences are sumption here that ex-vessel recovery always attributable to changes in the quantincation of in- succeeds because of debris-quenching in the dividual events in the SCET, as outlined in suppression pool leads to a substantial reduction Appendix B. in the frequencies of the C release males, with an accompanying increase in males D3 and D4.The Notable differences for PDS ID are the marked large reduction in the B release males is due to the decrease in the frequencies of several of the C re. assumption that early containment failure is far lease modes due to the significantly lower value less likely than in Reference 3 because the initial used in this report for the failure probability of - RPV rupture precludes later vessel failure at high ex-y-ssel recovery (see Appendix B), and the de, pressure. This contrasu with the quantification in crease by two to three orders of magnitude in the Reference 3, where high pressure vessel failure is frequencies of release males B4 through B6 and assigned a nonzero probability. L B10 through B12, which involve large breaks in the containment. This latter decrease can be at- The differences between the SCET and Refer-tributed in part to the exclusion of vapor suppres- ence 3 results for PDS lilB parallel those for PDS sion failure from the SCETs and the subsequent I A. This is reasonable because PDS 1118 is the early overpressure failure prior to vessel breach, LOCA analogue to PDS l A. as discussed under PDS IA above. Also, Reference 3 includes entainment isolation fail- The differences between the SCET and ure as a contributor when calculating the proba- Reference 3 results for PDS IllC parallei dose for bility of early containment failure, even though PDS ID, As above, this is reasonable given that the failure of containment isolation is included PDS lilC is the LOCA analogue to PDS ID. 31
~ _. . __ -_ __ __ __ _ _.. _ ._ _ _ _ _ _ _ _ _ _ _ _ _ _
Like PDS lilA above, there are major differ- Effects cf Venting en the
. ences between the SCET and Reference 3 in the quantification of PDS IllD. These differences are Containment Release Mode detailed in Appendix B. The major difference is The SCET results for each of the three venting the exclusion of seismic RPV rupture from the strategies and each of the two sensitivity cases for initiating plant damage state in this report, in Ref-the post-venting reactor building environment erence 2, seismic RPV mpture contributed 80%
are shown in Tables 20 through 25 and summa-of the Class IllD core melt frequency and was as-rized in Figure 9. sumed to include coincident failure of the dry-well, which automatically bypassed the SCS and The primary effects of the ORNL and SNL cal-the suppression pool. culations on the present analysis are to modify the quantincation of reactor building fission product retention. First of all,it is assumed that the The differences between the SCET results and RBSYS is failed by the harsh reactor building en-those in Reference 3 for PDS IVA are due to the vironment that results from primary containment use of 5.0E-01 as the probability that the early failure during a severe accident sequence. This cordainment failure results in a large break (ver- tends to lessen the severity of the release because sus 9.0E-01 in Reference 3), and the different it allows time for fission product depos! tion in-quantification of reactor building fission product side the reactor building.
- retention, as outlined in Appendix B.
NOTE: The ORNL report assumed that the RBSYS would continue to operate follow-As explained in Appendix B. Reference 3 ing c ntainment failure, assumed that the vessel failed at low pressure for However, the failure of the refueling floor stJ-I'DS IVF. Appendix B explams why this assump-t,on was considered invalid in the SCET, which ing, coupled with the increased probability of i
$bal hydrogen burns inside the reactor build, results in a much higher probability of late con.
ng, are competing effects, justifying a conserva-tainment failure due to high temperatures in the tive requantification of reactor building fission SCET than in Reference 3. Also, as discussed in product retention for the sequences involving ear-Appendix B, credit is given for the use of attema-ly containment failure. For these cases, event RB tive luject,on i systems in establishmg a coolable is taken to have a failure probability of 1.0 (reac-debris bed following vessel failure. This leads t tor building DF of unity). For cases of late ther-
- a nonzero frequency for irlease modes D1 and mal failure of the primary containment, the D2, Likewise, the failure of long-term contam-failure location is assumed in Reference 3 to be ment heat removal is quantined based on the use =
the drywell head. The assumption that the refuel-of attemative systems (other than RilR), produc-ing floor siding falls due to overpressurization as ing a failure probability that is less than umty, a result of the release to the refueling floor also leads to a requantification of event RB; the failure Th: quantification of the SCET for PDS IVO is radically different from that in Reference 3 be- Reference 4 also discussed the effects of both cause of the assumption, used in the SCET and containment failure and venting on the atmot explained in Appendix B, that the vessel does not sphere inside the teactor building. ORNL mod-fait at high pressure, This leads to successful ex- eled the Shoreham secondary containment < vessel recovery because of debris-quenching in . building using the MELCOR computer code with - the suppression pool, thus shifting all releases each of the six main levels of the reactor building from release mode C to release mode D. divided into two subregions. The mass flow rates 32
T;ble 20. Shoreham simplified event tree resuL., with Filtra (optimistic Ril assumptions) Release Mode Conditional Release Probability Mode Release Mode (9) Frequency I A1 Recovered in-vessel, no CF 61.9 2.20E-05 A2 Recovered in-vessel, vented 0.1 2.18E48 A3 Recovered in-vessel, late CF 0.0 1.59E-10 Subtotal- Recovered in-vessel 61.91 2.2 l E-05 B1 Early small-slow CF, pool + Rx-B DF 0.1 3.31 E-08 B2 Early small-sAw CF, Rx-B DF only 0.0 4.54FcO9 B3 Early small-s.vv CF. no DF 0.0 1.04fbO9 11 4 Early large-slow Cr, pool + Rx-B DF 0.0 4.44 E-09
!!5 Early large-slow CF, Rx-Il DF only 0.0 5.8 I E-10
!!6 Early large-slow CF, no DF 0.0 2.92fblo B7 Early small-mod CF, pool + Rx-B DF 0.5 1.68 E-07 !
B8 Early small-mod CF, Rx-B DF only 0.2 5.41E-08 11 9 Early small-mod CF, no DF 0.1 1.8tE-08 B10 Early puff CF, pool + Rx-B DF 0.4 1.43E-07
!!II Early puff CF, Rx-B DF only 0.0 1.62E48 1112 Early puff CF. no DF 0.1 l _.96E-08_ _
Subtotal- Early release 1.30 4.63E-07 Cl Late vent, controlled relcue 5.1 1.80E-06 C2 Late vent, uncontr911td release 0.1 3.10E-08 C3 Late vent with DW (ali(OT), Rx-B DF 0.0 1.37E-08 C4 Late vent wLn DW lalt (OT), no Rx-B DF 0.0 7.36FA9 C5 Late OP CF with pool DF 0.0 2.37E-12 C6 Late OF CF without pool DF 0.0 2.48E- 12 C7 Late puff CF, with pool DF 0.0 2.6SE-13 - C8 Late puff CF, without pool DF 0.0 2.68E-13 Cy Laie OT CF with Rx-B DF 0.0 8.24E-11 CIO Late OT CF without Rx-B DF 0.0 4.50E-I l Subtotal- Late release 5.20 1.85E-06 D1 Recovered ex-vessel, vented 9.8 3.48E-06 D2 Recovered ex-vessel, vented NG release 1.3 4.60E-07 D3 Recovered ex-vessel, no CF 20.5 7.30E-06
.D4 Recovered ex-vessel, late CF 0.0 6.50E-10 Subtotal- Recovered ex-vessel 31.56 1.128-05 Total 100.0 3.56E-05 33
Toble 11. Shoreham simplified event tree rvsults, with containment venting through existing equipment (optimistic Ril assumptions) Release Mode Conditional Release Probability Mode Release Mode (%) Frequency A1 Recovered in-vessel, no CF 58.6 2.20E-05 A2 Recovered in-vessel, vented 0.1 1.991M18 A3 Recovered in-vessel, late CF 0.0 2.211MN Subtotal- Recovered in-vessel 58.62 2.21 E4)$ 11 1 Early small-slow CF, pool + Rx-B DF 2.9 1.08E4)6 B2 Early small-slow CF, Rx-B DF only 0.1 5.43fM18 11 3 Early small low CF, no DF 0.0 4.85E49 B4 0.3 1.321M)7 Early large-slow CF, pol + Rx-B DF B5 Early large-slow CF. dx-li DF only 0.1 2.86tW18 B6 Early large-slow CF, no DF 0.0 9.08E-09 117 Early small-mod CF, pool + Rx-B DF 4.1 1.561M)6 11 8 Early small-mod CF, Rx-Il DF only 1.0 3.64E-07 B9 Early small-mod CF, no DF 0.2 5.88E-08 B10 Early puff CF, pool + Rx-Il DF 1.9 7.20lM)7 Bil Early puff CF, Rx-!! DF only 0.4 1.511M)7 B12 Evly puff CF, no DF 0.3 1.(ME4L' Subtotal- Early release 11.34 4.27E4)6 Cl Late vent, controlled release 0.0 0 C2 Late unt, uncontrolled release 0.0 2.211MN C3 Late vent with DW fall (OT), Rx-3 DF 0.9 3.511M17 C4 Late vent with DW fail (OTK no RvB DF 0.0 7.51 E4N C5 Late OF CF with pool DF 0.0 1.76E4N C6 Late OF CF without pocl DF 0.0 2.57E4)9 C7 latte puff CF, with pool DF 0.0 2.18E-10 C8 Late puff CF, without pool DF 0.0 2/ A 10 C9 Late OT CF with Rx-B DF 0.0 5.61E4)9 C10 Late OT CF without Rx-B DF 0.0 3.02E4N Subtotal- Late release 0.99 3.74E417 D1 Recovered ex-vessel, vented 0.0 0 D2 Recovered ex-vessel, vented NG release 0.3 9.83E-08 D3 Recovered ex-vessel, no CF 28.7 1.0S E-05 D4 Recovered ex-s essel, late CF 0.0 1.09F418 Subtotal- Recovered ex-vcuel 29.01 1.09E-05 , Total 100.0 3.76E-05 34 L__________________________________ _
T;ble 22. Shoreham simplified event tree results, with no containment venting (optimistic Ril assumptions) Release McJe Conditional Release Probability Mode Release Mode (9 ) Frequency A1 Recovered in-vessel, no CF $0.3 2.20E-05 A2 Recovered in-vessel, vented 0.0 0 A3 Recovered in-vessel. late CF 0.1 2.21 E-08 Subtotal- Recovered in-vessel 50.33 2.21 E-05 B1 Early small-slow CF, pool + Rx-B DF 9.1 3.99E-06 B2 Early small-slow CF, Rx-B DF only 0.2 7.63E-08 B3 Early small-slow CF, no DF 0.0 1.22E-08 B4 Early large-slow CF, pool + Rx-Il DF 1.0 4.22E417 11 5 Early large-slow CF, Rx-Il DF only 0.1 4.66E-08 11 6 Early large-slow CF, no DF 0.1 2.71 E-08 B7 Early small-mod CF, pool + Rx-B DF 10.2 4.47E-06 B8 Early small-mod CF, k. x-B DF only 0" 3.78E417 B9 Early small-mod CF, no DF 0.2 7.35 E-08 B10 Early puff CF. pool + Rr-B DF 2.3 1.01E-06 Bil Early puff CF, Rx-B DF only 0.3 1.51 E4)7 B12 Early puff CF, no DF 0.3 M)E417 Subtotal- Early release 24.64 1.08E-05 Cl Late vent, controlled release 0.0 ., C2 Late vent, uncontrolled release 0.0 0 C3 Late vent with DW fail (OT), Rx-Il DF 0.0 0 C4 .2te vent with DW fail (OT), no Rx-B DF 0.0 0 C5 Late OF CF with pool DF 0.0 1.81 E4)9 C6 late OF CF without pool DF 0.0 2.65E-09 C7 Late puff CF, with pool DF 0.0 2.25E-10 C8 Late puff CF, without pool DF 0.0 2.7 I E-10 C9 Lele OT CF with Rx-B DF 0.0 1.88E-08 C10 Late OT CF without Rx-B Di 0.0 1.OllM18 Subtotal- Late release 0.08 3.38E-08 D1 Recovered ex-vessel, vented 0.0 0 D2 Recovered ex-vessel, vented NG release 0.0 0 D3 Recovered ex-vessel, no CF 24.7 1.08E415 Di Recovered ex-vessel, late CF 0.2 1.09E-07 Subtotal - Recovered ex-vessel 24.91 1.09E415 Total 100.0 4.38E-05 35 1
T;ble 23. Shoreham simplified event tree results, with Filtra (pessimistic RB assumptions) Release Mode Conditional Release Probability Mode Release Mode (ek) Frequency AI Recovered in-vessel, no CF 61.9 2.20E-05 A2 Recovered in-vessel, vented 0.1 2.19E-08 A3 Recovered in-vessel, late CF 0.0 0 Subtotal-Recovered in-vessel 61.91 2.21 E-05 B1 Early small-staw CF, pool + Rx-B D 0.1 3.31 E-08 B2 Early small-slow CF, Rx-B DF only 0.0 3.58E-09 B3 Early small-slow CF, no DF 0.0 2.01 E-09 B4 Early large-slow CF, pool + Rx-B DF 0.0 4.44E-09 BS Early large-slow CF, Rx-B DF only 0.0 0 B6 Early large-slow CF, no DI- 0.0 8.74E-10 B7 Early small-mod CF, pool + D' -B DF 0.5 1.69E-07. B8 Early small-mod CF, Rx-b DF only 0.0 1.27E-08 B9 Early small-mod CF, no DF 0.2 5.95E-08 BIO Early puff CF, pool + Rx-B DF 0.4 1.4' E-07
-Bil Early puff CF, Rx-B DF only 0.0 0 B12_ Early pff CF, no DF 0.1 3.58E-08 Subtotal-Early release 1.30 4.63E-07 Cl Late vent, controlled release 5.1 1.80E-06 C2 Late vent, uncontrolled release 0.1 3.10E-08 C3 Late vent with DW fail (OT), Rx-B DF 0.0 1,17E-10 C4 Late vent with DW fail (OT), no Rx-B DF 0.1 2.09E-08 C5 Late OP CF with pool DF 0.0 2.38E-12 C6 Late OP CF without pool DF 0.0 2.498-12 C7 Late puff CF, with pool DF 0.0 2.68E-13 C8 Late puff CF, without pool DF 0.0 2.74E-13 C9 Late OT CF with Rx-B DF 0.0 0 C10 Late OT CF without Rx-B DF 0.0 1.29E-10 Subtotal-Late release 5.20 1.85E-06 Di Recovered ex-vessel, vented 9.8 3.48E-06 D2 Recovered ex-vessel, vented NG rafree 1.3 4.60E-07 D3 Recovered ex-vessel, no CF 20.5 7.33E-06 D4 Recovered ex-veswl, late CF 0.0 8.03E-10 Subtotal-Recovered ex-vessel 31.56 1.12E-05 Total 100.0 3.56E-05 36
TCble 24 Shoreham simplified event tree results, with existing equipment venting (pessimistic RB assumptions) l l Release l Mode Conditional Release Probability Mode Release Mode (91) Frequency A1 Recovered in-vessel, no CF $0.2 2.20E415 A2 Recovered in-vessel, s ented 0.0 0 A3 Recoven !in-vessel, late CF 0.0 0 Subtotal-Recovered in-vessel 50.21 2.20E-05 B1 Early small-slow CF, pool + Rx-B DF 2.5 1.08E-06 B2 Early small-slow CF, Rx-B DF only 0.0 3.60E-09 B3 Early small-slow CF, no DF 0.1 5.56E-08 B4 Early large- slow CF, pool + Rx-B DF 0.3 1.32E-07 B5 Early large-slow CF, Rx-B DF only 0.0 0 B6 Early large-slow CF, no DF 0.1 3.76E-08 B7 2arly small-mod CF, pool + Rx-B DF 3.6 1.56E-06 B8 Early small-mod CF, Rx-B DF only 0.0 1.45E-08 B9 Early small-mod CF, no DF 0.9 4.08E417 BIO Early puff CF, poo1 + Rx-B DF l.6 7.20E-07 B11 Early puff CF, Rx-B DF only 0.0 0 B12 Early puff CF, no DF 0.6 2.55E-07 Subtotal-Early release 9.72 4.27E4)6 Cl Late vent, controlled release 0.0 0 C2 Late vent, uncontrolled release 0.1 4.29E-08 C3 Late vent with DW fail (OT), Rx-B DF 0.0 0 C4 Late vent with DW fall (OT), no Rx-B DF 15.0 6.58E-06 C5 Late OP CF with pool DF 0.0 1.76E4)9 C6 Late OF CF without pool DF 0.0 2.57E4)9 C7 Late puff CF, with pool DF 0.0 2.18E-10 C8 Late puff CF, without pool DF 0.0 2.64E-10 C9 Late OT CF with Rx-B DF 0.0 4.65E-13 CIO Late OT CF without Rx-B DF 0.0 8.61 E-09 Subtotal-Late release 15.12 6.63 E-06 . D1 Recovered ex-vessel, vented 0.0 0 D2 Recovered ex-vessel, vented NG release 0.3 1.18E-07 D3 Recovered ex-vessel, no CF 24.6 1.08E-05 D4 Recovered ex-vessel, late CF 0.0 1.31 E-08_ Subtotal-Recovered ex-vessel 24.92 1.09E-05 Total 100.0 4.39E4)5 37
T;ble 25. Shoreham simplified event tree results, with no ventmg (pessimistic RB assumptions) Release hinde Conditional Release Probability ' hiode Release Mode (9) 9 Frequency A1 Recovered in-vessel, no CF 50.3 2.20E-f,5 A2 Recovered in-vessel, vented 0.0 0 A3 Recovered in-vessel, late CF 0.0 L ( s Subtotal-Recovered in-vessel 50.28 2.20lM15 B1 Early small-slow CF, pool + Rx-B DF 9.1 3.99E4)6 B2 Early small-slow CF, Rx-B DF only 0.0 5.00E-09 B3 Early small-slow CF, no DF 0.2 8.36E418 B4 Early large-slow CF, pool + Rx-!! DF 1.0 4.22E-07 11 5 Early hirge-slow CF, RA-B DF only 0.0 7.92 E-10 B6 Early large-slow CF, no DF 0.2 7.28E-08 - B7 Early small-mod CF, pool + Rx-B DF 10.2 4.47tM)6 11 8 Early small-mod CF, Rx-B DF only 0.3 1.40E417 B9 Early small-mod CF, no DF 0.7 3.12E-07 1110 Early puff CF, pool + Rx-B DF 2.3 1.01E-06 Bil Early puff CF, Rx-Il DF only 0.1 4.89E418 B12 Early puff CF, no DF 0.6 2.42E-07 Sut. total-Early release 2464 1.08E-05 Cl Late vent, controlled release 0.0 0 C2 Late vent, uncontrolled release 0.0 0 C3 Lme vent with DW fail (OT), Rx-B DF 0.0 0 C4 Late vent with DW fail (OT), no Rx-B DF 0.0 0 C5 Late OF CF with pool DF 0.0 1.8tE-09 C6 Late OF CF without pool DF 0.0 2.65E-09 C7 Late puff CF, with pool DF 0.0 2.25E-10 C8 Late puff CF, without pool DF 0.0 2.71 E-10 C9 Late OT CF with Rx-B DF 0.0 8.09E-13 C10 Late OT CF without Rx-B DF 0.1 2.89E-08 Subtotal-Late release 0.08 3.38E-08 Di Recovered ex-vessel, vented 0.0 0 D2 Recovered ex-vessel, vented NO release 0.0 0 D3 Recovered ex-vessel, no CF 24.7 1.08E4)$ D4 Recovered ex-vessel, late CF 0.3 1.31 E4)7_ Subtotal-Recovered ex-vessel 24.96 1.09E-05 Total 100.0 4.38E415 38
S , , , , r ,~ NV opt. RB NV pess. RB EE pess. RB i - - , , l$ i ; .! & y, 4
'l& EE opt. RB W f;,"' , L hh y , .
jh.
.+
Fi opt. RB Fi pess. RD
,e N -
, ,p . w - g .a
!_ 3 ;,4l
)
h[_i le '
'o $ ,5 .
6 ,' M, , gw em
~
,, /
w / E < 1
/
0 , i , i i
/
i Release modes A EB I IC f~~lD Figure 9. Release mode sensitivity to CV strategy and RB assumptions. of steam, nitrogen, oxygen, and hydrogen were tion that all reactor building equipment fails foi-calculatcJ using the IlWRS AR code. The venting lowing venting is not unreasonable. In addh..- calculation was performed for an ATWS with personnel entry into the reactor building to make main steam isolation vahe (MSIV) closure from repairs or align alternative injection systems will 25% power; the assumption is made that venting certainly be prohibited. under these conditions leads to a loss of injection and induced core melt. The ORNL analysis pre-The effect of these assumptions on the SCliTs dicts a global hydrogen burn that results m a peak is to bin sequences into mote severe release teactor building differential pressure of 6 psid modes. Thus, rather than rete.uc mode A2 the 1 and a peak temperature of 12(FF.The final equi- vented, recovered in-vessel end state is binned librium temperature inside the reactor building is into release mode D2 (recovered ex-vessel, on the order of 200 F. Ilecause safety-related vented, noble gas release), based on the assump-motors, such as ECCS pump motors and MOVs, tion that venting lea i , to failure of wssel injec-in the reactor building are generally qualiiied for tion and eventual core melt wite n - lable debeh long-term oraration up to an ambient tempera- bed once the molten core reaches the uppression ture of 65 C (149'F), and short-term exposure to pool. For acquences where venting intough the temperatures up to 1(WC (212 F), at 100% rela- existing lines leads to a late mitigated particulate tive humidity, the survival of equipment in the release due to deposition in the reactor building reactor building following venting is doubtful. (release mode C3), the resulting end states have Also, considering the likely deleterious effects been rebinned into release mode C4 (no reactor from the predicted hydrogen bums, the assump- building retention), because the RilSVS is 39
assumed to rapidly transport the fission products tne ECCS pirnps during s enting), implying emin to the refueling Door, from w hich they can esit to er core melt and thus an earlier release 01 signife the environment through the failed siding. cant radioactive material to the ensironment. N m, there is a good probabilny that the We of Applying the ORNL results to the Class 11 th, containment failure wdl be larger than the plant damage states is somewhat more problem- break that results ".hn the sentilation ducting atic. For these sequences, the pressurization of failt And findy, ven mg prosides something of the containtnent is more gradual. Consequently, a controPed blowdot n into the reactor building, the question arises as to how the amounts of along wit t the pos bility (although an unlikely steam and noncondensible gases released into th- one) of chwing the mts once containment pres-reactor building after full power operation conc sure has been rdieu J (reclosure of sent velves pare to the quantities calculated by ORNL for the has not been roodeled in the SCETsh 25rfe power ATWS case discussed above. As for the ATWS.no hydrogen should be generated un- The results of the Sandia annhsis' support the til the onset of core melt; therefore, no hydrogen ORNL calculations for Shorehan$. Shoreham, un-should be release ;into the reactor building at the like La Salle (the subject of the SNL analpis), time of venting. Ie amount of steam released does not have ECCS pumps k,cated in individnal into the suppression pool is likely to be corrpara- room wiWin k mm NHding ad m a ble for the two cases, although there is uncertain- RBSYS rather than an SGTS to process the reac-ty as to the behavior of reactor power over time for building atmosphere, so equipment surviv-for the ATWS case. I or the purposes of this re-du follow n venting would appear to be le ss port, it is assumed that venting the containment likely at Shoreham Also, the !ower failure pres-for Class 11 plant damage states results in a mass of k % 0 m Sho d m flow rate of steam into the reactor building in ce (0.5 versus 2.0 psigh counted with the tendency cess of the RBSYS exhaust capacity of of the RBSYS to mix rather than nlter the reactor 11(O sefm. This is assumed to cause failure of the building atmosphere and the high connectivity of refueling floor sidmg, similar to the 259 powei the Shoreham reactor building and refuel'ing ATWS case. For both cases, hydrogen is released Door, lessen the probability of fission product re-during core melt, with hydrogen burns outside the tention in the Shorehan reactor building. primary containment deemed to be probable. Therefore, given that vessel injection will be in summary, the effects of venting on comain-lost at some time after venting, there is a question ment failure depend an the environment pro-of what is to be gained by venting the contain- duced in the reactor building. The most optimistic ment for Class 11 plant damage states. The answ er case for existing equipment senting results wben appears to be time. If the containment is not using the assumptions from the front end of the vented,it will fail, most likely at a pressure that is Shoreham PRA. which yield an 889 probability more than twice the pressure at which venting of core damage being arrested either in-s essel or would take place. This translates into a more en- ex-vesset This falls between the noa enting and ergetic release into the reactor building and a Filtranenting cases (75% and 939, respective-larger effect on NPSil for pumps that take suction ly). Iloweser, if the pessimistic reactor building from the suppression pool Vessel injection is assumptions are used, the existing equipment likely to be lost so(mer than in the case of venting case produces a 759 probability of recovery, (assuming inadequate NPSil is not a problein for which is identical to the na-s ent case. 40
EFFECTS OF CONTAINMENT VENTING ON RISK Decause Shoreham did not have an approved nate nh;st early health ef fects for TW sequences, o acuation plur, at the time of tiiis anal) sis, cumu- but u ould probaNy not signific ant!,. aff ect the 'a. lative risk measures such as total population dose tent risk measures. could not be calculated and there fore, only dose-setsus-distance probabilities have been devel. The not :"cu desirable s.tuation is the esist-oped. Figures 10 through 13 display the ing equipment sei,t with the optinastic assump-pmbability of exceeding . specified dose (either tions on its perfonnance. nis case reduced core 5 rems or 2(O tems) at various dist,u;ces (frem melt frequency by 14% , but resulted in an insb;- 03 to 10 miles) from Shoreham for each of the nmcant duedon in M den mmparn{ to W nwnung case, un iennore, if the pwmhuc thrte unling strategies and teth of the reactor assumptions about the effects of venting on the builW.ng sensitivity cases. As expected, the most - or Mdig aw inged is k mp bemgn consequences are associated with venting sis, ex sting equipment venting increases conse-through the Filtra system. This system reduces quences by approximately a factor of five over the core melt frequency and mitigates ti e conse- the reventing case, quences of most severe accidents by controlling and filtering the resulting release. This decreases NOTE: Because only dose-versusslistance the probability of exceeding a specified dose prot: abilities were calculated and not the (comp.ced to the no vent case) by approximatel) population doses that factor in evacuation, a f actor of five. As discussed previously, consid- the cifect of the timing of any reler , cannot cration of evacuation would be expected to elimi- be judged. 6 i i i i i i i r- i 5 -- y e Filtra Cs a O Ex equip
$ 4 -
# No vent -
2 EE o- 3 - RE
%E EE
~
~
= \ o"-~~a O g ~~e g
f 3 _ NA% _ N .- -. , 1 0 O 2 4 6 8 10 Distance (miles) 00"0 Figure 10. Dose vs. distmee (>5 rems-optimistic RB assumptions). 41
i i i i i ' i i 1.3 q -
.2 e r 1.1 -
E e Filtra , 1 { 0.9 - o Ex equip -
@ 8 No vent -
j 0.8 - L e- \ - W6 0.7 - O
.*8 -
{} 0.6 -
~
o L .E 0.5 - gr - w 0.4 - e
-E -
o 0.3 0.2 # e.1
\s _ e
, t ,
w _. e 2 4 6 8 10 0 0-0641 Olstance(miles) Figure 11. Dose vs. distance (>200 rems-optimistic RD assumptions). 10 , i i , , i i , 9 -
, =%g e Filtra E 8 -
o Ex equip e
,o g 7 - a Novent .
^ -
6 - E e -
!=
RE 5 4 - gs. x 5 3 * - S 2 - " N ,_ - ,
-e-1 0
. . . 1 0 2 4 6 8 10 o-osas Distance (miles)
Figure 12. Dose vs. distance (>5 rems-pessimistic RB assumptions). 42
_ _ . _ _ . _ _ . _ __._.._.______._.m._ _ ._ .. _ _ _ _ _ _ 8- e i i e i i i i e-
,1 7 -
I e Filtra 6 ' - o Ex equip m No vent y .6 , s8 jj 4 - g.@ ms R 3 - 5 . N j 2 - 1 . 1- -3 -- k,e _ o
' ' ' g
-i 0 ~'*4 4- ' --- b; '
e I o 2 .4 6 8- 10 Distance (miles) oosse Figure 13. - Dose vs. distance (>200 rems-pessimistic RB assumptions).
;; The increased consequences for the existing por suppression failure (given as 3.1 E-05 in Ref-equipment venting case result from eativ over-
. crence 3). This results in a decrease in the l pressurization failures (likely to occur in the wet- frequencies of several of the B release modes. Be-
.well) in :he no-ven' case generally' being cause the B release modes generally result in se- l replaced by unscrubbed releases as a result of vere offsite consequences, the exclusion of these ventirg. Because of the assumptbn that both the events was examined to ensure that the calculated -
wetwell and drywell vent lines would have to be ~ consequences were conservative. , opened to mitigate the containment pressure rise,
; the vented release would not pass through the suppmssion pool. Some existing equipment vent. When these two events were added back into ink .unariot r"ty also be accompanied by over- the SCETs, an increase of approximately;100 to temperature failure of the drywell head seal. ~- 200% was scon in the probabilities of release 1 Again, this replaces earlier wetwell releases, a modes B4, BS, and B6. In addition, a slight in.
- fraction of which would be scrubbed by the sup-' crease of approximately 1% occurred in the prol>l 4: pression pool. Furthermore, late overtLaperature abilities of release modes B 10and B 11. Ilowever, -
. failure of the dr/well always results in pool by- ther: was no effect on the probabilities of the re-pass because the failure location is assumed to be lease categories into which these release modes
' the drywell head, with the fission products being Lare binned because these particular release modes
- released directly to the refueling floor. are not significant contnbutors to their respective release categories. Therefore, the exclusion of va-As mentioned earlier, the failure of vapor sup- por suppression failure and very early overpres-pressiva and overpressure containment failure sure containment failure was found to have no -l
- prior to vessel breach were excluded from the : discernible effect on the consequences of a =
SCETs because of the very low probability of va. release. 43
- -- y , _ s r m ..-, e - -,
l l CONCLUSIONS A simple calculation to grossly estimate the This likely represents an upper bound because co,t benefit of the Filtra system can be performed the consequences associated with the Mark I con-as follows, using the (Draft) NUREG-ll50 anal- tainment of Peach Bottom are probably more se-ysis of Peach Bottom as a basis (Reference 6). In vere than those associated with the Mark II Appendix C.3 of Reference 6, a containment fail- containment of Shoreham because of the relative-ure frequency of 5.3E-06 per year (core melt fre. ly high susceptibility of the Mark I to liner melt-quency of 6.6E-06 minus the zero consequence through and early overpressure failure.This gross bin on Table %9) produces a risk of 61 person- estimate of $4.9M for the risk benefit can be com-rem /yr for Peach Bottom. Therefore, a normal. pared to the estimated cost for installing a ized risk of i 1.5 million person-rem per core melt Swedish-type filtered vent system of $30M (Ref-accident is calculated. However, the Shoreham erence 8), which is an approximate estimate that and Peach Bottom sites have different popula- may not include the cost of any design, licensing, tions and weather patterns. Using Table C-1 in or soil analysis perfonned by the utility. Reference 7, which generated consequences for evers comraercial nuclear plant site in the coun. Under the pessimistic assumptions about the try using WASH-1400 source terms, a ratio of effects of venting on the reactor building, the 1.22 can be determined for Shoreham/ Peach existing equipment has no benefit in preventing Bottom consequences, producing a Shoreham core damage and actually increases plant risk by normalized risk of 14.0 million person-rem per increasing the consequences of a severe accident. core melt ace'. dent. Then, using the base contain. In summary, venting should only be considered ment failv;e frequency for the no-vent case of as a severe accident manrgement strategy if RPV 1.1E-Of per year (core melt frequency of injection can be maintained af ter venting has oc-4.4E-0.' minus the no-containment failure curred. This has two facets. First is the effect of moded. ti.e Shoreham base case risk is estimated venting on the suporession pool (possible NPSH ut 1.44 person-rentyr. l A reduction factor of five problems) and the systems that rely on it as a in risk produces a Filtra case risk of 31 person- source of water. Second is consideration of the rem per year and a benefit from the Filtra installa. effects of venting on the reactor building tion of 123 person-rem per year. Therefore, the environment and the equipment therein. Unless total cost benefit can be grossly estimated as post-venting RPV injection can be ensured with a reasonable probability, containment venting Total cost benefit using existing equipment should not be pursued.
= [123 person-rem /yr (risk benefit)l These conclusions are summarized in Table 26.
($1,000/ person-rem)(4(Lyear plant life) See Appendix C for a discussion of reactor building environments and Appendix D for back-
= $4.9M. ground infonnation on the Shoreham PRA.
44 l l l
Table 26. Summary of conclusions Optimistic Assumptions Fessimistic Assumptions Filtra venting (or Lower core melt frequency Lower core melt frequency filtered vent through Filtered releases Filtered releases hard-pipe) (either Filtra (either Filtra or suppression pool) or suppression pool) Existing equipment Lower core melt frequency liigher core melt frequency venting Reactor building Reactor building and suppression decontamination factor pool bypassed (suppression pool bypassed by drywell venting) No venting Higher core melt frequency liigher core melt frequency Suppression pool and/or reactor Suppression pool DF (no reactor building DF buildirig DF) 45 -
REFERENCES
- l. Scaence Applications. Inc., Probabilistic Risk Assessmeru, Siwrcham Nuclear Power Station, pre-pared for the Long Island Lighting Company, June 24,1983.
- 2. E. T. Burns et al., Shoreham Nuclear Power Station Full Power PRA, PRA Update: Supplemental Containment System Implementation, IT/Delian Corporation, prepared for the Long Island Lighting Company, Febtuary 1988.
- 3. 2. T. Mendoza et al., Containment and Phenomenological Event Tree Evaluation At Full Powerfor the Shoreham Nuclear Power Station, Science Applications international Corporation, prepared for the Long Island Lighting Company, February 1988.
- 4. S. R. Greene An Assessment of the Shoreham Nuclear Power Station's Secondary Containment Sc.
vere Accident Mitigaria Capability, June 26.1987.
- 5. S. E. Dingman and A.C. Payne.1r. HWR Reactor Building Environments After Containment Fail-ure, S AND-88-1515C, December 1988.
- b. C. N. Amos et al., Evaluation ofSevere Accident Risks and the Potentialfor Risk Reduction: Peach bottom. Unit 2, NUREG/CR-.4551,(Draft) Volume 3, Parts 1 and 2, February 1987.
- 7. D. C. Aldrich et al.. Technical Guidance for Siting Criteria Development, NUREGICR-2239, December 1982.
- 8. R. O. Schlueter and R. P. Schmitz " Filtered Vented Containments," Fourth Workshop on Contain.
ment Integrity, Arlington, Virginia, June 14-17,1988, NUREGICP-0095, November 1988, pp. 57-76. 46
_J _ .e . J .* _4 O- A.A..a e4a+ e.A a4.e,ae 4 d. 2.. AJ 4,aaa. 441_mA_,2.m. .,hA-.# _%-mus 4 E., -A,a p4 4._4w. ~ czwAm..w._ l l f APPENDIX A QUANTIFICATION OF THE SHOREHAM SIMPLIFIED ACCIDENT SEQUENCE EVENT TREES I - 1 A-1 '
CONTENTS APPEN DIX A ACR ONYM S . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A-4 TR A N S I ENT INITI ATO RS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ., A-5 LOSS OF DC BUS INITIATORS . . ............ .. ... .. . .. . . . A-8 LOSS OF OFFSIR POWER INITI ATORS ...... . . . .. . . .. . A-Il LOSS OF SERVICE WATER INIT! ATORS ... . . . . .. . . . . A-14 INSTRUMENT LINE BREAK INITIATOM . . . . . . .. . . A-17 111G11 DRYWELL TEMPER ATURE INITI A7 ORS ,.,. .. .. ... . A-20 SMALL LOSS-OF-COOLANT ACCIDENT INITIATORS . . ... .... A-22 MEDIUM LOCA INITIATORS .... . .. .. . .. .. .. . ., A-25 L ARGE LOCA INITI ATORS .... . . .. . . .. .. A-27 ANTICIPATED TRANSIENTS WITilOUT SCRAM ,. .. . A-29 RPV RUPTURE INITIATORS . . ..... ... . .. . . .... . .. . A-34 APPENDIX A FOOTNOTES .. .. . ... . . ,. .. .... . ... A-36 REFERENCES . . . . . ... ... ... . .. . . . .. . . ... . A-37 FIGURES l A-1. Shoreham transient-initiated sequences . . . . .. .. ... . . . ... A-7 A-2. Shoreham loss of a de bus initiator . . .. . . ... . ,. . ... ,. . A-10 A -3. Shor ' :.. LOSP sequences . . .. ..... .. .. .. . .. .. .. A-13 A-4. Loss of service water sequences . ... .. . .. . . . ... . ... . , A-16 A-5. l Instrument ine break sequences ... . ... . . ... .... . . . . .. A-19 A-6. Manual shutdown - high temperature in drywell . , . . . . . . . ... .. ...... . A-21 A-7. Shoreham small LOC A sequences .. ... . . . . ...... . . A-24 A-8. Shoreham medium LOCA sequences . .... ,,. ... ... . . .. A-26 A-9. Shoreham large LOCA sequences . . ... . . .... . .... . . A-28 A-10. Shoreham ATWS sequences . ... .. , ..... .... . . A-32 A-1 1. Shoreham RPV ru;iture sequences . . ... . .. ... A-35 A-3
APPENDIX A ACRONYMS ADS Automatic Depressurization LPI low-pressure injection System hiAAP hiodular Accident Analysis ARI alternate rmi insertion Program CRD control rod drive NISIV main steam isolation valves CS core spray ht'ITR mean time to repair CST condensate storage tant PCS Power Com ersion System DBA design basis accident PRA probabilistic risk assessment ECCS Emergency Core Cooling System RBCLCW Reactor Building Closed-Loop Cooling Water System EDG cmergency diesel generators RBSW reactor building service water EOP Emergency Operating Procedure RBSYS Reactor Building Standby Ventila-ESF emergency safety feature RCIC reactor core isolation cooling FWCS Feedwater Control System RilR residual heat removal GT gas-turbine generator Reactor Pmtection System liCTL heat capacity temperature limit mactor pmssure vessel IIEP human error probabilities SBO station blackout ilPCI high-pressure coolant . . injection SORY stuck-open relief vahe IIPI high-pressure . . injection LER Licensee Event Reports LOCA loss-of-coolant accident TBCLCW Turbine Building Closed-Loop LOSP loss of offsite power Cooling Water System LPCI low-pressure coolant njection TBSW turbine building service water A-4
l APPENDIX A QUANTIFICATION OF THE SHOREHAM SIMPLIFIED ACCIDENT SEQUENCE EVENT TREES The following simplified accident sequence included. All footnotes listed in Appendix A are event trees are grouped by categories and an common to the entire appendix and are dermed at explanation of each factor in the event tree is the end. TRANSIENT INITIATORS The following is a description of the factors Cm. Reactor Protection System (RPS) me. included in the Transient initiators category (see chanical failure probability is the commonly used Figure A-1). value from NUREG4M6062,The total RPS fail-ure probability is 3.0E-05/ demand. Ilowever, be. T. Initiating Event (Reference A-1, cause the electrical failures are automatically Table A.1-5) is the sum of individual initiators, recovered because of the attemate rod insertion including the following: (ARI) system, only the mechanical portion (1.0E-05)is retained.
- Loss of Condenser Vacuum (Tc) 0.4/yr
- Turbine Trip (Tt) 5.77/yr U. Failure of all high-pressure injection
- MSIV Closure (Tme) 0.16/yr (llPI) includes high-pressure coolant injection (flPCI), reactor core isolation cooling (RCIC),
- Loss of Feedwater (Tf) 0.05/yr and injection through the control rod drive (CRD)
- Inadvertently Opened Relief 0.07/yr-P"*PS-Valve (Ti) e Manual Shutdowns (Tms) 4.30/yr* 1.E. F(1.E.) P(U/1.E.) TU
- Loss of Offsite Power (Te) 0.06/yr Tc 0.34 9.0E4)3d 3.lE-03 Total (T) 10.8/yr Tt 5.77 4.8E-03' 2.8E-02 Tmc 0.16 9.0E-03f 1.4E-.03 Q. The probability of feedwater being Tf 0.05 9.0E-038 4.5E-04 unavailable (this includes the possibility of Ti 0.07 9.0E-03h 6.3E4M feedwater being tripped and subsequently Tms 4.30 8.8E-031 3.8E-02 restored)is a weighted average of the individual Te 0.06 9.0E-03' 5.4E-04 conditional probabilities of Q for each of the initiators listed above, namely: Totals 10.8 7.2E-02 l . = 6.7E-03 1.E. F(1.E3 P(Q/1.E.) TQ b
. a re f the reactor pressure nssel Tc 0.34 1.0 0 34 (RPV) to be depressurized is taken to be unifomt Tt 5.77 0.15 0.58 f r all initiating events and is quantified in Refer-Tmc 0.16 0.3d 0.05 ence A-1, Figure E.4-5 as 2.4E-03. This value Tf 0.05 0.14' O.01 Ti 0.07 0.99' O.07 e uld not be verified using the documentation Tms 4.30 0.(Ms 0,17 available in Reference A-1.
