ML20154H872
| ML20154H872 | |
| Person / Time | |
|---|---|
| Issue date: | 12/12/1985 |
| From: | Advisory Committee on Reactor Safeguards |
| To: | Advisory Committee on Reactor Safeguards |
| References | |
| REF-GTECI-A-45, REF-GTECI-DC, TASK-A-45, TASK-OR ACRS-2379, NUDOCS 8603100392 | |
| Download: ML20154H872 (59) | |
Text
'
~l ft*RS-4399 7DR0303Vlo DA.TE ISSUED: 12/12/85 ACRS CECAY HEAT REMOVAL SYSTEMS SUBCOMMITTEE MEETING MINUTES l,,
DECEMBER 2-3, IS85 i b J 't WASHINGTON, DC i:.112., ri $$2 PURPOSE: The purpose of the meeting was to: (1) discuss the AFW system reliability, and (2) review the status of NRR's resolution position for USI A-45 Shutdown Decay Heat Removal Requirements.
ATTENDEES: Principal meeting attendees included:
ACRS NRC D ard, Chairman U Dennig J. Ebersole, Member J. Wermiel H. Etherington, Member R. Frahrn C. Michelson, Member L. Rubinstein G. Reed (3rd only), Member R. Hernan I. Catton, Consultant A. Marchese P. Davis, Consultant A. Thadani A. El-Bassioni UNITED ENGINEERS AND SNL CONSTRUCTOR 5 E Ericson F. Cook S. Hatch J. Mulligan W. Cramond PICKARD LOWE & GARRICK, INC.
K. Fleming A complete list of attendees is attached to the office copy of these Minutes. The meeting was convened at 1:00 pm, December 2, 1985.
MEETING HIGHLIGHTS, AGREEMENTS AND RE00ESTS AFW SYSTEM RELIABILITY
- 1. Mr. Ward noted that recent events are showing that the topic of AFW reliability is an important safety issue, hence today's meeting was scheduled to obtain relevant information on this topic.
8603100392 B51212 7 PDR ACHS -
5 DHRS Meeting Minutes December 2-3, 1985
- 2. R. Dennig (AE00) discussed the experience with startup and chal-lenges to the AFW system that were known to AE00. In response to Mr. Ward, Mr. Dennig indicated that the peculiarities of each plant's AFW system and operating modes make it almost impossible to generalize regarding the operating modes used and operating experi-ence with MTW/AFW systems.
AE00 discussed the data they have available on the AFW operating demand estimate for the three PWR vendors. The data requires interpolation partly because the AFW system is not consistently classified as an ESF. In response to Subcommittee questions, Mr.
Wermiel (NRR) said he would address this point in a later presenta-tion.
Key points noted regarding the vendor data included: (all date based on 1984 experience)
' For W plants, there were 130 demands for 34 of 37 plants licensed in 1984 This breaks down to ~ 3 + demands / year /
plant.
- A. Thadani noted W plants trip MFW upon reactor trip so AFW is almost always called upon. J. Ebersole questioned the wisdom of this design.
- The CE plants saw 22 demands from 8 of 12 licensed plants.
This is an average of a 2/ plant / year.
- The data on B&W units was sparse. Based on ESF actuation reports, AE00 found only 3 demands at 2 of 7 B&W plants.
- In response to Mr. Ebersole, Mr. Wermiel noted that histor-ically the AFW system did not receive the status of an ESF
DHRS Meeting Minutes December 2-3, 1985 classification because of the early focus on post-LOCA cooling and the lack of appreciation of core cooling challenges from transients.
- In response to Mr. Ward, Mr. Thadani noted that the 10-4 unreliability criterion is designed to credit recovery poten-tial upon failure. Mr. Ward indicated that his quick look of the data cited above indicates that for a large portion of the time (99".) it looks like recovery must be relied upon to meet the NRC SRP reliability goal.
- P. Davis said INPO has a significant study, just complete, on AFW reliability. Mr. Dennig said his quick-look at the Report indicates NRC is capturing most, if not all, the data INPO has compiled.
- The Subccmmittee expressed concern that the Staff has little good operational data on the reliability of a very important system (AFW). Mr. Dennig noted that one of the problems here is that there have been very few complete failures of the AFW system so reporting requirements are not always " tripped".
- 3. The evolution of the NRC Staff review of the plant AFW systems was detailed by J. Wermiel (NRR). Key points noted by NRR included:
- Staff criteria for AFW design prior to issuance of the SRP (1975) specified " good engineering practice." Plants reviewed against the SRP (1975 to present) have provided a " safety-related" system.
- A review of the AFWs in plants licensed prior to the TMI-2 accident (1979) was undertaken t'y the B&O Task Force, and both a deterministic and probabilistic evaluation of reliability
DHRS Meeting Minutes December 2-3, 1985 was performed by the Staff. Results were published in NUREG-0611 and NUREG-0635 for Westinghouse and Combustion Engineer-ing plants, respectively. B&W plants perfomed their own AFW reliability studies (BAW-1584) which were reviewed by the Staff.. In addition, all plants were reviewed'against the criteria of NUREG-0737. Items II.E.1.1 and II.E.1.2, in order to improve AFWs availability and the plants were found accept-able. In response to questions from Mr. Ward, NRR said the Staff accepted the reliability studies in BAW-1584 with~ the proviso that additional requirements for B&W AFW systems would be forthcoming in the future (e.g., power diversity require-ment).
- All plants were reviewed against criteria of Generic Letter 81-14 to improve AFWs seismic resistance and plants were found acceptable with the exception of Oconee. Oconee is still under review.
- SRP Section 10.4.9 was revised July 1981 to incorporate numerical reliability criterion, 10~4 to 10-5/ demand based on knowledge gained during B&O Task Force review. All PWRs not reviewed in NUREG-0611 and NUREG-0635 and licensed after the TMI-2 accident submitted a reliability study utilizing a comparable approach. These plants (12 total) comply with the SRP reliability goal. Current NT0Ls also comply with the SRP reliability goal. Mr. Ward asked if any plants were required to backfit modifications to their AFW systems vis-a-vis 81-14 requirements. NRR replied in the aftemative and indicated some plants are still in the process of upgrading their AFW systems.
- Staff review of AFW reliability in operating plants has continued since completion of the above actions. Currently a
SHRS Meeting Minutes December 2-3, 1985 CRGR package is under development. This will be addressed in a subsequent discussion.
- Mr. Wermiel noted that the following plants (Crystal River 3 Prairie Island 182, Fort Calhoun, ANO 1&2, and Rancho Seco) are under consideration for further upgrade'to meet the SRP reliability requirement per the CRGR Package above.
- In response to Subcomittee questions, NRR said all AFW systems have fairly consistent set of technical specifica-tions, but they did not know if this system is on the "Q-List". NRR also said that the older systems were not "QAed".