Tc 0.06 1.00 0.06 V1. Failure of all low-pressure injection Total 10.8/yr 1,28 systems (LPI) includes core spray (CS), Qm 1.28/10.8 = 0.12 low-pressure coolant injection (LPCI), and A-5 l
coolant injection through the condensate and con- lines,the above value of 5.0E-01 was used for the densate booster pumps, optimistic assumptions concerning the effects of venting on reactor building equipment, while a
,1.E. F(I.E.) P(V/1.E.) TV value of 1.0 was used for the pessimistic assump.
" "I# " #4" *#"I """" " "E *P Tc 0.34 1.38-05a 4.4E-06 ture f the ventilation system ducting.
Tt 5.77 6.3 E-06' 3.6E-05 Tmc 0.16 6.33-06' l.0E-06 Tf 0.05 1.3E-058 6.5E-07 QU. IIPI availability is quantified in Refer-Ti 0.07 6.3E-06h 4.4E-07 Tms 4.30 6.2E-06' 2.7E-05 ence A-1, Figure C.1-4. Since Reference A-1 Te 0.06 6.3E-06f 3.8E-07 gives no credit for alternative high-pressure makeup or turHne-driven sources, the CRD P""I* ""' " I** ##" "*" '"E 1, 7.0E-05/10.8 = 6. tank (CST), are the sole remaining source. Be-W. Failure of containment heat removal in-
'"
- I '*'""""E" 'I "* E" '
cludes failure of the Power Conversion System mnt @n Refennce A-1, u q. appannuhauhe model used did not account for isolation of Reac-(PCS), the residual heat removal (RilR) systern
" E ** P ""E "'
(including the RHR service water system), and the RCIC system in the steam condensing mode in
' I" ' P"*E "# "
an acc ntWsc oN6 pump eManh."E n-conjunction with the Rl!R heat exchangers and jection would require the operator to mstall elec-the RilR service-water system. (This last operat-ing mode typically has an insignificant effect on ""' I*"E* I E"" * "". " int r ek. This is not directed by procedure and is the overall availability of containment heat not considered to be a likely operator action; the
" * ""b) operator would most likely attempt to use low-pressure sources first, such as LPCI, core spray, RHR failure = 1.lE-04h and condensate pumps. As a comparison, Refer-ence A-1 gives no credit for manual alignment of PCS available = P(MSIVs open)
+ P(PCS operable) e tespravsucti ntotheCST,anactionthatiscov-ered by procedure.Therefore, this report has reor-
= IE-03 1+ SE-03) = 6E-03 dered events QU and V2, placing QU after V2.
This is reasonable because the RPV will be de-
= pressurized by the time Filtra actuation occurs [the (1.l E-M)(6E-03)
Shoreham Modular Accident Analysis Program W = 6.6E-07 (MAAP) analysis indicates that the suppression pool heat capacity temperature limit (llCTL) will FL Failure of the Filtra system to operate is be exceeded approximately three hours into the quantified in Refesence A-1, Figure C.1-2 at sequence).^4 It is assumed that the operators will 7.2E-03, ovecride the RBCLCW isolation interlock only if CRD pumps are the only available injection CV. This is the failure probability for venting source. For this case, a failure probability of through the existing vent lines. Event CV is quan- 1.0E-01 has been assigned if venting has oc-tined in Reference A-1, Figure K.1-2 at 1.0E-01 curred, based on engineering judgement, and 1.0 for sequences initiated by a transient. if venting has failed, based on the fact that failure to vent leads to containment failure, which is as-V2. Failure of RPV injection given that Filtra sumed to fail equipment in the reactor building. successfully operates is quantified in Reference For the pessimistic reactor building assumptions, A-1, Figure C.1-5 at 5.0E-01 for sequences initi- the CRD pump motors are assumed to be failed by ated by a transient.Two sensitivity cases are used the environmen: in the reactor building, leading to in this report for venting through the existing a failure probability of 1.0 for event QU. A-6
l 1 1!)ljl j w o t r e N,:ca 1 mFC t o uf M s- e t o t L 6 a =. 1m.F Fs t "A L a -
"e Po 1 .. ekc ' T u w h. T
- T a
- s W T
- T =. Tok W . = = W =. S o . . T T k o z3 C T o t T T T E
C a t. uss P A t w 7 a. Ft Sc . k.
- F a I z I z k. n ** m I a, *
. =.
,: A
- 1 1. I. . DI
- n me .. =. r
- I I . D I h.
D D A ? x I I A
, , o t e o . 5 s 4s0 c ee - ce,iret et e eC
. 8
- 7 e d e ' r s r r . 9 , 9 e e s T.
r .e. e
- ~c- e - -ee - -e -
- e e.
o- -e e_
- - - e t
. e ,3 a a i s s e eee eee e E t
7 o t
* ' E e t t r e c 7 aMcE o r e s e 9m . i. * - s. - s. e
- s. -ms. we.m ee.m. 8 e. 0 c. t. e. . s. . 1 v. J d. 4 e. e. S.
1 em 5 s ' s 4 e e 5 1 1 4 s . 8 . 4 e a s e e e 4 e e 1 t s e t I 1 i 9 n o rc OT C u..D J M e o E c. c. em t 1 i A f I L.L
- t t,
...sn e s
e g
,g e .}
V
- n. i.v o o ,
Ceee Ii t ,;, l 2 3 ti pte L s Y $
. e. .
C.. F 2
..r m 2 N t 3
L L L t 7
.i..
e C
~ s
- r tR.. s.n.. r W N xe w C . .
a l E L - I n
. s 4 2 I 2 4 a
.t
- t. a 3 9 9 s s u L- C L-le..
ia i, c e. Ft L
,il W -
E S 1 7 Nfyt C C. i 1L C D. i l c s. i nto i n. = o. L..t g
- 3 n
p.
.iu e
- f. r ,
1 t g u m # g . 1. w =. . u. 5 L.. r..r D a g y s e c
. # n e e
!u.
P. 1 v e u q Le s e e s d 3 c t e t.v X eg i a 1.e 1 r in t F U i L s n t i i e 1 i I L t9 s P. H 2 n r a r J t m hsm
.s M
e a C w._ h c.u r o e r w . e. - o r t_ h
. S u
M. F C . 1 n - u , A u e
.t r e
u
.t.
Trte s. e T m.1 i F g f4 l -l.; lIlIlllljlli fll1, ;ltI.lll 1ll lll l L
LOSS OF DC BUS INITIATORS The following is a description of the factors in- The updated pobabilistic risk astessment (PRA) cluded in the Loss of DC Bus initiators category cites Reference A-4 as the source of these num-(see Figure A-2). bers; however, documentation supporting their derivation could not tw found. A brief descriptior Td. Figure A-2 quantifies the accident se- of this event is presented in Reference A-4, quences initiated by a failure of a 125 V de bus, pages b281. specifically the Division 11 de bus. The frequency of de bus failure is estimated in Reference A-4 as U. Because only a single ESF de bus is 3.0E-03/yr. His number is the result of a loss- available, only one llPI system is available. The of-de bus frequency of 6.0E-03 and a 50% oper- de power dependencies of the FIP! systems are as ator recovery factor. Additionally, there is a follows: RCIC supported by Division I and ilPCI contribution from other initiating events coupled supported by Division II. Therefore, because the ( with a coincident failure of a de bus. This yields initiating event assumes a failure of the Divi- k the following: sion 11 de bus, only RCIC is possibly available for llPl. Using the quantification of the failure of Loss of a de bus initiator (Td) 3.0E-03 RCIC and llPCI to operate during a station black-out (SBO) from Reference A-1, Figure E.4-4, Other initiators plus failure the failure probability of event U is calculated as of de bus = (10.8)(1.7E-04') = 1.8E-03 7.9E412. Total (Td) 4.8E-03 X. A value of 9.4E-03 is used for this event based on the value used in Reference A-1, Cm. RPS failure is quantified the same as for Figure 3.4-50. Figure A-1. V1. The quantification of event VI is based D1. This event represents the conditional on the potential availability of one of two CS probability that the second 125 V de bus will fail, pumps, three of four LPCI pumps, and the con-given the failure of ths first bus. The probability densate system. The failure of all LPI is therefore of this event is calculated from the results of the calculated from the values given in Reference 125 V de power system quantification presented A4, Figure 3.4-44 as in Reference A-1, Appendix B.3, as Probability of one CS train failure = 9.7E-02 Probability of both buses failing 1.7E - 07 baW M No W Uains = MM
= = 1.0E - 03. failing Probability of 1.7 E - N Grst bus failing Condensate system unavaliability = 1.0E-02 l0. The availability of the feedwater system, Probability of event VI = 9.7E-06, given a failure of the de power system,is quanti-ified as follows: W. The availability of containment heat re-moval depends on the availability of either the One emergency safety feature (ESF) RFIR system or the PCS. This event is quantified de bus available P(Q) = 0.19 based on the values presented in Reference A4, Figure 3.4-44 and uses the following system Neither ESF de bus available, P(Q) = 0.5. unavailabilities:
A-8 ( I
Probability of main steam isolation CV. This is the failure probability for venting valves (MSIVs) being closed = 1.0E-02 through the existing vent lines. Event CV is quan-tified in Reference A-1, Figure K.1-2 at 1.0 for sequences initiated by loss of a de bus. with no Probability of PCS not available = 5.0E-03 details of the quantification given. Probability of PCS not available NOTE: This is probably due to the loss of for containment heat 125 V de power to panel ill1*PNL-VC2, removal = 1.0E412 + 5.0E-03 = 1.5E412 the control room panel from which the vent valves are operated. Probability of RHR failure = 5 AE-03 Tk value used for the probability of LPI V1 failure is taken from Reference A-1, Figures Total unavailability of 3A.6-4 and K.1-1. As before, use of the pessi-containment heat nustic react r building assumptions leads to a removal = (5.4E-03)(1.5E-02) = 8.lE-05 value of 1.0 for venting through the ex,isting lines. FL. Failure of the Filtra filtered vent system QU. The failure probability of IIPI is quanti-fied in Reference A-1, Figure C.I.-4 at 1.0 for se-to operate is quantified in Reference A-1, Figure C.1-2 at 7.2E-03. quences initiated by loss of a de bus. A-9
ill d r O I ET CP NI ER 1 F I F 1 F UC f- C FC F- C V x Q QS - - - U U c EE k k W W k k M W k k W W O O k e C SD o o T T e o T T o o T T T T o T T E C N ES US S QA F A F A F A W EL k k I I k k I I k k I I D A k A T SC o o I I o o I I o o I I I I o I A E 3 7 7 9 4 8 8 0 5 9 9 1 0 7 8 8 e C 0 0 0 0 0 0 C- 1 0- 0 1 1 o- 0 C- C N. - - - - - - - 4 - - - - - E E E E E E E E E E E E E E E E E E US 8 6 6 7 9 7 7 9 3 7 7 8 2 7 0 0 0 QO 8 5 5 7 3 3 3 8 1 8 8 1 9 7 4 4 8 ER SP 3 1 1 2 8 3 3 4 7 2 2 4 8 8 2 2 4
- 1. -
s1 Ope U Rmv Q Cue Pht 1 1 I i 1 9 Ie 2 E Pv V 9 Le 9 n U 2 M
.g 3 3 tn 3 ni e 0- 0-ot r Cnet L E F 0 eil ovwi 2 8 M -
n F 7 2 I l t n g g
,ta la - g g
t a v N g _ n eoel o e oHmF C e t. g - A 8 g l 6 a f-l I l 1 Y E Pe L O F J S os. 3 ts 9 V - lr eP H X lp ee Fd E S r. o t _ e I ll ia U i t We F i n s u 1 b 1 1 9- e - We d v G E a - Fe 0 n 0 f u o s w Ies 9 k ul i
.B i v a D L- m D
iCF a i DD D. h 1 e r e 5 o h
.r C S Sh lu M -
E C Wc e Ma i 0 9 . F 2 L - I ' A I ss ul e r D
.tia T u ivCF g DD i U F ig
LOSS OF OFFSITE POWER INITIATORS The following is a description of the factors in- of repairing a failed EDG is estimated to be zero; cluded in the Loss of Offsite Power Initiators therefore, this event only includes the possibility category (see Figure A-3). of restoring offsite power. The values used for the ] Te. The value for the loss of offsite power nonrecovery of offsite power are listed in Reference A-1, Table E.4-1. The value used for (LOSP) initiating event is documented in Refer- 30 min is 0.55. ence A-1, Figure E.4-1 as 0.062/yr and is the sum of the probability of the random LOSP initi. U1. Failure of RCIC and llPCI to operate ating event plus the transient-induced LOSP,e.g., during a station blackout is quantified in Tt plus LOSP. Reference A-1, Figure E.4-4. This value only ] covers the (L to 2-hour time frame. Although the Shoreham PRA estimates that the Division I and DC. This event represents the failure of Divi-sion i and Division 11 safety-related de power Division 11 battery life will be at least six hours, buses. The probability that both de buses are un. the availability of the llPI systems at two hours available is quantified in Reference A-1 Sppen. (event U2)is reassessed on the ever.: tree because dix B.3 at 8.3E-06, f p ssible battery depletion, room cooling re-quirements, and high suppression pool tempera-tures. The value used for event U l is composed of DG. The failure of both Division I and Divi-the following: sion 11 emergency diesel generators (EDGs) is quantified in Reference A-1, Appendix B.2, at 8.7 E-03 RCIC/IIPCI independent failures 7.7E-G4. Although Shoreham has three emergen-cy diesel-backed 4160 V ac power buses sup- RCIC failure 7.9E4)2 ported by two EDGs each, only Divisions I and 11 are included in this event because they supply llPCI failure 1.lE-01 power to the llPI systems (llPCI and RCIC) and the ADS system indirectly through the battery RCIC/HPCI dependent failures 1.0E-03 chargers. With no ac power supplied to the Divi-sion I and Division 11 emergency buses. the time Total llPI failure 1.0E412. dyring which HPI and ADq will be operable is 82. This event considers the possibility of limited by battery life. Division ill supphes pow- restoring ac power at two hours and is inchUed er to RilR pumps C and D, which would possibly on the event tree because of the time-phased be available during SBO sequences. nature of the availability of IIPI. At two hours, there exists a small probability of repairing a GT. Shoreham maintains an onsite, black-failed EDG as v, ell as the more likely possibility start, gas-turbine generator (GT), which, al- of restoring offsite power. As for event B , the though it is not covered by technical probability of event B2 is quantified based or. the specifications or maintained by Shoreham's values listed in Reference A-1. Table E.4- I and maintenance department, could be used during a s composed of the station blackout event. Failure of this event is quantified in Reference A-1, Figure E.4-3 at Probability of not restoring offshe 0.33 if de power is available and 0.37 if de power power between 0.5 and 2 hours = 0.35 is not available. For this analysis, a uniform value of 0.35 is assumed for all cases. Probability of not repairing a failed EDG by 2 hours = 0.75 B. This event represents the probability of not recosering offsite power within half an hour I*" ' (30 min). In this short time frame, the probability P(B2) = (0.35)(0.75) = 0.26. A-ll 1
U2. The continued operation of the llPI sys- bility exists that LPI may b available Mfore ac tems at two hours is checked with event U2. As power is restored to the entire plant emergency ac discussed in the description of event Ul, the power system. Because a third emergency ac llPCI and RCIC systems are susceptible to failure power division exists with dedicated EDGs, and because of possible battery depletion, high room the LPCI system is partially supported by this di-temperature, and high suppression pool tempera- vision, this system could be used during an SBO ture. The quantification of this event is based on sequence. Reference A-1, Figure E.4-6 quanti. Figure E.4--4 from Reference A-1, and is esti- fies the availability of LPI during SBO at mated at 0.1. 3.2E-01. B6. This event accounts for the probability of W. For sequences in which ac power is re-covered to Divisions I and 11 before six hours, the restoring ac power in the 2- to 6-hour time frame. As for event B2, credit is taken for recovering ei. probability of containment heat removal failure is
~
ther offsite power or the EDGs. Based on the taken as 6.6E-07, the same value used for trans-probabilities listed in Table E.4-1 of Reference ient-initiated sequences. If ac power is not avail. A-1, the conditional probability of not recover. able to Divisions I and 11 after six hours, but LPI ing ac power by six hours is is available through RHR pumps C and D, which , are supplied from Division ill, a value of 5.0E-03 l is used, taken from Reference A-1, Figure E.4-1. P(NR @ 6h)/P(NR @ 2h)
=: 0 024/0.144
= 1 67 E-0~,' ep a ty mat Ntm fans to actuate is taken as the usual value of 7.2E-03.
X. Event X represents the probability of not CV. Failure of the existing vent system to ac-depressurizing the RPV. Because the Shoreham tuate is quantified at !.0E-01 if ac power is avail-Emergency Operating Procedures (EOPs)in- able to Divisions I and Il and 1.0 otherwise. s"uct tne operators to inhibit ADS during SBO sequences, this event only considers manual de- V2. Failure of LPI is quantified at 5.0E-01 if pressurization. Although this event is placed on ac power is available to Divisions I and 11 and 1.0 the event tree at what appears to be the six-hour otherwise. For the case of venting through the ex-time frame, it addresses the question of depressu- isting hnes with the pessimistic reactor building rization at the time of HP/ failure. Additionally, assumptions, venting is assumed to fait equip-after six hours of IIPI operation the suppression ment in the reactor building, giving event V2 a pool is assumed to reach its llCTL, at which time failure probability of l.0. the operators are instructed to depressurize the
,RPV. Reference A-1 quantifies this event in Fig- QU. Unavailability of the CRD pumps is ure E.4-5 at 2.4E-03. In addition, depressuriza- quantified as before at 1.0E-01 if ac power is tion is assumed to take place at six hours with a available to Divisions I and 11 and 1.0 otherwise.
probability of one. For the existing vent case with the pessimistic I reactor building assumptions, the CRD pumps are V1. Because of the assumptions made in the assumed to be failed by the reactor building evalurtion of the onsite ac power system (i.e., environment, giving event QU a failure probabil-only Divisions I and !! were included), the possi- ity of 1.0. l A-12 l l l
o ,
!!.!. n i i i i . i ei-4il- <I-i I
;$$ tis $silb$litb$tiib$liib51tib58tilb$8Si11b5SMlib$5 iiiii??????i????8??*?i?*???ifi??i%ii??i*iff??i??????
lGldl uthButMHthBBE!!MH!HIHHlHAMBIDHH i s
!..........c.. ... ............. _.w..
i, . _. i, a u e ul ve e u 2 u
- m u k
spji
;a w w n
o e w r q- l > e
*- .n- L r# Y s
- .# .# ,9 [
;d N... Ii . . , , , . .
Jb *
,1 Jb Ub l l .b 5 9 2 9 9 i l9e gli e a 3< a d $ "9
#y 4 J Iyu u
La W p
..'! d
-1 11
". W.
A a 6 d 9 d d
- s. s' y
9 w L _. A A - .s JJ4 11 Ji d i ! 9 9 9 Ei d d 1 a u a
- - . 9 5 f
.,8 [ .j a, y
-)
.{
1 4 - A
- inW l &
.it si e
;3.3:
y I 3!I 5
- s ! s.
k.A ! d , i l
*h}h 3:# :l 1: ["
4 a f 9 B-Ei n ? I a u
. a a
- t g . V 5-05 " ;
N 9
- v *i" .i :
.i J - W
!!i!!; D [
3 n s an 2I , n 1 n s:. E i 5w 128 I 3 Z 31 a O 3:- : < O $ u}!! # n I H 4 9 u. A-13
LOSS OF SERVICE WATER INITIATORS
- The following is a description of the factors in- maining 50%, the assumption is made that 26%
eluded in the Loss of Service Water Initiators of the RBSW failures will also affect TBSW (and
-. category (see Figure A-4). therefore would not be recoverable by cross-connecting to TBSW). Therefore, A loss of reactor building service water (RBSW) will result in a loss of cooling water to P(S) = (5.0E-01)(2.6E-01) = 1.3E-01.
- the following heat exchangers:
Q. The availability of feedwater is based on
- RHR-causes a loss ~of suppression MSIV availability. The failure probability of pool cooling. 3.0E-01 is calculated in Reference A-5, Table
~
SG.5.
.BCLCW---causes a loss of drywell R. This is the probability of failure to recov-air cooling, a loss of cooling to RHR er RBSW, calculated as P(R) = exp(-t/MTTR).
pump seals, and a loss of cooling to the CRD pump oil coolers. The value of 6.0E-01 is for t = 10 h with an as-
;, gg g discussed in Reference A-5, RBSW must be rc-
- Reactor Building Standby Ventilation covered within 10 hours to prevent high contain-System (RBSYS)-loss of service-ment temperatures and subsequen_t Emergency water flow to the RBSYS water chill-Core Cooling System (ECCS) degradation.
ers will result m a loss of area cooling in the reactor building and a loss of U. This is the RCIC unavailability for times control room air conditioning, ' in excess of 10 hours. The value used is from Ref- . erence A-5, Table 5G.5. IIPCI is assumed un-
- EDG-causes a loss of EDG engine available for operation after 10 hours because of coolers. high suppression pool temperature.
Ts. The initiating event includes contribu- X. The probability of failure to depressurize tions from ioss-of-service water suction and fail- is increased after 10 hours without drywell cool-
- ure of RBSW during a shutdown caused by
_ ing based on degradation of the safety relicf valve another trancient. The value of 2.4E-03 used is (SRV) solenoids. The value used is from Refer-taken from Reference A-5, Appendix 5.G-3. ence A-5, Table SG.5. Cm. This is the failure probability of the RPS - V1, With a sustained loss of condensing wa-due to mechanical faults. The value used ter to the RBSVS water chillers, temperature in (1.0E-05)is taken from Reference A-2. the area of the ECCS pumps in the 8-ft elevation of the reactor building vill eventually exceed the
-S.- nis is the probability that the operators environmental qualification temperature of the will fail to recover from the RBSW system fail- RHR and core spray pumps, in addition, pro-
. ure. Service water can potentially be restored to longed loss of cooling to the RllR pump seals will vital components by either restarting the RBSW- eventually result in pump failure. Because of system itself, or by cross-connecting turoine these limitations, it is assumed that, for times in .
. building service water (TBSW) to the RBSW sys- excess of 10 hours, only the condensate pumps tem.The value used for the failure probability of are available for LPL The value used is taken this event was calculated by assuming, based on
~
from Reference A-5. Table 50.5. - the discussions in Reference A-4 and Reference A-5, that the operator will be able to recover W. The failure probability of containment RBSW immediately 50%of the time. For the re- heat _ removal has three contributors: A-14 4 g. m
unavailability of RHR, unavailability of RilR/ ing SCS actuation. RilR and core spray pumps RCIC in the steam-condensing mode, and un- were failed earlier in the sequence and, based on availability of the main condenser. The values the notes to Figure C.1-5 in Reference A-1, used are from Reference A-5, Table SG.5. makeup to the condenser hotwell from the CST will be unavailable, resulting in the eventualloss FL. Failun of Filtra to actuate is quantified in of the condensate pumps once the hotwellinven-Reference A-1, Figure C.1-2 at 7.2E-40. tory is exhausted. Tius may be overly conserva-tive. For cases where RBSW is restored or CV. Failure to vent through the existing sent transfer to TBSW is successful, the failure proba-lines is quantified in Reference A-1, Figure bility is taken to be 5.0E-01, as quantified in Fig-K.1-2 at 1.0 for sequences where senice water is ure C.1-5. For venting throup, the existing lines unavailable. No basis for this quantification is with the pessimistic reactor building assump-given in Reference A-1. Loss of TBSW will tions, the assumed failure of reactor building cause the Turbine Building Closed-Loop Cool- equipment leads to a failure pmbability of 1.0 for ing Water System (TBCLCW) to be ineffective in event V2. cooling the station air compressors. If the station air compressors fail, a loss of motive air to open the vent valves could result. A loss of RBSw OU. For cases where service water is failed, would result in a loss of control room air condi- cooling wa'er to the CRD pumps from RBCLCW tioning, causing control room temperature to in- is unasallable and Reference A-1 assigns a fail-crease.This could conceivably cause a failure of ure probability of 1.0 to high-pressure injection. control room panel lil1*PNL-VC2, from which if service water is available, event QU is quanti-the vent vals e. are operated. fied as in the case of transient-initiated se-quences. Again, use of the pessimistic reactor V2. Low-cressure injection systems are as- building assumptions for the existing vent case sumed unavailable for long-term makeup follow - leads to a failure probability of 1.0 for event QU, A-- 15
L b
- LI I " ' ! L *F skii j f' . ri 7 fF if t Ij,;= :! I i!l 1 !! hI fttji;
- !t !tbL dl itkIIi 4
D 0 E' T1 CP d I ! F 1 f g F 1 F 1 DR IC WS f f e C-m f e s C n f w C w
- f w
Cp - F-m
@- t w e e c N xu FE k k k e e k o e k k k , e k e o k e o e s e SD o o o l T o Y T o o o y Y o T T o T T T T T EC -.
N-EN y s W F A F A A F A F A w E Sh k o e k k I I k o I I r k k o g I k I I k I ? D n T o I I I I ae o g I o I I o I I I I a E 3 0 e 1 2 4 7 S 5 2 2 3 3 5 7 9 9 e e e C N E. tS 0-E s 1 F t E 1 E S 1 E e-E e 0-E O E c E 1 E e 1 E-e p g g 1 E 0 E-0- E 0 E
- a79 e E e
4 cE 4 E e F e OO 6 0 1 4 0 4 9 1 3 2 3 t a 9 1 8 t. 0 2 7 3 g 1 4 0 9 S. 6 0 1 4 ER O. S. t. _ t. SP 2 4 4 4 S 2 3 2 3 e 7 g 1 4 s 5 t s 1 1 2
- 1. I i st g Ope I L
ev C Gee Pa l t D g t _L 1
.e.
sen j P s- 4 telI nrl V r. e 2 9 epeW 9 CafP c n. F n 2
.g 3 3 3 tn. 2 ni e 4 9 et r L l- 4 cnet E i r F 0 D o owvt N
eti F M 7 2 7 2 Z m 7 l 3 7. e e t.tav lt are W Q E e-E E ne 9 4 0 4 a E I CeHeeP m. ' G. m A L 4 a 1 l i - l c-Ie- t r Pv W a to g h t L n os.. o-te v e " e teP. t pf - o es o_ FO i i t Ie-
.t c-U E ev te o h
t c. 2 . s oe , e trw c oS n ttB tsf 4 ee Fr R u e e q
.- s 1
-t 0 r t e M e t
F o v S 2 a h- 0 .w t 3
.c e
e8 i oo9 tf 1 o vr e F e tn T S F- s te o O er S_ f Ft t 1 o
. ss
.e s o
.r a u L Snl Pc il M
C = Re Me F =. . i 4 f._ee y g A s S T e stu g r ef a u g L 7 i F - h - C
l l INSTRUMENT LINE BREAK INITIATORS The iollowing is a description of the factors in- P(Q) = P(Leg A failure) cluded in the instrument Line Break initiators x P(Leg A-initiated category (see Figure A-5). FW trip) + P(Leg B failure) P(Leg B-initiated ITV trip) A break in one of the RPV levelimtrument ref-crence legs would result in the generation of erro-
= ( 5)RO) + (RS)(R1) neous upscale signals by the differential pressur transmitters associated with that leg. The erro-
= 5'M '
neous high reactor water level readings would OR. This is the probability that, given one leg prompt the operators or the control systems to re-is already unavailable, repairs or tests perfonned duce RPV in,jection (e.g., feedwater flow), creat-on the intact reference leg alto cause a failure of ing an unstable operating environment for the reactor, that leg. With a failure in both refermee legs, llPCI and RCIC will be prevented from operating by the Level 8 trip of the HPCI and RCIC tur-Tr. The initiating event is a leak sufficient to bines. The value of 2.7E-05 is taken from Refer-drain one of the reactor water level reference legs. ence A-4, Figure 3.4-45. Because there are two reference legs, the initiat-ing event frequency of 1.8E-02 per leg calculated Cm. This is the failure probability of the RPS in Reference A4 was doubled, giving 3.6E-02 due to mechanical faults. The value used as the initiating event frequency for the simplified (l.0E-05)is from Reference A-2. event tree in this report. U. This is the combined probability that HPCI and RCIC will fail to provide high prersure Q. This is the probability that continued coolant makeup. The value used (2.0E-02) is tak-power operation will not be possible because of a en from Reference A4, Figure 3,4-45. loss of feedwater, if reference leg A has failed, a high-level trip of the feedwater pumps will occur, X. This is the probability that the oper stor because tw< af the three feedwater level;ransmit-fails to initiate ADS if RPV water level cann< t be ters are associated with the A reference leg (the determined. The values used (5.5E-03 and trip logic for the main and feedwater turbines is 2 3.0E-0!) are taken from Reference A 4 out of 3), if reference leg B fails while the B-side Figure 3.4-45. level transmitter is supplying the level signal to the feedwater control system (FWCS), the false V1. This is the combined failure probability high-level signal will cause the FWCS to de- of LPCI, core sprr.y, and condensate pumps to I crease feedwater flow. This will result in a scram provide low-pressure makeup. The values uad when Level 3 is reached unless the operator (7.0E-04 and 5.0E-03) are taken from Refer-succeeds in taking manual control, if reference ence A-4, Figure 3.4-45. leg B fails with the level signal to the FWCS be-ing supplied from the A-side level transmitters, G L. High drywell temperature will prompt power operation wili not be affected. The initiator the operator to initiate RPV depressurization as frequency is split equally between the two refer- directed in the EOP. This could cause flashing in ence legs. The failure probability of event Q is the unaffected reference leg and a subsequent loss then calculated based on the assumption that the of accurate vessel level indication. Should this level input to the FWCS will be from the A-side condition occur, the operator is directed by the level transmitters 90% of the time. Therefore, EOP to initiate RPV flooding. Failure to do this is failure of reference leg B will only affect power rnodeled by event GL. The values used are taken operation 10% of the time.Thus, from Reference A-4, Figure 3.4-45. A-17 l l
l W. Failure of containment heat removal is a V2. Failure of low-pressure injection follow-combination of failures of RHR, RilR/RCIC in ing SCS actuation is quantified at 5.0E-01 in Ref, the steam-condensing mode, and the main con- erence A-1, Figure C.1-5. For the existing vent denser. The values used are calculated by com- case with the pessimistic reactor building as- , bining the values in Reference A-4, FiEure sumptions, a failure probability of 1.0 is assigned. 3.4-45 for these three components. FL. Failure of Filtra to actuate is quantified in QU. The failure probability of high-pressure Reference A-1, Figure C.1-2 at 7.2E-03 injection is taken to be 1.0E-01 for cases where CV. He value of 9.96-02 used for the proba. venting has occurre . and 1.0 if venting fails. A bility of failure to vent through the existing vent failure probability of 1.0 is used for the existing lines was calculated using Reference A-1, vent case with the pessimistic reactor building Figure K.1-2. assumptions. 1 A-18
1ljll liljl Ilijlji 1lI i lI' l1 I . so c s ri uw n F 2 2 ca + M v es re sc =. h. m.
- s n s e e T m =. k. =. - Mt 1
W r T e T r s w m a er c T Te a =. k s
"w e
r e T r
'O T
.t r
T wam r r e T C r T a c ns us au e w1 r1 , A s m A S
=
sa =. k s k. =. s . a h. =. 1 1 2 maI A r A =. m. =
=
r I r mmr A T A e a e e o n a 9 s . e r e r c s 4 Me- e 4 - - s i
. e e e .
s e
- - e e + e- - -
r e s e - - -a - - s a s s s. .
. 4o r m t
n s. aaw r e m se eae. e . es N e .-.. ww m e t s s e a e s e e e o e e e m . re sr s. n S. i 4 e. 4 s. 4 7 e a s
. e.
3
- t. e.
m s
. s.
4 o. s i a a
. t. s.
s 7 s
- n. c.
s a a s . .
- a. a. s m. s.e T s
s. S s
.i L *
- c. s o%n u e
,l
-a d
..sa
- i
..tx
..i.v a
v E t s c r n " e c n. 1{ Ig 1 t .e . I n aa on er s e e-e e c a. s. ar z # w w vv w e
# 7 7
2..
- 4
- e. e. .i.t a c .e m "
a s . n .s e t a e + e m.a t
. w a
m e m s n a , e s r v s r u m ~ 1 e s to.v y e e e-n.P rn - r e a s.o r c s n o s , r o s r ** e w c
' n e
u q
.w s a
s o-e e s sa.se mr s. a a e sy t k e a e r r.e.a .s s b t.r.* eIE ne e- e
.e n
.r.e.
oe1 w i l t n e
- a. e
- r. <na.
i m
..i a
.r.e u r
rp. ee oi m M I t s n s
=
- u. n = .
.u. T 5 e.r = -
a , A e r u i g F
>i3 ll lllt Jl{l lIlll\t
HIGH DRYWELL TEMPEPATURE INITIATORS The following is a description of the factors in- X. Failure to depressurize the RPV is quan-cluded in the High Drywell Temperature initia- tified in Reference A-1, Figure E.4-5 at 2.4E-03 tors category (see Figure A-6). for the first two hours after the initiating event, assuming that both Divisions I and 11 of emergen-During normal power operations, a loss of dry- cy 125 Y de power are available. This value could well cooling may occur, followed by a rise in dry- not be verified using the documentation available well temperature for which the EOP requires a in Reference A-1. plant shutdown and initiation of drywel! sprays if the drywell temperature reaches 296 F. V1. This is the unavailability of LPI systems, (i.e., LPCI, core spray, and condensate). The val-Th. The initiator is a loss of drywell cooling ue used is taken from Reference A-4, Figure prompting a manual shutdown. The value of 3.4-55. 9.3E-03 is taken from Reference A-4. Fig-ute 3.4-55 and is based on two Licensee Event G. This is the probability that the operator Reports (LERs) in which drywell cooling was will fail to initiate drywell sprays. The value r.ned lost and drywell temperature exceeded 212 F. is taken from Reference A4, Figure 3.4--55. j 1 Cm, This is the failure probability of the RPS L. Given that adequate drywell co'oling is : due to mechanical faults. The value used initially not available, event L represents the con-(1.0E-05)is from Reference A-2. ditional probability that long-term stable cooling is not established by flooding the RPV to Level 3 Q. This is the probability that feedwater is as directed in the EOP. The value used (1.0E-03) not recovered immediately. The value used is tak- is taken from Reference A-5, Table 5F.2. en from Reference A-4, Figure 3.4-55. W. In References A-1 and A 4, because the U. This is the probability that neither HPCI resulting end states are a small contributor to the nor RCIC will be available following a loss of Class 11 core melt frec .ncy, event W (failure of feedwater. The value used is taken from Refer- containment her removal) and subsequent ence A-4, Figure 3.4-55. events were not quantified. A-20
Feecwater iPI Fails ' Fail to LPI Fails Drywell Fail to SEGriNCE SEQtENCE SEuuENCE Hanuel Shu RPS DESCRIPTION tdown Caus Failure unavail. depress. Heat Initiate PHOB. CLASS ed by Hi D RPV Removal DW Coolang M Temp. Fails TH CN O U X V1 G L 8.76E-03 ok ok 4.51E44 ok ok HJDEdlE_ j JJpE-03 4.61E-07 ID Th6L 7.OiE-05 ok ok 3.58E-OS ok ok y s',> L3 J 2E-9? ,
- l 199E-9' 3.69E-09 ID ThGL fLDDEdla_
1 3gE-03 6.61E-O' .sk ok 2Md5 5.02E-11 ID Thot.fV RE&23_ l l s Md3 1.59E-09 ID Th00X LDOE45 9.30E-08 ATWS ThC 1 Figure A-6. Manual shutdown - high temperature in drywell.
SMALL LOSS-OF-COOLANT ACCIDENTINITIATORS The following is a description of the factors in- G. Failure of adequate drywell heat removal cluded in the Small Loss-of-Coolant Accident during the sequence may lead to inaccurate vessel Initiators category (see Figure A-7). level indicadon. Two systems are available at Shoreham for drywell heat removal drywell air coolers and the containment spray mode of RilR. in small loss-of-coolant accident (LOCA) se. As discussed in Referenet A--t, Appmdix B, the quences, the RPV either remains at pressure or depressurizes slowly so that mitigation is possible drywell air coolers are conshiered to be unavail-able for two reasons. First, cooling water supplied through any of the llPI systems. from RHCLCW will de isolated during a LOCA. Second, the drywen fans are not qualified for op- ) S2. The initiating event is a small oreak in the eration in a harsh environinent like the one that reactor coolant system for which RCIC is ade- ' wouM ex in containnu nt h,Howing a LOCA. quate to maintain coolant inventory. The frequen-The f anun' pmbaEty uu d is, theref ore, that as-cy is taken from Reference A--1, Figure 3 A-10 at wiated with manual initiation of containment 8.0E-03/yr. sprays by the operator. The ialue of 5.0E-02 is taken from Reference A-4, Firure 3.4-53 (nmall ' Cm. If the operator does not take action to in' LOCA). sert control rods based on increasing temperature m the suppressica pool, drywell pressur will O. Failure of eyes G will resolt in a high slowly increase, eventually resulting in the gener' drywell temperature that could cause water in the ation of an automatic scram signal.Tiurefore, the vessel level referenee legs to flash to steam, probability that control rods fail to insert is taken Event 0 is the probability that the operator fails te to be the RPS mechanical failure probability of detect this flashing. Regardless of whether the op-1.0F-05 used previously, erator does detect the flashing, EOP will re-c, aire flooding of the RPV and maintenaace of Q. The probability that feedwater fails to be vessel insentory at Level 3 (event L below). recovered immediately is given as 2AE-01 in Therefore, success or fai'are of event O has little Reference A-4, Figure 3.4-10. effect on the outcome of the sequence and has been omitted from the event trees in this report. i U. IIPI failure is the combination of IIPCI This is consistent with the treatment in and RCIC failure. Event U is successful if either Reference A-5. 1-lPCI or RCIC is available and fails if both flPCI Given that adequate drywell cooling is L. and RCIC fail. The failure probability of 9.0E-03 ~ initially not available, event L is the conditional is taken from Reference A-4, Figure 3.4-10_ probability that long-tern. sidh ,ooling is not established by flooding the RPV to Level 3 as di-X. Failure to depressurize the RPV is quan- rected in the EOF The value of 1.0E-03 is taken tified in Reference A-1, Figure EA-5 at 2AE-03 fmm Reference A-5, Table 5F.2. for the first two hours after the initiating event, assuming that both Divisions I and 11 of emergen- W. Event W is a combination of RilR, RCIC cy 125 V de power are available.This value could and RilR in the steam-condensing mode, and the not be verified using the documentation available main condenser. If feedwater is available, a value in Reference A-1. of 2.2E-06 is used. If flPI is available, the value used is the average of the values calculated from V1. LPI failure is a combination of core Reference A-1, Figure 3A-10 with both RCIC < spray, LPCI, and condensate pump failure. The and flPCI available. If IIPI is not available, a fail-value of 6.3E-06 is calculated from Reference ure probability of 1.6E-05 is used as the value av A-4, Figure 3 A-10. sociated with the success of dther core spray or A-22 l
- - _ _ _ _ _ _ - _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ . _ _ \
LPC1 Because it is extremely unlikely that nei- IlPCl/RCIC operation, the only high-pressure ther core spray nor LPCI will be available, the system available for injection at this point in the case where or.ly the condensate pumps are avail- sequence will be the CRD pumps. Unavailability able is ignored in calculating the failure probabil- of the CRD pumps is quantified in Refer-ity of event W. era A-1, Figure C.1-4 at 2.9E-01 for a LOCA with failure of Filtra to actuate, and 7.9E41 for a FL, Failure of Filtra to actuate is quantified in LOCA with Filtra actuated. Ilowever, Reference < Reference A-1, Figure C,1-2 at 7.2E-03. A-1 did not consider the loss of cooling water to CV. The fen. . nrobability of the cu. sting the thrust-bearing oil coolers of the CRD pump vent system is quantified in Reference A-1, Fig- and speed-changer gear-box coolers supplied by the RBCLCW system. This loss would isolate ure K.1-2 at 1.0E-01 for sequences:mtiated by a LOCA' g ,-reiated loads, such as the CRD pumps, automatically on low vessel level (Level V2. Availability of LPI systems is quantified 1) o high drywell pressure (1.69 psig). For this in Refererac A-1, Figure C.1-5 at 2.011-01 for r fxu, it is assumt ' that the CRD pumps will be sequences initiated by a LOCA. For the nisting used only as a last te art with a failure probability vent case with the pessimistic reactor iuilding as. of 1.0f3 OJ if venti g has occurred,or 1.0 if vent-sumptions, a failuie probability of 1.0 is used. ing has not o,cun.d. Once again, for the existing vent case with the pessimistic reactor building as-00, Because the RPV will eventually be de- sumptions, a failure probability of 1.0 is assigned pressurized either by the initiating ever;t or b3 to operation of the CRD pumps. 1 1 1 A-23
[ .~ * . E C N ES S S
)S F s I M 7A fL F S k k k I I k D k k o o 7IS k 1 I o I D k k o o I I k D C I T t'o I I e I I I A SC o o o I I o I
.I 3 8 9 o t1 4 7 3 a 0 0 5 g 5 o 1 2 m 7 0 0 e 9 E
C
- - - - 0 0 0 9 -0 0- t- 1
- 1E- E0-Eo -0 t- 1-1 E-t E- 0
- - -1 *- t E E E E E E N.
E E E1 E 7E: E5 E E E E 1 M 3 0 1 GEe 6 0 9 i 2 9 5 0 U8 e " 4 4 QO ER 7 0 2 5 1 0 0 6 S. M4 3 1 9 g 5 5, 6 0 6 2 1 2 4 5 1 8 s 1 4 e
- s. a. 8 0 1 0 SP 5 1 2 2 9 3 3 1 1 1 i l
1 O 9 0 ei - - - C a 0 E E S v 0 0 0 e 9 CMn U 0_ t I1 1 1[
,wte 1 i t ne 0 c neer 2 E:
IlCp E i e V O t yenD 0 D. pF S he . 2 2 l~
.1 a 3 3 3 tvr 9 4 n t 4 -
E otl L E E D Cni F 0 0 eF 2 2 oV 2_ . n e 7 7 Z
.l B 5 l 0 9 a 1e C t E tev 1 W E E n o m s 0 0 9 oim 2 1 E.
CteP B 1 1
?
e g 3 3 3 ot n 9 9 ta 1 - 4 - E i u1 . E E lt t Q 0 C li Dc Q 0 e . en C FI i .a 1L 2 l l s c C l t l a - - e avi 9 E E w eoa 0 0 yHm r eF 0 0 ^ cR 5 5 6 I e 1 A 3 E 1 Q Pe V L 3 P L
. 3 .
os 9 s te - e eV X E c lrP 2 n i ae f 4 e FD 2 u q
- e s
e A l I l P e O E C HF O L l l e .1 a t a 1 c e m Q s drv E e o m Fe M t a h 5 e r e 0 o
.r -
Sh 1u M E h Pc1 C 0 S 8 ee 0 MF
- i. .
3 7-lA 0
- A lC aO 2
S E Q r e mL Q S L u g i F
'a4 t E
MEDIUM LOCA INITIATORS The following is a desenption of the Iactors in- The value frorn Reference A4,l'igure 3.4-9 was
- aded in the Me Sium LOCA initiators category used in collapsing these es ents.
(see Figure A-8). G. Failure of drywell coohng is the same as The medium 1.OCA is similar to the small that dscuwed for unalli OCA sequences. LOCA but diffen from the large LUCA in ti at . L. Failure to mitiate stable drywell cooling rapid RPV depreuuritatioa does not occur; n quan%.ed m de same nmnner as for Ow unaH therefme, high-pre.sure injection is requued, or the reactor must be Opressurized to allow injec-b A 'C4"C"'" tion by low pressure setems. Failure of vapor supprewion (D) does not appear in the es ent trees W. The LOCA success criteria lishd in both ( d Rhm A4 M m id&- for medium and small LOCAs because contain-k n condenser as a means of containment ment prewurita, ion occurs at a much slower rate, heat remosal for medium LOCA sequences. allowing time for operator action to depressunic W h M Mh hn the RPV to the suppression pom or initiate con-ences do m.elu& it as a siuble path for mitigating tainment sprays to reduce wntiarnent pressure ^ the pressure and temperature rise in the contain-ment (as does Reference A-5). Consequently, tue S1, The initiating event is defined in Refer- main ;ondenser is included as a success path of mee A4 as a break between OM and 0.1 tt2 for event W following a medium LOCA in this re-a water line and between 0.016 and 0.08 f 2t for a gort. P(W)is then the pn&t of the failure prob steam line. livent S i is quantified in Refer- abilities of a) the lloolean combination of the ence A4, Appendix A at 3.Oli-03/yr. failure probabilities of RilR and the steam-condensing mode of RilR/RCIC, and b) Die un. Ci . rhis is Lt.e failuu , robability of the RPS availability of tae nnin condenser. The value of due to mechanical faults. The value used 1.llM15 is calculated from the values in Refer-(I.011-05)is from Reference A-2. ence A-5, Table $C.I. U. The probability that ilPCI is not initially FL. Failure of Filtra to actuate is quantilied in available is taken nom Rettrence A4, Reference A-1, Figure C.1-2 at 7.211-03. Figure 3.4-9. CV. The failure probability of the existing vent system is quantified in Reference A-1, Fig-X. Failure to depreuuriic the RPV is quan-ute K 1--2 at 1.0b01 for sequences initiated by a tified in Reference A-1, Figure !!.4-5 at 2.411-03 g for the first two hours alter the initiatiag event, auuming that luth divisions of emergency 125 V V2. Availability of LPI systems is quantified de power are available. This value could not he in Reference A ., Figure C.1-5 at 2.0!!-01 for verified using the documentation available in sequences initiated by a UX:A. For the existing Reference A-1. vent case with the pessimistic reactor building as. sumptions, a failure probability of 1.0 is used. V1. In Ref-rence A-1, failures of core spray, LPCI, and r ondensate pumps were modeled as OU, The CRD pumps are assumed to be inad-separate events. For simplicity, these events have equate for mitigating the loss of coolant inventory been collopsed into one for this report. There is an during a medium LOCA. Tins is consistent with inconsistency between the LPCI failure probabil- the assumption that RCIC alone cannot p": vent ity med in Reference A-1 and Reference A4. core uncovery. A-25
. j ' ll l ll s
p O I ET CP NI l F i r Ep e C L N L F- c- V J M 1 Lc - t 3s t w wa S 0 e e 9 3 C EE k k t k 1 1 k k t t 1 1 t SD e o s 'S e 5 S e e S S S S S E C Ds Js C C C S 2 r B I I F s I 9, 6 I I t I I k k I I I ? N Eh S h. e I I e I I o e L I I I A E , e 9 0 4 7 e d 9 0 t 9 7 8 C N c - 0 1 0 2 H c e C- 1 s 0 0 C E E E E E E E 5 e E E E X e. 9o Fsa SP r w5 2 2 2 3 e. 5 4 0 2 5 3 1 e 3 1 C e. 1 T s. 2 1 5 2 9 2 6 s 2 2 0 7 1 9 6 t C C. 3
'l]
t t J DA *._
.e w eS C s t
i i e C C
.s e I ~ -
~ 2 E tncl el'L V ? eee Cef d - "2 " L
.g 3 3 tn mi er L - 1 ct tnet F &
ei1 L evw1 e F 4 N L l ee g t.t n 1 v 1 W n se eee e CtteF S g e g 3 ot tai i n 3 ltu Ds el L E li 2 en e Fi c E. 1 1 l 2 1 e s 0
=tw li 9 E
=ae ee o yH r ooF Q D t S
1 s r 8 8 w r e d $ e I l l 1 E u Pe V q L 2 e F 2 s
> M 1 A
- e. 3 C s s C- O l eV X E L lorP at pR D F e ( m 6
L u id e e m Il Cl U m Pe a HF h t r o e 5 h
.r u S Snl Pe l l e
t C d f *e e .
*F M 8 3
A eA uC
'. e r
l O t S u
& dL e
s e _.- l. ig F
>i J9 ?
. li f\l! lil!
LARGE LOCA INITIATORS
'1he following is e tiescription of the factors in. Reference A-1, a value of 1.111-4)2 was used. No cluded in the Large LOCA Initiators category explanation for this dillerence could be found.
(ue Figure A-9). V "' . liased on the discussion in the llNI. re-new of the Shmeharn PR A Weference A-5), the The large LOCA is characterited by rapid RPV probability that the condensate pumps will f ail to depressuritation; thus, llPCI and RCIC will be unavailable, provide ade(luate makeup has been taken to be 6.0F4)l. This contrasts with the value of
' " ' " " """ '^ " " ^
A. The large LOCA initiator is quantified in Reference A-4, Appendix A at 7.0E-4M/yr. Both W. Successful containment heat removal re-water and steam line breaks are treated in the quires initiation of RilR in the suppression pool same event tree. cooling mmle within 25 hoats.The main condens-er is considered unavaihible for containtnent heat Cm. This is the failure probability of the RPS removal because of the fission product inventory due to mechanical faults. The value used likely to exist inside containment. The value used (1.0E-05)is from Reference A-2. was taken from Reference A-5, Table $C.I. D. Failure of vapor suppression is discussed FL. Failure of Filtra to actuate is quantified in in Reference A4, Section 3.4.2.1. Failure of the Reference A-1, Figure C.1-2 at 7.2F40. drywell floor seals was considered to be the dem-inant contributor to event D but was not probabi- CV. The failure probability of the existing vent system is quantified in Reference A-1, Fig-listically evaluated. Reference A-4 used a value ure K.1-2 at 1.0E-01 for sequences initiated by a of 1.0E44 based on engineering judgement. LOCA. V'. For success of V', one core spray pump V2. Availability of low-pressure injection must be available for coolant injection fellowing systems is quantified in Reference A-1, Fig-a large LCX'A. The value of 3.6E-40 for event V' ute C.1-5 at 2 0 Fall for sequences initiated by a was taken from Reference A-4, Figure 3.4-8. LOCA. For the existing vent case with the pessi-mistic reactor building awamptions, a probability V". LPCI is assumed to have only one loop of 1.0 is used. available for injection because the first loop is av somed to be disabled by the initiating event. 'the CU. The assumption is made that the CRD value used for event V" of 1.PE-01 was taken pumps alone cannot prevent core uncovery dur-from Reference A4, Figure 3.4-8. Ilowever, in ing c large I.OCA. A-27 ,
iI'lll\j 1lj ;l1 l 1l ilI 1 N O I ET CP NI E51 1 l 1 F F: F- F C-UC f d M QS W W EE k k w k k " W k W V D SD e o A A o O A A O A A A A "A E C N ES C S JS D 7A EL k k F I B I k k F I S I k F I B I I I I I w T SC e o I I e e I I o I I I I A E 4 3 6 0 6 0 1 2 9 9 1 s e 9 C 0 C 0 1 t- 1 1 1 0-C 1 o- C 0 N. E E E E E E E E E E E E E E E U 7 e 3 C 3 1 2 0 3 3 0 9 0 0 Q 0 5 6 5 2 5 0 9 2 1 6 1 1 EN t. SP 7 s 1 5 2 2 5 2 8 2 8 1 7 7 1
)
s1 J Ece F ev t Q Cue Pn U l e 1 1 i irD . 0 O ee Ft .s 2 4 - - fts Y E E E 0 C IAne d P LsCp or 0_ f 2 M R
.i e 3 3 3 tvr 0 0 C n t - - -
etl L c E E Cni F 0 0 0 eF 2 2 ov 2_ -- N e 7 7 I l 4 t - s g C t.te l _ - v s l W E n g e*wa CteF Q p _ M e i s C n sl
- e emepi d
- E ntuF V O e P D C L 2
s
- c Il - .
Cl
- E s Pa V 3- e LF 1 c n
1 e u q y s s e e e ll o r rpa o "V A CSF C O L s 4 e g s re ls 0 r a ernl ppo D E l epi e 0 Vu F c_ m S a i h e r e 5 o r 0 h S lu M E S P 1 i C Q f a . F 9 M - A eA e r r e A u g L i F y;tac o
!Ir (,ll
ANTICIPATED TR ANSIENTS WITHOUT SCRAM ne following is a description of the f actors in- for at least Iwo minutes. This will af feet later cluded in the Anticipated Transients Without events, e.g., event X, w hich depend in part on Scram category (see Figure A-10). suppression pool temperature. Ta, The initiating event frequency used here P. This event represents the failure of more is the sum of individual initiators listed below than one SRV to reclose as pressure decreases (from Reference A-1, Figures 3.4-14 through during the sequence. This event is quantified in 3.4-19): Reference A-1, Figure 11-8. The value used is that calculated for sequences initiated from high e Turbine Trip With Hypass reactor power (>259 ). and Feedwater (Tt) 3.86/yr
. U. This event matels the failure of all high-
- MSlY Closure (Tme) 0.86/yr pressure injection sources, including feedwater, llPCI, and RCIC. The CRD pumps are assumed
- Loss of Feedwater G f) 1.26/yr to be inadequate for mitigating ATWS sequences because of the relatively low flow rates that they e Loss of Of fsite Power (Te) 0.06/yr are capable of providing. Event U is quantified in Reference A-1, Figure 11-9. Although the calcu-e inadvertently Opened lated values show some variance with the initiat-Relief Valve (Ti) 0.08/yr ing event, reactor power level, and outcome of event P above, they are clustered around 1.0E-Ol; therefore, this value was chosen as a Total (Ta) 6.12/yr representative approximation.
Cm. This is the failure probability of the RPS ML. This event models operator failure to due to mechanical faults. The value used control vessel water level at the top of actise fuel (1.0E-05)is from Reference A-2. (TAO before reachmg 260 F in the suppression pool. Because operator action to lower power by C2. Tc iis the probability that boron will not lowering the vessel level to TAF is only required be injected lo cor. trol reactivity within 30 minutes if boron is not injected, event ML is only quanti-following the initiating event. In Reference A-1, fied following a failure of event C2. Lowering the this event was divided into three distinct time pe- vessel level will result in a decrease in fission riods to account for the increased time for opera- power, according to the Chesal-Layman correla-tot action provided by the use of highly enriched tion, and will allow the operator to reduce power boron at Shoreham. For simplicity, these three to within the calculated limits of the SCS. With events wcre collapsed into one by multiplying the the reactor at normal operating pressure, low eiirg three probabilities together to obtain an overall the level to TAF will reduce fission power to ap-failure probability. The value used is that asso- proximately 15% of rated power. This is withiri ciated with MSIV closure at high powei. it .s tal- the calculated SCS capacity of 18%, but exceeds en from Reference A-1, Table 1h1, and includes the SCS design capacity of 8%. For the existing contributions .bm hardware tailure probabilities equipment cases, the 6-in lines are inadequate (see Reference A-b \ppendix G) and human er- for the espected heat and pressure loads. There. ror probabilities OlEPs) from Reference A-1, fore, event ML is removed from consideration by Appendix D. Because Reference A-1 reported a setting its failure probability 10,0.0. Event ML is very high probability of failure to inject boron quantified in Reference A-1, Figure 11-14. The within the first two minutes, the assumption was value used here of 4.0E-02 is a weighted average made that boron injection will always be delayed of the values given there. l l A-29
1 X, 1:ailure to depreuurire the RPV is a com. limiting makeup flow to the reactor so that toron bination of two events quantified in Reference is not 11ushed out of the core. The event is quanti-A-1, Appendix 11. The first asks whether the op- fied in Reference A-1, Figure 11-16 for the fol-erator acts to inhibit ADS, the second whether the lowing four cases: operator manually depressurires the RPV based on symptoms for w Mch the liOP requires depres- e Case I-llPI available, suritation. The Iailure probability of event X is RPV at high derived by combining these two evems. From preuure: P(Il m l .0lM13. Reference A-1, Figure 11-12, failure to inhibit ADS is primarily govemed by the availability of e Case 2-Manual high-prenure injection. If IIPI is available, in- depreuurization, LP ventory control is considered adequate and the systems controlled: P(lini.OlM11. probability that conditions would exist for initiat-ing ADS is small. On the other hand, if IIPl is not
- Case 3-Automatic available, it is more likely that low of coohmt in- depreuuritation: P(ll>5.0lMil.
ventory will result in a low veuel level, and con-sequently there is a higher probability that ADS e Case 4-ilPI available, will be initiated. Failure of the operator to manu-RPV ilow pre ure P(ll>5.0lM13. ally depressurire the RPV is quantified in Figure 11-13 and Table 11 3 of Reference A-l. Ilecause F hi C 3 s mumed not m mcm the auumption is made that boron injection is b d denurirmion is fu nmre like. delayed for at least two minutes, supprewmn pool Om dmmmirahn by ADS. ten.perature will always be >l40"F at this point in the sequence. Therefort, there is a significant W. Failure of containment heat removal is probability hat the operator will depressurire the quantified in Reference A-1, Figure 11-17. A val-RPV as directed in the !!OP. If bomn injection is ue of 1.0!!-02 for the failure probability of event delayed beyond 30 minutes (failure of event C2), W is conservatively auumed for all sequences, the suppression pool llCTL will be exceeded and even umugh Reference A-1 calculates a value of the operator will be almost certain to depressurite 1.0!MM for dmse sequences initiated by a turbine the RPV, based on direction given in the !!OP. Th a p widmo closure of the MSlVs and no multiple failure probabilities used for event X are summa- mWpen relief valves (SORW). riied as follows: FL Failure of Filtra (SCS) to actuate is quan, Manual tified in Reference A-1, Figure C.2-3. A value of IIPI Bomn ADS Depreuurir-5.551!-03 was calculated and has been founded Asallable injected inhibited' ation Failv" X g g b Yes O!rN3 5.21!-0 2" 5.2 F,-02 f.KIM)2" 1.8lM)2 CV, llecause the existing 6-in. vent lines are Yes No 03N93 No Yes 0.9 I AFr02 131!-02 assumed to be inadequate for mitigating an No No 0.9 131!-02 1.2iM12 ATWS sequence, event CV has been omitted. V1. The probability of failing to provide ade- QU, llPI failure late in the sequence after the quate low pressure makeup is taken to be uniform time at which SCS would be actuated by rising for all initiating events and is quantified in Refer- containment pressure has been conservatively as-ence A-1, Figure 11-15 at 1.0P,-(4. sumed to occur with certainty for all ATWS se-quences. This was done to further simplify the H. Success of this event requires the opera- event trees and is in line with the quantification in tor to ec.ntrol vessel level above the minimum Reference A-1, Figure C.2-4, where the average level necessary for adequate steam cooling while of the calculated failure probabilities is 8.61 Mil. A-30
V2. LPI failure following SCS actuation is sequence with multiple SORVs (TuCmPUll), the quantified in Reference A-1, Figure C.2-5. For rapid pressure increase in containment is as-cases w here SCS is successfully actuated, an av. sumed to collapse the drywell floor seals (these crage of the values in Figure C.2-5 w as used. For seals are internally pressurized with N: to cases where SCS fails to actuate, low-pressure 60 psig), thus causing suppression pool bypass systems were assumed to continue to be and overpressure f ailure of containment. For the available. TaCm sequence, the calculated value of 1.5E411 was used. For other sequences for w hich event I is NOTIk in reality, for cases where SCS has quantified, a value of 3.0E411 was used. failed to actuate, LPI would eventually be lost because the SRVs would reclose as a re. NOTfD For cases where venting occurs sult of either high containment pretsure or through the existing f>-in. lines, a departure high temperature degradation of t'-e SRV was made from the treatment in Refer-solenoids, causing the reactor te ,cpressur- ence A-1, Appendix K. Instead of terminat-d ire above the shutoff head of the LP injec- ing the avent tree af ter event W (as was tion pumps. This is teflected in the model by done in Reference A-l) event I was in-binning end states for which core damage cluded after W ;o model the probabilistie occurs before containment failure into acci. subdivision of containment failures before dent sequence class IC. and after core melt. This is consistent with the models used in Reference A-1 for se-N. Unavailability of suppression pool make- quences where Filtra fails to actuate. up is quantined in Reference A-1, Figure C.2-7. For simplification, a weighted average of the val. For si;nplicity, two events included in the ucs calculated there was used. ATWS event trees in Reference A-1 have been omitted from the event trees in this report. The S. Failure to actuate containment sprays or two events are failure of the automatic trip of the containment flooding is quantified in Refer- reactor recirculation pumps and failure of reactor ence A-1, Figure C.2-8. Ilecause event S was pressure control (i.e., failure of the main condens-calculated to have a failure probability of 1.0 for er and SRVs to limit the increase in reactor pres-eli sequences, it was omitted from the simplified sure associated with the initiating event). esent trees. Reference A-l quantifies both esents at 1.0E-N. Taken in combination with the low probability of
- 1. Containment failure before core melt is RPS failure, the resulting sequences tecome in-quantified in Reference A-1, Figure C.2-9 and significant contributors to the overall ATWS core Tables C.2-8 and C.2-9. For the very high power melt frequency.
A-31
mg..sa. n . m.. no... " u. e.n. wr..J....i. .
,, e . n.
e.Ia,i,.q.i.!
, ,.f I
T& CW Cf P U BL M ._
,L s_ ant-er i Bak-ei 1.
.m.
M [+00 L L121:02 SI
'8 *"
APERTURE CARD Figure A-10, Shoreham ATWS sequences. Also Availat>le On Aperture Card A-32 _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ . _ _ _ . _ __m._ ___ _. _ _ _ . . _ _ _ _ _ _ _ _ _ . _ _ . _ _ _ _ _ _ _ _ _ _ _ . _ _ _ _ _ _ _ _ _ _ _
LPI P;tle h . felle to Comt. host No Cent. Cint. IP meseop Ctet. fatte DftEN2 SfFJPM3 tentN a &;7el tweevel Pelle V:ntina vte eer-*es. Paine 4 3f ere Cad 89c6. M ae Nt rtitee feine fe'W in). . . , . _
~
V8 H _e FL V8 W I 0 SE*00 ok 4.47E-06 ok
, 3. 4 7E-07 ok 18 M -0 8 8.882-08 IVS g, A.AWi 0.0 4 -08 IW t.98E-09 1C 8 N *0 I g.itt-lO IVA haE-ed 4.8tE-05 10 9.4W44 ok
, 0,40E-09 ok IE N48 1,03E-09 IVE L D.DE-e t IW g, 4,ett-08 gg, _ , 4.04E-40 ft L3: " *o' 4.daE-ti IVA
, t.saE-Os IVs IS M-t i s.soE-40 tVA l' eBF et gp
, 4,ggg.gg
,,no,,e, , 1.04E-88 3C 11 o$'41 4,46t-te 3YA 4.88E-09 ok
, 3.8tE-00 ek it 20E-of g,ggg.gg gyg g ang e, f*EDE*0I 0.34E-00 !W g,ggg g , l . S M-10 IC d"*1 e.4aE=tt IVA
, 8.90Ea$7 IVS ls . w -e t gyg g,g4g.gy 8 N ***
, _ e n,,,, , , 8.04E-C7 !W
' p or-es r I'I'E*0' IE
'i"~** 6.pse-60 !YA K :0d - .
B. TEE-10 20
- 8. 07E-04 IA 4.9BE-08 ok
, 4.9tE-OS ok IE N ~Oi 3.08E-08 3YS 2 00E"Oi
, ,u ,
- 8.Deta+9 IW g,c ,
-- 8.ltE-10 3C
' * # ~0 8
- 9. car-tt VA E -od 6.0tt-10 10 2.72E-07 ok
, 4.04E-09 ok LLE24di-- 4. 8 4E-08 !YS yge, AskC6 ** I E.46E-t0 tW
- - -s.ner.es , t.tBE-st 2C is.ene-ci 4,g4g.gg gy3
, t.53E-10 !YE IS * *01 6.66E-ti !YA
"#*i
,_eng.e3 6.47E-14 IW gg 1.15E-89 3C 3 # '* I
- 4. 95E-13 1/A 5.17E-07 ok
. r. t.98E-09 ok 9/0408'O287-o I .- ,
_ _ _ _ _ - _ . - _ _ _ _ _ _ _ - _ _ _ . _ - . _ _ - _ _ _ . - _ _ _ . _ . _ _ _ - _ _ _ - _ _ . - - - . _ _ _ _ _ _ _ _ - - _ - . - _ _ . - _ _ - . . - . - - - J
~
1
,1. nce -e i LkQL$d 1.snr-gg ,_
J J et-cn ,. 1 l LCDL.De t.nne-cr __
.1 LSDE:04
- 4. eor-er L aat:St L99L_G2 L
L 20L-21 1132.:9 A 1.PDF-B2 Figure A-10. (continued). j
l E- * + - - ' L eat-et s.17t-ce
~
_ IVe m _e, > a* + i . 0 = 4. :# g gg m t.192-it IC 10" ~0 ' e.soe-le !YA
- 4. t it-ce IVA
,gne.cl .. N Aff *01 1.16t -04 !#
a ew-es 3.4st-60 IVA 8.90E-11 %
- 7. 64t4s IA
~
, s set-07 ok
**~0' i . i nt-0 IVA E :EN ~01 0.582-07 !YS 1 Saldi 4.181-07 !#
Agtums ._ 0.est-03 2C
~
D 3.722-00 !YA
, 5.7er-08 IV6
'3 0M di 3.44E-Og IVA
,gog e3 LOM-M t.04E 45 IW g g_g , , 4.Det-it IC
'3 * ~01 1.378-18 !YA k.098-10 !D
_ r -- i.01248 ok 1' - -- 4.84249 JYA gs - i.itt-ce IV8 ME~0 ' 4.742-09 !YA 8 N "Oi 7.86t-09 3F j, , 4.00E-10 1C
'3 N "Di 3ga
- s. pat-gi
, t. IRE-li !YS is 0--e ' -
m e, n ~4'
. 0n-in
- 7. en-2vA Al'ERTORE
, , ng, , 1. tot-iS 2C CARD 13 N -Oi 3.96t=14 IVA ry
'8 ~ 4 '
4.ese-Ce Also AWM'N
# " ~0' t . 0m-Os 2VA Agrture Card 1.78E-08 IF
'gg e3 , 3.8M-10 3C
'8 W *ol g.geg.lc IVA 8.052-82 10
, 0.83E-40 !YS
# d '
3.782-i0 IVA
- 2. m -gi 3.ittair IW I g g.e , , t.ast-it 2C I ME~C 8.SSE-it IVA
, 7.03248 ok ti . EM-e t 1.842-08 IVA E # ~0' f O.03t-00 3YS
'# # ~04
- 4. 33E-08 IW f mg, , 0. 8 E-10 IC
# ~0' 3.921-i0 IVA
. 35t-08 IVS d"d' B.772-09 !YA gew 2_ m-o f 4,ggg.00 IW j e . c e-e s , 8.02t-10 IC 13 # ~01 4.382-11 IVA S.42t*li ID
, e. 4e-0. :C A-33 l _fto9 c tg(c 2 8 7 - a g. -
RPV RUPTURE INITIATORS The following is a description of the factors in. Therefore, this event is assessed in Refer-cluded in the RPV Rupture Initiators categorv ence A-4, page 3-166 at 0.3. (see Figure A-11). R. Tb: initiating frequency of a reactor # ""I"I" * * "" I* I I " E "" ' I'" N 'I' ence A 4. Appendix M) of 2.5 times the design pressure vessel LOCA is estimated in Refer-basis (i.e.,135 psia), the pressure suppression ca-ence A4, Section 3.4.2.5, as 1.0E-05 per reactor pa ty of tk suppression pool is esthnated at
. year. This value estimates the frequency of a me-four times the design basis LOCA. Therefore, dium-sized or larger LOCA occurring in the given an RPV rupture larger than a DB A LOCA.
Pressure vessel, ' only 10% of these ruptures would exceed the vapor suppression capability of the suppression
> S1, This event estimates the probability that system, the postulated RPV rupture, if equated to an equi-valent-sized LOCA, will be larger than a medium W. If the rupture is larger than a medium LOCA (SI). The fraction of RPV ruptures esti- LOCA but smaller than a large LOCA, the proba.
L mated to be larger than a medium LOCA is 30% bility of containment heat removal failure is taken This value is based on the conditional probability to be 1.1E-05, the value used for medium LOCA , that if a LOCA occurs it will be a large LOCA i.equences, if the rupture is larger than a large (see Reference A-4, Appendix A.1). LOCA, event W is assigned the large LOCA value of 1.1E-04.
> A. This event estimates the probability that, Fl. Failure of Filtra to actuate is quantified at
' given an RPV rupture at least as large as a design basis accident (DB A) LOCA, the equivalent the usual value of 7.2E-03. break size exceeds that of a large LOCA ( A). The CV. Failure of the existing vent system to ac-basis of this estimate (0.1) is given in tuate is quantified. as before, at 1.0E-01. Reference A-4. V2. The probability of LPI failure is taken te L. The most likely scenario for the RPV be 2.0E-01 for sequences in which venting is rupture exceeding the ECCS capability is for the through Filtra or the existing lines with the opti-RPV to leak initially before failing catastrophi- mistic assumptions. For the existing vent case cally. The result of this scenario is the relatively with the pessimistic reactor building assump-slow development of a failure in which the tions, a value of 1.01s used. suppression pon' is likely to provide adequate va-por suppressir.' * : the ensuing blowdown. The OU. Event QU is not quantified because the . slowly developing rupture is estimated to be CRD pumps are incapable of mitigating a me- ! twice as likely as a rapidly developing one. dium or large LOCA. l l l l l o A-34 Y a.- . .. :, - . ,. - -. . a . , .,--.-_.-._..---.__,---__.--__.__._.:__-:__.:
ft 5 Lit!tLT!f Iij!h>:>s l:iki 't6 . l teiE f };tIAt! ,,I t ' l! ii 'f ;ii[;t Lt ' 66>t[t. N O I ET CP NI ER LC eS V EE SD 1 S A = Rt_ R _ E C N EE _ US A A D BA EL SC t e s = r I B I I a a F I sI I I I I I I I I I I E e s s 3 e e 1 m r e 9 C 4 s i 1 e- s 1 t o c- 0 EN. E- E E E- E e E E E- E E Us WO 6 s s 4 e e 0 e o 0 0 0 s. s. 5 7
- s. 3
- s. s.
1 0 ER 5P 7 s s 5 a a S c a S 9 e. p1 m1 Pv ue u e o Dm 1 3t _ (
.IP * '
tee
.ssL a 4 n rl e
Cee pl DF V w* " - I k 8
.e .
ti s nwS . e-a C t cts F s n n oe a _. NV 7 7 l e S d t .te l e e- _ neevl see W F F CeHo F a n _ R t. i. i 1 . p - ee o
- rdrc
- _ deo V -
tep. ycap Nexvp u s
- s m
a eng kei r e e i u err Lou t
. e q
e ft E Vep bu e s. e P R P D. r 3 t u A p C u e- O i r rsenL u A O V tgee > - P pacrhg u E R - M R l - o. r t.
- t 1 .m a
h
. e r
e
- o r A t u C S
- h t r O >
S penL u e R 1h t s 1 1
'e r - s e s A o rlr t se ue u c ss t R e
e r eee a e E o u g RrvN P n. i t F Ywu _
,di I<lii a :l3j i 14,iJ 3 ' < 4;i! ,2 ' j ;i!! .i,.i.<i< i i i / ,'4 tiIltIl_
APPENDIX A FOOTNOTES
- a. From Reference A-4, Section A.I.3.2. .i . From Reference A4, page 3-47 and Figure 3.4-1.
- b. By initiating event definition.
L. From Reference A-1, Appendix 11.3,
- e. From Reference A-1, Figure 3.4-l. page 11.345.
- d. From Reference A-1, Figure 3.4-3. 1. Frobability that the operator inhibits ADS.
- e. From Reference A-1, Figure 3.44. m. Frobability that the operator fails to manu-ally depressurize the RFV.
- f. From Reference A-1, Figure 3.4-7.
- n. Average of the values in Reference A-1, j
- g. From Reference A-1, Figure 3.4-2. Table 11-3 for which suppression pool temperaturt is >l40 F and ilFCI is
- h. From Reference A-4, Figure 3.4-1. operating.
- i. From Reference A-4, Figure 3.4-1 and o. Average of the values in Reference A-1, Appendix A.2, probability of recovering Table 11-3 for which suppression pool the MSIVs within 15 hours. temperature is >260 F.
h A-36
REFERENCES A- 1. E. T. flurns et al., Shorcham Nucicar Power Station Full Power PRA PRA Update: Supplemen-tal Containment System implementation, IT/Delian Corporation, piepared for the Long Island Lighting Company, February 1988. A-2. U.S. Nuclear Regulatory Commission Anticipated Transients Without Scramfor 1.ight Water Reactors, NUREG-0460, March 1980. A-3. FausLe and Associates,Inc..MAAP Analysis to support Shorcham IWYA Power PRA, prepared for the Long Island Lighting Company, March 1988. A-4. Science Applications, Inc., Probabilistic Risk Asse.ument. Shoreham Nuclear Power Station, prepared for the Long Island Lighting Company, June 24,1983. A-5. D. llberg et al., A Review of the Shoreham Nuclear Power Station Probabilistic Risk Assessment, NUREO/CR-4050, November 1985. A-37
APPENiilX B QUANTIFICATION OF THE SHORE';AM SIMPLIFIED CONTAINMENT EVENT TREES n-1
CONTENTS l 1 APPENDIX 11 ACRONYhtS . . . . . . . . . . . . . . . . . . . . . . . ... ......... . .. 11 - 5 CLASS IA PLANT DAh1 AGE STATE . . . ... .. ..... .... ... . . . 11 - 7 CLASS IBI PLANT DAh1 AGE STATE . . . . . . . . . ... . . . . . . . .. 11- 1 7 CLASS IB2 PLANT DAhtAGE STATE . . . . ..... ...... .. . ... .... . ...... 11 - 2 3 CLASE IC PLANT DAhiAGE STATE ..... . .... .... .......... . .. ... 11 - 2 9 CLASS ID PLANT DAh1 AGE STATE . . . ........... ... .. ... ............ 11 - 3 9 CLASS il A PLANT DAh1 AGE ST ATE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 11 - 4 5 CLASS illl PLANT DAh1 AGE STATE . . . . . . . . . . . . ..... ....... . .... . .. B-47 CLASS IIF PLANT DAh1 AGE STATE . . . . . . . . .............. . .. . .. .. 11 - 4 7 CLASS lil A PLANT DAh1 AGE STATE .... .. .. ..... ... . . . ....... 11 - 5 0 CLASS illB PLANT DAh1 AGE STATE . . . . .. . .. .. . . .. ... ...... 11 - 5 5 CLASS illC PLANT DAh1 AGE STATE .... . .. . .. ... .. ... ... B-59 CLASS illD PLANT DAhiAGE STATE . . . . . . . .... ... . .... .. . . ... B40 CLASS IVA PLANT DAhiAGE STATE . . . . . . . . . . . . . . . . . ..., . .. .. .. .... 11- 6 6 CLASS IVF PLANT DAh1 AGE STATE . . . . . . . . . . .. ., .. .. .... . .. 11-68 CLASS IVG PLANT DAh1 AGE STATE . . . . .. . . .... . . . ..... . .. .. 11 - 7 1 REFERENCES . . . . . . . .. . ... ....... ......... . . . ... .. .. 11 - 7 3 FIGURES 11 - 1 . Shoreham Class I A SCET with SCS . . . . . . ..... . .... ....... .. ... 11 - 1 4 11-2. Shoreham Class Illi SCET with SCS . . . . . . . . ... . ..... .. . ... ..... 11 - 1 9 B-3. Shoreham Class IB2 SCET with SCS . . . . . . . . . .. ..... ... 11 - 2 5 B -1. Shoreham Class IC SCET without SCS . . . . . . . . . . . ... .. ........ . . 11-32 Be. Shoreham Class IC SCET with SCS .. ... .... ....... . .. .. . 11 - 3 5 B +. Shoreham Class ID SCET with SCS . . . . . . ....... .. . . B-41 B-7. Shoreham Class ll A SCET (no SCS) . . . . . . . . ... . .. .. 11 - 4 6 11 - 3
B-8. Shoreham Class 11B SCET (no SCS) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .... 11- 4 8 B-9. Shoreham Class !!F SCET with SCS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 - 4 9 11 - 1 0. Shoreham Class ll! A SCET with SCS . . . . . . . . . . . . . . . . . . . . . . . . . .......... 11-52 11-11. Shoreham Chtss lilB SCET with SCS . . ...................... ...... 11 - 5 6 B-12. Shoreham Class IllC SCET with SCS . ............... ........ . ... 11 - 6 0 B-13. Shoreham Class IIID SCET with SCS . . . . . . . . . . . . . . . . . . . . . . . . . . . ..... 11 - 6 5 B-14. Shoreham Class lVA SCET (no SCS) . . . . . . . .... ............. . ..... .. B-67 B-15. Shoreham Class IVF SCET with SCS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B-70 B-16. Shoreham Class IVG SCET with SCS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B-72 TABLES B-1. Early containment failure results for even: RB . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B-13 B-2. Percentage of event F1 failures . . . . ............................. ........ 11 - 6 8 B-4
APPENDIX B ACRONYMS ADS Automatic Depressurization System h10V motor-operated valves ATWS anticipated transient without scram N11TR mean time to repair CET containment event tree NPSil net positive suction head C1 containment isolation PRA probabilistic risk assessment CRD control rmi drive RilSW reactor building service water CST condensate storage tank RCS Reactor Coolant System ECCS Emergency Core Cooling Sy stem RilR residual heat removal llPCI high-pressure coolant injection llPI high-pressure injection SBO station Nackout LOCA loss-of-coolant accident low-pressure coolant injection SCS Supplemental Containment System LPCI hiAAP hiodular Accident Analysis Program SRV safety / relief valve B-5
_ _ _ _ _ _ . _ . __ _ _ . _ _ _ _ _ _ _ . _ _ . _ . ~ . _ _ . _ . _ _ _ APPENDIX B QUANTIFICATION OF THE SHOREHAM SIMPLIFIED CONTAINMENT EVENT TREES Although simplified over those presented in a discussion of the event quantification schane the original probabilistic risk assessment (PRA), based on the assumptions of the Shoreham the event trees used in the present study are still PRA tan-2. Events that are modified by the more i quite large, with some requiring three 11 x conservative reactor building assumptions are 17-inch fold-out sheets. This appendix contains noted in the text of the main report. , CLASS lA PLANT DAMAGE STATE The following is a description of the factors in- ing core melt. Although the Mark 11 containment cluded in the Class IA Plant Damage State is nonnally inerted with nitrogen, plant Technical category (see Figure B-1). Specifications do allow operation up to a certain power level (159 at Shoreham) for a limited peri. PD. The Class I A plant damage states are od of time with the containment de-inerted. The transient-initiated sequences involving a loss of value of 1.0lM)3 is given, without accompanying coolant makeup with the reactor remaining at explanation, in Reference 11-!, Figures A-2 and ! high pressure. Il- 15. RI. This event describes a failure to recover VF. Failure of this event denotes RPV failure i the sequence through low-pressure coolant injec- at low pressure. If the RPV fails at high pressure tion prior to vessel failure. Only low pressure sys. (upper path of event VF), a dispersive core melt tems are available for in-vessel recovery because exit is assumed to occur. Due to the unique con-high-pressure injection (llPI) systems are failed figuration of the Shoreham vessel pedestal re. by sequence definition. However. for these sys- gion, dispersal of corium onto the drywell floor tems to be employed, the reactor pressure vessel outside the pedestal is not considered likely un-(RPV) must be depressurized. As discussed in less the vessel fails at high pressure. Therefore, Reference 11-3, failure of early recovery is domi. the lower path of event VF is assumed not to lead nated by hardware failures that disable both the to tn overtemperature condition in the drywell Automatic Depressuritation System (ADS) and because the prerequisites for debris-heating of non-ADS safety / relief valves (SRVs). Of the cut- the drywell atmosphere and intemals will not be sets for ADS failure.219 are hardware faults that fulfilled. Vessel failure at high pressure can also affect all the SRVs. Therefore, event R1 is as- provide the driving force for an energetic puff re-signed a failure probability of 2.llMil. lease in cases of early containment failure. The value of 1.0lM)2 is taken from Referens B-3, Cl. This event models failure of containment Figures A-2 and D-15, and is based on engineer- , isolation or the presence of a pre-existing leak. If ing judgement. containment integrity is not maintained, an early fission product release is assumed to occur and DC. Failure of the pedestal downcomers pro-later containment failure is precluded. The value vides a path for steam escaping from the vessel to of 6.3E-03 is taken from Reference B-3, Fig- bypass the suppression pool.This results in a deg. ures A-2 and D-15. No basis for this value could radation of vapor suppression capability and in-be found. creases the probability of early overpressure failure of the containment. In addition, failure of H1. This is the probability that a hydrogen the pedestal downcomers creates a path for fis-bum occurs inside the primary containment dur- sion products to hypass the suppression pool, 11 - 7
eliminating the mitigatise effects of pool F1. Failure of this esent denotes early over-scrubbing. The value of 2.0E-01 is taken from prenure containment failure. The likehhood of Reference B-3, Figures A-2 and Ibl5. early overpressure failure is determined by earlier events in the CET, speciGcally 111, VF, DC,112, H2, The probability of a hydrogen burn oc. and VE. First, the auumption is made for class curring in the primary containment following lA that if early venting through the SCS is vessel failure is given as 1.0E-02 in Refer. succeuful, early overpreuure failure will not oc-ence B-3. hgurn A-2 and D-15. cur. Second, as discuued above for event VP, failure of the vessel at high pressure, per se,is not VP, The lower path of event VP implies that likely to cause containment preuure to eweed containment prewure at this point in the sequence the SCS actuation setpoint. Therefme, for Claw is leo than the Supplemental Containment Sys, IA, high preuure venel failure alone k awumed t be incapable of causing early overpressure fail-tem (SCS) actuation setpoint of 60 psig; there, fore, no early venting to Filtra occurs if no ure. For the condunation of vessel failure at high hydrogen burn or pedestal downeomer failure has prenure along whh pedestal downcmun failure, occurred, then the lower pathway probability of Reference B-3 assigns event F1 a f ailure proba-event VP is high because the only contribut5r to bility of 1.5E-03 in Figures A-2 and D-15, but a containment overpressurization is failure of the detailed basis is not given for how this value was veuel at high prenure. Although this does cause detennined. If the vessel fails at high prenure and a pressure spike in the containment, the a hydrogen burn occurs either during or alter coie Shoreham hiodular Accident Analysis Program rneh, Refuence B-3 um a failure probability of (M A AP) analysis indicates that it is not likely to 7.5E-02 for event R Again, the basis for this be large enough to actuate SCS Dd A probability cwid not be identified. Finally, for cases where of 9.0E-01 is given for this case in Refer- an early or late hydmgen burn occurs along with ence 11-3, Figures A-3 and D-15. On the other downcomer failure Reference B-3 uses a value hand, hydrogen deflagration or downcomer fail- of 5.7E-01. This value is apparently taken from Table A-18 in Reference B-3; however, the other ute will most likely result in containment pres-sure exceeding 60 psig, even in sequences w here values could not be derived similarly. the vessel does not fail at high preuure, in this PF, A driving force for a puff release (also re-case, a value of 1.0E-02 is used, again taken f rom W to e a enepic releM h for e Reference ti-3. For the case where the vessel quences invohing the lower branch of event PF. fails at low pressure with no hydrogen burn or A puff release is only of concern in cases of early downcomer failure,it is assumed that contain-overpressu e containment failure, so event PF is - ment pressure will be less than 60 psig ut this
, only quantified for sequences involving a failure point in the sequence, implymg a lower pathway of event Fl. For Class I A, the dominant contribu-probability of 1.0.