Mr. Ward asked NRR to determine if the older plants AFW systems are " operationally QAed". Mr. Hernan comitted to obtain that information.
- 4. R. Frahm discussed the application of AFW reliability requirements.
He detailed the methodology used in the post-TMI reliability estimates as documented in NUREG-0611 (Figures 1&2).
Results of Staff review of recent OL applications shows an unre-liability difference of about a factor of ten between a 2 and 3-train AFW system. The exception is for a loss of all AC power event where both have an unreliability of ~ 10 10-2/ demand (Figure 3). In response to Mr. Ward, Mr. Frahm said that recovery times are considered in the above unreliability values.
NRC's study of potential core damage accident precursors showed that for the AFW system, its unreliability was "10-3/ demand. Most AFW system failures are recoverable, i.e., if the system fails on demand,.the ability to recover system function is high.
DHRS Meeting Minutes Cecember 2-3, 1985 As a result of further discussion, NRR said it was not aware of any event where an AFW failure was not recovered in a timely manner.
Mr. Thadani cautioned that due to inherent limitations in the reliability analysis methodology, the.10~4 unreliability can not be taken as a "hard" value. The goal of applying the unreliability criterion, as NRR sees it, is to identify and correct AFW system vulnerabilities.
- 5. K. Fleming of Pickard, Lowe and Garrick (PL&G) overviewed a Paper he authored entitled, "A Systematic Procedure for the Incorporation of Comon-Cause Events Into Risk and Reliability Models." Key points of his presentation included:
- There are two mutually exclusive propositions about dependent events (comon cause events) that are often noted. These are:
(1) we don't know how to define it, model it, quantify it, or know exactly what causes it; and (2) we know how to defend against it.
- For modeling of common cause events (CCEs) (Figure 4), the key is the proper grouping of the model component subgroups.
- It is necessary to carefully incorporate CCE's into the logic model, perfonn the Boolean reduction and then convert to algebraic expressions. This assures many potential CCEs do not " fall through the cracks".
- The PL&G analyses shows that common cause events tend to dominate AFW system reliability calculations. Results shown (Figure 5) indicate that, based on experience data available, little is gained from power diversity. Mr. Fleming also noted that simple hardware redundancy is not a real panacea because the " human element" eventually tends to intercede (in a
DHRS Meeting Minutes December 2-3, 1985 comon-mode fashion) to reduce or eliminate the defense of redundancy.
Key conclusions of the PL&G presentation vis-a-vis AFW system reliability were:
- Current two and three train auxiliary feedwater systems (AFWs) have system uitavailabilities on the order of a 10'3/ demand.
Plant-to-plant variabilities result in a bound of 10-2 -
10-4/ demand. (This does not account for recovery factors.)
- AFWs unavailability is dominated by comon cause events.
- The principal benefit of a steam driven pump is its indepe'n-dence from electric power.
- There is a need to rethink the use of redundancy and diversity as a defense against comon cause failures.
- PL&G described their activity as part of a joint CCF (comon cause failure) reliability benchmark exercise. The purpose is to develop consensus CCF analysis procedures. Ten teams from various countries are participating (Figure 6). The problem being modeled is a loss of AFW at a six-train German KWU~
plant. First round results show a large spread in the point estimate CCF calculations (Figure 7). Mr. Fleming noted that the US Team (includes PL&G) calculated a lower reliability than seen above because most of the AFW trains are diesel-driven and PL&G assumed diesel engine reliability as very low compared to electric motor reliability.
- 6. Mr. A. El-Bassioni (NRR) provided coments on the above PL&G report. Key coments included:
DHRS Meetirg Minutes December 2-3, 1985
- The PL&G methodology is attractive and systematic, but has a number of limitations:
(1) CCF coverage is limited to treatment of dependency among similar components; (2) The failure dependency experience data is still sparse thus the PL&G methodology relies on subjective interpre-tation of data and its applicability; (3) The case study evaluation did not account for AFW support systems, and for recovery a:tions that might have im-proved AFW unavailability.
(4) Developing fault trees beyond the basic event level resulted in an increase in the number of cutsets by a factor of 5 which can overburden the analysis; (5) The PL&G methodology emphasizes the quantitative aspects of the analysis as compared to qualitative aspects (insights, vulnerabilities)
(6) The impact of CCF on AFW unavailability may have been overstated (i.e., a 99% contribution is too high).
(7) NRR does not agree with the report's conclusions about impact of diversity on system unavailability.
(8) The Staff recognizes that assurance of independence among redundant and diverse trains is the real key to AFW unavailability improvement.
DHRS Meeting Minutes December 2-3, 1985 In response to the above, Mr. Fleming noted that any analyst must make subjective judgements and he is putting these on the table.
He believes that the CCF failure contribution results from differ-ences in data interpretation.
- 7. L. Rubinstein discussed the status of NRR's actions vis-a-vis the upgrading of AFW system reliability for certain plants. He noted the following points:
- The Division of Reactor Safety and Oversight will be tracking resolution of this issue under the new NRR reorganization.
- The'AFW reliability study required in the SRP is used in conjunction with deterministic analysis to provide a relative compariscn of AFW reliability. It is not a stand-alone evaluation that can be directly applied to plant risk.
- The Davis Besse event crystallized an on-going concern with some plants' AFW systems as noted above. This concern lead to the development of a proposed requirement that would mandate that all PWR plants:
(1). demonstrate that the AFW system meets the SRP reliability goal, (2) assure AFW system motive power diversity, and (3) specify all details of current AFW system configurations
- post TMI fixes.
- The NRR Proposal has had internal Staff review and is being reworked to meet criticisms received. These criticisms center on the issue of resolving this concern in the broader context
s ;
DHRS Meeting Minutes December 2-3, 1985 of other on-going issues such as USI A-45 and the expected action items from the Davis Besse event.
- The Package is still undergoing NRR review and a decision on actions to be required (if any) should be expected on the time frame.of 4-6 months from now.
- 8. R. Hernan said that the topic of overseas practices vis-a-vis AFW reliability would be addressed by Dr. T. Speis at a meeting planned with the ACRS in 1-2 months following Dr. Speis' visit to the French Paluel plant.
- 9. D. Ward expressed some disappointment on the lack of a good opera-tional experience data base for AFW systems. He said that NRR seems to have a program to address concerns with AFW reliability and has a proposal under review. He solicited Subconsnittee com-ment. J. Ebersole suggested any or all means of getting water to the secondary system be endorsed. C. Michelson said the real reliability of the AFW systems seems to be an unknown. NRR doesn't seem to have any effort underway to improve their reliability data.
Mr. Ward asked if there is a real problem with AFW systems and is NRR properly addressing the problem. Mr. Michelson thinks there is a problem but the reliability data cannot be used to delineate a solution, given its poor quality.