for to the driving force of the puff release is venel fauure at high pressure. Ilydrogen burns are not VE, For those cases in which containment considered because they (io not provide a sus-pressure exceeds 60 psig (upper path of event tained driving force; the pressure spike following VP), event VE asks whether SCS actuation a hydrogen burn is of too short a duration. There-occurs. The value of 7.2E-03 used for VE failure fore, for Class IA, the probability of a puff release probability is from the quantification in Refer- is estimated from the probability tha' the vessel ence B-1, Figure C.1-2. Note that this value for fails at high pressure. event VE is not applicable to the Containment Event Trees (CETs) that are used for venting R2- Event R2 models the failure to establish through existing equipment. A value of 1.0 is ex-vessel debris cooling using low-prewure in-used for event VE for existing equipment venting jection systems. For cases involving early con-because the early containment pressure rise is tainment failure, Reference B-3 assumes that very rapid for the Class I sequences, reactor building equipment is failed by the harsh B-8 1
l l enuronment that ruults when the atmosphere in makeup water to flow from the CST. Assuming a j the primary containment is released into the reas - vacuum can be maintained using either the me-tor building through the containment break. Ad- chanical air removal pumps or the steam jet air ditionally, the followmg high-capacity low ejectors with motive steam supplied from the pressure systems with water sources and pumps auxiliary boiler, use of the condensate pumps located outside the reactor building are also may still not succeed because there is an uncer-unusable: tainty as to w hether the II 10 ppm flow rate { would be sufficient to establist a coolable debris e Service water pumps injecting through bed. Therefore, the condensate system is conser-the ultimate heat sinA connection to s alively assumed to be unavailable for euessel the residual heat removal (RHR) sys. recovery. tem: This source cannot be used be-cause the two motor-operated valves if an earh containment breach has not oc-(MOVs) that must be opened are lo' curred, then the failure probability of event R2 is cated in the reactor building. ne mo' based on two factors: the likelihood that adverse tot operators are likely to be disabled wpprenion pool conditions (inadequate net posi-by the post-containment failure atmo-tive suction head (NPSil) and clogging of suction sphere of the reactor bunding, strainers by core debris) will preclude operation of RilR and core spray pumps, and the probabili-e Condensate transfer pumps injectinX ty that the operators fail to align alternate injee-through the loopfill connections to c tion systems, in this case, service water to Ihe ther the RHR or core spray system: RflR ultimate heat sink connection. Refer. Use of this source would require oper- ence 11-3 quotes a value of 1.0E4)! for the prob-ator entry into the reactor building to ability that adserse suppression pool conditions open the loop fill valves. This uould will prevent the use of the Emergency Core Cool-not be possible under the conditions ing System (ECCS) pumps, and a value of pas ulated to exist in the reactor build- 3.0E-02 for the probability that operators fail to ing following early containment align alternate injection systems. The source of frilure. this value is given as the Peach flottom analysis done for (Draft) NUREG-il50 (Draft NUREG/ e Fire pumps injn ting through the ulti- CR-4550, Volume 4). Ilowever, Reference 13-3 mate heat sin 4 connection to the RHR inciues the diesel fire pump and the condensate system: Use of the fire pumps would system a.nong the alternate injection sources con-not he possible because a manual hose sidered to ! 'e available. In the present anPvsis, the connection would be requbed in the condensati system is assumed to be urmilable reactor building, again necessitating for the nasons dir.u.ed above. Additionally, the operator entry. diesci oc 6"k notor-biven fire pumps could supply water at a t ow rate m '500 ppm each, but The only other water source would be the con- use of this source s'ould requin operator entry densate and condensate booster pumps injecting into the reactor buildy to ediish a hose con-through the feedwater lines with suction from the nection to the reactor building service water condenser hotwell. This is a high-capacity (RBSW) system. The assumption is made here source; however, the storage volume in the hot- that operator entry into the reactor building fol-well is sufficient for only five minutes of opera- lowing vessel feiture will not be possible because tion at rated flow. Makeup to the hotwellis of the expected high radiation les els in the reactor provided by " vacuum dragging" water from the building following vessel failure and dispersal of condensate storage tank (CST), with a makeup core debris onto the drywell floor, along with the capacity of 1000 gpm. In addition, a vacuum possibility that fission products might be released must be maintained in the condenser to enable to the reactor building through containment pene-B-9
. rations, which may have been degraded by ele- where they were obtained. In Reference B-2, a vated containment temperature. Therefore, value of 1.0E-02 was used, based on a lack of service water is left as the only viable means of proper operator action. If this is taken as a base-injection should u.*ppre' ion pool conditions pre- line value for the case with no venting and no in-vent the use of ECCS pumps. s essel recovery, then the other two values appear reasonable because successful early recovery du)uhl increase the likelihood of successful con-Alignment of RBSW to inject through the tairunent heat removal, while early venting may RilR system requires that the ot>erator open two
~
MOVs that are normally kicked in the closed po. ads erwly affect NPSil to the RHR pumps, mak ing the success of event HR less likely. For the se-sition by a keylock switch in the main control ream. If either of these valves shouhl fail to open, quences without SCS, early venting through the ase of the ultimate heat sink for debris cooling SCS is not a consideration and the value used is detennined by the outcome of events R1 and R2. will not be possible. Using data from Refer, ence B-2, the probability of this happening is es- . FaHuw of late containment vent.mg is timated at 1.5E-4M/ demand. Maintenance errors could also affect ultimate heat sink r.vailability. quantified in Reference B-1,1iguie C.1-2 at 2EM for Filtra venting, and Figure K.1-2 at Maintenance unavailability is estimated (again 1.0E-01 for the existing vent.The fault tree used using data from Reference B-2) at 1.4E-4M. The to quandfy dw faHum of thm ng vent padiin-Reference B-3 estimate of 3.0E-02 for operator dicates that both the drywell and wetwell vent alignment error is probably conservative because Unes nmu actuated to sucMuHy whm du- i it is based on action occurring within two to four hours of vessel breach. As Reference B-3 mdi- contajnment pressure buildup following a loss of C *""""""'I"'"'"'*"*"l cates, the minimum time window for operator ac-tion following vessel breach is more likel* to be OT. For sequences where recovery fdlowing on the order of ten hours. However, because there vessel breach has failed, long-tenn radiative heat is a large uncertainty associated with this human transfer from the core debris to the drywell atmo-error probability, the Reference B-3 value has sphere and internal surfaces can cause drywell been preserved in this report. Thus, for Class IA temperature to exceed the thennal capability of sequences with vessel failure at high pressure, but containment.This debris-heating of containment with no early containment failure, event R2 is is assumed to occur only if the vessel has failed at assigned a failure probability of (l.0E-01) high pressure, causing the core debris to be dis-(3.0E-02) = 3.0E-03. For sequences where the persed outside the sessel pedestal region. If the vessel fails at low pressure, it is assumed that es' vessel has failed at high pressure (upper branch of tablishment of a coolable debns bed is not neces- event VF), drywell overtemperature is taken to be sary because nearly all dehns is expected to flow extremely likely, with event OT assigned a into the suppression pool; the accident is recov- probability of 9.9E-01. ered outside the vessel in the suppression pool s For this case, event R2 is assigned a failure proba- F2. This is the probability that late contain-bility of 0.0 to remove it from consideration. ment failure occurs due to thennal and/or pres-sure loads. The value used for event F2 is HR. The failure probability oflong-tenn con- sequence-dependent. For recovered sequences tainment heat removal is conditional upon the (in-vessel or ex-vessel) in which containment prior sequence of events in the SCET. If early re- heat ~moval has failed and the SCS (or existing covery (R1) was successful, a probability of equipment venting) is either not etuated or is un-1.0E-03 is assigned. if early recovery failed, but available, late overpressure failure is considered early venting was successful, a value of 1.0E-4)1 certain. For sequences where ex-vessel recovery is used. Finally,if both events R1 and VE fait, and late venting have failed, but no drywell over-P(llR) is taken as 1.0E-02. No basis could be temperature condition occurs (upper branch of found for these values in Reference B-3, which is event OT), Reference B-3 assigns event F2 a fail-B-10
ute probability of 3.0E-01 based on the time re- in order to reduce the size of the esent tree. For quired to exceed the ultimate pressere capacity of thermal failures with drywell temperature the containment. Finally, for sequences with fail- >1200T, Reference 11-3 estimated the f ailure ute of event R2, and overtemperature in the dry- size to be on the order el 1.0 it.2 For this case, well in conjunction with successful venting to event BS was assigned a value of 1.0. Filtra, Reference 11-3 assigns event F2 a failure probability of 5.5E-01. This is supposedly based HU. For sequences in w hich late thermal fail-on a high drywell temperature occurring within ute of the containment is prevented by ventmg u 18 hours and a failure to repair failed attemate in. SCS, event ilU asks whether the SCS relief jection systems that would provide some cooling valves open to depressurite the SCS, causing for the debris bed and containment atmosphere, imble gases to be released without hold up for using a mean time to repair (MITR) of 19 hours. decay. For Class IA sequences, Ref crence B-3 llow ever, calculating P(F2) from the cquation auigns event 11U a failure probability of 9.9E-01. P(F2) = exp (-t/M'ITR) SR Success of event SP implies that the re-lease is directed to the suppression pool w here F IVCS fission product scrubbing occurs. Failure implies that all or part of the release bypasses the pool. P(F2) = 3.9E-01 for t = 18 h. Reference B-3 eniculates the failure probability of event SP using Figure A-12 for sequences However. because 5.5E-01 is a conservative with the SCS available. This method was fol-value, and there is uncertainty in the time avail- lowed for this report, with suitable modifications able to establish debris cooling, it has been re- to the table to account for the explicit inclusion of tained in this ieport. events C1 and DC in the simplified CET. The re-sults are as follows: BS. The probability that the containment break site is large 2(1.0 ft2) s determined by the a if containment isolation fails, rate of pressurization and the thermal loading. P(SP) = 6.0E-01. The values were taken from Reference B-3. Ear-e if early overpressure failure occurs, ly overpressure failure is assumed to be due to a rapid pressure rise, for which Reference B-3 as. accompanied by pedestal downcomer siFns both large and small breaks a probability of failure, P(SP) = 1.0 for a small break 5.0E-01. Failure of containment isolation (Cl) is and 5.5E-01 for a large break, assumed to be equivalent to a small break, giving P(BS) = 0.0. Sequences with late containment
- II "'I fY " *"'P""'# I"II"'" ""# "'
failure not caused by drywell overtemperature re-sult from a gradual pressure rise or from pressure p[ spikes that are not sustained, For these sequences, break and a large break. Reference B-3 assigns large breaks a probability
- For cases w here late containment fail-of 1.0E-01. For late containment failures induced ure occurs due to combined thernutl by drywell temperatures in the range of and pressure loads, Reference B-3 es-800-1100 F, containment break size was judged timates that breaks in the drywell and by Reference B-3 to be uncertain and event BS wetwell v 'll be equally likely. Chang-wi s assigned a value of 5.0E-Ol. Ilowever, the ng the failure probability of event release mode for the late containment failure in- DW in Figure A-12 of Reference B-3 duced by drywell overtemperature sequences is to 5.0E-01 Fi ves the 'llowing:
determined not by break size but by whether reac-tor building retention of fission products is effee- - P(SP) = 1.0 for a small break live, so BS was not quantified for these sequences with downcomer failure. B-11
- P(SP) = 7.5W01 for a large RB is not quantified because the release is already break with downcomer failure. scrubbed in the suppression pool; in this case, the contribution of reactor building retention to a fur.
P(SP) = 5.0E-01 for a small or ther reduction in the release was judged to be in-large break with no downeomer significant. For cases of late thennat failure, the failure. failure location was determined to be the drp ell. According to the discussion in Reference N3,if For the existing vent case, Reference b3 venting occurred prior to thermal failure, leakage quantified event SP using the above method with rates would be lower than in the non-vented cases, calculations based on Figure D-24. Ilowever, this and a f ailure probability of 2.0E-01 was to be as-is inconsistent with the assumption in Refer- signed. Conversely, the non-venad cases were to ence B-1 that successful mitigation of a TW se- be assigned a probability of 5.0E-01. This is in-quence through manual venting requires use of consistent with the Class I A containment ev at both be wetwell and drywell vent lines. Venting trees in Reference N3. In these CETs, th railure from the drywell oypasses the suppression pool, probability of event RB is determined by whether implying failure of event SP for all cases where the break size is large or small, regardless of manual venting occurs following failure of con- w hether or not venting occurs. For thennal failure tainment heat removal. Cases of early contain- with a small break. P(RB) = 2.0E-01. For a large ment failure and late containment failure without break, P(RB) = 5.0b01. For the simplified CETs venting are quantified identically to the Class IA in this report, the assumption was made that sequences with the SCS available. P(RB) = 3.5E-01 following thennat failure Bhe average of the above s alues v cighted by the break RB, This event questions whether fission site probability). product aerosols are retained in the reactor build-ing, thus reducing the magnitude of the release to For cases of early containment failure, the fail-the environtnent. The quantification in Ref er- ute probability of event RB is, as mentioned ence W3 is based on the assumption that the above, proportional to the probability of suppres-amount of aerosol deposition in the reactor build- sion pool bypass, with the probability of a break ing is detennined primarily by the hold up time in the drywell calculated using Figures A-12 and (residence time), which in tum is affected by the D-24. !I the break is small(upper branch of event break location, magnitude of the driving force, BS), the constant of proportionality is set equal to and possible hydrogen burning in the reactor 0.5 For the lower branch of event BS, a constant building. The MA AP calculations performed in of 1.0 is used. If event PF succeeds (no puff support of Reference B-3 modeled the reactor releaset P(RB) is reduced by an additional factor building as a single, homogeneous control vol- of two. The results are shown in Table bl. ume. Based on this assumption, the dominant pa-rameter in determining the residence time tumed For sequences where ! ate containment failure out to be the break location. liydrogen buming occurs due to combined thermal and pt ute was not dete mined by the MAAP calculations to loads, Reference B-3 does not quantify even, 1 be a significant issue (see Appendis K of Refer- following the failure of event SP.This appears to ence W 4), be an inconsistency in the methodology. One pos. sible explanation is that these are late noble gas Event RB was quantified in Reference B-3 us- releases with particulates making up 510% of the ing Figures A-12 and D-24 by assuming that the release. The release mode is determined by the failure probability of event RB is proportional to break size and whether the release bypasses the the fraction of suppression pool bypass events that suppression pool. Reactor building retention has result in a break location in the drywell. For events an insignificant cffeet on the overall release be-with no pool bypass (success of event SP), event cause the amount of particulates released is small. B-12
1 I i l l Table B-1. Early containment failure results for event RB Cl Fails DC Fails Break Size Puff Release P(kfl) Yes N/A Small No 2.l E-01 Yes N/A Small Yes 4 ?E-01 No Yes Small No 2.5E-02 No Yes Small Yes 5.0E-02 No Yes Large No 9.0E-02 No Yes Large Yet 1.8E-01 No No Small No 2.5EM)I No No Small Yes 5.0E-01 No No Large No 5.0E-01 No No Large Yes 1.0 B-13
" *& " .L'
%...N. , 2.!.t;h,..
.u . a5.h C.gp llitt 'PJ','.Vi MA r.n,. .!7."'.
n i . "b'K. og,gv
..&r. A v.eggi.
PD Ri C1 HL VP 1 DC W VP YE PI P
, .or-ns Q.8hE-Of I
7_PSE-01 1.onw-of - 7 10E=cP a.ma t nor.en l I Figure B-1. Shoreham Class IA SCET wit. ?PS. bI Al'EUTi!ItE CARD Also Avellable On B-14 A Perhtre Card 1 m/ w
d.,,. n ~ , b%. w. w. ace n e.ce
.c e.--.i, e.i .
c..n. e.
. i 1 .. i.A i .o,,,
- o. in .cu.i,..
t
. to-.te.
6 .
. e
. i.
. .n u .n wo, . un .
a sin m i ~. ~ cn er as m ec se [ l vt ag k ***2 mua-o h..* !!!!;g gj it usrm __ 4 4 p..3-o,, : iu--" !. .. . 1 um naan- q:,53,
.::j 1,l omst o
. w-tin
'I-
! ? . por -c3 _ +
g.g e
, . p ii -e.
L.,4~,1 1 u zio :** 34'I !,l Lt.rct:c i . m tsr:rr---
$!g" 5l P LQadJ , :
-ctme1 L !!
"a -n [2anst%a-si az:21 - .l::3:
- ig: ; g; saa-u .rrsm e- F,
;p1y] 1 w4> v m -es igi 3e mm.
h.J. -un -.n. ;g
>-e ka**'-lua-o -usar- m : N,,,-.. i ij n rm j $3l c, 3 2 --n h,gg.quusi m .m 3:
g;
- msrar-
, u x--
uar.si l:g:jg, , . ,
- "J:!! L q,
o r, m .: t "x -s ' [. _ ,,-n : reran 1g r:}ga: e a*-e i n:st- ggt c o6 rjg 3 Ig!!; i
!i e" a,mes mn n.av--
l g, I ime%,1 -" om1.. ._ j:.jk,; <j . me-o l,
- p.
1 u.,m, % asi m- c _ ua:u_ :curiu. :;; ,: 8; s su-ei - - - - wesr m - E*o 3.
- ,-i mmm -- n:h.0: ,:i: ,, '. :}-os
[ I" --e i Loa-m mm. :li ! !!
,u er -e i 't:i'l::{1 w tin
,n I _ -u -
. se-ce Iu-o as
-e EER 01--{I2ZiUC ! 4!.J. n n (i.es-nr ,o m-n .* *'-o f l
- ncrin _ i;'M4. si.
4.lg:go i k m-n 1 nmm 1:lg.2 D 1 , ou < < rw rar- y ! 4.ntL-ss plo,,::: gag
== 2:!!g- e- . E.
i im-u k,utat_inami r na_ g.gg:. ; g; n:st_
- l unst._ -
- g:
tj. 3::!n:!s . ref .!! I n m-m 8. ! I 0: ct I Agg-gs _ _ f S.172-07 Cl m ' _ _ , cj jo40802 8 7 '05
l
- 3. 0 M -e 9
? ser-et i tar.cn B _ maf 41 i _ DNr43
- 1. 0 M MO E" D4 7 BSF 4 i f~
Is.nas-es
.3.e08-23
?_9eF+0S
- 1. DOE-02 E_7BF-01 a maf.e t ,
L Df18 -0 2 3_887 63 I t 000 81 L _ L G9 +63 i i ' la. car-es APERTURE CARD I I I i Also Ayallable On Figure R 1. (continued). AIXTture Card 1 __m- .. - _ _ __ 1
m
- _ - - _ . _ m
. - ~ g., , =~--=>
!az a.e-o, s.sie ,-os w-v s - -ei
- 5-.! -.4
" "" - . -og a ii .of-.2 , 7"BEnne "
I fu .nr -s u **
.8 18 "88-* '
k an-e t 5 anr-c i --- I' -s s 1, = $, _ and .
, - - . .e .
..--.. -i 3:ha,
. iga.O. . .
- .. . e a.-
-r--
...t
- g::
<=4'
- . .. a -
-ei s. .ggy a
;i.nnr.ne oo, -o, e....A..
q..or-y E M4 I n. 2.
- n onF-e f
{!!+ 3 m .c g - - -
.. o
. , - - - 4. 4 a s:.... o. ;! -.e. .
i
- n. g 6 ia == * .
go . r ier==, L: Of b
~~
, _ -ci d;fg- i I. . i g
, m -ei L. -e, ,
a -- >
>.--e,
@:*g:f"1 $
-11 i;
=m1 L:!g : ff m .e n ..w+ s.
<.. , w i .. :
- .- .- 4 , . + C EM " da .ns-ci f;lI g47
.. c, 3 . w -e t
. N n
-> :, , . + "' g i ' :5"
, :p L .e. . '-' -
a: : ; c.nas . :r.n t m _ -:t"3: -. L ( -
,, w e i , ==--- o: !!!di
>= .d
, ,-n a w-e < :7 - -ei , **g:- !;
. .e. ce l a ro*
- r - -ei m - -,
.i...l
.I .
g;tg it b.i:d. .l'. a w-a i .-
. .m.
s i,,, + . . . . ., _ ,, g ,
.-- m_.
- lga;;
9 7 l.
* = -ei
- n .u -ei l3.::;
}7I3;
- r. '
, , -o ,
fa--- ;, - -.s . M. n o,,,, i
*:15*:
.3 d
l 1
- ._, n . 3E- :
ia -' 6 c ei n. :. , I x_-, ;:,,a c= - - f I 6.4M-ia 3j t t_ser-es ts nar ci r_ :1_ non .ac g,%*'0* Ce ti N -ei :, - -et ..seE- CJ
. . .as-e i, 3 m -Bi
, - 4 g
' ow .
i. 3 F .- " a --- 8..oGE-.tt-.5 Of M a.- )*I, a m.= i l, *:15 1. 4.o98 i
-- i mm. w ,
P "'" :::!g n h m-e i nar n ---
':88 48 mm l*i - - . . . + i i
- . _ -1 3
i 6 m-e t ,-- *:k,g;ci0 s ,.x _- !:. . . +
....d. 83
._ 4:n!'" B ll.08F* C i
_a . .om -c t
- l
"~"
\ 7/O'-/dFO28 7-dY
J-T Ko9-et LSQ LggE-et i
- 2. f BP-e t 7.20f 4s
- 2. 00F41 1_eer-es ,
s_w n_ tan t ear-en SI APERTURE CARD l Also Available On ' ' ' " ' Aperture Card usa 2 uer-e,
- n. set.es _
am Figure B-1. (continued). Il-16
g pg -
9 -
.. ~
43E-09 C2
!n _802-01
~
'Y BDE h g*, f gE 1.ser- t 03 h
L.R9E-ch 3, t eE+ 4 C6 I ns . noE-c t 3, g gg - 4 Ce 13 . e 0E-e i -. I 3, gg- g C7 y.per-et li BOE41 m.nne-ct 3 gg- 5 Ce e moE-e t N "-C 1 - 0 3 . .t h o .1 . N.E ti.oon-ei m m -ot ' ;g- l!
,..eg.*
a E 00E*0I I t . BQE =e t a O 7_ w e i , ;
.. - -0o I?
- t ne~ i 41
= m -c i I:.w: t .g ,y O E a"-ei knoE-et m m .m= 0:ggg+g Lh
- 8. ..64E 4e n
32 Ii 00E+00 3 D' l7_not-ei et - ,c c - i tG
- f t M
E,98E* t . SE- , 1 u mor -c t . -Ci l I' #*O $g _ non-e t is:nnr-et * *jl h 3 DM-23 3. -13 03
; 4.DaE-44 CS l***[$lg
~
M coE-et 1 7 enE-29 11.ooE-et ::.nor-ci , ,ygg-[g n noE -e i ,,,, _ ' N -01 ! 0 Tr _ --c i 4.05F-04 s
, gy -06 C I 43E-10 C ts. ear-es I -
- n -o t .3 -Ce C 5t_ ant-et 'I~RC f',II M
. 38E -07 h
03
,.36E-00 a
- 1. 0M *21---4 7 ,og-es 0.00f*00 3
, oy ,og g , ggE * : 3 3 2 C6 0.771.
. Set a Cl I :n unz-c i ,.seE- 0 C2 I' *
!$.50E-0i :3 M -D' r, gE- ,
m::R.:.4 $M
.eM-,
I ri.nor*ac 7,ggg C6 11.00E-0t I
'i noE-o' H.0SE- C7 7 nor-es 7 unr-c ' s ser as ce 1.7eE-it Ce n.noE-et ' " -*' 9.4er-19 C40 0.00E*00 W1 i onE+on Y"1tnrrdY"" S E . oM-e i
"' .. [lf.E--
- 4. g
' norat
- n.r.nt-gt -
.- 0
-g gg 4,g g.00E+0 87 Il DGE+oo m. m -na g *0l 4 $
B0
- "-t E.neF-et '3.h4E-00 8.756-00 b .i
'w: RM-e t :i mnt-at g , gg-og g j
- 4. 0 M-0 7 El 4.0st-00 02 f , ~ eeg -e y , 0.00E*00 03 11M-e1 ri nae +ao R 97E- 04 1.23E- C4 8.deE- Cl I
58_act-et
- u _ mot w t
- 3 noE-c i 6.4SE- ,0 g '. [8E"* .
g 1;4 3 00E-03 6.25E-te D3 g
- t.mw +oG 0.00f+00 C8 3.4tE-id C6 L1,&QE*01 -E coE-e ' 6.seE-16 C7 7 ree-es m_ - -ci Ce a.gg-te
' N -ci
~e aer-et h6eE- C0 1.SSE-Os 01 rr_cDE-Ci 3.05E-07 02
.00E+00 *t
- -'=in ocE*co ,47t-08 53 486-10 32
' 00E**'
l [-r _ w r-e s et gur,cc '.N $ 0 90E*00DS 0.voE*00 Cl ge*en 1 I 0.00E*00 C6 17.fst-e1 m.caE-Q' O.00E*00 C6 ls.nSE-et ' O . 0 0E +00 C7 ti.00E 4 1 M BOF-C i 0.00f*00 C8 2.0 71 *06 03
, t.077-10
""-"2
{ !* rer-es et 1r5g,cc k' . kh 8:8lin El 7 wM-es I ra _ cce -c i O N car-et I
#1 eeE-ei m_tior-at
!*$06+
g, 0.00E+00 C8 00 C7
,,,'4 6.30E-06 bl lm cer-e t _ 6.37M-06 B)
_ n con ca ,
'EM 34E-04 87 lg coE-et p p 4.b6E-04 Se
+
gE *e 5 s g, 9fo VoBo287-of .- _ y
l CLASS IB1 PLANT DAMAGE STATE The following is a description of the factors in- ference is presented. Because TeQUV (low pres-cluded in the Class 181 Plant Damage State cate- sure) sequences are slightly larger contributors to gory (see Figure B-2). this plant damage state than TeQUX (high pres-sure) sequences, the reactor is more likely to be de-PD. Class 181 plant damage states are station pressurized at this point in the sequence. blackout (SBO) sequences where core melt oc- Approximately 55% of the dominant intemal con-curs in the short term (<6 hours) due to a loss of tributors to the Class 1B1 core melt frequency are coolant makeup. TeQUV sequences, thus the failure probability of VF is taken to be 5.5E-01 for this report. F11. For these sequences, the dominam fabure mode for loss of all core cooling is a loss of elec. DC. The probability of pedestal downcomer failure is taken to be 2.0E-01, as for Class I A. trical power. Therefore, recovery during the early stages of core degradation is quantified by ex-H2. The probability of a hydrogen bum fol-amining the probability of recoveting ac power. I wing vessel failure is taken as 1.0E-02, as for The value used in Reference B-3 was obtained Class I A. by considering both intemally and extemally ini-tinted sequences. For this irport, only sequences yp imtiated by mternal events were included. Using tuati n setp int of 60 psig is taken to be the same Table 4-4 in Reference B-3, with externally initi-as for Class ' A. ated sequences excluded, P(RI) is calculated by dividing the sequence frequency at one hour f i The probability of failure of earli .enting VE. the onset of core melt by the sequence frequ' throuch the SCS is considered to be unaffected by with recovery. This yields the conditional non- SBO liccause Filtra is provided with its own back-covery of ac power probability, which is taken as up battery system for MOV operation and instru-P(RI), as shown at the bottom of the page. mentation and control. in addition, the actual
.. venting of the containment to Filtra does not re-Cl. Reference B-3 assigns containment iso- quire electrical power. Therefore, P(VE) is taken lation failure a probability of 5.lE-03 for se- to be the usual SCS failure probability of 7.2E--03.
quences with the SCS available and 2.lE-03 for sequences without SCS. No basis for this differ- F1. Early overpressure containment failure is enee is provided. A value of 5.lE-03 was as- quantified as for Class IA. sumed for ;his report. PF. Reference B-3 assigns event PF a failure H1. The probability of a bydrogen bum during probability of 6.7E-01, apparently based on a core melt is taken to be 1.0E-03 as for Class I A. probability of 3.3E-01 that the vesseldoes not fail at high pressure. For this report, a value of VF. Reference B-3 gives the probability that 4.5E-01 is assumed for the lower branch of event the vessel does not fail at high pressure as 3.3E-01 PF, based on the probability that the vessel fails at for sequences with the SCS available and 5.3E-01 low pressure 55% of the time (see discussion un-for sequences without SCS. No reason for this dif- der event VF above). P(R I ) = 9.5 E - 08 + 8.0E - 09 + 3. l E - 08 + 2. l E - 08 =+5.1.6E - 08 l E - 01 1.9E - 07 + 2.2E - 08 + 6.0E - 08 + 4.l E - 08 + 2.4E - 08 B-17 I
a % e*% Citts No In- Cent. HE Burn FPY Qote Ptasstel 19 Bum Cont. No Etely OP FellWP PVff No Em ISL PD3 vesse! leenotten During CN het rail DCs Pett Aftop Posts. < Cent. e Aftse A feeleste vests' 8estetter Pellere et High MPV 50 pang Vsnt via PV tretch iteceva Presa. Fellure SCs PD Rt CI Hi VF DC H2 VP VE Ft PF ft2 f 13.00E-0
- ene-en 3.00E-O e _00!-c i 3 Dof-0 I
li cet-2 m .coE-c 7.to*-e3 4 tbF-02 7.EdF*02 9 . DOF -o f Lant.en Lans I
!?!JQL:Di Si i Figure B-2. Shoreham Class 181 SCET with SCS, APERTURE CARO AISO Avidlable On Aperture carg
- .n_
. vi. . u, ., .r ect C.n .
. .i wo.aAo u.- .
- o. in 7 C s. .
uoe. wJ. . C.s..*. . , . .h0m ..,t.. i.. m.
~ .niao .
.m.. m.r 7 11.
$M __
Y' l 07 F3 88 tt) ap .W) 4.90f a0 4 A4 Mi nM-es ;7.PSE-03 4.96F-04 A2 yg* g5FE E 0.00E*00 41 3.$W*06 A3 " *** 4.57f-02 Ot e } k -06 ll*93 2 ' 0.006600 C1
#~ O . 00t +00 CR E12e ,rpr:or-- 3;g-0j g
.rSF-04 03 M i cor-e2 w:--es 6sro 1. PSP-08 0.00E+JO
.-C.
DE b* g ... 0.71-0.
- 0. 00r.
- 00 C.
Cl la .M-ci :n uns-c1 I 0.00f*00 C2 LA.$2Et90 g C3 e MOE-O I~ l $4F-06 4y .00 C4 I . 97t-10 D3 #
- 5. car-ci l l . 79F - 11 C6 3,ygg.it ce 51.ccF-Dj 7.for-es rr pF_c t j;ff*,I g
.R 20E-ei WC a;hN! I0
- t. bee-04 3
--{ i . een -e n ;? PD'*C3 4.67f-03 2 DS455 9.00e+00
(,t4f-06 04
' 4. 84E -06 C1
* " *
- iuef.co m r:ci 8:8n38 8 e
wr:n- ;;g :: a I 2.46E-07 D3 u=~1 4 4.. ,74e-0.4E -06 C6e bE2f-03 1 00F-01 PF EMc t ;
- n.nor-et Ce
- s nor-ci g ggr:Et--"- j;8"~j!1. b_-o kk n 1 74i-04 0a c i C1 rix-c i .'6 .O 365-C7 00 C li.cBE+0S __g.80.E.*00 Cl
,pSE.-61 ' SE
. tee-05 03 dt ecF.c2 ;t w-es . . i sf -0 7 D2 n1lLrfgc u 00f+00 e.4oe-to 4 y 3.SSE-Ce C1
- n .w-c r 0.00E+00 Cl
~
0.00E+00 C2 b 20E +cc
, a F12 F-e t I ,* .h"h h
.82f- t D3 l, , , , m-1: lit.ll i a
? teF-os li 00E-01 g(di M ar-ci l;h*h h
- 1.riar-o t ' g;h h0 4 . 3eE -01 Of Il 0M MET ~ET"~~ f hk !!
E W "UI 4.39E-09 B4 [t enF-di - ., m m - .c t 9.446*to BS
- 3. 44E + t Q B6 4.85f=47 87 i EM'0 I E # ~0 I M::E-c 1 hh-0 4,36E-07
.g gg 810 i . DOF -01 r ynirq***"" 0, o D3 Mt DeF-c? ;7 70F-03 {1l75E-05 7eE-07 D2
- i curic: 0.00t+00 D3 1.27f-09 D4 I 6. 44E -0 9 Cl
- s . v:E -c i C.00t*00 Cl
~
C GE+40 Ci i . ocr +oo a 1 sar-ci h.gy*[ 781 - 4 D3 h I. 13 eep .e t , EN1 E.3Pf- 2 C5
- 7. PSF-03 li.00E-Of 6.326 *- 2 C6
- i;g5g.cg ,
L gSE-oi :3 w-o i { *, h *,$g h80 W.7~c t 3. tee-02 Dt 4.3tt-03 J2
' l 1.33F-05 C1 0.00E*00 Cl lt tSF-ei I :H H3E -c I 0.00t*00 C2 11.02L+cc g,g79.g4 g LULdL-- 4.s tE -0 4 c4 1 3.095-04 D3 7 1 co!+02 3.106-06 D2 VWO80287 7 66 JT
1 07-0 l D#""DU 3) *E-8 eJnSla ple3 aanpady 80 alqct!uay os;y GHVD 3MO.LH3dV IS I . 5 IP-sw u 59-NQU ' y io-nt s 59-d99 1 W BGE'E I WJUG g N57'T s w sus i w-en , WdM '4 l - - - - _ _ _ _ _ _ _ .
i7 20E-01 :n_ car +co I',.lig Ne*g h
!,,o,, ,
- n 8:!E3: g umoa , m -e i 1:m : uDS g ef a# 4, set-10 I mgg I.00E*00 C8 a coe-et , .e7t- 9 go h
7 2ee-on U QL-Al :7 cm -c i -Q*: { gj . a tor-e t :3. ncr-c r j;jjjy 0.00t*00 C8,
. t . a nr +oo yg y 3.36E-00 Ba h 3 unt-a' a .m-e i n . car ma
.g-:.iJ.Ii -C
,,og-te se li.00f*00 87 aen co=..xre, a g,71g g;
'"*-a* 8. m-e i 31,.
- 'n{ll{
g 4.u.l.0. 4.3 . o 13 I- 'O'
!?.ncE-os eg g ygen- l; % $ y g.m-c? gCl t,,c,., ,-
==o . 0:5 08
. . m +eo , ser m g
.3.not .
I
- 3. .
0.00E&00 m.3E-i. C1 ti- *CU C6 m cop.ot i
- 11. c of-01 gg3E-0 7_20F 43 'y py e t g , g y *,, Q u na w -.-- s =r+
---- M r-at 1:.!.n 4,m 06 ie-04n c'1O
[a C4 g.36MF d' :E _ M-C t g ', u y.og A m +oe :, mr.c, ..-E+ 4.SSE=06 C3 C4 g.34E-OS 03 It _ nef-e2 ;7 -- m .m 6.,.m m -t egj h.4 4.g6E-t
" U E*01 .0 +0 C2 3.cof-e2
, eop .o t i u m se , :m-e , umg g
!.itt-43 I
-1 N +"
i i .00E *00 C8 g gg-cg 8.g6E=l 3 C6 7 W -63 i D of -o ' :7. scr-a t g',dl I j E7 f.20E-01 :s scr-ci \ 6;.ly;] I' W '00 :1_ sc r -c z j',gotm3%;j0 l{O 3 E#* In w-oi ;, , g g7_c y
- 0. 00t! *00 87 lt.acr+oo mr 72 -0 38 U ^ # ~0 '
Is nor-at .' " " ' k 8 9.0eE-oe 03
- 4. 30E -06 I' W"' 4. = -0 83.
a~~ : , =r +c c L:m-]8 a
" r- -
le ecF-o. i
'"* u:M*0:
0.gCE*00 il C2 li.8aftoo :.s . --a n l;4'gl$ C3 ca s.nns-c2 6.00E-13 63 I
' "* ' 0.00E+40 C5
- 13. coE-o t i CS 7 non-en 11.002-01 j.86E*g3 C7
'n snr -a t ' g,38'
- j g
.A toe-d ( - - - a mor-c i M:l, cja
- 3. 79E -01 03 3.77E-03 D2
^*"
57 N-03 :n nae +co g', h $ 0.00E+00 h Ct a . noe +oo I 0. 0cE *00 03 I? . non-03 1
" " - 1 0.00E+00 CS
- 11. boE =o t I 0.00E*00 C6 it BoE-of .g,ne, og 0.0 C7 7E-04 4
- I w-o f { ', t 06 D$
O. -06 Cl g . ger-e t i N *C l '. 0 0 ii 00f *00 I C3
,7 }-0 d
- _, _ nor -c t i
. .e-0.
.. et'*06 M
D2
' N -02 '
,,p .o , V. OE* D3
'- :t_cor+ac g ,4 g g.c0 gg g4 3 . Set - 10 Cl
" I 0.00E+00 Ct la_gce-et i O.00E+00 C2 it.noE+oo i nor-a t 3t. . 3..t.-E-os. C.3 4 Ile J U
p{) , 916 V080287-07 -
L3CL-AL_ 2.nonna_
- 3. RSE -C i I
l LML i f LQQE-02 i I LL.00E-02 1 con-on M. ice-ei 7.20f 43 I n . car-o t 5.70E41 1 en9-03 E.DcF-01 1.$$f=02 1R11f 42 . In oce+on u =-es n . nee d o l l n.sor-et 1 f . cover G. ode +00 M 10E-41 9 . 90F -d i Figure B-2. (conunued). l l w ~. ,,
1.82E E 03 '* 7_20P -c3
~ Q Jag-gt M1 $kth5 43 C
,po E -e i ___ ,g EE-c1 I' ice *IA E7
- 3. DOE *t4 CS EM'M UE0E ghf 4.3ME*t0 Ck0 S
.fm -m bar* ' m cdi:ni:.'
, ;4. - - o n.4 61 Jct-e t .Y 55r c t
-jjj- [ y
- 4. u -0. .7
, a.seE-o i us; . . E -0.
e-0. .. m w .,, , 4 .s0 o " -e i r u m d :. w -0.:ng+gsti u m. 4, a
- m -0. a o.m-u _aonn- ---- - } m4. m j;
!k 44E g k-
' m,-m
'~
, mi !!:Hi3 Cl i '
11 AQL*00 TY5l"*l5I I . ? ttE-i t D3 6 l C5 7.2ce-c3 { Q ew-31 T"'
't cor-et 7*3 3 E-M
.r g m ,
lgl33E,gg>
- g. gag g3g 4
gg g 11.33E -01 :3 -ar-oi af6g . . c CF-d L N -0 On 1 h'EE-08 4 02 C YBU"C ' .g.36E-Ce g ey !0 00 2 aatea m er g g; g; at nor-or , 1 D3 it. w -os g.34E-07 .3EE-09 g2 gg3 jo
. 0 I 4.17E- 1 Ci I E " 'O ' 0. 00E +00 Ci
[3.30E CIt.coF+0o C.00E+00 C2 3_ npo , 2.84E-09 C3
- 1. 44E -0 9 C4
! 9.12E=t3 03
- 13. 0 0P-01 3 *3 0.00E+00 C5 7 w -03 i
i t . nor-e t
- 0. tee-14 gg , .g,1
-e i it . nof +do vnm-g:.=00E*0g dlCe g;,
81 y y y.a 0 . 0*JE- 4 0 92 g _ cor-o g 3.3M-$1 83 4.L71-10 54 312Ehn ner-m j;y;{ y $} 0.00E+ B7 5 "~a ::Ing: n m m-- APERTURE
" "r-" w -e i m .m, .m :11!3:
- 5. 09E -0 9 ili B12 CARD 4.3eE-07 f43
% ~ -a' :7 m -os
' " "_- 8:ena! 8:
=:g:ta g; Also Ava.nable On
'. .o~ t ,
it_GDF+oo
$:nta: 81 Aperture card
- 3 m .ci S . 75E -09 C3 4.7tE-09 C4 0.00E-tg 03 7_2ne-os t_oor-et ' 7 2.22E-'4 l CS
- a. toe-of pg.og G.SSE* 1 CS 3-
- 3. 42E-' t
#t nor-on Cg0 j;47IQl h3 0, +4s Ct 03 "a"*> g . 97E-OL
- , .e~3 m m m,,,
Ojopj jj l' .00E+00 Cl 17 M -03 I O.00E600 0% 11_ooF-ot ' " " ' 0.00E*00 C5 1 O.00E*00 CD IL QoE-e t 3 gg.a, 0. 00E +0 0 C7
- 0. 00E +00 CS d,i . ocF-o2 2.78E-06 03 U * *03 2.77F*08
- i.naz+ca j'%'jj Q2 0
- 0. 00E +0 0 C1 17 w-03 I
- 13. 00E-6 t 1 m m-a t h:h 0.00E+00 o! kg C8 Li.noE-oi 3- mm, 0. 00E +0 0 C7 0.00E*00 Ce
;" "" ' g . 04E-06 84
;o _,e, mc ==- 3:!ual n 0.00E*00 1.03E-03 87 i gJCE-01 .. g n . ., i 4. 9eE -04 88 lo corme e.4aE-04 Be 0.0CE+00 I$ -2l 9 tov 080287-08 ;;~ l
l l CLASS IB2 PLANT DAMAGE STATE The Iollowing is a description of the f actors in. VP. The probability that containment pres-cluded in the Class IB2 Plant Damage State cate- sure fails to exceed the sent actuation setpoint of gory (see Figure B-3). 60 psig is the same as for Class I A with the excep-tion of sequences with vessel failure at high pres. PD, Class IB2 plant damage states consist of sure and no hydrogen burn or downcomer failure. sequences in w hich core melt occurs in the long For these sequences, Reference B-3 gis es a value term (>6 hoars) due to a loss of coolant makeup. f 1.0E-01 instead of the usual value of 9.0E-01. No basis for this value is given, but it seems rea-R1. As for Class IB 1, the failure of in-vessel cur at a much later time than in Class I A. Thus, it
"#I recovery is qumtified based on the probability el.
b &ly that containment will br at a higher pres-not recovering ac power prior to ve<sel failure. sure when the vessel fails, with a corresponding Table M in Reference B-3, with extemally initi-higher likelihood of exceeding the actuation ated sequences excluded, gives P(RI), as shown on the bottom of the page. VE. The failure probability of early senting Cl, Reference B-3 states the probability of through the SCS is taken to be the SCS failure containment isolation failure as 5.lE--03 if the probability of 7.2E-03. SCS is available, and 2.lE-03 if venting is through the existing lines. For this report, a value F1. Like event VP above, the probabilities of of 5.lE-03 was assumed for all cases. early overpressure containment failure are in. creased over the values given for Class I A. Again, i H1. The probability of a hydrogen bum dur. this is reasonable based on the long period of time ing core melt is taken to be 1.0E-03, as for Cla< before core melt and vessel failure. The values i IA. used are taken from Reference B-3, Figure A-4. VF, PF. Reference B-3 uses a value of 2.2E-01 Using Table M in Reference B-3,459 f r the probability that a puff release driving force of the internally initiated Class IB2 sequences will exist. 'This is consistent with the value of i have the RCS at low pressure. Therefore, the 7.8E-Ol that Reference B-3 uses for the proba-probability that vessel failure occurs at low pres-bility that the vessel fails at low pressure, llowev-sure is taken to be 4.5E-01. Reference B-3 used er,it is ine nsistent with the low-pressure vessel a value of 7.8E-01 for sequences with the SCS available, obtained in the same manner but in- failure pr bability of 6.lE-01 used for sequences without SCS. For this report, PF is assigned a fail-ciuding externally initiated sequences, and ure probability of 5.5E-01 based on the value of 6.lE-01 for sequences without SCS. 4.5E-01 used for event VF. DC. The probability of pedestal downcomer R2. As for Class 181, ex-sessel recovery is failure is taken to be 2.0E-01, as for Class IA. quantified based on the probability of not recov-cring ac power within 10 hrs from the onset of H2. The probability of r, hydrogen burn fol- core melt. Table 4 -4, with externally initiated lowing vessel failure is taken to be 1.0E-02, as sequences excludeu, yields P(R2), as shown at for Class I A. the top of the next page. P(R I) = 2.2E - 08 + 2.1 E - 08 + 1.4E=-8.9E 08 - 01 2.9E - 08 + 2. l E - 08 + 1.4 E - 08 B-23
P(R2) = 5.6E - O' + 4.0E -09 + 2.7E - 09 = 1.9E - 01 2.9E - 08 + 2.1 E - 08 + 1.4E - 08 Reference B-3 uses a value of 1.lE-01, calcu- tion is made with extemally initiated sequences lated in the same manner but including externally excluded, the result is 1.1E4)l. For sequences initiated sequences. with failure of events R2 and OT, but with succes-sful hte venting to Filtra. Reference B-3 assigns H R. Tha probability of failure of long-term event F2 a failure probability of 1.0. This is sup-containment heat removal is taken to be the same ported by the Reference B-4 MAAP calculations, as for Class I A for sequences with successful ex- which indicate that, even with successf"' late vessel recovery, venting, drywell temperature will slill reach 800*F approximately 14.5 hours after vessel fail. VL. For sequences with the SCS available, ure. For late overpressure failure, Reference B-3 the failure probability of late venting is taken to uses a value of 1.0. This inay be overly conserva-be the usual SCS failure probability of 7.2E-03. tive; however, the MAAP analysis for a long-The existing vent failure probability is taken to be tenn SBO indicates that drywell pressure reaches 1.0E4)1 if ac power is available,(i.e.,if event R2 approximately 80 psig at the time of vessel fail-
. succeeds). If event R2 fails, event VL is assigned ure, when the 'vs ell sent to Filtra is assumed to a value of 1.0 txcause operation of the existing be opened. If no venting occurs, pressure will vent valves requires ac control power. continue to rise until the ultimate capacity of the containment is exceeded. Given that no ac power l OT. Reference B-3 assigns event OT a failure s , available at the time of vessel failure, I
probability of 2.3E-01, apparently based on the coupled with vessel failure approximately 1.5 l value of 7.8E-Ol assigned to VF. Again, there is hours after the onset of core melt, the assumption an inconsistency for the case without SCS, for can be made that ac power will not be restored m which VF was assigned a value cf 6.lE-Ol. For time to prevent late overpressure containment this report, event OT is quantified as for Class I A failure. Therefore, a value of 1.0 is used for this because vessel failure is included as a top event in report. the SCET. BS. Break size is quantified in the same man-F2, As for Class 181, Reference B-3 gives ner as Class (A. no basis for the late containment failure probabil. ities used. For cases where long-term contain- HU. The probability of the Filtra system fail-ment heat removalis lost, late containment failure ing to hold up noble gases is taken to be 9.9E-01, is assumed to be certain. For sequences with as for Class I A. successful early venting to SCS, but subsequent failure of events R2 and OT, Reference B-3 as- SP. Suppression pool bypass is quantified in signs event F2 a failure probability of 1.0E-01. the same manner as Class I A. This corresponds to the probabihty of failure to recover ac power 24 hours after the onset of core RB, Reactor building retention is also quanti-melt, calculated from Table 4-4. If this calcula- fied in & same manner as Class I A. B-24 l
.SCS htiw TECS 2BI ssalC maherohS 3-B erugiF I
1 24 f0J.1 t 4ebe e ia-fa7 n 2G-Eb0.t so- w .7 unaL u m _ tmt
.o-,em i
_ -+ w waL se- - 7 ic alt F i ~ t' l rgp tP EV 9V 2R CC Pi l iH IC tR DP acs eruu er cen cov seoe gwt es.rn ,e ei.vtnoc tnev etes oe vra .a.on n R eetfA e < ese*P n m te erwtise ,re v o c en -e eH ffuP PWlleF PO ylraE oN metfA tteF acO llef teH fO gniruD M 3BI e
.tnaC nruS fH letseesP seoD VP8 nPwS 2R ne t. tele a! tessev tneC -nI eM sealC ~'
-~~-~- - - - - - - - - - - - - - -_-__ - _
T a: e oa ?.
-+=
ac 3u ee me< i ca
- 'd
~
O <b o c.
< x< g-s <
I lm3 N l :s:sa=nunna=usassasshnz=unnstsshs=nasuu=ensassasshusase== :~naas=nnantu s=cassa:=nc g i 3 evnnnni enne
~..finetenern-i... ne._i......i...i....g.itf__r nre tnen ne n te.: r rrnev .. ii.iinigtit.ifgfiff..__..
t . g
. .,. ,gs.:.i....ner rrrrnverven. N L I J Mi@@!5@@*5@ME'ES._@.l.S,.Ms_US,_454.5.M..@..EHiiRIM.S5__5_15t@S_M.
= . . __ ._ _ _ . - . . _ $._5M.5..%S_SS_@sE545..,S
_ _._ . m
- ---- w-<----,
3 _ u - _ _ - m
? ? 2 9 9 Y i E- E
? ? ? ? ? ? Y ? ?
h h M a a a b a a a M b
. l l l , g,$ .
g. F m P P. F . i 9 , lE F. E- E .M 19 1 _q w .a w w w w - w w
,. w =
? %
- ~n ? ? ? ? Y. ? ? ? ? ? ?
! ja:. I h a a a a a a e a a a b b @ d i n e P
=
F
.M e
N 9 M e M r _ t _ F M e _ . - - = * - [ at I . , Y y y ? Y ?
?
I l l ! l I I - I
# . !! r - m m- !!
I 245' ' 5 5 5 5 5 i . 331 I I I I I I _2 _a _e _. _- _e l . . 2 . . . _ . . . . I
-J ,3 %
T g I 2 2 2 2 ? 2 ? ? 2 2 i f I
$j3 '. ]
i M [ e a a 3 % % e M I- ] a
- l [
n e _d _: e - e e e e 9 _d _: e - _t _: _t P _ q _ _t. lM t
? ? ? ? ? ? T T
.[s.g g ? V T T I
i' :38 I I I I I I I I- I :I. I _a - - - n - -- n n. n _. e i og e , e e e m a e j- gf3 $ I I I I I I I I x n - - ? - & - A __ a - = - a
~. * - a g
[ % E 3 3 1e 3 3' ;
. a. b.
- b. o .
J _ e l' ;; _ _ _ i ' Y T' _ T' _ . _ _ T' . _ _ _1__ . _T _ __ T _. _ _ _C uT :___________.
- s -~
u.at-a -- I uer-as _ J
, LDDL-Q 2
,L iCL+2 L a ser-es I _ ; ' f _, ,. UYs*$L 4
t .002 42 _
* $ $ . id ) L . V, ' l~ t . QGE +G3 h.tI) : ,; , ,e i.nce-on I
un,4, 1 SI l APERTURE CARD Also Available On l Aperture Card l Figure B-3. (continued). B-26 1 . - - , , L. ..
m t 4 t i3 j asucabanaanas!!hasa=nass=c=hansnessasa=nasaucuh= sus == ish=a=acassansshnacssansasasacas=assnasssa
..........-..-.......iv.......-.
1 veervrvrevertrettee8.v..f...........--,.................. e errevreeerrveerrevvverereverererererettervervivreverter errrrevyerreverreverve ........--.fie.. ...-....verterever- - Q x=_.m.,m..
~, enusp,ennamnu=r,-=wnsernar=====
.- e.. . --- ye-- e m. .. 8- ,---, n-- -------------
= _.--. . . .
_ . ._ - . = . nun ==ppuuspungnugn-s.p..2
... ..... . . m. .2 .. e - - - -. _ _-- .---
pen.
. .. -z-.s.=..sungppg88pupsyg.
-........e,..,..
...-_s
. . N
. - - - - - - - - - - - - m - - - - -
! ? ? ? ? Y ? ? ? ? ? ? Y ? ? ? ? ? ? -?
b 5 e 5 e e a a a a b b a b h h a b m r a a e e m ,rg m, a o e a a r, E, a m
,F
{ * "' r .N 9 .M
~
<F .E l9 N E N M 9 9 9 E E. %. E. %. v E
. a. E
. a. %.
~
v %.
~
v E 5 6,
~
, E. h h % % . %
.% H e b A 3 % % b b b E E p o m o a o a p a e o % a e a n b{b 1 a 5
$. ! ? #
M. b .* $ A 0 kk N
'6 ;
4
; ; ; 9 h I h I h g
9 N M
~
N N h N
- o. E. E. E. E. E. E. E. ? b' o
5 o C
% 5 % % 5 %
9 9 9
- C
.I __ ? - _t _t _ . .
o a o a o o a o o a o - a o . T % V ? ? ? S T ? ? ? ? ? ? v 5 % % I h a 5 a 5 % % % % y e 1
- o e y e e b
a
-9 y b
a e s.
- : E _d _:y Ld b _t. _: -
_t _r " _: t ~ _t .* 1 _9 I I $ T $ _I $ I i i - %. ., %., %. %. a "8 %. O _2, 5 - = =
? _2 r e
-f r
.=
. e e e e e
._ :1 e
I E k k
% 5 % % 5 % % % - %, 5 e q q, n <3 n n n n
," ?J S .t
* .t
- t _t .t a - a n n - a 5 b % % % M %
o a o o o a o I
~
. . o . -
a , = i.1 = = , = = g
'l 4
"1 - .
4 e _-
=
g- .
- % pm 24D E_70E-01 E.90E-01 i oef-02 Lv.1-2L t . nee-e t
._ I 2 *SS-01 1.0SE+00 l
I- OE# . g gg,gg i 1.00E-02 LEGL-AL
- n.e0E+0a E 7.PoE-03 JJ9Lt22.
4.soe-Ot t.802-02 0 doe +00 m
!! fDE-03 e_GOE-01 I I
Figure B-3. (continued). _~ --
b 20E-01 !i . ece -e t ggr c i s.17E-13 CP m-si n5rar- f:Higi ': C
- 5. eM -10
, 7.iEE-0. .Ci 10
[ d.J2E*0i..TI K*TMH*1.96E-to O M *i0 82 B3 500f*01 1
- 7. t rE -Og 64 ti ccM-o f ,,7 25Pqt- 3.968-10 B6 3.9ee-10 Se It 00F-Oi 7.052-0) 87 p=jr:n-- _ 3. 3.92E-06 9W-0 8 Be 89 E 00M -C i 7 . 06E -0 7
;1 00E-o f yy gg--- 0.00E+00 Sig et 7 . 84E -06 Sit 3.1:'E-06 D3 dM 22-02 17. 2e r-o 3 3.I32-08 D2 MRa 0.00E600
.2.27E-10 C3 D4 I i 7. 3SE-09 CL is.goe et i Df E-C 1 0.00E*00 Ci 11 ^~"any ,co 0.00E*00 C2
'5T TgT'-- 4 . 73E-0 7 C3
- 2. 56E-01 C4 1 0.00E*00 D3 11.00r+0c sT ZZE-U t jj
_7 " *0 I 'E"63r.cn 2.68E-12 C7
,^,n.., 2. .E-i2 C
- s . 5 2_c g 3 . 43E -09 C9
--trs e-u i:n!4:
7.7EE-06 81' D2 1 1.SPE-07 Cl Iq . ges.o t i :s acr-at 1.82E-07 Ci
- 3 scr.-c i -0
-16.30E-07 c4 ~l t ^ nee _op ;0.00E*00 D3 i7 . 2c g-0 3 i 0.00E+00 D2 PW+c a 0.00E+00 D3 O.00E+00 D4 1 0.00E+00 C$
ig,geg.of I ;E H::E-01 0.00E+00 Cl 0.00E+0.; C2 j,,eop.og
- 3.5cF-01 0.00E*00 C3 0.00E+00 C4 1_ . _ . 0.00E&00 D3 11.00F+oo i : 1 C Cf"
- a 0.00E+00 C5 7.2$f-03 jg egg.o, 0.00E+00 Ca
- 7_ soe-a t 0.00E*00 C7
- .3 sar-e t 0.00E*c0 Bf 1 00E*0 % 2_so .c2 3.SaE-09 32 8.66E-11 83 CI v1
!.Sof-of , 1. baE-C e sa
>a - -Si e x-c, i:;#4 n APERTURE 4.00E+00 B7
=i=a LaLOOF-O f x==~'
'f:See07PJ4:
n 810 CARD
;E 90"*01 : g _ s y.c i -
l1.86E-07 B(i 3.40E-06 812
^00E'0' -{ 7. 2eg o3
- 1_ c cr +c a 0.00E+00 9
D3 Also AvaMAe N e.see-tt 1.84E-09 D4 Ct Aperditri t.ars, l Ig gog.ot 'E5cE-a t 0. 00E+00 CL 0.00E+00 C2 ig geg,U 1.1SE-07 C3
- 3_. scr.c ,
- 6. 37E -04 C4 I 0. 0CE 600 D3
[t eey,co :1 :::Etco 0.00000 C5 7.20F-03 1, j g3p.og 1.20E-il C6
- 7. 5 7-c i 3.33E*t$ C7 e nor-ot 9.99E-13 C8
' M K -Ci -- l2 0
- f e -c i 3. 8 2E-0 4 0 3 . 92t'-09 02 0.00E*00 Cl di u,.e3 2.812-06 D3
- 3. m _e3 2. sae-C e 02
- 1.c:E+ca .0.00E*00 03
;2.04E-to 04 Y C.00E+00 Cl 17.2t?-o1 5 0.00E*00 D3 li.00E+0, i 15 UZ N i ti .ceE-o f
,"--g_,i 0.00E*00 C7 0.00E+00 C8
.4'1 geg.o, 3. M -06 03
!y- .o3 3.SeE-06 02
^
- 1. :: C E + 3 0 0.00E+00 03 2.87E-to 04
.) 0.00E*00 Ct 17.2cE-os 1 0.00E*00 03 11.ccE606 i " C CE -01 l; h @ $
d 00E-01 'Y, c cg.c t 0.00Et00 C7 0.00E+00 C8 t . 82E-Otl Bt i A 00E-Oi :2_1: g c s 2. t BE -0 5 82 lo ocr+oo 5.72E-06 B3 0.00E*C0 L SOE-03 87 I E 00E-O i ; 4_2y _c ; 1.66E-03 88 LCCE tco 1.13E -0 3 89 0.00E+00
!!-27 9/0Vo8287-N -
l l CLASS IC PLANT DAMAGE STATE Re following is a description of the factors in- H2. The probability of a hydrogen burn fol-cluded in the Class IC Plant Damage State cate- lowing vessel failure is also the same as for gory (see Figures B-4 and B-5). Class I A. PD. The Class IC plant damage states are an. VP. Because these are ATWS sequences, sup-ticipated transient without scram ( ATWS) se. pression pool saturation is likely and steaming of quences that induce a loss of coolant inventory the suppression pool will contribute to the early with the reactor remaining at high pressure. Core containment pressure rise. Based on Table A-18, melt occurs before containment failure. Class IC Reference B-3 assumes that the vent setpoint is subdivided into Classes ICI and IC2, depend. pressure of 60 psig will always be exceeded at ing on whether the SCS is failed or available, this point in the sequence and assigns the lower respectively. branch of event VP a probability of 0.0. R1. Because these are sequences for which VE. The failure probability for early venting high-pressure injection has failed, the RPV through the SCS is taken to be the usual SCS fail-would have to be depressurized to allow low- um pmbability of 7.2E-03 for sequences with the pressure systems to inject. However, for ATWS SCS available and 1.0 for other sequencas. ' sequences the operator is directed to inhibit ADS F Beer.ise of the addmonal pressure contri-to prevent boron from being flushed out of the l core by the low-pressure systems. Based on this, t.wn fmm suppmsdon pool steaming, cady mpn ssum faHum tk contaimnent is nmd R1 is assigned a failure probability of 1.0. more likely than fer Class IA sequences. If the reactor vessel has failed at high pressure with no Cl. t he containment isolation failure proba-hydrogen bum and intact pedestal downcomers, bility is given as 2.1 E-03 in Reference B-3, Fig-ures A-5, A-6, and D-18. No basis for this value Reference B-3 assigns event F1 a failure proba-could be found, bility of 5.0E-01. If high pressure vessel failure is accompanied by a hydrogen bum or downcomer H1, failure, the probability is increased to 9.0E-01. The probability of a hydrogen bum dur' F nally,if there is both a downcomer failure and a ing core melt is the same as for Class l A: the val' hydrogen bum, early overpressure failure is as-ue is taken from the Reference B-3 figures cited sumed to be certain. above. PF. The probability that a driving force exists VF. The probability that the RPV does not for a puff release in the event of early contain-fait at high pressure is given in the above figures. ment failure is quantified as in Class IA, with a llowever, the probability assigned varies with the value of 9.9E-01 assigned. venting s;rategy. The reason for this variation v uld not be found in Reference B-3. Therefore, R2. No basis could be found in Refer-for the probability that the RPV fails at low pres- ence B-3 for the values assigned to ex-vessel re-sure, a uniform value of 1.0E-02 was used for covery for Class IC. When the SCS is available, both Classes ICI and IC2. This is reasonable be- the failure probability is taken to be 3.0E-03, as cause the RPV is at high pressure by sequence in Class I A. If the SCS is not available, Refer-definition. The values given in Reference B-3 ence B-3 assigns failure of event R2 a vaiue of ranged from 1.0E-02, with the SCS available, to 1.4E-02. For the present analysis, a uniform val-2.lE-01 with no venting at all. ue of 3.0E-03 has been assumed for scouences where the vessel fails at high pressure. If the ves-DC. The probability of pedestal downcomer sel does not fail at high pressure, the sequence is failure is the same as for Class IA. assumed to be recovered in the suppression pool B-29 1 1
and event R2 is assigned a failure probability of "The conditional probability of [ late) con-0.0, as in Class IA. tainment failuren, ... strongly dei dent on the availability of CRD flow-H R. Failure of long-term containment heat No reasons are presented as to why CRD flow removal is quantified in the same manner as should be available for Class IC, but not for Class Class I A. I A. Theref.ne, the Class IA failure probability of 5.5E4)I is used in this report for both Class ICI VL. For late venting through SCS, the usual and Class IC2. value of 7.2E-03 is assigned. For the existing
- 1. ate overpressure failure with failure to vent to vent, the value of 1.0E4)1 from Reference B-1 is Filtra is assigned a probability of 5.0E-01. For used. For Class ICl. a value of 1 A is used (the Class I A, a value of 3.0E-01 is used. This in-SCS is assumed to be failed).
crease is not analyzed in Reference B-3 but 3 seems reasonable based on the additional contri-OT. Refeience B-3 is inconsistent in its quan' bution to containment pressure from suppression tification of drywell overtemperature. If the SCS pool steaming. For sequences with failure to vent is available, event OT is assigned a failure proba~ throuch the existing lines, Reference B-3 assigns bility of 9.9E-01, which is reasonable given the a pmbility of ME4)1 m ime overpressure faib likelihood for vessel failure at high pressure. For ure.This decrease in the failure probability seems sequences with venting through the existing lines unjustified because there should be no difference or no 5enting OT is assigned a failure probability between laiture to vem through SCS and failure of 1.0E4)2. The reason that venting through the to vent through the existing lines. Therefore, SCS has such a large impact on this event could 5.0E411 was retained as the late containment fail-not be detennined; theretore, a value of 9.9E-01 un pdability for these sequences. is used uniformly for this report. BS. The probabilistic subdivisian of small and large containment breaks is quantified the F2. No explicit basis is presented in Refer. ence B-3 for the probabilities assigned to late same as Clasa IA. containment failure. For sequences with the SCS
. W hohl up of noble gases by Filtra is as-available, a pmbability of 3AE4)I is assigned to signed a fai ure prob bility of 9.0E-01, which is late overtemperature failure with venting to less than that of Class I A. This appears to be in-Filtra. In Class IA, a probability of 5.5E-01 was consiskut we the saturakd conditions in the assigned to this same event. dne reason for the
*"PP"'***n pml and the establishment of a steam dif ference might be that control rod drive (CRD) heat transf.er cycle to Filtra with CRD flow. The pumps are apparently assumed to be available for fa lure probability should be,if anything, higher Class IC but not for Class IA (see Table 4-2 in for Class IC rather than lower. Because of this in-Reference B-3). Coolant boil-off from CRD consistency, the value of 9.9E4)I used for Class flow would pmduce additional steam. This addi-IA was retained for Class IC.
tional steam could establish a heat transfer cycle to the Filtra system, preventing thermal failure. SP. The proMility of suppression pool by-This is supported by the MAAP calculations of pass is calculated as for Class IA, using Figures drywell heat-up timing presented in Refer- A-15 A-16, and D-27 from Reference B-3.The ence B-3, Table 4-2. As this table shows,if CRD resuhs are as follows debris cooling were available for Class I A, the , time to reach 800'F in the drywell would be eb
- If containment isolation fails, tended from 10 hours to 28 hours. In addition. P(SP) = 6.0E4)l.
Reference B-3 makes the following statement concerning the relationship between CRD flow
- If early overpressure failure occurs, and late containment failure: accompanied by pedestal downcomer B-30
failure, P(SP) = 1.0 for a small break P(SP) = 5.0E-01 for both small'and and 5.5E-01 for a large break, large breaks.
- If early overpressure failure occurs, with no pedestal downcomer failure, As f r Class IA, venting through the existing P(SP) = 1.0E-01 for both small and vent lines is assumed to lead to a release that by-targe breaks. passes the suppression pool.
- For late overpressure failure with NOTE: Here are inconsistencies in the val-downcomer failure and a small break' P(SP) = 1.0. ues used by Reference B-3 in the above fig.
utes. These inconsistencies were corrected
- For late overpressure failure with prior to using these figures to quantify event SP.
downcomer failure and a large break, P(SP) = 7.5E-01. R B. Reacter building retention of fission
- For late overpressure failure with the product aeroscis i: quantified in the same manner pedestal downcomers intact, as Class IA, with identical results.
i I l 1 l l B-31
i J
~~ !ct...rea ms
=
..r.n-c.,. ww win acw ==,g...
u ocv. ir ni
= w, are.- rev e.n
...< = ,, i v o,.
re n... j a
.ov im.ium si.o < gig reum .. nie arg,y
- v. cons .o;gi.
]
no ai er m a oc = v vi ri y H I I l l i; . I l i.ecE *cL i
, - - > \
l u unw memo t oce-en -
- 9. OnE 41 u
a . ace +c s SI APERTURE I CARD Figure B-4. Shoreham Class IC SCET without SCS.
- Also Availabic On Aperture Card
No . Cant. Lgte Late Large h Me 89 SFF'JNCEfgf7. ENC] l ous. p..op.a w__. n N.f.f. 3. m.Ean= == . vNo.ns la. s ..O.v .e =
. in Cm. r as . M. o Sto ruir %sm m mm . !ue C
Reesvery Removal sie SCE C1 Petture Detak O p p.i.I.0
*etts _ , , ,, _ ,
Dr Ra - Mt' A. Of Pt es W Sp - fe
*~ * ' * - - : 0.00F+00 41
,ri cor-o3 -0.00E*00 43 M QL'08- - - - U g((ha il%{j$ y r D h0"'
!!:M
;0.00f*00
- BA C
DI " - -
%,4g y--- - - DF-El il -* f EMM -- - - -
0373f-d1....j[ % f $ 1
-- - - - W J.14E-0 03 M QE-ca !0.00e+00 Da OW" DM *Da - .I:!!!*$$ 83
~~ ; .00E+00
- - - - I gn g ; .006600 l ' 'MQh2Q.1. Sche i ,0.00E*00 C2 g gg - j0.00E*00 C3 A90M '; h .*$$ $j M.eoE-ol P- N d'. N 1 coE+Co M2E-o f 4aE-07 C7
._- W**E.o t :
4dE -0 7 CS l ME*N 3 !10E. R [0 k .h
.. 4+ -0$
m { b'dI :2 mar-a t
. f. .
=06
-04 SJ Is. ace-of 8. *04 84
'- -"-*"'IMgE-0.1 rgmy.c , -- --. 4 -06 38
.47 -96 Se 1 - < l .6 -02 37
' I 00E"O I ag.ag ( :n . nor-a t uar-" m en-o , ., m .aa s:Mai *it42 1 8.90E-03
. 0.00E*00 y _. e nE.o s , 0.00E+00 Da N" 0ho+00 0.00E+00 C4 1
' " *"' 0.00E+00 Ci la noE-et r O.00E+00 Ce
!B %op-o f , ,g.3g 0.00E+00 C3 0.00E+00
~~
C4 3_ .eg.e3 0.00E+00 D3 . E.Bar-of 0 it cQE+oo Ilcof-ot ny.a g M .00E 600 C7 0.00E+00 Ce I-13GKh !'h0f 0
.t. car-et
- 0. 00E +00 04 0.00 +00 Da
.00 +00 C 13 oor-03 I D r-ci .00E*00 C
!9. mor -o 1 i 0.00E+00 CE LE.SoE-of 3 m.a t 0.00E+00 C3 0.00E+00 C4
- 7.79E-04 03
' q ~ ~ a p .m
' -~
;0.00E+00 02 (t.noE+on g gga 0.00E*00 C I -- --
- 6. +00 C
'"-C' le mor-o f r O.0 +00 CG IS.SoE-Oi w 0.00E+00
.00E*00 C3 C4 s_oor-o3 i
!E . cof-o f I
- 54: a i _ ooE+co i t Sof-o t . 33P -OS G8 g .00E-10 C7
'Y. aor-a t
- 8. -to Cs a nor-e t erigg ET"'""" g*.at %0
- 3. ace-06 f0 sk f
,t oof-of i -04 83 I a mor -a t 2.0.W.E-01g 0
It_Sof-ot 5_3 1
.7 36
.7tE-08
. 7F-03 1 ;
,t.60E-of --
. SE-04 GG i i
a . w-a r g.a t . . SE-04 Se [1_ doe =of ., m ,,o 1 3.S2E-04 2 __ 0.00E*00
" ~ 0
! t . coE +oo " 'O 0.00$*00 04 0.00E+00 C L
~ I
;' " - ' 0.00E*00 i
le.nor-at I - O.00E+00 1 Mo#-o f '3^gg.at 0.00E+00 C3 3 00Ed3 0.00E *00 c4
- 0.00E+00 D3 1
* " " ' 0 . 00E +00 C8 la. cof-o t r O.00E*00 CS
_f.oortoo y oE-et no,y , 0.00E+00 C7 0.00E+00 Ca e_noE-ot a sar-a t 0.00E+00 Cs 0.00E+00 C40-
'I sor-o i ll$N0 0.00E+00 D
C4 BEE-03 I 0.00E*00 Cl
~
9tosO8c287-12
_m ..... . _ . . . . . -~ .. _ _ _.-..._.m . _ _ _ I l l 1shn L.AQEt9L LDQE-01
.B.20E=di i
1.nican.__ ,J a totL+0e z.no 1 a.mn 1
,a.ma n nan.aa om.
- 1. 08N .0c E. Ann.et __
1 b . 1 l l 1.RGE:en "
- 1. . A.RAE6 SI i APERTURE CARD- -
I Also Available On i
- Figure B-4. . (continued). Aperture Card
1 l I Mc E -Q i .. '- ' O.00E+00 C3 EE*$ I W F-01 _l' j',jf h
).93E-02 CXJ M t SSE-c2 r c.00E*00 08 CEUEu ',
,C4
- 0. 00E +00 e ecE-e t i U ' 0' O mcar-J g. OE.gg:go C3 er-ei 3.13F-0 U
18 eof 41 i C4
.1. 0 0(+, l g gt-et ngg i j 9
...=e-et reac g:gg gl g;,
I 0. 8 go E +00 S3
'I 00E*00 MMZ 'ME-Og g ME-Q i .. , g gm ; .
- , 4. 3SE -05 96
-0 8
'O 00E*00 97 I 00E'M - 3_ccE-c2 l; (
E oe t 3 .0 8 808-C' ruera%..u-Gi;gjj.gj ! .)
- 0.00E+00 1 3
- 1. DCF -g , l0.00g+*00 0
00M0 'T*!"M c l*$$ggl 0.00E+00 1 I ygggg, 0. 00E +00 Cl y s er-o e ___y', ., 7 ,c , - !:0 3,0.00E*00 C4
,0.00E*00 D3 I ngygc o 0.00E+00 CA I 00E*C0 DI :7.ECF-OI I .I- E-Ul
.L M *01 :3. scE-O L OO
- 0.00E+00
- 0. 00 E MO 0.00E+00 O
kO 3 I3 - I ME-c i . f18 M 2 E** i 0.00E+00 0.00E+00 Cl C2 4
'Y~T'Y 3 f~~"" llll l 0.00E+00 D3 h
g _ eng.c, 0.00E+00 D2
,' I 0 0 E *00 *
;t_ m eca l'.0 M
'0.00E+00 C1 1 0.00E*00 Cl le _ ecE-e t i M ' ' I0.00E+00 C2 d.12E*Ri :3 ~scE-e t 0. 00E MO Q
=
I . , _ 8::IO:
- 0. 0 0E +0 0 8s C8
- d. ~"E* '
iM . ne r-e t i O 00E+00 Ca MEtSS.--- li 00E*0' M E-ci ; l*lll'9l g [g
, - #~ t e-e i g:gg:gg g, 1 0 8
'i.coen w _ M e.00E+00 . sa-C e el 2.47E-07 83 '
s = "-O' . me,41
-c -
1:i.'.14: 4 E -0, 0.00E+00 87 t.i. .I8
,1 ceF+ao 1
3 ~~ c7 S.28F-C4 4.80-06 99 89 , ,, E I [, mara t 1' E- ' A: hk 9.87E-06 912 0.00E+00 D3
' 00 It GoE+ea . ' . ' ,'-
O.00E+00 04
- 0.00E*00 Cl I
' " *2' 0.00E+00 Ct I n . ecF -o t i O.00E+00 C2 i 15 Edf-et :3 er-c i 0.00E+00 C3
- 0. 00E +0 0 C4 0.00E+00 D3 I *a" 0.00E*00 C3
' 1. 0.00E+00 CE t 00E+' o mQ li o$f-et 1
0.00E+00 C7 7 mg.cg 0.00e+00 Cs e . com :3. 33.o t g;g +00 0
. . .n-0 3 es 0.00E+00 02 t Sag.e, f t _ oer+ce gg y 0 0.00E*00 Cl 0 I 0.00E+00 D3 i t _ nM+ed !
-C L 0.00E+00 C8 13,0cf g
~'-
- 0. 00E M0 C4
' 0"-a i uwc e :8u4: S i ccE -c i :an08 81 0.00E+0u Cl 9 I 0.00E+00 CL le eer-et F D 0.00E+00 C2 in net-St :3 sz-c i l;lyQ 0 $
7.80E-05 D3 0.00M+00 Dt
+0 lt o cE +o e nm {0
~
- 0. 0* +00 Cg
- a a* *0 e
, ! , m -m i
I 7/oYO60287-/3
-. _ ___ -- -- h3 3 ____ _ _
2.com.ca _ ugt-c t J 0.0cE+00 l r-
,_e2,.co
.2_ a cF-o f i 00R-c3 E
SI APERTURE lume -- CARD Also Available On Aperture Card i ee .oo
- 1. 0 Cf 42 0.00E+C0 2_fof-os 6
Figure B-4. ' ?ntinued). B-34
" ~ ' -
f t. tup ,g t ' 30.00e+00 ca 1 t aaE-es rf1 tar-a g 0.0g +00
, :Ter-0. Os my- .3w-50 gs teor _ _
m.0 t 18"HA nne-e t nar-oi n-= - bb Es e..0E-0s. itO
!!coF~of _ ,'f*18fjC h hE Lant:C' v,
l'. eg-Os li.nor-O n - uor-c i 3338? E
- a. n-04 er II Dof_-ot E0 ggC * $"O a . ou +
h st MC c sanon f 0.l'lhh
, ll
+00 *
- 0. 600 (32 DMt .cof +eo DMfod 0.00E+00 ,J
- g. 00 M 0
1
- n me-c i 0. 0a.04 11
!* **#~* I
$a_ gor.oi 0. 00E +co C2 Laare o'
_ _ , N-o I - - h,[nm 0.0 h m i Is.cor-ot i m,m m 8:
.. ,. u. -ei nur:n $':
.8 *0E +00u:n
- 0. 0vt00 C7
, ,,e t
, -4i 0.00e.40 0.00 00 <0 W r-ai -0.00 . *00 ',
[ -~ - 0.00 +00 N 0.00i +00 C gg3 n6c i O . 00l '+00 C 0.00E+00 C1 t
- n mor-et a w -oi
!:$NIN b 0.00E+00 38 L #'O' r 0.00E+00 It_oof+0a : n . txu sca 0,0 +00 0.0, +0g ,4 i Cl r.r-e--
=
ta.ao m i 6.00t.+00 0.0 . 00 C2 L10E43
- 9. -
___-.i -- :gn:n a g 0.00E*00
,7 EfBWn '
Is.cof-of I O +00 i a6etoo ie_oor-,t 0 +00 C7 r_w * - 0.0 E600 C8
.g.30E-01
- 3 " * '
0$ +00
- 0. +00 g,10 g I_l toaf+0a :1 - -nu j', 4 00 53 er+ m-3.sae:a%,_ , :::st 4.4dE-0 0.00E+00 07 mor-o i l t _ nor +0c c , _o, 3.jqg0 . Meg 3.aQE:2' 4. 05 Slj ls_norant .r EM9R'"""" l'dg-06 g g
1 0.00E+00 03 li M'0' 0.00E+00 C3 g'ge 3 m +aa 0.00E+00 31 0.00E*00 04 l 0.00E+00 C1 0.0 +00 C
" -*' - - 1, . 3 -o1 0.0 +00 .
,-o i , - -e i *:en:g C1 2 - -os I
8.0a +00
- 0. 0= +00 l8.
C.
- i am *no
,_,og,,o
!s noF-ot 1 0.00E+00 Ce si 00E-ol :7_ nor-c i 0.00E*00 C7
- 4. 64E +00 CS 3.goE-01 3 " ' " ' 0.00E+00 Ce O.00E+00 0
,1. QDF-C_ i 0.00E+00
- {L eas +eo 0.00E MO 03 l
6.00E+00 Cl g , gg-Os D3
- 0. +00 02 W M t.ons6do i m +ne 0.0 g,SAE-08
+00 0.00E*00 Ci M'*" t _ oof +0c 1 0.00E+00 03 l
t
!5. hof.-01_..
1 oor-of
's . ow -a i Il'hNh
! 0,0 +00 h!
C7 s _ now-s i l O.0 +00 Ce 0.00E+00 03 t . nW2 9. +00 De o 0.0 +00 J +an _ 0.00E+00
'O^0 0.00E+00 Cl I
- 0. 0M+00 03 i.00f+Qo 46ci $'.!N Cs 15 aa"A-ko E-oi en i-
...N.:n. 0 .iii-
, canch i--ei 1:Mia !!
ta McE+ed 0.00E+00 0.3af-04 B7 I8 GiE~of 4 nor-o t y to 0af tco
,' 'T"Thr-a t [%$
O.00E*00 ll t l \ HoVo30287-/ - . _ . .
.~
% '" ∈, c-l.W: a='4.;iD...,it ' ='a>'. ais:+: +:l;,:
"'ht
- .c. m.is .....
~; m .
&h PD Al C1 t. VE DU HI VP VI FI ! E8 8'I I
M.$Isd JJthi
?.roboL .
l u:en _ A 3;r-ni AJKtao bCEC r--- tuncks t un
?_201 03 4.00P 02 222kDL AJakR1._._ _
A.00Etto _. XJDkL
, SI r--
, l l APERTURE i tuakt CARD ug=* B-r,. Shoreham Class IC SCET with SCS.
Also Available On Aperture Card v, .
- . . - . - . - ~ . . ~ . . _ _ . . _ _ . _ _ , . . _ ~ = . . _ _ . = .
_ _ _ _ _ _ - . _ - . . ~ . . _ . , - _ . . . .. . bJ aM
- y C t.
u.t . N. ( .,t4. . .D.v se a., ' se 3 h.e, H.p . , .tu.pp, .No.,006 u .o .f.l3JENLF R f. u .W.8FN Cf f, g,.,.,i n. .u
- o. i .c t.. . t .J... ...
- i. .
. <t. o, r. .. ,o .
00F400 4
-h -t-" bu, ;"
- n[:n
- n. ; t,u_ _ ..or., an
- = ,: M .. l, n..,i::I
(--nami g 6me .n , n:a4 -
- n,:81 !r ut-n - n u r g a g:n !;
; 6. I af -0 5 ,13
- b :ec uwe-o c
.a. .._
-W i: t.:H';n, C1 n.
Luv uen n:st _ j g!..e:n,j,
- n:n d hg,.gy MLkD1 EIUt u:ar.q!k' 2.2L-2L -. y:G1 ){; f' y M & 21 E g G C g ,$ $ jgg r am a w at= n.,n:n : got.-04'} l
,gt_ n 64 heen-c:mo _ g ;n:n :.oor-ce
...,... lt, i
u mon:sc ::n:n ;; imai- --
~wmt u ::m = 6!.!hot!!!!l 6 l!