Dr. Catton said the NRC's operational experience " data bank" is poor. More and better analysis of the data is needed. He was also puzzled by the apparent inconsistency in the fact that some plants' AFW systems are considered ESF systems and some are not. He was generally supportive of the work being done by C. Fleming of PL&G.
He believes it has good promise for modeling of comon-mode fail-ures.
DHRS Meeting Minutes December 2-3, 1985 Mr. Davis suggested that INP0 and EPRI/NSAC be invited in to a future DHRS Subcomittee meeting to discuss the work they have underway / completed on the topic of AFW reliability. He is comfort-ed by the fact the NRC is pursuing fixes for the plants' with AFW systems that are the " outliers" vis-a-vis system reliability. He believes the onus for assuring a " reliable" AFW system must be with the Utility operating the plant.
USI A-45 RESOLUTION EFFORT
- 10. A. Marchese provided a report on the status of the USI A-45 resolu-tion effort. Figure 8 provides an outline of the presentation topics. NRR noted that the Point Beach and Quad Cities " rough drafts" were reviewed with the DHRS Subcommittee in May 1985.
Significant additional analyses were required by SNL after detailed NRR Staff review. Figure 9 lists the Staff review issues. In response to Mr. Davis, NRR said they did not receive any coments on the Reports from the parent plant Utilities.
D. Ericson (SNL) overviewed the revised analysis status. The internal analysis was reworked to include: (1) broader treatment of comon mode; (2) additional treatment of actuation; (3) long term blackout; and (4) re-examination of human factors / recovery.
Figure 10 lists the current status of the remaining plant analysis report schedule. The next analysis reports to be issued in draft form will be for the Cooper and Turkey Point plants. It was noted that the planned Report on Trojan is temporarily on hold due to internal manpower problems at SNL. In response to Mr. Ward, Mr.
Marchese said the SONGS and Grand Gulf plant analysis were dropped due to budget and scheduling problems with the A-45 Program.
A draft sumary Report (probably in letter format), that pulls together the conclusions of all the Plant Analysis Reports should
DHRS Meeting Minutes December 2-3, 1985 be available in January 1986. Figure 11 lists the status of other supporting topics for the A-45 Program.
- 11. W. Cramond provided a detailed presentation on the revised Point Beach plant analysis Report. In response to Mr. Ebersole, SNL said they did not consider core melts initiated by containment failure.
Mr. Marchese said containment initiated core melts are more likely for a BWR and will be addressed in the Quad Cities Report dis-cussion.
A number of changes / additions to the Point Beach Report were made as a result of NRC Staff review (Figure 12). In response to Mr.
Ebersole, Mr. Ericson said SNL did not consider environmental qualification in their analysis.
The significant Point Beach vulnerabilities were descrioed. Eleven vulnerabilities were identified in the internal plant analysis (Figures 13-17). In response to Mr. Michelson, NRR said that generic reliability data was used, to the extent possible, in these analyses. Key vulnerabilities identified included:
- Failure to switchover from emergency core injection to recirculation.
- Comon mode failure of safety systems pumps.
- Comon mode failures of safety system valves.
- Failure of the CCW pumps.
DHRS Meeting Minutes December 2-3, 1985 The seismic hazard analysis results are given in Figure 18. In response to Mr. Michelson, SNL said that relay chatter has not been considered in this analysis. It is being addressed in another Program. Mr. Ward asked why the seismic analysis was not related to USI A-46. SNL responded that they believed they had the re-sources to address the issue. SNL noted that the modifications proposed to address seismic vulnerabilities reduced the CM prcb-ability by an order of magnitude (6.1 x 10-5 to 5.0 x 10 Figure 19).
Mr. Michelson asked if the impact of concurrent failures of non-seismic equipment were accounted for in this analysis. SNL in-dicated they did, but on a limited basis. Further discussion resulted in a request from Mr. Michelson that SNL specify all the major assumptions in their analyses. NRR agreed and indicated they would look into this.
Discussing the plant fire analysis, Mr. Michelson asked if inadver-tent actuation of fire suppression equipment was addressed. SNL believed they had, but NRR ccmitted to research this question and respond at a later time.
Based on the internal analysis, a total of 7 modifications were proposed (Figure 20). These were arranged into a set of four Alternatives (Figure 21). Alternatives 2-4 are arranged in order of increasing modifications up to the addition of an add-on DHR system (Alternative 4). In response to Mr. Ward, Mr. Ericson said Alternative 4 would address all the external and internal event threats. The add-on DHR system decreases the core melt (CM) probability by a a factor of ten (3.6 x 10~4 to 3.6 x 10-5),
Given the omission of a number of CM contributors (LB LOCA, ATWS, Event V.) SNL estimated the above CM probabilities represent from 40-60% of the total CM probability.
DHRS Meeting Minutes December 2-3, 1985 Details of the value/ impact analysis for the proposed Alternatives were discussed. The source terms used were based on WASH-1400 (RSS) values and two fractions, (0.1 RSS and 0.3 RSS). The frac-tional values are based on SNL's best guesstimate as a result of the on-going source term work. In response to Mr. Ward, SNL said the values of population dose used in the impact analysis includes the core melt probability.
Figure 22 shows the results of the value/ impact analysis based on the central-value figures. None of the Alternatives show a cost-effectiveness ( < $1000/MR) based on consideration of offsite costs only.
SNL discussed the results of sensitivity analysis of the benefit of feed and bleed cooling, feed and bleed with the PORVs blocked open, and use of secondary blowdown. The results show a significant benefit for use of feed and bleed.
For the case of assuming no interdiction or decontamination (!&D) given a CM event, then all the Alternatives show a substantial cost-benefit (Figure 23).
Mr. Ward asked if there is any reason to consider I&D. Mr. Mar-chese indicated that there is a question of the realism of the CRAC code calculations and that the degree of ISD expected is not really known. This issue needs to be addressed by NRC Management. It was noted that the $1000/MR is an estimate of all costs associated with a CM accident. The "true" I80 costs are only indirectly considered via the $1000 value. Mr. Ward observed that the 180 estinates do rot include the economic penalties associated with ISD activities.
Mr. Davis observed that the offsite costs are not well considered in this analysis.
DHRS Meeting Minutes December 2-3, 1985 Mr. Ward asked if SNL believes the analysis is mature at this point. Mr. Cramond said he believes the report is a 90% " mature" given the initial constraints 9 "cd on them, i.e., they have captured all the major CM contributors due to DHRS-related events.
- 12. The results of the Quad Cities plant analysis were reviewed by S.
Hatch (SNL). The format for the presentation was similar to the Point Beach presentation above. Key points of the presentation include:
- The initial insights potentially affecting DHR capability were noted(Figure 24). Four major changes were made to the Report analysis (Figure 25). These changes resulted in increasing the CM probability by a factor of two.
- The most significas.t vulnerabilities identified from the internal analysis, fire analysis, and seismic analysis were described (Figure 26-28). The Subcommittee had a number of questions, mostly centering on the details of the above analyses. In response to Mr. Ward, SNL said that they did not explicitly consider the downside risk of any of the modifica-tions proposed.
- A set of five Alternatives were developed to address the above vulnerabilities (Figures 29-30). As with the Point Beach Report, the Alternatives are arranged in order of increasing expense with Alternative 5 being an add-on DHR system (with a 2-hour RCIC dependency). J. Ebersole suggested that NRC evaluate use of a " fire-pump" type system for long-term cooling. NRR (W. Minners) suggested that enough information will be obtained from the A-45 Program to determine if such a fix would be worth pursuing, without benefit of any further study.
DHRS Meeting Minutes December 2-3, 1985
- The impact of the Alternatives on CM probabilities were shown (Figure 31). An Alternative 6 (add-on DHR without 2-hour RCIC dependency) reduces the CM probability by a factor of 10 to 2 x 10-5 This Alternative was dropped from the draft Report.
- The results of the value/ impact analysis was detailed (Figure 32). One of the Alternatives (Alternative 4) falls below the
$1000/MR figure. As with the Point Beach analysis, the value/
impact analysis shows any fix is cost-effective if no inter-diction or decontamination is assumed (Figure 33).
- Concluding remarks noted for this Plant Analysis included:
(1) Station blackout sequences are not important.
(2) Quad Cities has good physical redundancy in DHR systems.
(3) Electrical failures' dominate internal event analy-sis.
(4) Seismic events dominate special emergencies.
(5) Mods were found which meet the $1000/MR criteria.
(6) Many scenarios will affect both units.
- 13. A presentation on a cost analysis for a dedicated primary blowdown DHR system was given by representatives of United Engineers and Constructors. Recall that ACRS Member G. Reed has proposed a dedicated blowdown system for all PWRs to assure core cooling.
DHRS Meeting Minutes December 2-3, 1985 UE8C detailed the approach used in the' analysis. The approach was to compare the blowdown, and RHR systems against a baseline of an added AFW train. The base plant used in the ana' lysis is Point Beach. In response to Dr. Catton, UE&C said they would provide cost estimates for building the above system assuming " industrial grade" instead of " nuclear grade" quality.
The UE&C design criteria used in their analysis is given in Figure 34 Figures 35 and 36 show a diagram of the UE&C blowdown system and the differences from Mr. Reed's proposed system. Figure 37 shows the proposed high pressure RHR system. Mr. Reed noted that the UE&C version will not be as versatile as his system. UE&C said that they had not tried to optimize the costs of these systems at this stage.
The cost comparisons were given for the three proposed systems (Figure 38). UE&C noted that the main " cost driver" for these systems was the cost of piping.
UE&C said that the high pressure RHR and primary blowdown systems are significantly more expensive than an add-on AFW train. They also suggested that the blowdown system introduces control problems due to the potential hydraulic loadings involved. Mr. Reed again indicated that the primary blowdown system will be able to handle almost any accident situation.
14 A. Marchese addressed the issue of how A-45 will address the CE/PORY issue. Key points of his discussion were:
- All 5 PWRs in a A-45 program, including St. Lucie (CE), will be analyzed with and without bleed and feed capability and the change in core melt frequency and risk due to addition of bleed and feed capability will also be characterized. l
DHRS Meeting' Minutes December 2-3, 1985
- Costs and other impacts of adding bleed and feed capability are being addressed.
- The bleed and feed approach is one of the Alternatives being considered in A-45 program. This Alternative will be ranked with others based on value-impact analyses.
- Feed and bleed will be compared to other alternative measures on a value/ impact bases for improving overall DHR reliability.
- In response to Mr. Reed, Mr. Marchese indicated that NRR's initial position vis-a-vis CE plants without PORVs was that as a minimum PORVs would be required. He indicated that the A-45 resolution position may specify something more or less compre-hensive than the addition of PORVs.
- 15. The A-45 resolution schedule was discussed. This schedule calls for a final contractor summary report by May 1986 and a draft regulatory analysis issued for internal NRC review by July 1986.
The resolution position Package would be submitted to CRGR in October 1986. In response to Mr. Davis, Mr. Minners (NRR) said the fann of the resolution position is not known at this time.
- 16. D. Ward noted that he believes the NRR Staff, SNL, and the Util-ities involved deserve a "well-done" for their work and/or coopera-tion with the overall A-45 resolution effort.
- 17. D. Eticson overviewed the results of the sabotage analysis for both
, Quad Cities and Point Beach. The discussion was held in closed
, DHRS Meeting Minutes December 2-3, 1985 session because of safeguards information presented. The analysis was restricted to the sabotage threat by an insider.
The SNL analyses concluded that various alternative fixes are cost-effective to implement, if the sabotage threat is considered likely to happen.
- 18. The next DHRS Subcommittee meeting on the status of USI A-45 resolution effort will be scheduled for late March 1986, with a brief status report scheduled for the April full Committee meeting.
19 .- The meeting was adjourned at 5:10 p.m. on December 3, 1985.
NOTE: Additional meeting details can be obtained from a transcript of this meeting available in the NRC Public Document Room, 1717 H Street, NW, Washington, DC, or can be purchased from ACE-Federal Reporters, 444 North Capital Street, Washington, DC 20001 (202) 347-3700.
1
- 1. METHODOLOGY (NUREG 0611) ,
SRP 10.4.9 REQUIRES AFWS UNAVAILABILITY RANGEToOF 10-5 10-4 F;ILURE PROB / DEMAND, HOWEVER CAN USE OTHER MEANS OF DECAY HEAT REMOVAL, NUREG 0611 REQUIREMENTS ASSESS RELATIVE RELIABILITY COMPARISONS OF VARIOUS AFWS DESIGN (PURPOSE) z -
NO SPECIFIC COMMON CAUSE EVALUATION USE EVENT TREES (INDUCTIVE LOGIC)
FAULT TREES (DEDUCTIVE LOGIC) 0611 DATA BASE TO OVANTIFY TREES
- ANALYZE LOFW, LOOP, LOAC 0611 HAS RESULTS FOR OPERATING PLANTS AFWS EVALUATION INCLUDES HARDWARE FROM THE WATER SOURCE TO THE AFW N0ZZLE ON THE SG, SUPPORT SYSTEMS ANALYZED QUALITATIVELY N0 HIGH ENERGY LINE BREAK OR EXTERNAL EVENTS STAFF REVIEW (ASSISTED BY BNL a SNL)
- DOMINANT CONTRIBUTORS FOR EACH INITIATOR (MAINTENANCE, AND HARDWARE FAILURE OF PUMPS, VALVES, ACTUATION LOGIC)
SINGLE / COMMON MODE FAILURES 5
- 0611 RESULTS ARE POINT ESTIMATES WITH LA'RGE UNCERTAINTY
- IMPROVEMENTS IDENTIFIED (AUTO SWITCHOVER, ACTUATION LOGIC, ALTERNATE WATER SOURCES) e O
9 hG
/ II, STAFF REVIEW 0F RECENT OL APPLICATIONS .