Ja 1 N^"E - 0.000 00 ),
-l, ror-c3 ggg 0 gg Og jj g..ee.8.
Luv men ..
,u:ai m ur- Uj:ia d:
g::gg:gg :
..o ... Jg F
uux-- La.a . _{t-.ne" mat i P:n :9 i en tgn:n g, uno -umac F):!n:'. ei n',,st -8 gg:g; )a usi I Emenuucu' 2 C' uut-o:lN.e.ig! h,
- a. -
e.eir-8 e b H u es umus' mm: .. H:.n.r;*8 4 1 L
'I
.sa- o ,:
I D C' 4t - t ll g-g.n *B.72- t g, smm. , m a c u n:n a ' ..ga g og um1_J am~ , Lai-hmum El2 umm - - by!! a:,: il s.sor-e e g t , :mem m:m c ea:n ,f,, f ' ual. ummi mux _
"jyiji
.,u of .,
ll e-mm1_ , - m~~ ~
- Ia:n st-as-1-a=mc a.:n.!e I.o..:.
I:nl
- l u-%~, m m.
0.00 00 )4 un.:n o.on og .. n n:rm 2:nl:8. E uc'n H s. set-n un u . . e. c; 8.gg:n ox.x 83
~ 8 2 , , =:s> .- 8:ni:n u o o. -o a . n e-n nua, g;gg:gg g wat , n:at_ cf.
~ a:H.i:8'3 p . .- gi
[""~ , iii!!!!: si B-35
~
9/04cFo287 -/ -
j . t. t C 3__ J 1# m E*.L J. L2;Lt2L , a sittan L E E:11
? _20E -0 3 k.uust-.- -
.. . m .ee da LQCLtSD b -ea SI APERTURE CARD l l 6 Figuro B-5. (continued). Also Availabic On
/.perture Card 11- 3 6
c . LL.12hd AJEW y t-et N I!'N *k! ! i
,"$j '
ti Du *ff .at'*0
;y _ py -es 0,li a gp?a 0
'.0
'*0
.I
'4 i l-0 0 uns-1.00E-02 le. mar.et 1,ML-Li TtB E- e i - ' ' .s.h' og
*I .de e
'a 4
.y '. -b dI 3
\
n.0 . ME DEU_I: .* i jj get-of 'Y t'r at 'J a
+
1 1 g gar-et a gz-c t _ .e$ ka
- .oo : A It.eeben ,',:r.1 1
- .00
[*,0 ) l [ f.eef-ei e -8. i4
's.Ent-et '
r? r-*Y""'"
'l:h i!a
*0l II.cet.ee Op .40 hM-e . :n a a-c r - ' ,;f3 f.*N !!
I 00E*0I .et .-44 e
If EBE*0i *T"R5 ';;'d i '. I 1 m ,0 :' .4 m Li 00E*0# I, 10 ,.!) '
[7_r ,.03 U fL*2: p, th '.i) [p i
;..io :.0 44 I
i {
- n.u:r-ci in.nor-ei i '
g3,,g {-, [0 0 3
- i n - t y er-en .
. .g0 4
08 gl y o ,0 ; M cE-21 4 ,g g ..gE}n ;i.00 : 00 i 2.IDI*EI ' g r.e, ll;jl lg d A'I TE![ b i ' ' ~ 0;$(h6 !"O
- 1. nu-e t I *$kN*.h!
i,.0st+00 11.0e2-03 I '. ',[ it. net-en i G E-01 ' tEg5 g,ror-ei Ter Yt*""" lh.:,0! h$ g . n : 0. 0$ $4s 1.ne2%uy.e, runcc ii. ci' 00 ii.00 00 ft
)3
,i 00i 00 14 c . 00 m i .l . d. 00 :1 te.nor-ni i n:& ;t L:St :!!
l un-o g,,, Tff t -" m:g:
'i g0 y i
> = e -" rsen m. e m 40 .S :.
na-" u'gt.o . goc.8029 m -c1 H:48c:!8 1.
...or-e. w .1, mg:og j a.ucia___n mmr- -
s
.401.g ?,}0 mr--- ",;'"j':gi uu-o :uer-n g {. . 44
, n 3 j, - y
.ei t
- e u .ee T -tr m - 3:l;ggg\ t:0 ::,
a.ut-e 8.
- . u .ei ".i>
c:,0:gt 1.x = i
.00 :+ 1 I-
- f a"-O' ,.,>0 :.06
- a. nee 3 n x +== ,., ,0. :.8O 0 Le' SI' i W:) l' 1
- 3. eor-en ta.snt n % . u -ei ux-=>
" " - = '
1 [:g:E'g y l!I:!! t i 0.00.t0 a.
- 1.scr+tr .:
'].0 .y li8 li BM
- 0 I }g_eeret
' '# ' O
- 7 sar-ni 1 O';.00 58
- a. ant-e. :s unt-c i ii ,
'. 0 ' .
O i 'i.st .03 13
<f noE*08 i,90 +05
;?.nor-c3 00 ]2 <3 M or+ca it.
' . 4E-07 >4 uu.o 1, ,, _ , , i
- j. 00
]
- s = -=' li. O i. ' S 1
18.eM-ei dg ent-e.
,,,g;r- n i I
- .0L:.! :' li ;t 8
p u mt-ci I 9, ;! 2J*06 0 e h.tBE t3 1
- u n. b 0.
_s unt-t t la nor-et y gg,3, : 3. su -c i
'g'[*y .
'3 C
4.4 - {4 i . DM*02 i .k3E 17_2SE-03 :i.ncE+03 c . 00 t00 ".0 9:3 4.00f-il 34 i g. ] g I ' duett
.. - .- _ .- 910V0802 87-/(s_
u La3
! I l u tu a__.
, , i l
! ; use i !
l } I i
- i. .
1 l .na:act i I I I l I i i 1.cana___ i il.121_2A_ uw . I l t.cor-es usung
. ti. ace-l
't JL2cE
- ca I
Si l Al'ERTURE CARD .lso Available On Aperture Card t " -os uursa unuce . u nt-n1 na-n Figure B-5. (continued).
- e. ~
' d g g-c3 **** '"snu t rnard *.47f
'? CR I o
[
!!.nl: .08b5 tht- 4
- 4 , '7 7E - 4 C6 LEEE-EL l .1.UM14 d=Ed'El ~DLG1 i I ' hh *. ' s h$
&nui 07g-te ce ammu 4:nE:l! Et0
- 3.30E-os g r
1 au'-nest.- Lut:u -3.30f-ce 3 4 pR-D i gyg-[ { ; , e, a ast-Os ir A MHb--- MEU J
*" ll h h
.seE-Os
( i0
- 1r&Q i n;mrj- [$ *$$ .
)
"d a m st - i G Risi :8:!81:88' il m mo .ifi:n ?!
i g D"EilT 000k!0 II;f ' dJQh21 0.0cg+00 g g-gs E l a C !; hog Ch h!
,0.00E*00 MEL-12 i 'wted E9L*9i I""
D EE-Ei !!:!!I!!!I I$ Og J.J C 0i l *g g MEU M EE .E.L - 6.00f+00-.* l ?68-04 0 g ;O' i d;NI N doma w "" * . M a-at n:mi I Esd ";'jh:$f
$:U!:!' e l0.h0E*00
- !]
33
' ;0.008+00 dJEL*Al )#
mEHL I 'N 0 - i DOUL !! 0.0%:n t} l ," ' " ' 1J2E21. I; wt-n 3mB r-* $;,gCE+00 i jj r- if!;;$%: il
$E 2,w-r " " 1 - h t-01 . $$i I !!!!!!!! l;'
dJCL*11 mMi .. *
. 88I+000 g{0
- }
MD0 ' "'QBC SN $3 i d SEE~ I [B.BdE-01 hl 3.gDE gG
. B4 g g p ]* g6 e$
i m un % x-a ! gh agg.tL,3:, MR*QL--DUL-c l ' . ' ) U 'O.00 +00 Q$
.? 2 t-c3 0.00 +00 v2 WFTh ;hlhh hh h I 0.00E*00
"E-" ' :E.D;E-DI Cg u n-Di amsi !;:gj{:nl3 l I
M,EE*UU O .' 0 0E + CS M&E 0.00E*00 C8
'd.121*Ei H E-ai f 0; bJE*Qi M EE-01 5 h;hh' s N ;O g r.nt l e . 0n-Os )
EMO -07 L ' h:0.91gI E O8 3
- k n " u.v e-cs m E.cc 4 7. E-10 32 1: Mitt 31 0.00E+0 C1 (ZJEkQ3 !
TT5Ua 1
- d JEE*01 1
- 11. c a t-0 2 ME"~o f . _
ll h h
!O,00 +00 h
33 j gg,g3 l0.00g+00 )2
~ ,_----4 , *
!0.00E+00 C1 SE*03 r- CIG1 +
K'S i M EE-Qt h',Ohh 8.40f-06 h 6L i ddEhEQ;E-c i .. . . ll$h!kl $$ LQJ C1!20 0.00E+00 8,3tE-04 B7 E UDE~01 {;l[,2!l !! 1 EE01. '." .' f L ERt)0 t 0.00E+00 B-37 9/ovogc 267- /7 l
CLASS ID PLANT DAMAGE STATE The following is a description of the factors in. PF. Ilecause the probability of vessel failure cluded in the Class ID Plant Damage State (see at high pressure is small, core debris is not likely Figure 1640. to be dispersed outside the pedestal region at the time of vessel breach. Theiefore, the lower PD. Class ID plant damage states involve a branch of esent PF is assigned a value of loss of coolant makeup with the reactor depressu- 1.0E412. rized to s200 psi. These sequences include com-R2. The value of 2.4E-01 used by Refer-mon mode failures that disable low-pressure ence 11-3 for failure of ex-vessel recovery could injection systems. not be substantiated and is not derived in Refer-ence 11-3. Ilowever, it seems reasonable that the RI. Recovery of low pressure systems has al-probability of using low-pressure injection s)s. ready twen treated in the front-end analysis. No tems to establish a coolable debris bed should be additional credit for in-vessel recovery following lower than for Class I A tecause low pressure sys-the m,tial i stages of core melt is given. Thus, event tems are initially unavailable. Also, for sequences RI is assigned a failure probability of 1.0. w here the vessel fails at low pressure, most of the debris should flow to the suppression pool Cl. Referena Il 1 assigns failure of contain- through the pedestal downcomers; very little de-ment isolation a probability of 2.1E-03. No basis bris should remain in the pedestal region directly for this value could be found. undemeath the vessel. For this report, a value of 9.0E-03 is assigned to the lower branch of event H1. As before, Reference 11-3 gives the prob- R2 for sequences in which the vessel fails at high ability of a hydrogen burn during core melt as pressure. This value was calculated by examining 1.0E the dominant cutsets for low-pressure coolant in-jection (LPCI) and core spray failure in Appen. VF. Since the reactor is depressurized, the dix J of Reference 14-2 to determine what probability that vessel failure occurs at high pres- percentage of the cutsets would be nonrecover-sure is small. Accordingly, Reference 11-3 as- able in the short temi. Sixty percent of the LPCI signs the lower branch of event VF a probability cutsets and 60% of the core spray cutsets were of 9.9E-01. judged to be nonrecoverable. Either LPCI or core spray is adequate for successfui ex-vessel recov-DC. The probability of pedestal downcomer cry, so the failure probability of event R2 is the failure is taken to be 2.0E--01 as for the previous product of thest two percentages, combined with classes. the probability that the o[vrator fails to align al-temate injection (i.e., RilSW, which is assumed H2. The probability of a hydrogen bum fol. not to be failed at this point in the sequence), lowing vessel failure is taken to be 1.0E-02. as which is taken to be 3.0E-02 as in Class I A. This before, gives P(R2) = 9.0E-03. If the vessel fails at low pressure, the assumption is made that the se-quence is ree vered in the suppression pool and VP. Reference 11-3 assigns event VP the event R2 is assigned a failure probability of 0.0, aame values used for Class IA. as was done previously. VE. The failure probability for early venting HR. Failure of long-term containment heat to Filtra is the SCS failure probability of 7.2E-03. removal is quanIifeed as for Class i A. Even though there is a higher probability than in F1. The values used for early overpressure Class IA that RilR will initially be unavailable, failure are the same as for Class IA. the Class I A values can still be used because i 11 - 3 9
y - - _ _ _ _ _ . _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ event ilR asks about long-term measures. As dis- separate event tree. These trees wcre then used to cussed in Reference B-2, failure to establish quantify early venting a.id early containment fail-long-term containment heat removal is primarily ute on the main CCT. The later CET events are due to operator failurt rather than system hard- then quantified according to the initiating plant ware failure because enough time is available for damage state. This tends to blur the actual pro-repairs to be made or for altemate heat removal pression of events in the sequence. The only dif-rnethods to be used. ferences between Classes I A and ID are the probability that the vessel fails at high pressure VL. Failure of late containment venting is and the effect on event R2 of the initial unavail-quantified as before at 7.2E-03 for tne SCS and ability of low-pressure injection systems. This is 1.0E-01 for the existing vent. evident in the SCETs but not in the CETs devel-oped in Reference B-3. OT. The probability of drywell overtempera-BS. !!reak size is quantified as for Class lis, ture is taken to be 9.9E-01, as before. HU. P(HU) is taken to be 9.9E41, the same F2, Late containment failure is quantified in as Class l A. the same manner as Class I A. Because the SCLTs la this report explicitly include the early pheno- SP. The probability of suppression pool by-menological events, Class ID sequences are es- pass is identical to Class I A. sentially equivalent to Class I A sequences for the , later events (with the exception of R2). In Refer- HB. Reactor Suilding retention is quantified ence B-3, these early events are developed on a in the same manner as Class IA. i B--40
ci... so oo. n
.. w...i,
...r
.c..,,.o.
e., e o or . uuui,,, b i .i e,<Py..t g;
..... .i oc. .. o .. ..--m r.o e.
e ,, .
..i..
. e.
v.c,o
;gi.
i,
, . o. %i, a e g.n o,.;,g<
a . e.. , . PD mi C1 __ kt VP DC H2 WP nf Fi PP Rt I lidQkt; 7 PSF-t3 uers n.nor-ni onces;
\
~
l la.ca-c, u2cs] 7 mnE.es t . 0 0E -0 2 7 eier.02
.R ff0E-oi uness _
AJDEs) i i i i luces; Figure B-6. Shucham Class ID SCE1 with SCS. S1 APERTURE CARD Also Availab)e On Aperture Card
~ - ~ _
Cin N ew N. D M 92NC8 ot tk.*. Dict ey bleup
- i. L ioDrep-
. a h.*!1 on . e. p..I:s 1. ma s oni s ee. Paca. cuse y n.Pell.
t.w i. N.
. vi dnd ven e on rc ate ai.
n wr. eren erneues [ m 4 Of Ff 88 Hi op 84 9 00t+90 3pg ' - e. net-04 Di 31' f.77f-0e
' .0*t-ce*
!n.act-on i DE-c' ;kNNS du
.s . set-AL. Tur - 'i8
.4t-Os .4
*--ii.ect-on
. er-Os p
. o 84 - 28 N -"" n m1=- L8M'h !!
i
. co :- ig f,gge a r- DM t:aIi *-de !:
&JQL-Al -- -
'rnr:3T--- 7 e y0J '8 r
- a. >s ,a
!,m-ei i ox-= 1 :: R Qce
?.rer-es it est-ci .o1 g g- j g
.LBCE-o f :t uor-a i 8- '
i8 di) *~AI e.* - e f
% one-on ;L2DE-0L r-*snra e
8 g;$0['$$, e, . .. : i 1 1
.ese-07 .:
1Sak2L ;n nor-ea .ao
.ag.-os :
rg nor c, -e p i .it S 4 ,. 1
- n. net- e 1s
- ' " ~
e . air 0 CS
? for-es is.nne-on i .anl- fi ce i fut-el U7
- n nor-a t
'.e8 I*-- ,
ce R.bOE-gi nm A.ao,-0
, . SC l-e l
ie 0 T6 G.09-0{ r P nr-at ,
" :31:8:
2' " 'D' i m taE-et i N $:r it har-ei f3_no,.g, j t- 7 j. l f.50t=
.S 4 H com,... j
]U:%.'os
;7 # -03
'I"1BD u c ',% 00 4
' 4.90 O.ipw' --
;4
- " ""- ' -:1 1 .ag-g q n tor-en n.a -os :-
y ,o,, g ,. ga t .o -,4 u..e::
- a. -- aOs g:
itaaeni ;i.nor-ei n z-m :gi:. : s L20hna 'r sur-o , i .- a ;-- U dt n anr-na i .. i n i;s
., ,or.c, 0!. :o
, il il l l 11 BSE -e i 5.nnE+nL , *0 I- 19 l ll.30!= i ! 13
!E OhIdi
- l1.bnt-ci --
.n.unE+ct
,g7 j g
, .s0 !-Vi l 11
>i.00E-01 n.nnE-nt .00 908 4
. eer-ci . Det e-0-c e. l,eo t.i .oof-ci 733rgg-- 11
. 0l+00 g l
Hl. cat-es ;7.2eres it antiac i
- 0 0E 'f'fl 38 it.ON- )4 I
la aaE-o f i 2 " '
*l h!
4E-M f C ' E W *01 '
- 5_ snt-a t '
,;[j -[j
>- - - e2.*es. ,.- 4, o.
g
- 7. tor-e s i.ent et [
. r-ei o...} 28 m _ __
.7M-ig ,10
% .nor-ci ]8E !$ 4, .y t i .75. m-0e di i
a 0-4 l 9/o Yo80?87- /8' _ _ _ _ _ - - _ - _ - - - _ - - - i
1 1 I 1
)
- 7. ent-es .
1 R3F.e1 1.Def EF AsSSL*L f eDF oi 7.90F.01
,LADL-JL f.70E.01 1.un:o 1
1 , . x a.et SI APERTURE CARD I I i Also Available On Figure B-6. (continued). Aperture Card IM2
.7Pf-06 1 innt-ci. deme-ei eg F*r e t
.l***
3,Stt*0) i84
.> t*$a - 3 b8' ' ** ?
l1 eat-et l y; a rnr-en 1 m sn o
'jl:$$
' '2'*"'
I. 4 0 I a tor-e t it.LBE-ci 3~7'r.rg 4.44E $ 1%I _R . eSt -e t i l'f*00 id i
- 11. t eE -e i I i t . e bt-e t 66 .S 46 hf- --
4
?_tSF-en g ,7.g, ,
l R _ RSF =e t 3 = 7. t t - a. 1e. .5
}
- o
,.1
. 40 +00 'l ttf+et :r. u -cr 0
>3t *]
kat - $ pI t.ebt-et n n e-ei 4 . lb,l,
- n,,e n . 5,r .
4, .w.v, . .a, ,, . ,
- 1_ r - E Z l [
g i m, -e i .. ,0 -41
*
- r-ci mm--- 4.;;; . 4 . -i) gg .l':
.24 -07 l es, e, c, *ct-c3 -
a.cu.c h
;** :*j 3' 1, 1
' u ,-e, -x-o 3.tl :ta t n lease-ei u n a ent e,
- l,me-e, mm
- m. I
.. .$.L:. st.04 u sj i
, ::181:08
, ,er-e, a m t-ei ,m,.m o...... . :t
. .ne-ei um ,i -
;t.o
,.:,43;- [
t
- < ur-o :7ha L l._e9f-c3 I.. ,,,-e, , - <,h. h m m t-ei m -m o. .
. ..e-..
0- -.0. . p. 4
- r-e' :, ,er e, ;u c=u- 8,: , , ,L;p, .,
du i . 46t -i l ! UE E;"E '
.t aSt et i it Eat ei ., v.7 tg l kh.-
'f
)
['ci
~~
u z-ei M: *. i ij s_ent-ei i
~ 1 E E" E 2 k
- i. 7 [. f. ' k. .
7 ter-e3 l it e F-ei uge r 6 i ta D$f-ci 3 ,,r-tt U . f>5 : ..t 20 L. 7 o . fel*.o. R4Cl9 ti.eBE*De u ugry i.DH : i
. 0 81 I
'- 54 t_eer-ci r . 1 E ESE-ci 3, gu ; . gi' b.- i e,6 00 8
.0 :+,i 17
'I ^ 0 * :r r:r-cr
_act-et I !~0l $ iemr e. .0 8. a E ESR-ei :r_ m .ri .64 ;-00 I l.97 !*D0 l
.96!-07 ).'
l1.eet-et ;7_fSE-53
. 96f-00 It .00f *0f' {:.
_1 .7,,"..
. d tf -: k'
. Set
- I -i gag. 1,
!n.e$F-ei r T Err ri y
,, g ,y
. e a nor-ci m .o s.
'a est-et 3. "< >
y 9.0 64 .
, ser-c, it.enF-ei i a ese-e, n n.u 1 E R" E- .
i.IN see-sN,agi ce 2 . 9 11 *le I;8 a.tSt-e, ,g, 4,3t :- t l d.de :- t }40 0.77':- 1 13 0.60 !-0$ it.cor-en )t MS E *e 3 :1 r-r,-.
" ~~
0.0 '*00 ?3
>4 f.gE-06 e eer,ee i 4-f*0h y I? ref-e9 I -' I n m e.e. ,
ii.eef-ei m="" ]*ll]*he 4:00 0 "r. 0.00E+0 j7 7 r;ge t o .a m :- m .pgg.0 g,gge -;;
- i. ee,-e, i
'2-
' g:tl gg-c itEg el.356-08 dot-ca
- E 21 in est-et i i utsn-e t u,, . o
- . p 4 1
IL. SE*0e 53 ti. cot-e? l i. SF- at
;7 *S!-t3
'r*E*FT'* - l.7.f+go
.).
O E-it 0 33
>4 4.70Ealt
'4g g
- u un,- i t.10E-68 H0408c287 - / 7
% ew===
7.20E-t3 7 tDE-tf J .RQL-11 LQQL-t
- i_D0f,6D
- 7. toe-c3 P .DDE-c i _ _
1.2DL:21 _ 1.00E-t!
- 9. tot-ci 1.0bE-dr SI APEllTURE CARD ,
7_rer-es \lso Available On 2.n::-ci Aperture Card L0kL:LL 2.ibE-C3 9.90E-ti F!gure B-6. (continued).
.. - ~ _
m m __m_._ m. __ _ _ _ _ _ _ _ _ _ __ _ _ _ _ __ _
__ ._ _ _ _ . - - - - ~ ~ - 1 ( ? c_ - 4 1 it.60t-ti
. . .. .e 3 fi_scr-ci J a.=*-- J y LA 0bE-Bi I in nur-et j'Ih *. l 5 6
7 tor-c3 . li Bef-ei _
- E nnr-ci
,' gg .
pg . Jtg m.not-ci
- y tor-on
$! I i0
. hy
- f I
w eer-e' 0
- m rm-- ogg: c it geg.* id LL.t0L bl ggg g g ; ,
ll l5
. 3 01:.00 l IL buf-of T Bhr ET'""" ;0[!-ylj ~
f E.eot-Di
;1_ geg.e g hni i .
M 3 nr + x '
.; 0f( *ih
, l
' _cer.et
,i .o ,-na >>
;7 sM-t3 1,il :+N )#
'r*Bor + x li.i>D *0' g4 3
lb.O !*
- I i.
~
lt . hM-D i :s mor -ci l, != , I 4, ta is ; E.hDF-el 3 MDE*CI a hM-es (:- j I l. 6 4 -13 L.0ZE*El
.b.g ..
l :5 13.eet ci I y por.e3 l {
@ 11 :r.cor-Di 9 t
e_ ant.01 t 09 l 5 m -DL - - - l 0
- 1. nar -o i .71:, .o . ) i
" ,g4:.a .L ' ,
-)
, _ gg.e3 I u
.77 := ;
A 7 . la hoF-el I . M -nn h, g_goF,e{ 33,gg7,gg g.:-4 ;f 4 j l:$F' u nt.ar___o ,o,.e, i fr nu ,x 1,00 3 4.s $.
. u
- .ll
- l i,4 i
- t,4 - :l 13 ger.et T""*. T'WJr-ri *== ,'.4 -
g < (R.hof-ei --
,9 3 Y p.cbt-03 23 BDF 'D I .- , 4f.
;4
. )
I ' i y_pey.c3 13.bDF el I DKion I
,g i- t
, G 11.001-01 ,! 97 '* ll
- 7 nor-ci o y og.Bi ii, ;.- i ;e 53
', :. n ,e
"_DF
__ . --- D - '
.' 7:= M 0 li.00: +0i i 6
- 1 eM+ee gr g ; 8 %,, , y E BM*Ci Ii nnt-ci h10 64 pSr trr"*"" -
o, n ..u l( f". I l6 f BDF+BD is.nnr-nr !;l :h ll 1.191-of .e .
-De I
- s _ f nr-c i *
- 1. mc.r-c i ; j
,'yl*t
'l 00E*0# '
. set - 0
- . ml U gge3 :n.ontene ii,00F+00 '
3
" I le ant ei
. . _ ii'~ mar-ci '
y _ gg,e j . r
' 0 M~0'
- 7_geg_ei 9,, ,; 1
]
s
'9_ yg,e3 1
u ent-ei I
- 11. eeE-e i
- t _m anc " ,,$b
- p. :-
1
!g 6 n ME-c i 0 27 t, t ;- 0 46 Mg,g g :3 Eri-ci y $ 40 il;371 4 4gg. 3 g ar EEr-ai .74:-04 5 do nM+oo g '-OS W
. 0 :*
i_nnF.DP ;hf.*g0 0 if,20E-e3 af"Bar4ct e eM.oo ,
,,~,,,.e, i d B OE -e i I n x-m : ,0 :.h 0 :.
6 11 eu-ei ua-o , A0'!*00 i 4 00 :
,0 0 n
{i DD?-ef ., ni - >
;y pg.e3 3_ ng.gt p 0 . 0. :
7
- 24 M 400 ft.PDF-03 I 4. .
i UE-c 4. 0 :. O n eir-en 1 ), 0 4 6 LLter-ei _ L BGE-01 a, o! +0 ,
.[1 I),00 *00 :$
. 20 -0 4
, :. es-ei u nr+oo m icr-c,
- n:: t3
- o e*-e i
- lit::t .,
rnor--- tt m oor.ee 0.00dI::i!:i i po4080287 -- 20 - B-43
CLASS llA PLANT DAMAGE STATE l The following is a description of the factors in- lost only as a result of containment failure or cluded in the Class llA Plant Damage State cate- venting. Therefore, a 50% probability for a puf f gory (see Figure 11-7). release does not seem unreasonable. PD. Class llA plant damage states are trans- BS. The probaSlity of a large containment lent-initiated sequences involving a loss ,f con- break is taken to be 1.0E411 using Table /.-19 in tainment heat removal. The RPV is initially intact Reference 11-3 and assuming a slow pressuriza-and containment integrity is challenged prior to tion rate. core melt. The SCS is, by definition, not availaHe. SP. The probability of suppression pool by-pass given early containment failure is calculated VE. Failure to nnt containment to Filtra ear- as for Class I A using Figure 4-4 with input values ly in the sequence has a probability of 1.0 because from Table A-19 (both in Reference B-3). The the SCS is not available, by sequence definition. results are as follows: F1. Overpressure containment failure occurs
- Given a small break. P(SP) = 1.0E412 prior to core melt by sequence definition, so that event Fl is assigned a failure probability of 1.0.
- Given a large break, P(SP)- 1.lE-01.
PF. Reference B-3 assigns event PF a failure probability of 5.0E-01, but no basis could be RB. The failure probability of reactor build-found for this value. The RPV is at low pressure ing retention is calculated the same as for Class for all of the dominant contributors to Class llA I A based on the data used in quantifying uent SP listed in Appendix J of Reference B~i; llowever, above. The results are as follows: because of the loss of containment heat removal, the suppression pool is saturated at the time of
- Given a small break, P(Ril) = 5.0E411 containment failure, in addition, closure of the SRVs prior to containment failure (as a result of
- Given a large break. P(Rii) = 1.0.
drywell pressurization) would cause the reactor vessel to reprc'surire, flowever, this has not been These values are reduced by a factor of two if no modeled in Reference 11-1: vessel injection is driving force exists for a puff release. 11-45
i vivv;ves,ven,
. II!!!WbEVasVII sw.;.ww.wws .
8 Ej g i Y I I fy D E E E _a _t _o _: g h h h a
& 5 E 5 !
===
- t
$f I
1, . . si: l e A og
!E B so 3 '
A
- e
- - m f*5 5 C
3 Ta g - J 23 i W k NE ! h, s . A sh!! t I cal E 3u _. .a a e j5.vg , a) a
~
s b a s- e E a
}E u
E 2 1 B-46 c
CLASS llB PLANT DAMAGE STATE The following is a description of the factors in- cal to Class llA above except the initiating event ciuded in the Class 11B Flant Damage State is a loss-of--coolant accident (LOCA). The SCS category (Figure B-8). is again assumed not to be available. All other CET events and associated failure probabilities PD. Class 118 phmt damage states are identi- are exactly the same as for Class llA. CLASS llF PLANT DAMAGE STATE The following is a ck a. :an of the factors in- ment, Table 4-2 in Reference B-3 indicates that ciuded in the Class llF Flant Damage State cate- drywell temperature will reach 800 F 20 hours gory (see Figure B-9). af ter vessel failure. Using the methodology of Class IA, PD. These would otherwise be Class llA or 11B sequences except that early venting is as- P(F2) = exp(-20 h/19 h) = 3.5E-01. sumed to take place through either the SCS or the existing vent lines. B S. Venting is assumed to prevat overpres-sure failure, so event BS need only te quantified VE. Early venting has a failure probability of for cases of late thermal failure. Ilowever, be-0.0 because it takes place by sequence definition. cause break size does not affect the end state re-lease mode assignment for these sequences, the R2. Reference B-3 assumes that core coolm, g explic t quantification of event BS is, for simplic-is lost at some time followmg ventmg and there-sty, omitted. fore assigns R2 a failure probability of 1.0 (Note that venting plus successful RFV injection results HU. Reference B-3 assigns Filtra hold up of in a recovered state and these sequences are ac- noble Fases a failure probability of 1.0E-02. counted for in the front-end calculat ons for core melt frequency.) SP. For sequences vented to Filtra, overpres-sure failure is assumed t.ot to occur. Thus, the HR. Reference B-3 assumes that low-pres- suppression pool is bypassed only for those se-sure injection systems will be unavailable follow- quences involving overtemperature failure, in ing ventingt consequently, failure of long-term which case the break is assumed to be in the dry-containment heat removal is also certain well head region and the suppression pool is al- [F(HR) = 1.0), ways bypassed. For sequences vented through the existing lines, the assumption is made that both OT. The vessel is not likely to fail at high wetwell and drywell vent lines will have to be pressure because the Reactor Coolant System opened, which leads tu bypass of the suppression (RCS) pressure is expected to be low for se- pool, quences initiated by either a transient or a LOCA. Therefore, Reference B-3 assigns event OT a RB, Failure of reactor-building retention is failure probability of 1.0E-02. quantified for cases of late thermal failure only, for the reason discassed under event BS above. F2. Reference B-3 assigns late containment For this case. . ilure probability of event RB failure a probability of 5.0E-01 following failure is, as before, t en to be the average of the values of event OT, but supplies no basis for this value, given in Reference B-3 for large and small For Class IIF sequences with high debris entrain- breaks,3.SE-01. B-47
. .1 O 98 fil
- k$ 5 8 I; E E E 2 8 5 E E E IIff9EfifIffII II'
!...s
! t !' t t.ew..w,..
ttttttEl ij ; ; ;
'I
] _$ _I
_g2 t _. 1 3 - 1 I
- .I. a
't I
- 5. .
N
,I a
2'1 ; C 36 t O at.b 5 11)! m it D a si r f O e) - a Ok g M E$ $
~
0 g.s .5
- s ..
u
- 64. a
- . 5 s"! j B y e E
iE y E c . E B-48
I. f i{ : il.il i. i. !
!r t!' i .
E C d DS JS WA F 1 nE t s s 4 S(2 D oC e e c c E 9 . t s s ' s , C N R, 0 .
- o eee*a.
E eE e r " r 0 o 9 4 s. t r a. m. r r8 0 o C s o : B04 -
"5 P5 O 9 o s
- t .
n e
- i R
It l
- Sn e -
re g 4t 8e r N -
,3 1 .
d . e p.l* e e P _ pe* u p S SPfO e rp tu 2 1d p 9 1l K Fe o H M N ,1 n i. e g t kn S ree aeC S Lr B n e e e r u t eint ta t 2 e-
- e. e F
- t. ntF o
Ct w _ s e r mS e T e-o .D vp 0 r Oeo o t c_ t l e o
.te t l o te nr -t f +
ee e I F C6 F e C c - _ t _ ly S er ee C sv 2 S eo c R 8 ht we i rR E N w I e e t fs F L fe ul P S Po R F I I s e r s rou utl a t tre eeF l t F G Ft eW m PPP a O 3 e h e r y e o l .i h r wS at C E ne S n V EetS oCe N V n
~ .
9 F B I I S 9 e r D eD eP P u s g 1 C N' i F CLe
CLASS Illa PLANT DAMAGE STATE The following is a description of the factors in- downcomer failure are the only contributors to cluded in the Class lil A Plant Damage State cate- early overpressure failure,lloweser, even a gory (see Figure B-10). hydrogen burn in conjunction with a pedestal downcomer failure is not calculated to lead to a PD. Class tilA plant damage states are initi- pressure that approaches the ultimate pressure ated by RPV failure. The ECCS is assumed to be capacity of the containment (see Reference B-3, insufficient to prevent core melt. Ilowever, ex- Table A-18); consequently, early overpressure vessel recovery is possible. containment failure is assumed not to occur IP(FI) = 0.0]. Rt. Because ECCS is insufficient to prevent core melt,in-vessel recovery has a failure proba- PF. Reference B-3 quantifies event PF at bility of 1.0. 1.0E-02. Based on the discussion for event F1 above, event PF need not te quantified because it Cl. The probability of containment isolation is only televant to sequences with early overpres-failure is given as 2.1E-03 in Reference B-3, sure containment failure. Figures A-8 and D-20. As befoie, no basis for R2. Reference b-3 quantifies ex-vessel re-this value could be found. covery in the same manner as Class IA. For this H1. The probability of a hydrogen bum dur- report, the assumption is made that core melt oc-ing core melt is given in Reference B-3 as cuts with the vessel depresari:cd. Thus, molten 1.0E-03. corium is not dispersed from the vessel at high pressure, but rather flows out of the vessel to the VF. The probability that the vessel does not pedestal area beneath the vessel, from which it is f.:il at high pressure is given as 9.9E-01 in Refer- directed to the suppression pool through the ped-ence B-3, However, because the vesselis already estal downcomers by the corium ring installed failed a' this point in the sequence, asking wheth- around the ped stal region. The debris is assumed er it fails again does not me.Le sense. Therefore, to be quenched in the suppression pool and event for this repon, the assumption is made that core R2 is taken to have a failure probability of 0.0. melt occurs with the vessel depressurited. 'I hus, the lower branch of event VF is assigned a proba. HR. Failure of long-term cc.ntainment heat bility of 1.0. removalis quantified in the seme manner as Class IA. DC. The probability of pedestal downcomer failure is taken to be 2.0E-01, cs before. VL. The failure pre >!!y of late contain-ment venting is taken to x 7.2E-03 for the SCS H2, The probability of a hydrogen bum after and 1.0E-01 for the exit.ung vent, as before. core melt is taken to be 1.0E-02. OT, Vessel failure at high pressure following Reference B-3 quantifies t "ent VP iden- core melt is not possible, so core debris will not VP. be dispersed into the drywell outside the pedestal tically to Class IA. region. This implies that debris-heating of the VE. The failure probability of early venting drywell atmosphere and intemals does not occur. through the SCS is taken to be the usual value of Therefore, event OT is not quantified for 7.2E-03. Class lilA. This is a departure from Refer-ence B-3, where event OT was assigned a failure F1. Because high pressure vessel failure probability of 1,0E-02 (the probability assigned following core melt is not possible with a pre- to vessel failure at high pressure following core existing rupture, hydrogen burns and pedestal melt). B-50
. _ - . - . - . . _ _ = . - . - - . _ - . _ . - . - _- _ _ _ _ - - . ._ _. ..
F2. No basis was found for the quantification temperature due to containment heating by core of late containment failure in Reference B-3. For debris dispersed outside the pedestal region. this report, the quantification for Class I A is used. Note that late thermal failure is not a possible end SP. Suppression pool bypass is quantified state, only for the case of late og erpressure failure. The quantification is the same as for Class IA. As for B S. Event BS is only quantified following pre us cases, undng brough We eMng Unes late overpressure failure, in which case the Refer. is assumed to result in suppression pool bypass. ence B-3 failure probability of 1.0E-01 is used. RB. Reactor building retention was not quan-tified for Class lilA in this report tweause it is HU. Event IIU is not quantified tecause it is only applicable to sequences with early overpres-only applicable to sequences with high drywell sure failure or late thermal failure, i l 1 B-51
i dli% *'.
'. .'.T. , 9. g"t,U *" O' ' "'" '.II III,I a?....
;,;g Y'2*'Yb
",I.'.,W'
, 4,,
. e;;i ". ,IvI.A."E
.Y.. '.'i "v.d",I er .a
]
eo ai _ c: wi v, oc _ _ ~. n . J~ l ri 1 m sunes _ A ger-et 7 rer-e1 SI ' - -e ' APERTURE _muna-- CARD a Also Available On Aperture Card Lanr.:st I Figure B-10. Shoreham Class Illa SCET with SCS. 11 - 5 2 L
w.,.. C* N Lgt. t.t. . M. St*UCNCE St ** **NCE i: f.f.. .H .zt.z: - n hSPt en N.N.
- uppis n.een.ni i .nmee. eL &
n ..vnn n .n.t . v.i. tins & cent. t .O..v .e resiwe.
-o. in er . .. rl.w.
are.i...e s 7 13.
..: e iste.
>F RR _ M VL Of FS l 88 HU SP__ M
- i
- f. 0 0 t .
must 1:20:!l
'.0g l1 l.40 g 1 cer.es __j D E-01 b!g )ge 3 uxeu naac c; 3:ht-ot:g
- 4. '.0 : 3
'g,ggg,g y*2*IEE~I' - ---
1.00!.0 a E Il3 -_ .~1 llI8l$$ $3 i
---* 6 f.0
;g i li. net-et r . MEC' 01.g! h 9 Li. Steri - 3 s3g-et r 9. C :.00 43 JDE *00 i:lUog luir-u . ..a - ' mr-" ,
M$b'EI- d+0 EE01 usEdi l: E:l8 (gg f og gg uenu usa g:gg:lg 'ji,
. ge-.+1g }3
.ZJ2et2 g g{fg- -
fg 0
. _ _ m .1t u. 48 luce OnfL unei
- ii . nor-e t
":40:o! gl
e nuan .og'H: g (3
- :e. g gi unen_ luis H - -n n "-", - ' :1n 8. c :
u,u, : !gg:gg g; uneen nar-r : 1gg:gg i, e--o i - :1.J::a
"" k e,-e,
.4.
u i 4.ae= > : 18l:28 in nr:n--- :3p:og 'ig
. 40 :. 0 11
[1. 0M-e2 .1
. 00il+ 0 ga u n n> naruo :381:.2 31
.go :+o6 a
- 48;88 lEl EAo
[un'=-tu.eu " _'*' m . ,-m . >. - =
- lgl-2.
lg$$c. j unengaesi m # -" :4.. :... c.
"a-e s rm-n . -
;gp:gg g; luces > nasc
- gg,:gle ej, var-u 1.,,-,,
- 281.:i :
1. in I^00
'3. ant-ci .g $.g. 0
;, ,,,.e,-- : ..
.g[i j
!!g.h
,.. , , 1 ~ -ei e, . < ,
une-" s m-n . jjlgl
- lt
- . 11
- g. l}[-
uom%,_., .g!qig: m m. i. h L_. - i m .. . , oxs. -l2 "l 8i l l est.en u.nen urar-- a::: m n i _ - -
- l:g:
.. j:$.0
- . 8 c.
8j w n-u , 1 --"
.go :.. e.
un-en a e n-e i 1. mm .,r :o g; T. DOE-01 u u =n= ne :'j o!g! *i. L g l gh
.00'!+ C .
20ene I-
,_oug, .00t.00 00 C-.
80egi :s . unr-u .! C ;i
...c.000 c4
.00g.. D3
[22Ren1 y at.an g 3
...t... =
=- '
, e r-=> ::al!! !.g l u x -e,...la. sot-os . .00t.00 ca
~
9104080287- 2 l _
l Lm 12sLAS I i 2SliDL lati-if I LaQEan
, ADLt" LtDLt2L 12.eu_.