REVIEWED 17 PLANTS
- 2 TRAIN AFWS (I, E., ANO-1, CR-3, DB-1, RANCHO SECO)
- 3 TRAIN AFWS (1, E., CATAWBA, SEABROOK, MIDLAND, WATERFORD, SUMMER) 2 TRAIN AFWS UNAVAILABILITY LMFW ABOUT 10-3 TO 10-4 RANGE LOOP ABOUT 10-3 TO 10-4 RANGE LOAC ABOUT 10-1 TO 10-2 RANGE 3 TRAIN AFWS UNAVAILABILITY LMFW IN 10-4 TO 10-5 RANGE LOOP IN 10-4 TO 10-5 RANGE LOAC IN 10-I TO 10-2 RANGE i
i l
l l
l l
l l
l l
l*
SYSTEMATIC PROCEDURE FOR SYSTEMS ANALYSIS
- 1. SYSTEM FAMILIARlZATION
"'*"'^ #*
- 2. LOGIC MODEL DEVELOPMENT * **** * * ,
- 3. BOOLEAN AN ALYSIS * '"'^ M
- 4. ALGEBRAIC MODEL DEVELOPMENT *#"d"~
- 5. PARAMETER ESTIMATION ='/'% *"
- 6. SYSTEM QUANTIFICATION
- 7. RESULTS INTERPRETATION l
l
- CRITICAL STEPS FOR COMMON CAUSE ANALYSIS
- r. - -. -
~ <_._, ..
t
COMPARISON OF CAUSE TABLES FOR THREE AUXILIA '
FEEDWATER SYSTEMS IN NORMAL ALIGNMENT .
EVALUATED USING MGL MODEL ,
e le ys or-Wen hos a gen ators per train)
Cause Frequency Cduse Frequency cause Frequency P3 4.2(-4) P3 4.2(-4) P2 8.0(-4)
V4 3.6(-4) V4 3.6(-4) V4 3.6(-4) 4Y3 2.5(-5) M3 9.6(-5) M2 1.9(-4)
H2 (Pt + T) ,
1.1(-5) 4V3 2.5(-5) 4Vi (P1+M) 4Vg (Pg + T)(Pg+M) 1 4.7(-5) 1 2.7(-6) 4P V21 2.B(-6) (P1 + Hg )(Pg+M)
C 3 9.B(6) 2.3(-6) C 2.3(-6) C 12V Y21 2.3(-6) 1.9(-6) 12V V21 1.9(-6) 12V V21 1.9(-6)
Others
- 2.0(-6) Others - 4.0(-6) '
Others - 1.0(-6)
Total 8.2(-4) Total 0.1!-4) Total 1.4(-3)
Comon Cause/ '
i Total *996 Comon Cause/
Comon Cause/.
Total *997 Total i . . _ _ . _ *958 -
>w -cr mt wo m
't
. - _ _ - . - - - - - - - - h
'.r...
- ......:m...:.n
. . . . . , . . m .;;.;.4 n,y ma....
- j. _ y _ }~. ,is
.u..,., p.ym
,n e: 4 & e.7.=
. a . ::',; 't. m ..:.9 .R:e%.;. , ,., .Q& %un
. . = . ..
~
~
y.}z:9_Y: .
- ~:'1"*:
~
PRELIMINAFI ~15ESifLTS OF THE
~
~.
. ~ .. ... ~
CCF-RBE BENCHMARK
~
.? S TECHNICAL APPROACH . ~ . f.
..y 2._~e .r
.e g __
TEAM QUALITATIVE QUANTITATIVE - --
7;-
A. BELGIUM FMEA B. DENMARK BETA FACTOR SYSTEM FAMILIARIZATION BINOMIAL FAILURE RATE C. FEDERAL REPUBLIC SYSTEM FAMILIARIZATION BETA FACTOR, MARSHALL-OLKIN OF GERMANY-1 (GRS)
D. FEDERAL REPUBLIC SENSITIVITY ANALYSid MARSHALL OLKIN OF GERMANY 2 (KFA) .
E. FEDERAL REPUBLIC SYSTEM FAMILIARlZATION MULTIPLE GREEK LETTER -
OF GERMANY 3 (KWU) '
"' 9 :"
F. FRANCE GENERIC CAUSE ANALYSIS G. ITALY MULTIPLE GREEK LETTER ~.
SYSTEM FAMILIARIZATION BINOMIAL FAILURE RATE 4 l' H. SWEDEN SYSTEM FAMILIARlZATION
.U'M:
- l. UNITED KINGDOM MULTIPLE GREEK LETTER FMEA/ CHECKLIST J. UNITED STATES SYSTEM FAMILIARIZATION MODIFIED BETA FACTOR, CUTOFF : dj ;
MULTIPLE GREEK LETTER - tM
- . sp. te
...y .
1 R .. s .
y :'.]: .;j:~J.7'g, g-.;
o m arew m N, ,
. . , mm, w . m
/t!
d CCF-RBE BENCHMARK PRELIMINARY POINT ESTIMATE RESULTS Independent Failure Contribution
'r. . . .
BELGIUM i 3 0-I DENMARK
-' 99 i
FRG-GRS $$
l FRG-KFA $$
I FRG-KWU .l @
~~l, ['
FRANCE -
g- G-4 -
~l !
ITALY - l-
$ l SWEDEN -
--- l-- O O ,
..g.
U.K. l O O
-l U.S. -1 $$
I. . .
ID'I.. . ID* ID'Y
'tt' x . - -, % i I
b
PRESENTATION OUTLINE PLANT ANALYSIS REVISION ACTIVITIES AND SCHEDULE.
SUMMARY
OF REVISED POINT BEACH PLANT ANALYSIS RESULTS.
SUMMARY
OF QUAD CITIES PLANT ANALYSIS RESULTS.
PRELIMINARY ANALYSIS RESULTS OF A DEDICATED PRIMARY BLOWDOWN SYSTEM.
CONSIDERATION OF HIGH-PRESSURE RHR SYSTEM.
CE/PORV ISSUE. -
SAB0TAGE ANALYSES.
PROGRAM SCHEDULE.
~
%1 N
- \
l STAFF REVIEW ISSUES ON SANDIA-PLANT ANALYSES FOR P0 INT BEACH AND QUAD CITIES
- TREATMENT OF COMMON MODE FAILURES
- OPERATOR ERROR FOR FAILURE TO SWITCH OVER TO SUMP RECIRCULATION
- BATTERY FAILURE RATE (INCLUDING COMMON MODE)
- TREATMENT OF FEED AND BLEED
- FIRES IN AFW PUMP ROOM AND 4160 V SWITCHGEAR ROOM
- INTERNAL FLOODS INITIATED BY HUMAN ERRORS ,
(INCLUDING FLOODING OF MULTIPLE AREAS THROUGH COMMON FLOOR DRAINS)
- UNCERTAINTIES IN SEISMIC PORTION OF ANALYSIS, INCLUDING RELAY CHATTER FAILURE MODE l
- STUCK OPEN RELIEF VALVE FAILURE FREQUENCY
- LOOP INITIATOR FREQUENCY
- TWO UNIT CORE MELT EFFECTS D
- INSENSITIVITY OF AVERTED PERSON-REM TO SOURCE TERM USED
- SOME INCONSISTENT TREATMENT OF POINT BEACH AND QUAD CITIES k (E.G., STATION BLACK 0UT SEQUENCES)
E_
/ .
PLANT ANALYSIS STATUS / SCHEDULE r'
QUAD CITIES DRAFT SUBMITTED SANDIA MANAGEMENT APPROVED SENT TO CECO FOR COMMENT / INPUT PUBLICATION IN JANUARY POINT BEACH DRAFT SUBMITTED IN SANDIA APPROVAL PRDCESS SENT TO WECO FOR COMMENT / INPUT PUDLICATION IN FEBRUARY COOPER DRAFT SUBMITTED IN SANDIA REVIEW PROCESS TURKEY POINT DRAFT IMMINENT .
ST LUCIE INTERNAL ANALYSIS AT FIRST DUANT SPECIAL EMERGENCY IN PROGRESS TARGET DATE FEBRUARY FOR DRAFT ANO 1 INTERNAL ANALYSIS IN PROGRESS SPECIAL EMERGENCY IN PROGRESS TARGET DATE FEBRUARY FOR DRAFT TROJAN TEMPORARILY ON HOLD i
i 1' o N .
t SPECIAL TOPICS --
, i
SUMMARY
/ GENERIC CONCLUSIONS TO DATE - 1ST IN PROGRESS DRAFT JANUARY
/ 2/
f enftfMm ;
VALUE-IMPACT CONSIDERATIONS FOR NORTORIA - INREVISED PROGRESS DRAFT FEB SS STRUCTURE OF GENERIC V-I TO SUPPORT REGULATORY ANALYSIS - IN
SUMMARY
OF VALUE NEASURE AND V-1 STURCTURES -DRAFT IN PROGRESS ASAP l
lN N
,A '
'O . - - _ .- - --_
CHANGES FROM ORIGINAL POINT BEACH ANALYSIS TO CURRENT ANALYSIS
t 2. LONG TERM STATION BLACKOUT DUE TO BATTERY OR CST DEPLETION CONSIDERED.
i 3. CONTROL CIRCUIT FAILURES INCLUDED IN PUMP AND VALVE LOCAL FAULTS.
- 4. COMMON MODE OF APPLICABLE PUMPS AND MOVS CONSIDERED.
4 5. TEST AND MAINTENANCE DOWN TIME INCLUDED.
- 6. AC AND DC BUS INITIATING EVENTS CONSIDERED.
- 7. BATTERY FAILURE PROBABILITY CALCULATION IMPROVED.
- 8. ADDITIONAL SENSITIVITY ANALYSIS ON PORV BLOCKING INCLUDED.
- 9. ADDITIONAL SENSITIVITY ANALYSIS ON CRAC2 MODEL INCLUDED.
- 10. CHANGES WERE MADE TO THE SECONDARY BLOWDOWN MODEL.
- 11. LOSP INITIATOR FREQUENCY CHANGED FROM 0.100 TO 0.086. ,
- 13. MODIFICATIONS TO SEISMIC LOGIC MODEL ESPECIALLY FOR THE RWST.
s T' . .
MOST SIGNIFICANT POINT BEACH VULNERABILITIES - INTERNAL
- INTERNAL VULNERABILITY 1: FAILURE TO SWITCHOVER FROM EMERGENCY CORE INJECTION TO RECIRCULATION SMH{Hj 2 2.00E-5 TMQH{Hj 2 1.40E-6 T0HjHj 3 _
9.90E-6 l 3.13E-5 INTERNAL VULNERABILITY 2: STATION BLACK 0UT DUE TO BATTERY FAILURE T3MLE 4.83E-6 INTERNAL VULNERABILITY 3: STATION BLACK 0UT DUE TO DIESEL GENERATOR FAILURES TiMLE 5.49E-7 8
%y =
j SN. '
l .
MOST SIGNIFICANT POINT BEACH VULNERABILITIES (CONT.)
i
},INTERNALVULNERABILITY4: FAILURE OF ECC RECIRCULATION DUE TU
] RHR PUMP C0 CLING FAILURE CAUSED BY A VsLVE FAILURE SMH{Hj 2 9.60E-6
- T30H y2 M 'i.75E-6 TMQHjHj 2 6.72E-7
)
1.50E-5 i
INTERNAL VULNERABILITY 5: FAILURE OF ECC INJECTION DUE TO CCWS FAILURE CAUSED BY LOSS OF COOLING FROM THE SWS THROUGH THE CCW HEAT EXCHANGER S2MDy2D 3.80E-7
~
T30Dy2D 5.78E-7
. T2MODy2D 8.26E-8 1.04E-6 M
.N, .
MOST SIGNIFICANT POINT BEACH VULNERABILITIES (CONT.)
yINTERNALVULNERABILITY6: COMMON MODE FAILURE OF SAFETY SYSTEMS PUMPS SMHjHj 2 4.00E-6 S MDy2D 4.60E-6 2
S MXD y 2 1.20E-7 TyMLE 2.18E-7 T MLH y 2 2.03E-6 T MLE 6.61E-6 2
TMQHjHj 2 1.80E-7 T MQDy2D 3.22E-7 2
T0H{Hj 3 1.98E-6 T30Dy2 D 2.25E-6 T4MLE 1.18E-6 T MLE 5.80E-7 5
2.42E-5 p%
MOST SIGNIFICANT POINT BEACH VULNERABILITIES (CONT.) .
COMMON MODE FAILURES OF SAFETY hINTERNALVULNERABILITY7: SYSTEM VALVES SMH{Hj 8.00E-6 2
TMQH{Hj 5.60E-7 2
T0HjHj 3.96E-6 3
1.25E-5 INTERNAL VULNERABILITY 8: FAILURE OF THE LOW PRESSURE INJECTION SYSTEM IN THE RECIRCULATION MODE SMH{Hj 1.43E-5 2
TMOHjHj 9.99E-7 2
T0HjHj 7.07E-6 3
2.24E-5 INTERNAL VULNERABILITY 9: FAILURE OF THE AFWS TURBINE DRIVEN PUMP .
T,MLE 4.35E-7
- TgMLE 4.25E-6
~
TSMLE 5.71E-6 1.04E-5 Yy .
s -
MOST SIGNIFICANT POINT BEACH VULNERABILITIES (CONT.)