AOLt
, JtLtt L12E21 t_nor-or J2Ltt t-,.
i_99247 LADERL IALtI 1Jaltto
, ADLtI anen SI APEltTURE l
CARD ! aeLv Also Available On Figure B-10. (continued). Aperture Card
__ nere - - -
- t. e.o,!,e,.,t8,.
i . w-m g at::' a 1 m-s
""--lu u - e i mm '
i; um, n m rr-- !:i!!!:i.! 8:1:
.g.gge.n.
ur-" ner=rrr :,,i:n 8'}. i R. Od' [3 m-e i 4, s;t .a f.{..ar+ hg
, u.m-n- mm.s 1:ni:g, *:
a,r-"
- 3 m-u n,,r-g;g;3,g 8me.. jg.
g; H ""-" T.uus 8:Hi:n er rum._ l':
.!:l.!.i:n..
i
. e.
niro
""*--tutt-e ;l'jj l's i Ku-.ul:lli:t 1
, . m -e
! '"-*Huse-n runn- ltil::: 8.
nr-u 1 lumen . _ nasc j!!!'f.:nl"i:t l:j.
-u--"
e i !!!!!!!! !j u.or.u n m. 8:itint
; m,., , E.
ym_ y _ ;gg;g g3
.. n
---lunstw, --a .. e....
... n nim.
!:n'*U. *i lua- , n "-" l.... ..
- a. m ... u --m **c:t* ,},
= c:
" gg.
m ,.y , ,s--m : .. ! !! o.r-e I unaf q..e..! ; lu.,-u ar-e u :.1:: . m m-u jj a an u t u .x...
- j;ggj+n j,.
o - -" t:.:.:,:... .
- o .e r o m 1-- . .
- gg:g
.. ... 7
, a. - -n r u r u .g:::
tn.nt-" im-u m,..., 8:.t.l,:... g.g :: $!:l H ' -a :, m-o ... n.i u m. . . ..!!.!! il t
. ':n..g..
o. ci n im-., l: 8,
! ""-" l o.,-m rnr:n-- ; l:l'te:n In: g:
9 W:ni:n 81 k.,
' . . m-e " ' " "
*.gn:' c.;
, m-u unz-n u --u 3:
i : o4 s;/1 s..,...
- 1 m-.c- g:!.8:.
n:8' ic'l " 14 H "--- :, . m-o n.m.
. . n.-o.t.
.... . ge,
,. i j;jjijjj gj , of l io. . a imo -. i , y. . . . ::.ti:: 1. -
- 11...r- i
- . . n.r-. t I...
g....... c., e. C
. m . 0.e...
e+00 On s i ::nl:: Si
"*=-" t o.,-n n --o I:niut n nn-u - g:ngu g!
li w o,_. , 4 - -.m num. 1:i:ltu.
....e.. gs o=
g;gg.g ,j i n .-. ,
;o,-o 8:Hi:n 8.
,_u g3
- n.i:::
... e... n
, j,,,,,.,, ,
m -m 8:!n ni El i 11-53 9yoyogoz g7-22 -
kne-e,_
.- g 4
l
,nen, i.eer.e*
uneen y cor-e n -
. m e.no i eeP-e, t.eer-ce St APERTURE CARD --
, ,e,-e, Also Available On utnce Aperture Card i.etE-02 2.feF=e3 Figure B-10. (continued).
B-54
_ . _ . . . . _ _ _ . _ _ , . . _ _ . _ . _ . _ .._-_ ._..._._._ . _- m______m._, . bant-c, f eet-ci [E Pr-n t
- C 1
- 1. 00E-e n EN -cI -
Qgo p0
;> m -ei a n t m ti.00 ggi:gl 0 [14 E 66E *04--
knor-e i - j g,.{ j;0%(+{0j,gl
- 4. Ot+be 7
-.m e o- >
3:0:::g 1.00t+0
-- -a.m.m, . m 8:m:::
1.00f*00 )3 all i
# 'E' 0.00t*04 )F ,
j [t__. ner -c,
- _ _ f EN] Fr.t u .l.00.+00 $
$-{[
I ja 1 l.00f*0 ' ;
's . r c,E -c i ii.Dofdef ---j t . hoe-e t i
).00t+0 '
= *=-IUEI i f:
. eM +oe i i
- - -7_ gag.e3 f . e M -o i r-~ O'.{0!.
- 1. c.
- r . zur-t 1 l h C.
i . ear-es - 23- *D' 4 l'88e.0 l.00E*00 2, .$0 11 nDE*DI l.000+01 35
> ')2 1.00'i+0 l.00'f*0 ) :(
--eg gen I
!E m -tu l.00!+0) 'll I f . e nE-e2 . -
f,-. ),00 je ;t dg*0 3 Edr -M ~ f* 1.40e 0 33
~
l d.00t-QQ. ,or-c, " ~
-. prix lNhbhl !
- 1.00f*04 c--. l.gote 0 M.
I . - - - - . 1.00t* 0 wn. m -cI 1.0 01'6 g II O * ' L- h J E E-EL :);lh hg ;f! m e.no
;13 dSE-ei m m-m 4:80i 1.
- ,t0 it:.
i, per-e, L1J2E* D i dt. m -01 1. Ol !
- C,
- 1. Of 0 C.
f.eet-et 1. 06 0 Ce D E- C 0 18GE+8
- 2. 0e.0 .C 1 =*-" :nmi- 8 'l
{nM-~ 1:8
- 1. b
!.:.g*.pn {
;, n ,..,
nm 0, +8.. 11
=-**
uore >
- a. m -u --uo0!':0 m ai:'4.8:
- ,m ,-m mm.m !!!E.yl 8 c 8:38s.00 5,l a.sx-st- <, m e, m m,.x - 8:8:l:0 !!
- 0. m .00 i
.00E*00 s. 4 m m-m ..0^0 i
0,t .*0.0 80 l
"
- E -"
' g' ,-ei n rrr- g:[gg!+gg ene.ee g lL AM-e t r x .u 8.00!*00
.. 0e.00 es J
( g gn li_Bof-of 0.00t+00 C, ' _ <r m-r t , u nt-e, s m-m - :00,.00 .
~ * " g0 w -- hg i _ oM +oe . , , _ 0 7
- f M.00n)a
-06 I t . ce-c2 % 62-0. )2 (7,,3g.e3 Lm u7 0.00t+00 33 i )
="ao bj5 l I,,e,-e, r s.0 u . m-n > - --
8:d0+00
- 0. 0,.00
- t !!
it.allE-o ' :E nnr-et g, v,6 C
...Og.00 .. )
fi *r-a* .. -0. n
- , --e ,
m m .=- };gg-0. igj jj e meae ', ,e 0.80e.00 es
-c,-- i m .m 0 t::m!:g Et w a -et -
a ne-o x-o ,..0. 00 e, 00
- s. m -0< .C, ,
' Lanae ' ' pf' -" r erm- 1: m 8; 'il g:m::: .,
, w= -e im m .m g 1.:.m00+n:
on-en n eM +en 4
. - - . Wofosozt7 -23
CLASS lilB PLANT DAMAGE STATE The following is a description of the factors in. F1. Early overpressure containment failure is cluded in the Class 1118 Plant Damage State cate- quantified in tha same manner as Class IA. gory (see Figure B-ll). PF. The probability that a driving force exists PD. Class 1118 plant damage states are initi-for a puff release is taken to oc 9.9E41, the same ated by a small or medium LOCA with a failure as Class IA. of high-pressure coolant injection (llPCI) and ADS. The RPV most likely remains pressurized. R2. Failure of ex-vessel recovery~ is quanti-Class tilB can be considered to be the LOCA ana- fled at 3.0E-03, as for Class IA. logue to Class IA. R1. Failure to depressurize the RPV domi- HR. Failure of long-term containment heat nates the failure to use low-pressure injection rem val ts quantified identical to Class IA. systems. Therefore, event R1 is quantified as for Class IA with a value of 2.lE-01. VL. The failure probability of late contain-ment venting is taken as 7.2E-03 for sequences Cl. Reference B-3 assigns containment iso. with SCS and 1.0E-01 for the case of venting lation failure a probability af 2.lE-03. through the existing lines. H1. The probability of a hydrogen bum dur- OT. Event OT is assigned a failure probability ing core melt is taken to be 1.0E-03, as before, of 9.9E-01, as for Class IA. VF. The vessel is considered to be likely to F2, Late containment failure is auantified as fail at high pressure; therefore, the Class I A fail- ' for Class IA. ute probability of 1.0E-02 is used. DC. BS. Break size is quantified in the same man-The probability of pedestal downcomer ner as Class IA. failure is taken to be 2.0E-01. H2, The probability of a hydrogen bum fol. HU. Event ilU is assigned a failure probabili-lowing vessel failure is tden to be 1.0E-02. ty f 9.9E-01, as for Class IA. VP. Event VP is quantified in the same man. SP. Suppression pool bypass is quant;^ied as ner as Class l A. for Class IA. VE. Early venting to Filtra is assigned the RB. Reactor building retention is quauified usual SCS failure probability of 7.2E43. as for Class IA. B-55
- N- ^ C1..e H as P.O 1 N. t. ly O' 11w na m .h2.n-....i,
.. 2.c u, .n o.a in en.
..i n.t.
m.y Do... u oc .,t o Ma..B.o
.- n e c.n.t . P,.P.
a - n. . H. i . n..m.n. -v
- e. n. . u.. i. v.cooi e.a . ....
PD Rt C* Hi YF 0: Ha vP VE Pi I i l ! 7 rar-c3 ___ i i l i i i l l
< ee..ei 1
I 1 s nes-es S1 l t' " ' APERTURE CARD , ,
=m i
4 l Als0 Avail 3ble O u ,1 Aperture Card tu er-ci i ,, l Figure B-11. Shoreham Class IllB SCET with SCS. B 56 wm
-v-u.
Puff Fo Ex- Cont. No Late Dvem- Late i. ease No Supp. No FC SEDUENCE t Reisees vse e.e l host he7,ev el venting temp. in Cont. Coet. Ho n g Jp Pool te Retention Pf408. SElkJENCE CLAe4 l hosevery Pellues i vie 5.5 DN Seesk by Filtes Bypesees Paste i PF R2 M4 l VL 07 F2 B5 HJ DP Re
- 7. bee-OL At h3gg. U E~05
- 7. 44E-04 42 "I'E3Eff e k',h
== mv w 1:iN 4 ! By IJ0E*C3 d 4.seE=0{ I :CC' N 41 2,kSE-0
- 1. aE -04 CH 19.9W % 'M2E.-01 :3. E c.r-c i 1.73E +05 C3 4:32F-96 j4
-- 4.17E-04 3 it .e cE-c 2 r-1.17E-Os 52 17.POE-c3 Nbyg-- 0.00E+00 33
- 8.48E-09 )4 I 9 . 62E -09 C6 eg 'g37:3 g i.67E-09 Ca dJOL-U 15 00F-0 9 f.65E-07 C7 L oog-03 pg .25907 4.7tE-Os C3 C
I 1.7st-11 D5 wr- 1 3 . 4'iE - 19 Cb 13 . 0 0 E-c i r Nh![ .3,4$*- e C6 2.RC 11.008-01 ,m, o g 3. eW- 3 C7 3.esE- 3 Ce U N_ g,3
- 1. 46E 3 l hf0 ti,DtE-0 2 ;7 ?oF-c3 i.48E-03 2
_._ m 3rg3c 0.00E+00 3
- 1. 06E -06 4
- - -J 4.40E-06 C1
.964-0e Cl la.ncF-ci I A*3 W.D I .64E-04 C2 IE.EcE-01 *
-04 C3 g3g-c3 NU EI e. .3 De. - -0 5 C4 I 2 . 23E -O f4 03 r
"..c .m- 4.StE-ow CS 7 20E-03 dJShQ g 3 4.Sif=09 Ce 2E-01 N "E-on 4.79E-10 C7 4.7eF=10 C6 AJ2E-C i ;3, m -c i 2.0ff-06 CS i 1 =08 C10
-n!-01
~
g,g3g_eg , W At 1':.i]n-1 I e--06 81 ei ha e m 5 r M'.OsE-05 ita; % $a l D3 u.eeE-c2 ;7 PSE-03
.Cof-07 02
- CUE +CC A .00E+00 D3
.-+ a.e6E-10 04 C
- 3. 26E-a g la noF-di i I e
LE EDE-01 me r.c y : -- 1. t Wace C3
.3.._c aE-c 3 6 C4 I
- i.2tE-09
.eSE-12 D3
~
e*
**+* C ' 3.19E-13 CD 2,JAE-C 3 bJJkgf ,_f~1.00E-01 3.192-13 C6 1
g- 3.646-14 C7 3, tL4E-14 Ce ULS0k ti_ .- !3. s o r-c i 1.52t-10 C9
- e. tee-fl C10 r-- 3.99E-09 81 it.co!-ci NE r-- 3.3EE-10 B2 1.itE-1C B3 e.ece-01 --
3.06E-09 B.
--fi .ctE-c i '-
.. m c, 2.2tt-10 es
=
2.22E-10 Be 3.95E-07 B7 g age.c1 _
.l E 00E-UI a. car-ci l', j j 8f'O y gg;r-gt , 3.98E-07 Bl0 MJC 1.64E-05 Bl2 D3 2
17.20E-03 NU 3 0 1.19E-09 0 D3 D4 1 4.94E-to C1 19.902-2 I'" ~ '
- 2 08 2 2E41 , gy.wr--- 1. 75t-06 C3 gg .c3 - " ' " -
9.41E-09 C4 g 2.StE-12 C3
?.20E-03 11J.GE:ql. 1' 1tE-01 '. ! f3 5 . 37E - 14 C
?
g c33.c, e (1. ROE.Ri 5.37E-14 Ca j ny e1 2.30E-10 Ce - 1 24E-O CtC
,_ " mnr:g1
- " " - 3 e2t-02 DL
~ 4 02E-03 02 adL L-C1 I f.2tE-06 Cl
'8 'UE~UI J.J9F-07 C1 1.E . 90 E - 01 I 5 . P 44 -05 C2 15 BOU.C i 'y*F r_ct 4.298-05 C3 2.31.5 05
^ -
C4
- 2. e85?+0 4 1
0'S hM Ch
' -'l 7 . 20E =o 3 m . --
2.10E-Os 8h D4 p- - e.708-09 Ct 3 ',.r.c.' 3. sef-09 Ci i j le.90E-01 1 3.34E-07 C1 1 IS.50E-01 r 3.0eE-07 C3
.i
-_ l
, lunt(
M 2E 03 J i
- 1. BSE -0 3 1
.2.DC E -01 LEEC2
,1 coEeo MD 1.ocE-cn l
l l u.usd l 2.cc
- 7. 2er-c s ua:ar__ l M.70E-01
.9. 90 E-0 t unta M Qf.:2 t . c oE-e n
. 0 CE *Q, r--
asa S: APER': URE - CARD
- s. ace-1 AIso Ava ilaoic On 1 Figure B-11. (continued).
.n
_ . _ . . . . . ~ . . _ . _ . . - . - . ~ . _ . .__._._.-~..m....._. _ ..m. .m _ . . ____ _ -. i . -,- 1- 3. nne-c ' I t.eet-07 C4 I 4.43E D3 7 sor-es gu 0.pt, g gg_ E-Gi ' .7.EDE-OL . h3 -
-of - - b= ::
'q.00 +00 r0k t Ii oM*00 iz.una-ar [14'jj g! g A*0d ~ 9.07E-10 94 In.non-of eg gyrgg--- g. 60 je ej 0 l/
I N '0I 1 man--m,w ac *.COE600L:S=3.; m -0
,n 150E-01 i t , mg1. a 1. - {9g*] l g Ni . c or-es 4. 65-04 JM
- 4. SE-06 32 iL20L-95 1 anr+na
---e 0. Ot +0C 33 i H : set-ce E-Os h"4 t
- n mor-a t 4.4 g, -Os -07 C.
C2 Mgg-g i . --fi~EM- 4. n- Cj e !l,%NGf
.aoE b3
'7.202-03 is non-ci i
- 11. ocE-0i n o.. . ;0.ae:a
- 8 gi-MDoi '. l'gll.
.9eE. 2 C8 g7 n_non-ci is nor-ci l$;j$f*$e
.es!-04
-====41. JDE-G K i ' f d', f"'kO g re mer:si k:$3I I" EA '
' " {5. 50E-01 3. unr-c i g '$ *,gf h)
.28E-Oe 03 N1.cor-as &-
o N -oS .8SE-08 D3
, : n . nnr +an .-
. E+00 DJ
- 3. tQ*aE a D4
; NIGE- C1 la . nor-c i 1
' . set-' Cl in --ei m e n_ :111h! ~
10 e4 i
- a. - -o, . :'=--" NO ! C 7.20E-os LLROL.At mnr-ci - ji:gg::i a
, 7g . e c!
ce n.a0E-01 a aw-o r - g', j;i"; a 0 0 . 0 01 '+00 1 - .oo ;._--
- . iq--n:: ,,-
uns ei - - 3.k ga4 u CE + ... - -c. ::;y:t ;
- 0. 0fH :+00 1 --a nco- y :; et 4 U SE*SI 3. 0 lR BBE-oi i1.une-a1 ;l[p -07jg
- 4.0.5-08 03 Mi nor-en ;7.aor-es 4.10E-OS DE
.u. mar +nn 0.00E+00 D3 90E-10 04
* ~
I
M '-G1 -kk hk' 5 -09 C3 not-o i A nna-ai 4* ** I' i 9l U$
i h' -15 b- . 13.nof-o f 11 r mn c.3 -
*e 7.2er-oS . LLOGE-Si
~
nne -a i 6 :7 '-- lg 1+a C . C7 4 2.0 ,
--A CS e noE-of
- E M-c i
(( O 3.0? :-Ca d t . cor-02 ;7. - -03 3.ca t-06
- 1. m ,oo .00E+00
.ntE-07 g34
.00E+00 Cl .g -
5 7 20f*03 I - 0.00E*00 Og
- n . one-a t i 11 non-og 1 O' S li Dof-of 0.00E+00 C7
's . nar-c i 0.00E+00 CS wi mu-a t .
'gggje
-04
- g
- i. I C1 la. not-o f , :n. w on 3. tee-09 Ct
.rs.noE-of n.tGE-07 C 7.
- n una-ai- 73E-07 C3 I'
g W -00 C4 03 - S N 7 sof=03 100C-08. 0.E-= g -
- 1_ oor.an
- O I - .
;m unw-a t ':,44g=00 E:: .
El
!sgos-of C2
- 3.sur-at g;h.E-i..
C4 -
, 1, . ,,E.0 < , . _o t u=:hl: t gg C.
B-57 h_ h
I, -..
- 7. Bas.a.
t . 6M ob. 4 a no, ei 7.20E=e3 2.GoE*es _, B 7AE41 i haE-03 l l. t.sor-e2 SI APERTURE- . :CisRD t i- Also Available On , ,,,,,, Aperture Card l; i. ann-e. /
- i. n-on l
1 l' V i p 2. Im-es ., _%p, [ .- i: [. - l, Figure B-11. - (continued). L
- l. n-58 l
u l
1 I ' lu9r.-n_- li = = -e i .r cr-c i g u nr-v nemc- 3:$;1:1:;it:q:, a, c
- 3. aer- ,0 SI
,a.eeE-ai n m r-- ;;;n :tt gg a =-e i s
- i . m-c i .s m .c ,
,2.22E- i:Hi:ltti
- 3. m -0. t 66
.7
[i.eer-of . 3.20E-09 08 dA'-' E.20E"C8 88 anE-ci e anE-ei .95E-08 g:g;+ : gl0,; a . m-O s _.n,en-r-1 44E-06 33 1.enE-e2 s.64E-04 >2
!y,p g-c3 0.00E+00 33
'_I G E+C3 -- : 1.$9E-10 34 i 4. 34E- Cg 1,., ,01 , nt E-c 1
_-.n ,smr-1:ial:e. it u a-v 4:ni:n et 3.00E-03 g,ggg.g3 g3 I ,, m g , 4.84E- 4 C6
"^~~~
l3.00E-ei 4.84E' 4 CS f'7 M-'03 I l_I : 0 0E *0 I 'E c E -c i glhf l (g W -01 :3.5;E-ci h80 f'.y**ff 3 66E-0S D I MN1 4.07E-06
.P2E-09 02 M u=u- !. . .e,-c i , n eE-ci .:181 4 : gl C IE.EDE-01 3.5.r_ci *O F
- 1.26E-07 3
't. w -a2 i . 26E-0 0
' y 2ce -e 3 0.00E+00
- n.orrtco g,g3g.12 04 3.79E-12 Cl I $ 69E- 2 Cl (q goE-ei r 53 EP 1 a.67E- 0 C2 is.BcE-01 3 yer.eu { j y
/ M9f".93 1 92E-t4 D3 7 I ,E*" 0.00E+00 CS ~ 13. 0 0E -e i i 42f*tB CS 7.2cE-e3 it esE-e t 7 ytg.cg .
. - -ei .- # s m -e , - g;; }g;;j ;; g=,
1 0.00E*00 8 ti .ccEtee _ ._ 7 _3 ;g.c 2 g 2g;t B3
- 3. 8E* 64 s.eef.et 1 s EeE-ei nrr=rr- ::gg:{0,; y IacEet 0.00E*00 87 fi ceE+ee 3 . g_ a 7.s3E-Ce as 4.17E-09 BS E~eeE-eg 3. 76E-C e 810
," W *Oi p 'r:5T**" ly 0}
4.10E-07 D3 M L*23 :1 caE+oo 1.34E-ki Cl I
"'" 1 5.81E- 2 Cl lo . noE-c i I 5.46E-* 0 C2 1.00E-e3 IM M -Oi :3. 5:E-c i j j g,g7g.,4 03 I 0.00E+00 C8 I'__*U 13.0CE-Oi I 3.42E*t4 C5 7 w -43 ti ecE-et 7 74, 6.72f%t e C7 C8 g.02t-tB
- -ei :m x-c i ,.m:t; ce,
- 1. 86E-0 6 hk
- 1 aoE-ot 3,00E-07 02 i .00E*00 --
0.00E+00 Cl 48E-08 k--> :, 2o,-e3 ,1 DUE*CD 0:::148 !3! t.00E-12 04 0.00E+00 Cl
.DQE+0e I 0.00E+00 DS I__
17.20f 03 I
.' U "C 1 0.00E+00 CS
- 11. e eE-e i 1
- 0. 00E +00 Ce ti.eeE-et 3 m g .c , 0.0wE+00 C7 0.00E+00 C8
_- 2.07E-Os D3 2.06E-10 02
^
l 7. 2w -o 3 .,_ mag 0.g
.00E+00 Cl
. nee
- no I 0.00E+00 D3 17.20E-03 I
* 'H 2 2E*C 1 0.00E+00 C8 13 ecE-ei i 0.00E*00 C6 11.D eE-c i n gy-ci g. +C
, 1.76E-06 Bt
, is eef-ef B2 g.gBE-06 0:0n;g a' 7_,37 c, i - +ee _
75E-04 87 IscE-et j I ^ # ~ N 2 0E -01 I 00E **0 '. -0 89
- 0. 00E +00 1 - -- - - -- -- - - -- -- - -_----- - --- - _W_O Vo8cz 87 - Z G
CLASS IllC PLANT DAMAG . STATE The following is a description of the factors in- PF. Event PF is assigned a failure probability cluded in the Class illC Plant Damage State cate- of 1.0E-02, as for Class ID. gory (see Figure B-12). R2. Reference B-3 assigns event R2 a failure PD. The Class lilC plant damage states are probability of 3.08-02, which is just the human medium or large LOCA sequences for which error probability associated with using alternate low-pressure injection systems are initially injection systems to establish a coolable debris unavailable. Class lilC is the LOCA analogue of bed in the drywell. Since Classes ID and IllC are Class ID. similar low pressure RPV failure sequences, the value of 9.0E-03 derived in this report for Class R1. Because low pressure systems are initial-ID will be used for Class IIIC also. ly unavailable, in-vessel recovery has a failure probability of 1.0. HR. Failure of long-term containment heat Cl. Failure of containment isolation is as-signed.a probability of 2.1E-03 in Refer- VL ence B-3. Failure to vent the containment in the long term is assigned a probability of 7.2E-03 for H1. The probability of a hydrogen bum dur- cases with the SCS and 1,0E-01 for venting ing core melt is taken to be 1.0E-03. through the existing lines, VF. OT. The probability of debris-heating of the Because the vessel is depressurized dur-drywell is taken to be 9.9E-01 as for Class IA. ing the initiating event and repressurization is not
. likely, the lower branch of event VF is assigned a probability of 9.9E-01, as for Class ID. F2. Late containment failure is quantified as for Class IA. The basis for this is given under the DC. The probability of pedestal downcomer discussion for Class ID.
failure is assigned the usual value of 2.0E-01. BS. Break size is quantified the same as H'2. . The probability of a hydrogen bum fol- Class IA. lowing vessel failure is taken to be 1.0E-02. HU. Reference B-3 uses a value of 9.0E-01. VP. Event VP'is quantified the same as For this report, the Class ID value of UF41 is Class lA. used. VE. Early containment venting to Filtra is as- SP. Suppression pool bypass is quantified the . signed the SCS failure probability of 7.2E-03. same as Class IA. F1. Early overpressure containment failure is RB. Reactor building retention'is quantified l quantified the same as Class IA. the same as Class IA. 1-l L B-59
u .. .. - _ e _ -_. _.. __ tri...cm , ..ai.n- 1.c n.eo.n n, u u in,rnci, m x , s rui p.-:. .i oc, r.n auu...n c:ne. P- m < . e.r cn . a, .oP u r:nu, c e. 4.ex i civ;Py F.i11urJ o.t Hig APV 60 p lg V;nt vi. PV >;;1h 30.h . Felluem BCS PO R1 CI Hi VF DC H2 YP VE et i I I lt.aua . I ! t I
' [SEE-o t ._
I w e-c, , i 1.00P-02 ltmem Si APERTURE use CARD Also Available On Aperture Card
-l l l Figure B-12. Shoreham Class filC SCET with SCS.
11 - 6 0 wna
4_ % ~.
- Puff No E,t- Cint. No Late Ovtra Lets LCeg3 No tur3 Me RB SF M MCE $2J'MCE i Melleet v e =31 Stolt Vt9 ting trup. in Cont. Cint. >tojouc Pasa is hsteMign P8406 , CLAS4
!- Meteveay Raeovet wie SCS DJ Feltwee Bestk by Ptitre Eyceses$ Peile PY' R2 661 VL 07 P2 ts 60 gu j no 0.00Ft00 g'b. .g 1
- --- - -~
l 8. 9 9E -0 4 Di {
- - - - ; 7.77E-05 02
- - - - ~ : 7. 0 SE-Ce Ct 1000-SA.. ~ l __
. ,;1-
**- ~- :3.14e-Ce C1 LSQE-C ' 3.1 tE -06 C2 ggg.-01 _
y,g g 2.5fg gg g j i .SeE-OS 03 O g;g-g t p -- --- 5.59F-08 D2 3mttca 'l3 8$.y M
-6.06E-10 Cl Wm~5y -
r - ---- I 2. 2 6E- 10 C1
;2.a4e-os C2 ts.921-My1".0M t . 1.aoe-On C3 MkC3 --+ j gar' )I!E~$! $$
l7.20r-c3 aJoe com-- 11:17t'i a g4g _ ,,p.g N - % 6. 53E- 4 C7 W S . 5 3E- 14 C8 MUI IGB
, ~
' O.57F-03 f h[ D3 0
t_i __0 0 2-0 2 6. 9 9E-0 6 8 j 1 5 N~0 T~M(a '~
- ~ ~ * . *
., 9.hhi Oy 36E =0 7 C$
C' 2.83F-07 Ct dd .2GE-QQ . S . ! CE -Q 1 .,'*** g
;2.30E-05 2.25E-06 Ca C3 u m_ _ _
I j;jjjj, gj rr
" wEgg
!6.22E-10 Cfl IJ0t92 (2JQCdl Iii. 0 0E. 11 f8.22P-10 C6 M~ U.g 6.StE-11 G7 a.sts-st Cs 1 90E-ci .MM1- 2.96E-07 C9
- 1. 60E -0 7 C10
. -06 7.08E-09 Cl IJED03 I
'E~ncE-ci 3.14E-09 C1
' .2C!'-0 tJ d -3.11E-07 C2 dJOE-d i 'Y"Y5m 2.50E-07 C3 1 C4 0 .15E-0 34E-077 )3 i t . ecE-c2 ,
8.17E-09 32 1 20E*03 _ , gsqrg_ ~ ~ --a 0.00E+00 '3 3 3.76E-il 34 r 4.70? Cl immu u"*"2 m -ci i: %9 il m-u - - u ac u ta LEU 2.3pE-t3 il 8 03 I3.ccE-01 M E-at [,y[lj he J.JQE*C 3 F 1
- 11. 0 0 E -01 rg,gg a i ~5.t4E-15 S.11E-15 C7 Ca l12E -M.- ;3 ECE-ci 2.19E-ti C9 t.lBE-it C10
, . BEE-40 81 1 5 3-2 ggggt 4,92E-10 B4
[1 Q9Ed1 % 3_ aa r.a r 4.07E-ti 58 1.07 East 66
. i.80E-08 e7 a.n3E-ot
^ ' ' UTC
- BE-0l 9
,3Jpg-g , a.00E-Og 010 U@E 3',!! !fj r 7. e2E-0 7 03
- 11. e cP -02 ;7 2c'-*3 7. 85E -09 02 Oce ca 0.00 0 00 03 6.89E-11 C4 I
7.13E-il Cl ME- ' 3.(7E-il CL l9 90E-0t. j5. !CE -G i 3. t 4E-09 C2 3 3ag.a t 2.82E-09 C3
,9 JQE-03 ' t3 .62E-13 36Ea OS D3 C4 7.2cE-03 l3.coE-oi j
- 11. c oE-01
!3 CCE*U 1 e'!h$ $
- 7. 7 SE-in C7
- 5.acr-at 7.75E-t$ Ce
- 9. 90E-01 3.3cg.a t 3.33E-il C9 1.790-11 C10 2 7n - 1:;n:n Bi S ocE-oS I
- 7M 07 Cl UL3W ' J'7eE Ce Cg li.20hti _L 7.70E-Os c2 t%5Che t ry gar g- 8. tee-0e C3 3.33E-Os C4 1
4.3eE-06 D3
! i.DCE-02 1.38E-07 D2
! LZJfg3 gg,c 3 0. 00E +00 1.00E-09 03 04 5
- 1. 2 SE -09 CL 19.90 1-2 bYM ' . e 5.SCE-01 . 4.44E-06 C3 l
~
Wo8o287-27
==w- g
- WJRL-E 7.20E-d3 l
i M8F-C3 l .h2LU91
,LDoe-c 2 Qg6L LD QE-21_._
u SE-2
- 2. oot-c; 7.2de-03 i _ S SE *02 a!,7ep.o g unra l ace-c2 ,_
n.coe.c: u se-ei
.ecesen u ni:nj SI APERTURE CARD -
1 Also Available On , una Aperture Card I Figure B-12. (continued). s w
f . . 3 TI"T e.3se-0e C4 i a.3rE-te os LW93 0.00Eo00 C5 ArcE-o3 -' M D M L 1 L..aE-t2 C6 L1.00Mi 7 Y3E-o t a.as'-te C7 _ g,0ne-t3 Ce 2 3QE:01 a seE-to
--CCEdi- 3'. isE-t 0 C9 Ct0 0.00E*00 et I.i ODE +00 ,gg{ j ll
.75E-t 84 0 ME'O E-3 C g'2. 2 6 0.00E*00 87 1umo-uacC::!n:i: 4.70E-Os Bij a.=E + u nw1_ *1:.;<g !!2 l
- . -05 : 03
* % E43 uno-- =
- 1. ue-0, f:Sil*
02 83 I 1.7EF-00 C1 T~M&T1 --j 7.7.ess-to 7eE-04 Ci [L.9QE-01 I C2 uosi n5r+ . C3 3 e . 2m 4E..O.c
- 3. 0 C4 s.90E-t3 03 i,3 1 . 00E-01 DWW 0.00E+00 CS 1
a.4st-te Ca L.12 6 2_ LL201-ci g r4s M 9'BSE-t4 G7 n.ncE-oi a.eeE-13 Cs DM I.$N!!' C!O g g.gi t.752-0 04 1.04E-08 02 i 1.7eE-09 Cl DGDi 7.86E-10 Ct kigf;21 'I~M F-0 I ' 7.78E-Oe C2
!l$s-8.002-08 03 d t e32-on 8.04E-10 02 M EE-03 WM6s 0.00k+00 03 4.38E-12 04 j 5.4st-12 CL
- 3. n aE-c i 2. 43E -12 C1
~ ' b gE-oi EFdl ~
2.4tE-10 C2 i ,$$1I$ h! 1 2{'.77E-403 C5
- 13. 00 E-o f i UUC0 0.00f+00 t.07E-14 C8
- 7. 20E -0 3 L1 20E*01 7'Y5E-ci 2 C7 gl97E-16 9tE-16 C8
,0.2QE .Qi N '
. - 40 0.00E+00 81 Il 00F *00 f 7. 'sm-c z 3jhN h net .e2E- B4
- s cE-ei u m g.gp-l0t gg 0.00L 'O 8'I
--M .0 W OO pg- .l , gl t.80E-08 at0
!S 5CE 4T*MF:ET*"" [t.$'.0l g lN set-07 03 d i.00E-02 _ ;7. 20E -c 3 1. 96E-09 02 n5Maa D.00E*00 03 42E-14 04 I _
- 78t!-t t C&
;s . ar-c i . 94E-t 2 C1 7.86E-10 C2 f.-11 'I~Y5ER I~~" !*,$h~k0 I
( 9.04E-14 03 13.00E-01 :1.2CEt03 0.00E+00 C5
- I 3,Jgg-g4 Cg
- 7. 2tE-0 3 11 0 0E*01 7*YCE-OI e*49E-18 C7 3 ggg-gg C3 9 McE-01 UI l 12 0 B.77E-01 03 M.i . 00E-en ;7.2cE 03 g.00E-03 02
'T,f 5v+ca 0.006600 03 7 itE-OS 04 I l - - . -
4 00E 60 Cl lf.20E-03 I 0. 00E MO 03 13.oSE-oi MW 0.00E*00 C5 1 0.00E+00 C4 11.SSE-01 ry*5Er-ci 0.00E600 C1 0.00E+00 CS M 1 0 0 F ~C E .- I .8 0 t I 7.0st-:3 C4 19.90E-of 1 i'"*' 3. tee-to Cl 3.itE-06 C2 ue I:z r i:11: 3 1 !! 9 1.00E-02 s . t a-0. m
;7 reE-o3 5.18E-10 D2 3,wg 0.00e+00 03 3.76E-12 04 I 4.70E-11 Ci fs scE-of P
- EC " I 2.10E-12 Cl M.McE-ot 2.07E-to C2 y 3 N-d i 1.06E-10 C3 e.9sE-tt C4 2.3ee-14 03 F I 13.00E-OL i ME ~ ' 4.60E-15 C8 l 4.80E-LS C8 y} .,
ME I I ' I II I I I 'I
l, .e, nna.-u__ Jt i . n oe 4s 9 i es,.ee 7.20E-03
/
E Olaf =O i i E.70E-dI tr_n L zuw SI APERTURG CARD I uso Available on ,, ,
. Aperture Card
. ..a,4 i
- 1. 0 0E-0 2
$ . iOf M3 L
Figure D-12. (continued). IM2
4. l726F-e3 *" t DeE41 bi I*I ami 2,mst--":l:::11 i', 9tE*it 9
,i_ocE-Oi
.,,gt
,.Sct- t 82 M _ ceE-O i *I i _ o cE-01._ my;g-= 1 07E-te et es
.07E-ta
.90E-00 67
, i _ c c E-01 .. .CEE-40 B5 t
.E "#"'
r-of
'- .06E-10 89 ggE.o i ..e0E-Os eig
, t . eeE-e i q ,w- ung-
. v.00E+oo at i
2.stE-10 hit 7.83E-08 g3
^0 I 17.2eE-o3 mg3 .
7.83E-ta ': t 3, 6E-12
.- d te ecE-ei_ 1 U E~UI 3. SE-10 CL C2 M C-0i nr yt--- t.$2E-1J C3 1:ni:1181 - we3 I S.98E-15 CS
\ La s E -' ' s. nee-is C8 1 2SE-03 ugq1.00E-01 n r.at j.y*,y ] _ e nef-e t .y g g _e g 3.33E-12 CS CLO
- .$ '.79E=tt
.75E-06 i g :i.CCE-Gi ; ,g4g.g7 Ci g
n.coE-oS I .l y.77E-10 ,e7E-ti Ci E'O '
~ 19. 30E -01 i l7.79E-C0 Ca IM ESE-oi ;gy g.c , 8. 2$2 -09 C3 3.36E-09 Cr4 6.00E-09 03 8.02E-tt Da (i eeE c2 i
,'7- 'UE-0 3
'T LEDea 0.g ,
5.47E-13 C1
- - - - - I -
' E'01 3.44E-13 Cl le.gof-of r 2.41E-11 C2 C3 is.EBE-ei -3,g ,.ci r . ,gg.932-il0 g; g:g
..ooE-e3 I
- '" E.ca 0.00E+00 Ca 13.00E41 i t.078-tS C6 7 2cE __ WQE-o i 7_g g ai 2.C U C 8.9 L *17 CS B.ncE-et 3 5,,-at R.k : 13 CS
- 1. 3 7E - 13 ClO
, 0.0CE+00 9:
>i 00E+00 2.,,E<z 3.9EE-11 B2
.01E-12 83 00 "O' ls. tee-01 . E*2 2:!h5!!
3.01E-19
!S 86 0.00E+00 87 H i.BSE+00 "y 3. S t E-0 9 88
= J is . oe E _e i 2.0ie-30 t.etE-Os es et0 ME-Ci pg. t i . 8 t E-0 9 3.87E-10 Sit Bit D3 3---- g.96E-08
_-0 3 - 0
- 1 CCE M-"
g,4ag.12 04 4.78E-12 CL
- l n.act-ai 7.94E-13 Cl te ecE-of i 7.87E-ti C2 Ma',:e t n f-3T- 6.3tE- C3 v.coe-e3 340E*68 C4 9.05E-t C3 I mg 0.00E+00 CS 7 208-03 Q M E-Qu.002-Si C :7_ g g .c ,
3.49E-15 0.70E-t7 Ca C7 2.9tE-18 C8 B toe-of a scr-o g 8.32E-13 CS 4.46E-13 C10 0.74E-04 Di
~ . e e, .c o gg;g g
. . . n -0. o3 it.00E-o2 8.99E*06 02 7.20E-03 Na 0.00E+00 C3
- 5.07E-10 04 0.00E+00 Cl 00E *00 - I 0.00E+00 C3
- 17. 90E-0 3 1 0.00E+00 C5
' ^ ' " '
L3. Gof-O f I O.00E+00 C6 LL.22E-o f :s . acr _c i 0.00E+00 C7 0.00E+00 Ce
- 9.76E-06 03 it.cof-of r 9.8tE*00 D2 i7.20E-c? 'T"MFTEg -
- 0.00E+00 03
- - - - - - - - - 7.itE-10 04 0.00E+00 Ct
. aof mo i 0.00E+00 D3 17 . 20M-o s 1
' "E 0.00E+00 CS L3.00E y1_,,J O 00E+00 Cs i1 cof-oi 3_aag.a1 0.00E*00 C7
- 0. 00E +00 Ce
, 0.40E-06 84 i L QE u ^ in -a t 0.964-08 R2 i_gyga 2.85E-06 h3 0.00E*00
- 0. 32E -4 4 87 y
-'8 00E*01 :4 m r-at *3 W O 40802 81 - 2.*
CLASS lilD PLANT DAMAGE STATE The following is a description of the factors in. Cl. Failure of containment isolation is not a cluded in the Class lilD Plant Damage State cate- significant concem for Class lilD because the se-gory (see Figure B-13), quence is either su.cessfully vented to the SCS or results in early overpressure containment failure. To remove it from consideration, esent Cl has PD. Class lilD plant dameje states are se-en ass ne a fanure pmba@y of OA quences initiated by a large 'dCA accompanied by a failure of vapor sur .ession. Reference B-3 F1. For sequences where early venting to the indicates that 80% of th Class lilD core melt fre-SCS fails, early overpressure failure is certain. If quency is due to seism.c RPV rupture. Ilowever,
, the SCS is actuated, the pressure rise is assumed because seismic RPV rupture is hmacd into Class to be mitigated, preventing overpressure failure.