INTERNAL VULNERABILITY 10: FAILURE OF THE CCW PUMPS
~
S2MD32 D 1.23E-6 T30D12 D 6.03E-7 1.83E-6 T INTERNAL VULNERABILITY 11: LONG TERM STATION BLACKOUT CAUSED 3Y DEPLETION OF THE STATION BATTERIES OR THE CONDENSATE STORAGE TANK TyM * (FAILURE OF DGS) * (BATTERY DEPLETION AND/0R CST DEPLETION) 3.56E-5 -
O
%g, F%
==.
A
,/
SEISMIC CORE MELT PROBABILITIES LEVEL (SSE) y E01 E02 TOTAL I-2 6.8E-6 6.3E-6 1.2E-5 2.5E-5 2-3 1.lE-5 2.5E-6 8.8E-6 2.3E-5 3-4 4.5E-6 2.8E-7 3.7E-6 8.4E-6 4-5 2.3E-6 3.0E-8 1.0E-6 3.4E-6 6.1E-5 SEISMIC RELATED VULNERABILITIES
- 1. REFUELING WATER STORAGE TANK
- 2. ANCHORAGE OF ELECTRICAL BUSES, TRANSFORMERS, INVERTERS, AND BATTERY CHARGERS
- 3. BATTERY RACKS INADEOUATELY ANCHORED
- 4. INSTRUMENT AIR SYSTEM FOR PORVS s
9 -
B h
MODIFICATIONS BASED ON SEISMIC ANALYSIS
- 1. PUMPS AND PIPING TO MAKE SPENT FUEL POOL BACK-UP WATER SOURCE FOR RWST
- 2. PROVIDE ADDITIONAL ANCHORAGE FOR BUSES, INVERTERS, AND CHARGERS
- 3. REPLACE WOODEN BATTERY RACKS WITH METAL RACKS AND PROVIDE ADDITIONAL ANCHORS
- 4. INSTALL SAFETY CLASS NITROGEN BOTTLE AND ASSOCIATED EQUIPMENT l
SEISMIC CORE MELT PROBABILITIES WITH MODS' 1
LEVEL E E01 EQ2_
TOTAL 1-2 9.6E-9 1.1E-7 1.9E-7 3.0E-7 2-3 5.1E-7 3.7E-8 2.2E-7 7.6E-7 ,
3-4 1.5E-6 1.3E-8 3.3E-7 1.8E ! 4-5 1.8E-6 6.7E-8 4.9E-7 2.2E-6
, 5.0E-6 h
,s k,4 .
M W H Q E O E = ~
H W H &
M N O E E Q o 3 p Q M < A M H
= C o o E H OW J W H H ~ J U H U m
- Z W M 4 U Z X H 7 O H Q A M C E H M m O m 4 O E H H EW X W m M E W Z 4 A A M J M
= 0 E H M Q E WW M & W W Z W ( m J E
> b WM 4 M U < J a J Q R Z R W J
< N J J WO H H N X W WM < Z m 4 W 4 > M N H HM 3 & O M
- C WO ~ E N H J Z m H > E C E EW m 4 U Q 4 m = ~ m Z E X H E > H d J U D m m 2 W J om m < A M W 3 U E A MM H E E m E W M ( W A > O m D Q W X WW D M A WC A W m > H M Mm A E 4 2 C J WW D E E H E 3 W HW O > > M C 4 W O H E M H J J M m
- MW < < WO ( M Q C E W = m > > m H o O W
W X E mQ o A M
E m E
M = m M H CW WH 4 X WW 4 H Z 3 Wm > U <
N N O H Z M H O ~ D E
( X H H W W J H MM M M 3 4 4 E > M M Q X mW W J U H M W W Q E b Q Q 4 W X O J Z M w a M A > Z W 2 WH U W ( 4 DW E U Q U M 4 J M H U W J E WO e V ( M M E > 4 Q WQ ~ E J m W ( W = D E A X Om 4 W H U H Q m E 4 W 4 W W
- EW M E Q < Q H M Q WE W EW E A E Z Q X Q A W X m E WM E w M WW X d m R A E m H Q m W 4 OM E H
- O 4 m J K WD O m 2 UW m J M mW A m N A Q = Z D 4 W Z Z W N
~ Q > H mH H H a L WW 1J 4 W X E J .
mW QWM 4 A 3 4 m D A
4 E
O J HW E D d 4 m J J @ J J O J J E J W J E m U J E E H
( mm ( w W 4 ( m 4 4 W m
H E H H J &W H >
Q U Mm M MQ M X M w Q QW E J W E E O E m X M Q 4 m mWm m mW m J mQ 4 e e e e e 6 e
M N M @ m W N b
PN 20
CONTENT OF ALTERNATIVES ALTERNATIVE 1: RWST LEVEL ALARM IMPROVEMENTS DEDICATED DIESEL GENERATOR ST4RTUP BATTERIES REDUNDANT RHR PUMP COOLER OUTLET VALVES RwST SEISMIC IMPROVEMENTS IMPROVED SEISMIC ANCHORAGE OF ELECTRICAL EQUIPMENT BACKUP AIR SUPPLY FOR PORVS INTAKE STRUCTURE SHIELD WALL EXTENSION AFW PUMP ROOM FIRE PROTECTION SEPARATION OF DC EMERGENCY POWER SUPPLIES ALTERNATIVE 2: ALL ABOVE PLUS ADDITIONAL CST ALTERNATIVE 3: ALL ABOVE PLUS THIRD INDEPENDENT LOW PRESSURE TRAIN INDEPENDENT DIESEL DRIVEN AFWS PUMP DIESEL GENERATOR EXHAUST STACK ANCHORAGE IhPROVEMENTS ALTERNATIVE 4: ADD-ON SDHR SYSTEM
,s ,
4 -
93 CU '
Teats 9.5 estest seams . Susenhet or Wasser.gespeCT IsraSuers (Central valeet Cheogo v-3 ametrato eseed en erreste offette Total W=t Seelyst e of Of f elte fest e and mite feste Se Case awested Avert ed attesentlee smelt emee sames T-t est stetters T-t aset pue18ese es. Prettabillty tream) eet te eetle sweetit per pree sette senette year p-ree IS ele 3 ($ els 9 A,
P v
- 's 'I*
' e 2 e """o 'I'm "e I'*
e 1 2.135-4 3183 e.64 8.25 -S.Me 23M e.M2 -e.849 833 2 2.4ee-4 3738 e.7% e.15 -12.ae2 39ee e.251 -7.e 22 2335 3 2.84-4 eles e.86 0.182 -22.277 57%e e.347 -15.734 3ege 4 3.338-4 4Ms e.94 8.e44 -43.355 33ee? e.04S -54.2%e 12504 s
S ,
J_ l
k Tasta 9.9 POtWF BB&G - VALBE-INPACT RESOLTS - We Interdicties or Secoatselsaties aseecties otratte v-t Amstrete er of reite coste v-t Ana lys t e aseed en of f e t t e and oes t t e cost s te core Averted Averted altersettee melt some pose V-t met sellare V-I met sellare Probahtitty p-ree matte natie sesefit per p-ree natie seeerit per p-res a
pe
($ slel ($ ale-6, j
AP. Ty ADe , VIR, NRW
, DPR , v!R, NeV, DFB, L 43194 S.64 3.35 17.705 175 0.940 22.636 56 1 2.138-84 C 125856 S.64 9.76 66.384 68 25.904 71.155 23 5 431848 8.64 33.49 245.908 IS 89.P47 258.759 7 L Se485 0.75 3.99 34.757 295 3.377 28.432 385 2 2.488-44 C 347331 S.75 5.81 73.484 38 1 9.227 77.339 64 5 584781 0.75 19.92 201.409 29 31.317 209.564 19 L 57963 S.05 1.35 S.639 435 3.026 15.231 321 3 2.875-84 C 3663e2 0.05 3.94 72.853 349 5.3e7 79.396 131 W 570530 0.05 13.58 330.075 43 38.174 316.618 32 L E3296 S.94 S.58 -27.832 1834 S.65e -19.987 099
. 4 3.323-84 C 394575 S.94 3.69 44.174 348 3.903 51.279 300 5 632914 S.94 5.79 387.329 38 1 6.535 334.434 86 metes L C, and 5 represset setteetoe beoed se lower, central, and upper source tere values. .