V in Reference B-1 and treated by a separate CET in Reference B-3, the use of seismic RPV R1. If early overpressure failure occurs, in-rupture as the dominant contributor to Class lilD vessel recovery is assumed not to be possible be-by Reference B-3 in its quantification of the cause of equipment failures caused by the adverse Class lilD CET appears to be an inconsistency. reactor building environment. For sequences that For this report, only intemally initiated Class illD are success'ully vented to SCS, low-pressure sequences are considered. systems are conservatively assumed to be initial-ly unavailable, as was done for Class IllC ('arge VE. Because of the failure of vapor suppres. LCCA). Thus, for both cases, the failure proba-sion associated with the initiating event, the pri- bility of event R1 is tien to be 1.0. mary system blowdown results in a rapid pressure rise inside corntinment. Thus, for Class IIID, car. PF. The probability that a driving force exists
!v venting to 'he SCS is considered before in- for a puff release is taken to be 1.0E-02, as for vessel reco ' y; if early venting succc-ds, Class IllC.
overpressure .ailure can be prevented. On the other hand, should early venting fail, overpres. R2. Ex-vessel recovery is assigned a failure sure failure is taken to be certain. In the early, probability of 9.0E-03 as for Class lilC. Refer-time-phased phenomenological event tree for ence B-3 uses the Class IA value of 3.0E-03 but Class !llD (Figure A-ll), Reference B-3 assigns nojustification for this value could be found. early venting through the SCS a failure probabili-ty of 9.lE-01. However, in the Class IllD CET HR. Failure of long-term containment heat JFigure 4-16), a value of 1.0 is assigned ' ' early removal is quanti 6ed the same as Class IA. venting fai!ure. Per the discur:bn c .e fe r-ence B-3, the value of 1.0 is basea on 80% of the VL. Late venting is not evaluated because it is Class IllD sequences being seismically induced, n t relevant to Class IIID sequences. resulting in coincident drywell failure and hypass OT. Event OT is assigned a failure probability of the SCS. The basis for the value of ) .Iti-01 of1.0E-O~' used in the early time-phased tree ceald not be found. F2. Only overtemperatu e failure due to debris-heating is of concern in evuuating event This report only considers internally initiated F2. Assuming that overtemperature failure is pos-sequences, so the drywell is assumed to be intact sible only if debris is dispersed iruc the &ywell at this point and the failure probability of early outside tne pedestal ret, ion by failure of the vessel venting through the SCS is taken to be the usual at high pressure, the probability of late contain-value of 7.2E-03 for sequences with the SCS rnant failure becomes the probability of failure to available,1.0 otherwise. repair alternate injection systems that could l l B-63 l
l l provide cooling to the debris t:ed on the drywell SP. For cases of early overpressure failure, floor. Using the parameters for Class I A P(F2) = suppression pool bypass is quantified as for 5.5E4)l. Class IA, assuming that downcomer failure oc-curs based on the failure of vapor suppression, BS. Break size is only quantified for early which is most likely due to one or more vacuum overpressure failure, for which the rapid pressure breakers stuck in the open position on the suppres-rise is assumed to give e arge-break probability sion pool downcomers. Reference B-3 quantifies of 5.0E-01, as before. Reference B-3 used a val- th'.s event based on the assumption that the seismic ue of 1.0, apparently based on the dominance of initiating evt I causes coincident drywell failure seismically induced drywell failure. and therefore, suppression pool bypass. HU. For cases where overtemperature failure RB. Reactor building retention is quantified does not occur, Filtra hold up is quantified at as for Class I A, again assuming that downcomer 9.9E4)l, .s before. failure has occurred. \ B-M l l
pas In-- Puff P'* En- Cont. No Late Dvee- Late large No Supp, No Ats SF9fNCF I " 3ENCE Claes No Early Cent, OP Fallu teep- Cont. Cont. Heldup Peel is Fetentas PHOS.
- CL ass re After vessel Rolesse vessel eteet wenting 1113 PCS Cont. W Isolette Recovery hal via SC5 in Du f a t tere Greek er Bypasses a ont via n fD'W Bec Recovery Filtra SCS F e s tare och Fasle PF R2 m vt. Of F2 BS HLI ## IM P5 VE CI F1 Rt 0.00E600 A*
I
? *"2 c1 01 I
bJ29D01 - ~- 9 54lE-02 D2 i l'SR' - . ese-03 cs - g I
- om-* 7 c' tn9c-s1 _
NE~S 1-- 3.9eE-05 C2 LDQDSZ_ Lo0E +02_ 3.19045 C3
$39E,.8h 5 b .SQL91- t,72E-OS C4 0.00E*00 0.00E+00 a
0 00E+00 Bt 2.29t92 _ _ ._ l 3.47E-03 82 I La u.90E*90_ L,ggg e gg_ M:02 -4e 9tE-OS 83 1 60E-03 B4
.3 AQCA L.
l.7eE-03 e5
%. Sot 9L MISI- 1. 76E-C4 56
, ggg 0 DDE+00 ST I I l I2.on+onJ r _., ..
banent , .0e4. ,, tog _g, _ 1.82E-c5 910 109001 911 2 92E-05 1.DC91 dEt9L 3 see_ce etz Figure B-13. Shoreham Class IllD SCET with SCS.
CLASS IVA PLANT DAMAGE STATE l The following is a description of the factors in- Table A-19 of Reference B-3, which indicates cluded in the Class IVA Plant Damage State cate- that rapid containment pressuritation leads to a - gory (see Figure B-14). 50-50 split between large and small breaks. For this report Table A-19 is followed and event BS PD. Class IVA plant damage states are ATWS is assigned a failure probability of 5.0E-01. This sequences in which reactor power is greater than is consistent with the quantification for previous the capacity of the SCS or the SCS is not avail- classes involving early overpressure failure. able. Containment failure occurs prior to core melt. HU. Event HL is not applicable because Filtra has no impact on Class IVA sequences. VE. Es lv venting to the SCS is not successful in prevenan : '*1y overpressure failme because SP. Suppression pool bypass is quantified us-reactor powei is greater than the SCS capacity. ing Figure 4-4 o' Reference B-3, with the split Therefore, event VE is assigned a failure fractions modific j appropriately for Class IVA. probability of 1.0. The:1sults are as follows: F1. Because early overpressure failure oc-
- Given a small break, P(SP) = 2.8E-01.
curs by sequence definition, event F1 has a failure probability of 1.0.
- Given a lage break, P(SP) = 2.0E-01, PF. Because the RPV is likely to be at high RB. Recctor building retention is quamified pressure when vessel failure occurs, event PF is as for Class IA using the above results for sup-assigned a failure probability of 9.9E4)l, as for pression pool bypass. The results are as folicws:
Class IA. Break Size Puff Release P(RB) BS. The Reference 3 3 event trees used a failure probability of 9.0E-01 for event BS (i.e., Small No 9.0E-02 large breaks were assumed to occur 909r of the Small Yes 1RE-01 time given that early overpressure failure has oc- Large No 2.8E-01 curred). This appear- 'o be inconsistent with Large Yes 5.5E-01. 1 l l l B-66
%se h
v5
$2 :: : : : s . a s i s I y i Y i k Y k k f h
a t o b B o b W W
=
N u N o 3
=
S E. % m N
%. I' %. D I
- 4 h - N - M b 1 I J T
, a 5 e y ,
3 :b d M. N. M 1,
' 3 n n 4 s d
. ! l l ! l
. - a a -
J
. i . . d.
a . i a 1
. 4 a j
i l 9' EI l
- J 5 -
2- . Iilb g d N A A 5 b
- O a 3:
t a al.A . W
, a
- s I
h m
- a 2 tj @
w
!um
- ,- o e p
A W i:3 a
!N : ! O LA $. ] fe
- o a
o
.c:
O ,g,. wU I y Z u I =t j $ . - e 9= 0
.y e v a %
- a.
e u u J ^ p CD
~.
LL B-67
CLASS IVF PLANT DAMAGE STATE The following is a desr W'the factors in- vent actuation. Therefore, given that low-cluded in the Class IVF1 " - ge State cate- pressure injection has failed, the percentage of gory (see Figure B-15). these failures that are due to the reactor being at high pressure is determined. For case of refer. PD. Class IVF plant damage states are ATWS ence, the necessary data have been organized in sequences with reactor power within the capacity Table B-2. of the SCS, Reference B-3 assumes thet the reac-tor is depressurized for all Class IVF sequences. The next step in the calculation is to average As discussed under event F1 below, this assump- the individual contributions to the overall Class tion is questionable. At the time of vent actuation, IVF frequency, weighted with the percentage of low-pressure systems injecting to the vessel are each contribution that is due to the reactor being assumed to fail (this is part of the definition of the at high pressure. This gives the fraction of all Class IVF plant damage state). Thus, steam flow Class IVF sequences in which the reactor is not to the Filtra gravel bed is assumeo not to be main- depressurized, as follows: tained long enough for the Fravel bed to become saturated. Class IVF is oniy defined for sequences Pinot depressurized) = [(4.32E-08)(0.0) with the SCS evailable. + (l.09E-07)(5.2E-01)
+ (l .68E-07)(5.2E-01)
VE. Filtra is actuated by sequence definition, + (4.99E-09)(6.8E-01) consequently, event VE has a failure probability + (6.74E-09)(6.8E-01) of 0.0. + (2.28E-08)(5.2E-0I)
+ (2.49E-09)(5.2E-01)
F1. As mentioned above, the Reference B-3 + (1.30E-09)(5.2E-01)] assumption that the RPV is depressurized for t.ll /3.6E-07 Cle.ss IVF sequences is quesdonable, as shown by the following analysis Using the A"WS accident = 4.7E-01, sequence transfer trees in Reference B-1 (Fig-ure 3.4--20), and the fault tree quantification of Thus, the reactor is likely to be at high pressure in low-pressure injection failure (Reference B-1, approximately 50% of the Class IVF sequences. Figure C.2-5), the percentage of Class IVF se. However, because the containment is vented to quences in which the RPV remains pressurized Filtra at the time of vessel failure, the assumption can be calculated. Class IVF is defined by the is made that, even if the vessel fails at high failure of low-pressure injection at the time of pressure, the SCS will mitigate any overpressure Table B-2. Percentage of event F1 failures ATWS Transfer State P(X) P(XV) P(X/XV) TVF Frequency Tl 0.0 2.0E-01 0.0 4.32E-08 T2 1.lE-01 2.lE-01 5.2E-01 1.09E-07 T3 1,1E-01 2.1 E-01 5.2E-01 1.68E-07 T4 2.IE-01 3.1E-01 6.8E-01 4.99E4) T5 2.1 E-01 3.1E-01 6.8E-01 6.74E-09 T6 1. l E--01 2. l E-01 5.2E-01 2.28E-08 T7 1.1E-01 2.IE-01 5.2E-01 2.49E43 T8 1.lE-01 2.lE-01 5.2E-01 1.?E-09 B-68
challenge to the containment. Therefore, the approximately 5091 of all Class IVF sequences, probability of early containment failure is taken event OT is assigned a value of 5.0E-01. to be negligibk. F2, Reference B-3 assigns a probability of PF. Event PF is not evaluated because early 1.0E-02 to late containment failure due to high overpressure failure is assumed not to occur- drywell temperature. No basis could be found for this value. Table 4-2 in Reference B-3 indicates R2. Reference B-3 assumes a failure proba- that drywell temperature will reach 800 F 22 bility of 1.0 for ex-vessel recovery, based on the hours after vessel failure for a Class IVF se-assumption that low-pressure systems fail at the quence with high debris entrainment and CRD time of SCS actuation. This assumption is overly debris cooling following vessel failure. Use of the conservative. Only low pressure systems with methodology of Class IA to calculate the proba-suction from the suppression pool would be vul- bility of providing some cooling to the ex-vessel nerable to failure caused by containment depres- debris bed results in P(F2) = exp (-22 h/19 h) = surization at the time of venting, so that the 3. l E-01. operator could still align attemate injection from a system such as RBSW, which is indepen lent of BS. Break size does not affect the release the suppression pool. Therefore, for this report, mode binning for cases of late thermal failure; event R2 is assigned a failure probability of consequently, event BS is not quantified. 3.0E-02, the assumed human error probability for failure to align attemate injection systems. HU. The probability that Filtra hold up of HR. Reference B-3 assigns a probability of n ble gases is not maintained is taken to be 1.0 to failure of long-term containment heat re. 9.9E-01, as for previous classes. Reference B-3 uses a value of 9.0E-01, but no basis could he moval This seems inconsistent with the quantifi-cation of event HR performed for previous f und. classes, where the failure probability was based on operator error to align citemate means of heat SP. Suppression pool bypass is not evaluated removal, not on the availability of systems that for Class IVF because the end states are either take suction from the suppression pool. For this vented to Filtra or involve overtemperature fail-report, event HR is quantified as for Class I A. ute, for which suppression pool bypass is always assumed to occur. OT. Reference B-3 uses a value of 1.0E-02 for event OT. This is consistent with the earlier RB. For cases of late thermal failure, event assumption that the RPV was depressurized for RB is quantified as for previous classes by aver-all Class IVF sequences. However, because the aging the values obtained for small and large vessel was de, ermined to be at high pressure in breaks, giving the usua' value of 3.5E-01. B-69 I l ___ _ _ _ _ _ _ _ m
r p SEQUENCE tete 7_ er ge No F12tre Supp. No AS SEQUENCE Puff Ex-vessel Cor t . Over- t1olaup Pool retection P5C9. CLASS Class TYF No Early OP Failura percovery Hrat teep. in Centaineen Break in PCS Cont. Peter to . Release t failure Cont. g Bfpasses Vent via RPV Failur Homoval Det SCS e Fonte BS HLJ SP fab VE F1 PF R2 m GT F2 , PO
. _ _ . .______ _ e yy_og og 11.goE-cf - - - --- 9.70E-02 C2
- s.sce-o2 ct
- 1. 0 3E-0 4 C1 LS2g-92 3 QM* -- f102E-C2 C2 gg_g g " C3 ggq 3' J .1DE-91_. _ _ _ __ ._ _ .. _g ME-El- 1. e 35-9 3 C4
@ a ccE+00 5
4 J CLt99 C -. Figure B-15. Shoreham Class IVF SCET with SCS.
CLASS IVG PLANT DAMAGE STATE The following is a description of the factors in. R2. Reference B-3 assigns event R2 a failure cluded in the Class IVO Plant Damage State cate- probability of 1.0, apparently based on the as-gory (see Figure B-16). sumption that makeup to the suppression pool eventually fails, leading to the eventualloss of LPCI and core spray when suppression pool level PD. Class IVO plant damage states are simi- , falls below the suction lines (due to steaming to lar to Class IVF with the exception that injection the SCS). However, according to the ATWS event is assumed to continue after venting, leading to twes m Reference B-1, suppression pool makeup saturation of the Filtra gravel bed and a release of is available for approximately 16.5% of all noble gases to the environment when the SCS is Class IVG sequences. For this report, because depressurized. Therefore, in contrast to ""#I r vessel failure ecurs at low pressure, de-Class IVF, the assumption that the RPV is depres-bns.is assumed to flow out of the vessel and be surized appears valid for Class IVG. Again, directed into the suppression pool by the corium Class IVG is only define ( for sequences with the ring and the pedestal downcomers. There should SCS available. be no need to establish a coolable debris bed on the drywell floor. Therefore, for this report, event VE. As for Class IVF, early venting to the R2 is assigned a failure probability of 0.0 to re-SCS is assumed to occur, giving event VE a fail- flect the assumption that core debris is quenched ure probability of 0.0. in the suppression pool. H R. For the reasons discussed above under F1. Early overpressure failure is assumed not Class IVf, event HR is assigned the Class IA to occur, because the containment is vented to the value f 1.0E-01. Reference B-3 again uses a SCS and the reactor vessel does not fail at high value f 1.0. pressure, OT. Event OT is not evaluated because the PF. As above, event PF is not evaluated if vessel does not fail at high pressure. As a result, early overpressure failure does noi occur. subsequent events are not quantified. B-71
a ,, kk EWW D G O, 4 - Si ** O y i i I ! g
'i1:- r I i
6 I i D t 4
! E !
l:. -
- r I
- 2 2 i
- s se t
. 5I Di'E s au I e
- : e 3:
- i l j h3 5
.e.
I l I !!I5E h ! ,
-l -,
_a cm 1 E , .. a - t t= $ L- 8 l g U i : M
) t: g o E5 U i s.
h.e C
- 64. IC
,. g
- u. 8 b
& >i hl ui r
. J a c .1 m B-72
REFERENCES B-1. E. T. Burns et al., Shoreham Nuclear Power Station Full Power rRA, PRA Update: Supplemen-tal Containment System Implementation, IT/Delian Corporation, prepared for the Long Island Lighting Company, February 1988. B-2. Science Applications, Inc., Probabilistic Risk Assessment, Shoreham Nuclear Power Station, prepared for the Long Island Lighting Company, June 24,1983. B-3. 2. T. Mendota ct al., Containment and Phenomenolaical Event Tree Evaluation At Full Power for the Shoreham Nuclear Power Station, Science Application International Corporation, pre-pared for the Long Island Lighting Company, February 1988. B-4. Fauske and Associates. Inc., ofAAP Analysis to support Shoreham 100'/c Power PRA, prepared for the Long Island Lighting Company, March 1988. , B-73
l APPENDIX C REACTOR BUILDING ENVIRONMENTS C-1
CONTENTS ACR)NYMS.............................................................. C-4 DISCl.'SSION OF REACTOR BUILDING E' !!RONMENTS COLLOWING CONTAINMENT FAILURE AND VENTING , . . . . . . . , . . . . . C-5 A N A LY! !S R ES U LTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C-5 Secondary Containment Hydrogen Burns . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C-5 Reactor Building Standby Ventilation System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C-6 Venting the Primary Containment . . . . . . . . . . . . . .......... ................. . C-6 COMPARISON OF RESULTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C-6 R EFER E N C ES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C-9 FIGURE C- l . LaSalle reactor building temperatures following a 4-in. drywell break . . . . . . . . . . C-8 C-3
ACRONYMS ATWS anticipated transient without scram ORNL Oak Ridge National Laboratory DF decontamination factor . . PRA probabilistic nsk assessment ECCS Emergency Core Cooling System PilSYS Reactor Iluilding Standby LOCA loss-of-coolant accident Ventilation system MAAP Modular Accident Analysis Program Si10 station blackout , MSIV main steam isolation valve SNL Sandia National Laboratory NRC U. S. h tear Regulatory CommMion SORV stuct-coen relief valve s C-4
-]
APPENDIX C REACTOR BUILDING ENVIRONMENTS DISCUSSION OF REACTOR BUILDING ENVIRONMENTS FOLLOWING CONTAINMENT FAILURE AND VENTING This appendix discusses the results of the anal- tions existing in the reactor building after venting yses performed by Sherrell Ps. Greene at or containment failure. Oak Ridge National Laboratories (ORNL) on the Reference C-1 was performed by ORNL at the effects of venting and containment failure on the request of the U.S. Nuclear Regulatory Commis-Shoreham reactor building,C-8 Of particular in-sion (NRC) as an evaluatu of Lilco's submittal terest are the environmental conditions produced for a 25% power license for Shoreham. The re-in the reactor building by venting through the ex- port focuses on Lilco's claims conceming the se-isting lines. In addition to the Shoreham probabil- vere accident mitigation capability of the istic risk assessmentC -2 (PRA) and the ORNL Shoreham secondary containment (tractor build-work, the (Draft) NUREG-ll50 and PRUFP pro- ing),in particular its ability la retain fission prod-grams at Sandia National Laboratory (SNL) were ucts released from the primary containment reviewed for information on the possible condi- during severe accidents. ANALYSIS RESULTS ORNL analyzed the following three accident SOCOndary Containment sequemes: Hydrogen Burns
- A transient-initiated anticipated tran. ORNL concluded that the failure of the Shoreham p sient without scram ( ATWS) with Modular Accident Analysis Program (MAAP) main steam isolation valve (MSIV) analysis to predict hydrogen burns in the reactor closure at time t = 0 (initiated from building was due to the MAAP code's underesti-25% power) mation of hydrogen generation rates during se-vere accidents. This would lead to
- A seismically-initiated, recirculation " * "'*" .w es mates ome reactm buMng ec ntaminanon factm m Calculanons per-line-break, lost-of-coolant accident ,
f nn y m ate at up to M oW (LOCA) with a 3-ft 2 coincident Shoreham in-vessel zirealoy inventory could be drywell-head failure oxidized under some circumstances, generating approximately 2l00 lbm of hydrogen, T'.;s 2
- A station blackout (SBO) sequence would be enough hydrogen to produce four , lob-with a stuck-open relief ^ valve el dellagrations in the reactor building excr ding (SORV). The major findings from the the refueling floor, or two global deflagratio ,s in.
ORNL analyses of these three volving the entire reactor building. This n'ees sequences are listed below. that hydrogen explosions in the reactor buil&ng C-5
might be more likely than was predicted by the Venting the Primary Shoreham PRA using the MAAP code. Conse-quently, ORNL concluded that the probability of Containment tractor building bypass was likewise higher. Vcuting the primary containment to the reactor building would result in immediate failure of the Reactor Building Standby refueling Door siding. Because the RBSVS ex-Ventilation System haust now rate is so low (only 1160 scfm), the ability of the Rh SVS to maintain a negative pres-Operation of the Reactor Building Standby Venti- sure ius& the reector building relative to exter-lation System (RBSYS) may increase the severi- nal esmospheric ptruure is extremely limited. ty of hydrogen burns. Because RBSYS operation Venting of the primary containment through the promotes a well-mixed atmosphere in the reactor existing vent lines would inject at least 3000 cfm building ORNL concluded that severe global hy - of steam and noncondensible gases into the reac-drogen bums would be more likely than localized tor building upon failure of the ventilation duct-deflagrations Also, for cases in which the prima. ing. Although Shoreham's reactor building does ry containment failed into the lower portion of the not have blowout panels, the refueling Door sid-reactor building, this tendency of the RBSYS to ing is expected to deform and begin to leak at a circulate the atmosphere from the lower eleva- pressure differential of approximately 03 psi tions up to the refueling floor would tend to re- (Lilco estimate). Therefore, venting to _the m. 'or duce the DF of the reactor building by building would be expected te result ir rad transporting fission products to the refueling overpressurization and failure of the refuating floor, where a failure of the reactor building floor siding. This effect is mitigated somm twt by would be most likely to occur, the resulting long and elevated release p .hway. COMPARISON OF RESULTS The ORNL results are supported by work per- case of a 4-in. diameter drywell break was ex.- formed at SNL as part of the (Draft) amined to determine a lower bound. NUREG-l'5u and PRUEP programs that are analyzing the La Salle Unit-2 Nuclear Power Because there are significant differences in the Plant.C-3 C-4 The SNL analysis concentrated on design of the La Salle reactor buildings from that estimating the temperature and humidity condi, at Shoreham, the results are not entirely applica-tions produced in the reactor building compared ble, llowever, because of these design differ-to the environmental qualifications of the equip- ences, the effects seem to be less severe for ment therein. La Salle than for a similar scenario at Shoreham Thus, they do indeed constitute a lower bound. La Salle is a two-unit site with the two reactor SNL is currently performing a Level 3 PRA on buildings connected by a common refueling the La Salle Unit 2 plant (a 1078 MWe BWR/5 Goor, which in tum is isolated from each reactor with a Mark 11 containment). As part of this work, building. Shoreham is a single unit site with the calculations were performed to estimate the reac- refueling Door largely open to the rest af the reac-tot building environment (temperature and pres- tot building. At La Salle, there are two floors be-f sure) following containment failure. Modeling low the ground level in each reactor building. The was performed using the MELCOR computer steam tunnels exit the reactor buildings from the code with the Unit 2 reactor building divided into upper of these two underground levels.The lower 17 control volumes. Calculations were made for of the two levels contains the Emergency Core several case: of containment failure, as well as Cooling System (ECCS) pumps, with the pumps venting through the existing 18-in. s ent lines. themselves located in comer rooms that are iso-Shoreham uses 6-in. vent lines; consequently, the lated from the rest of the reactor building. At C-6
Shoreham, the steam tunnel exits the reactor relatively low average temperature; however, building at the 78-ft elevation (38 ft above four of these volumes represent compartments ground) and there is only one level below ground, that are largely isolated from the remainder of the where all ECCS pumps are located. In contrast to reactor building. The other two volumes repre-La Salle, the Shoreham ECCS pumps are not iso- sent the basement raceways surrounding the pri-lated from the rest of the reactor building. Anoth- mary containment. These areas are protected er difference of note is that La Salle employs a from the steam blowdown environment because Standby Gas Treatment System which, as noted of the location of the steam tunnel leading to the in Reference C-1, provides greater filtration of turbine building. Reference C-3 indicates the the reactor building atmosphere and less mixing large quantitics of steam will not be drawn down than the RBSYS employed at Shoreham. Finally, into the basement elevations unless the reactor the walls of the La Salle refueling floor are calcu- building intemal pressure is high enough to open lated to fail at approximately 2 psig versus 0.5 a blowout panel on top of the steam tunnel, in the psig at Shoreham, event of the 4-inch drywell break, the Sandia cal-culations found that the steam tunnel blowout Figure C-1 shows the peak and average tem- panel was not opened. These results can be ex-peratures calculated to occur in each of the tended to Shoreham to predict that, given a severe 17 control volumes following a 4-in. drywell accident and containment venting through the ex-break. The average temperature over the reactor isting 6-in. lines, the environment in the reactor building is 170 F, w hich exceeds the 149'F long- building and, in particular, around the ECCS term qualification tempenture for equipment 10- pumps, would be more severe than the qualifica-cated there. Six of the 17 control volumes h.we a tion environment of the equipment. L' C-7
i i i i i i i i i i i i i i i i 300- -
@ Peak temperature 7 280 -
M Average temperature < c 240 -
\ :
{ 220 - ' ? ' p 200 - f 3 180 - ' -
'* ~
~
i o . , f
~
,20 1
~
i lll, hh 3513b2 3b3 3b4 3b5 306 3b7 3b8 309 3k0 3113k3 3 Reactor building model volume 1 ! 321324 3b5 331 Figure C-1. LaSalle reactor building temperatures following a 4-in. drywell break . s a C-8 i
REFERENCES C-1. S. R. Greene, An Assessment of the Shoreham Nuclear Power Station's Secondary Containment Severe Accident Mitigation Capability, June 26,1987. C-2. Science Applications, Inc., Probabilistic Risk Assessment, Shoreham Nuclear Power Station, prepared for the Long Island Lighting Company, June 24,1983. C-3. S. E. Dingman and A. C. Payne, Jr., BWR Reactor Building Environments After Containment Failure, prepared by Sandia National Laboratories for the U. S. Nuclear Reguiu mmis-sion, December 1988. C--4. T. A. Wheeler et al., Analysis of Core Damage Frequency: byert Judgement Elicitation on in-ternal Events Issues, NUREG/CR-4550, Volume 2, Parts 1 and 2, December 1988. C-9
,- - , - , , - - - - - - - - - - - - - - - - - ---,.--,---,------------------------------------w- - - - - - - - - - - - - - , - - - - - - - , - , - . - . - . , - . , - . , - - - , , . , - . - - - - ,-
W APPENDIX D BACKGROUND AND STRUCTURE OF THE SHOREHAM PRA o D-1
l CONTENTS ACRONYMS.............................................................. D-4 REFER EN C ES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . D-10 TABLES D-1. Qualitative attributes of Shoreham release modes . . . . . . . . . . . . . . . . . . . . . . . . . . . . D-7 D-2. Summary of the twelve release categories for Shoreham . . . . . . . . . . . . . . . . .... D-8 D-3
l APPENDIX D ACRONYMS ADS automatic depressurization system RPV reactor pressure vessel NfWS anticipated transient without scram SAIC Science Applications international FAI Fauske and Associates,Inc. LOCA SBO station blackwt loss-of-coolant accident PL&G Pickard, Lowe and Garrick, Inc. SCS Supplemental Containment System PRA probabilistic risk assessment SLC standby liquid control f kN s
1 1 APPENDIX D BACKGROUND AND STRUCTURE OF THE SHOREHAM PRA Because this report is primarily based on the To a large extent, the 1988 PRA update drew probabilistic risk assessment (PRA) of the on the system-level fault trees of the original Shoreham Nuclear Power Station, and because 1983 PRA. New functional fault trees were de-most readers are not expected to be familiar with veloped for ATWS and SBO sequences to reflect the structure .md terminology of this PRA, this changes to hardware and procedures le.g., the use appendix has been compiled to provide the neces- of highly enriched boron in the standby liquid sary background information. control (SLC) system, the installation of an Auto-matic Depressuritation System (ADS) inhibit The original PRA for Shoreham was com- switch to be used during ATWS sequences, and pleted in 1983 by Science Applications Intema- the installation of three Colt emergency diesel tional Corporation (S AIC).N It was a full generators in addition to the existing TDI diesels Level-3 PRA but it did not include external for emergency ac power). System level and fune-events or an analysis of containment venting, tional fault trees were also developed for the Fil-tra system and the existing containment vent in 1988, a 100% power PRA update was com- system. pleted for Shoreham.42 Several contractors par-ticipated in this effort. The front-end accident The overall structure ud terminology of the sequence analysis was performed by IT/ 1988 update mirrors that of the 1983 PRA. Acci-Delian.42 the containment response analysis by dent sequence event trees are constructed for each SAIC,N and the consequence analysis by Pick- postulated initiating es ent to determine w hich ac-ard, Lowe and Garrick, Inc. (PL&G)? In addi- cident sequences are the dominant contributors to tion, Fauske and Associates, Inc. (FAI) the total core melt frequency. The event tree performed ;he thermal hydraulic MAAP calcula. branches are quantified using the system-level tions necessary for an;lyzing the containment re- and functional fault trees. The definitions of the sponse to a postulated core melt and for accident sequence classes and subclasses from calculating the accident sequence source the 1983 PRA were maintained in the update. terms.N Extemal events were also included in However, accident Classes 11 and IV w ere further this update to the PRA. subdivided to reflect the operation of the Filtra system. The new subclasses were designated as There were several reasons why this update ef- IIF, IVF, and IVG. fort was undertaken. Foremost among these was the proposal by Lilco to install a Supplemental In general, the accident sequence event trees in Containment System (SCS) at Shoreham using the 1988 update are more complex than those in the Swedish Filtra filtered containment vent sys- the 1983 PRA. This is especially true for the tem. The main purpose of the 1988_ PR.A was to SBO, ATWS, and Class 11 sequences. The SBO analyze the effects of Filtra on the frequency and sequences are developed on time-phased event consequences of postulated severe accident se- trees that reflect the time-varying nature of the quences. Another reason for perfonning the PRA probability of recovering ac power. The ATWS update was to incorporate modincations that had sequences use transfer trees to model the opera-been made to the hardware and procedures at tion of Filtra and its effects on plant systems. Shoreham since the 1983 PRA that were expected Eight different transfer trees are used, each repre-to reduce the core melt frequency of severe acci- senting a different set of initial conditions (core dents, most notably anticipated transient without power, vessel pressure, boron injection, etc.). For scram ( ATWS) and station blackout (SBO) the Class 11 sequences, Filtra transfer trees are sequences. again used, only here each transfer tree represents D-5
a different initiator type [ transient, loss-of- containment analysis. Additionally, the S AIC coolant accident (LOCA), SBO, etc.]. assumption that venting (through the existing lines) would fait equipment located in the reactor All of the accident sequences that result in core building is also inconsistent with the assumption degradation,(i.e., result in an eat state that is used in the front-end analysis that the existing other than OK) are grouped into plant damage eqa.pment venting has no effect on reactor build-states in order to simplify the containment mg equipment. analysis. There are 15 plant damage states with a one-to-one correspondence between accident The containment analysis was also found to sequence subclasses and plant damage states. In have internal inconsistencies. For example. the fact, the PRA uses the two terms interchangeably. assumption that venting through the existing lines In the strict sense, the accident sequence classes leads to failure of all equipment in the reactor and subclasses represent the end results of the building is not reflected in the containment event accident sequence (front-end) analysis, while the trees, where several vented end states are desig-plant damage states are the starting point of the nated as being " recovered in-vessel." Failure of containment response analysis. For this report, reactor building equipment would likely induce accident class (plant damage state) V was core melt because of the resulting loss of reactor excluded because it is an insignificant contributor pressure vessel (RPV) injection. This would lead to the core melt frequency and, because it eventually to vessel failure rather than to in-represents an interfacing systems LOCA outside vessel recovery of the sequence. Another related containment, a type of accident on which venting inconsistency is the designation of several se-has no effect. Additionally, only internally qeences as " recovered in-vessel, late contain-initiated sequences w ere analyzed in this report, ment failure " Again, failure of the containment is assumed to fail equipment h>cated in the reactor The containment (Level 2) analysis performed building. As above, this would be expected to in-by SAIC N3 also used event trees to model the re- duce core melt and eventual vessel failure rather sponse of the primary and secondary containment than leading to in-vessel recovery, to a severe accident in general, two event trees were constrteted for each plant damage state, one Each end state on the containment event tree is to show the effects of venting through Filtra, the binned into a release mode based on its character-other the effects of venting through the existing istics (early containment failure, energetie release, vent lines. The quantification of many of the bypass of suppression pool, etc.) in accordance events was somewhat nebulous and could not al- with Table D-1 (adapted from Reference D-3). ways be reproduced. While the quantification of some events was supported by the MAAP analy. The release modes can be thought of as con-sis performed by FAl,m3 other events were quan- tainment failure modes (in analogy with tified based on engineering judgement, with no WASil-1400), although this is not entirely accu-other basis given. And finally, the quantification rate in the case of mode A1 and the modes that in-of the events in the early time-phased trees was volve venting without actual structural failure of almost entirely undocumented. the containment. In order to simplify the conse-quence analysis by reducing the number of re. Inconsistencies with the front-end analysis lease end states for which consequence were also found. For example, the front-end anal- calculations have to be made, each of the above ysis assumed that both of the existing wetwell and release modes is assigned to one of 12 release drywell vent lines would have to be opened to categories (11 for the cases without Filtra in-mitigate the pressure rise expected to occur inside stalled). A short description of the characteristics containment during a TW sequence. However, of each release category is given in Table D-2, this assumption was not carried through into the which is adapted from Reference D-3. D-6 i
-_-c _ _ - _
TablO D-1. Qualitative attributes of Shoreham release mWe , Release Mode Attributes A1 In-vessel recovery, no containment failure A2 In-vessel recovery, containment vented A3 In-vessel recovery, containment failed B1 Early small break, release through suppression pool g B2 Early small break bypassing the suppression pool, with reactor building retention B3 Same as B2, but with no reactor building retention B4 Early slow release through the suppression pool B5 Early slow release bypassing the suppression pool but with reactor building retention B6 Same as B5 but with no reactor building retention B7 Early moderate release through the suppression pool
^
B8 Early moderate release bypassing the suppression pool but with reactor building retention B9 Same as B8 but with no reactor building retention B 10 Early energetic release through the suppression pool B11 Early energetic release bypassing the suppression pool but with reactor building retention B12 Same as Bil but with no reactor building retendion Cl Late vented release with Filtra holdup maintained (not applicable) C2 Late vented release C3 Late vented release bypassing the suppression pool but with reactor building retention C4 Same as C3 but with no reactor building retention C5 Late over-pressure containment failure with release through the suppression pool C6 Late over-pressure containment failure bypassing the suppression pool e D-7
Table D-1. (continued)- Release Mode Attributes C7 Late large over-pressure containment failure with release through the i suppression Imol C8 Same as C7 but bypassing the suppression pool C9 Late thermal failure in the drywell with reactor building retention C10 Same as C9 but with no reactor building retention D1 Ex-vessel recovery with containment vented thtough Filtra (not applicable) D2 Ex-vessel recovery with containment vented through the suppression pool
- D3 Ex-vessel recovery with no containment failure or venting D4 Ex-vessel recovery with late containment failure, noble gas release mitigated by smallleakage rates Table D-2. Summary of the twelve release categories for Shoreham With Filtra Without Filtra Release Type of Csl Fraction NG Fraction Cs; Fraction NG Fraction Category Release Released Released Released
_Feleased 1 Early 0.4 1.0 0.4 1.0 2 Early 0.2 1.0 0.2 1.0 3 Early 1E-05 1.0 1E-05 1.0 4 Early 0 07 1.0 0.07 1.0 5 Delayed 0.1 1.0 0.1 1.0 6 Delayed 0.04 0.7 0.04 0.7 7 Lklayed 1E415 1.0 0.002 0.6 8 Late 0.09 1.0 0.09 1.0 9 Late 0.03 0.7 0.03 1.0 10 Late IE-05 1.0 IE-05 1.0 11 Controlled 3E-05 0.8 N/A N/A 12 Controlled 3E-05 0.01 3E-05 0.01 D-8
The actual grouping of release modes into re- ing such unpopulated areas as Long Island lease categories is shown in Reference D-3, Sound, w hich Shoreham borders, Table 5-2 (with Filtra) and Table D-2 (without Filtra). Ilowever, Table D-2 omitted release modes A2, D3, and D4 from plant damage states The result of this consequence analysis is a set I and Ill, and omitted release mode B 1 from plant of dose-versus-distance probability distribu-damage states IIA and 118. These tables were tions. Each curve represents a risk distribution in-used, with the omissions corrected, for binning tegrated over all core melt accidents for one the release modes in this report. particular type of dose (whole-body, thyroid, ete,) 'Ite whole-body dose curves were used in The release categories formed the input for the this report to represent the overall risk for each consequence analysis performed by PL&G.M venting strategy. It must be stressed that a risk This analysis calculated the dose that would like- analysis of this type is of limited usefulness be-ly be received by an individual,in a 24-hour ex- cause it does not factor in actions that would be posure period, from the radioactive releases taken to protect the population during an actual associated with a severe accident. This analysis emergency involving a core melt accident and ra, was performed separately for each release cate- dioactive release. The doses calculated in Refer-gory using the CRACIT modificcion of the stan- ence D-l also do not include any contributions dard CRAC2 computer code. The CRACIT code from long-tenn chronic exposure that would be calculates doses only for populated areas, exclud. incurred from living on contaminated land. D-9
REFERENCES D-1, Science Applications, Inc., Probabilistic Risk Assessment, Shoreham Nuclear Power Station, prepared for the Long Island Lighting Company, June 24,1983. D-2. E. T. Burns et al., Shoreham N.tclear Power Station Full Power PRA, PRA Update: Supplemen-tal Containment System implementation, IT/Delian Corporation, prepared for the Long Island Lighting Company, February 1988. D-3. 2. T. Mendoza et al., Containment and Phenomenological bent Tree Evaluation At Full Power for the Shoreham NucIcar Power Station, Science Application International Corporation, pre-pared for the Long Island Lighting Company, February 1988. D-4. Pickard, Los e, and Garrick. Inc.. Core bfett Accident DoseA'ersus-Distance Probability Distri. butions.100% Power Operations, Shoreham Nuclear Power Station, prepared for the Long Island Lighting Company, February 1988. D-5. Fauske and Associates,Inc., bfAAP Analysis to Support Shoreham 100% Power PRA, prepared for the Long Island Lighting Company, March 1988. I s D-10 l
l l gag.t.avm v s sucu*a atcutatoav coussios t a,tg g ;pp
- s. - e...%-,,.....,
mu2n BIBLIOGRAPHIC DATA SHEET NUREO/CR-5654 , an ox,,m ew, .e,. ' EOO-2632 2 tad asa suat Ni Containment Venting Analysis 3 c a n a s o a i gu n ,,e For the Shoreham Nuclear ~ power Station March 1991 4 m -a c A6898 ,wa sv.w 5 avt-aais. e n t;.areer William L Onlycan Technical Dana L Kelly
,v a m% 3 ,,
a e -n eo.- u c a w sc,,c,m:,w2sacos.vwtnoab u3.ruc .. ~ . ,,a- sa%-,,.,- . - - - - . . . - - EG&O idaho, Inc. Idaho Falls, ID 83415 e spessomo oncasaaiios e. ave aso acoa tss . .ue ,,, s. ... ~, ,, wum., ,,
- . ~ ~ o c a. - o, ~, ,, . . v s u ... .- , c,~ ,,-
Division of Systems Research Office of Nuclear Regulatory Research U. S. Nuclear Regulatory Conunission Washington, D.C. 20555
. e veuw. Tau sotts i i . ass u aa ,m ,.. , . .
An evaluation of the Shoreham Mark Il containment was performed to ldentify the effects of containment venting on core melt frequency, containment falicie mode, and offsite consequences. The analy;is was based on the Long Island Lighting Company's updated 1988 probabilities risk assessment of the Shoreham plant with the proposed Supplemen-tal Containment System (SCS). The SCS is a filtered containment vent system based on the Swedish Filtra system installed at the Barseback Nuclear Power Station in southern Sweden.'Ihe following three different containment vent-ing strategies were examined for their effects on plant risk: o Venting using the proposed Filtra system o Venting using the existing equipment at Shoreham o No venting. In addition, the consequences of containment venting were examined in cordunction with two sets of assumptions about the efrects of a harsh reactor building environment, produced by containment failure or venting through the ex-isting containment and reactor building heating, ventilating, and air conditioning systems, on the equipment located there. Specifically, the analyses studied the consequences when a harsh reactor building environmen, is assumed to have elther no adverse effect on equipment or to fall all equipment. u u , a o s o nc m m m., --, - -,,,,, . -o -,, -. ~, ~ ,-, ~ , .,, . - -+- . ~ - - Jnlimited
,4 .- . o m , a.
Venting "'""*' Filtra Unclassified Plant Damage State """"'"" Assumptions Unclassified Effects u * * " " "^
- ie +a
'4 4". 084 bi 3 )$ f 2 448
, . ~ - - . . . - . . - . . . . . . - . - - - - . - . . - .. .. . ~ . . . - - .- - . ~ - - TH7" 1UMENT WAS PRINTED USING RECYCLED-PAPER.
UNITED STATES ,,, c ,,, , ou, , , n , s , y , , NUCI EAR FIEGULATOFIY COMMISSION m $ u ce 6 'iss n
- WASHINGTON, D.C. 20555 ,, ,[l'[' , ,,
OFFICIAL DUSINES$ F EN ALTY FOR PRIVATE U5t. 6300 a
\~
k 1
- ._ .__ _ _ _ _ _ _ _ _ _ _}}