S
(
\ . j
~
l umAL QUAD CmES NSIGHTS NO REACTE MANUAL CONTROL OF SAFETY / RELIEF VALVE 8 '
NO REROTE CONTROL FOR RECIRCULATION PLAPS NO INTECRATED REROTE SHUTDOMA PANEL ONLY ONE 125 AND 250 VDC BAiImRY PER UNIT PLANT APPEAR 8 TO BE VULNERABLE TO UPSTREAM DAM FAILURES BOTH SAFETY SPREAD 1NG ROCN DIVISIONS ROUTED TNOUGH ONE CABLE BUILT WITH TM EE FOOT VERTICAL CABLE SEPARATION VENT STACK COLLAPSE COULD IhPACT 4180 VAC SWITCHGEAR f
CONDENSATE STORAGE TANK I8 NOT SEIS4IC CATEGORY I g .
kb m -
l- .
l l
)
CHANGES SINCE PRELIMINARY DRAFT i
s ADDED PUWP AND VALVE COlWMON IWCDE FAILURES 1
INCREASED CONTROL ClRCUIT FA1 LURE l
- ADDED SUPPORT, SYSTEM INITIATING EVENTS REEVALUATED OTHER DATA
- RESULTS CORE IWELT PROBABILITY INCREASED BY A FACTOR OF 2 Q't -.
I i
/
Y
.s MOST SIGNFICANT VULNERABlUTIES - INTERNAL i
- 1. LOCAL FAULTS OF TWO DIESELS- '
! 2. FAILURE OF DIESEL FIELD FLASHING l
- 3. DIESEL COOLING WATER FAILURE i
i
i l
k#
v .
!h -
e l .
l . .
l/
I INTERNAL EVENT MODIFICATIONS i
l 1. ADDITION OF A FOURTH DIESEL '
- 2. DEDICATED BATTERY FOR DIESEL 1 J
t
- 3. INSTALL ADDITIONAL DG COOLING WATER PLDP '
t
- 4. PROVIDE AUTO TRANSFER OF DC LOADS i -
l
}
.-4 -- . - --e -__ _ _ .- _ _ -
Il s
P FIRE ANALYSIS -
1 ,
VULNERAB1LITIES l CONTROL ROOM FIRES 6.7E-6 CABLE SPREADING ROOM FIRES 5.8E-6 1.3E-5 PER R-YR MODIFICATION l
REVISE PROCEDURE FOR SAFE SHUTDOWN PLAWP OPERATION ;
CONTROL ROGA FIRES 1.5E-6 CABLE SPREADING ROOM FIRES 1.3E-6 2.8E-5 PER R-YR i
l r .
9
/ .
SESIMIC-RELATED VULNERABilmES FAILURE OF BATTERY RACKS FAILURE OF 4160 VAC BUS CABINETS
~
SEISMIC MODIFICATIONS INSTALL lETAL BATTERY RACKS ANCHOR TOPS OF 4160 VAC BUS CABINETS I
I
C
/
DEFINmON OF DHR ALTERNATIVES ALTERNATIVE 1 FOURTH DIESEL GENERATOR AUTO TRANSFER DC LOADS .
ALTERNATIVE 2 i-DEDICATED BATTERY TO DG 1 '
TH IRD DG COOL ING WATER PLAP AUTO TRANSFER DC LOADS ALTERNATIVE 3 SAME AS ALT 1 PLUS -
ENHANCE PROCS FOR SSP -
NEW BATTERY RACKS ,
ANCHOR AC BUS CABINETS 7y -
g -
w 6
e DEFINmON OF DHR ALTERNATIVES ALTERNATIVE 4 SAME AS ALT 2 PLUS ENHANCE PROCS FOR SSP NEW BATTERY RACKS ANCHOR AC BUS CABINETS ALTERNATIVE 5 ADD-ON DHR SYSTEM b
,g -
\
/
i i i ! QUAD CITES RESUI.TS FOR ALTERNATWES AltornatIvo Base Case 1 2 3 4 5 Internal 1.8E-04 7.4E-05 1.3E-04 7.4E-05 1.SE-04 5.5E-05 Se l ani c
! 8.3E-05 8.3E-05 8.3E-05 2.8E-05 2.8E-05 4.3E-05
- Fire 1.3E-05 1.3E-05 1.3E-05 2.8E-06 2.8E-08 1.3E-05 l Internal i Flood negligible - - - - -
- l External j ,
Flood 9.8E-08 9.8E-08 9.8E-08 9.8E-08 9.8E-08 9.8E-08 Extreme Wind 1.4E-07 1.4E-07 1.4E-07 1.4E-07 1.4E-07 1.4E-07 l Lightning 1.7E-08 1.7E-08 1.7E-08 1.7E-08 1.7E-08 1.7E-08 2.8E-04 1.7E-04 2.3E-04 1.1E-04 1.8E-04 1.1E-04
. Change in core melt 1.1E-04 5.4E-05 1.7E-04 1.2E-04 1.7E-04 Note: all values are per reactor-year of operation l
~
I w
s i
- 1. Y -
~
f! '
SUMMARY
OF VALUE-lMPACT ANALYSIS i I DELTA NET BEN $ NET BEN CORE $
OFFSITE PER TOTAL PER ALT IELT ($XE-6) P-REM
($XE-6) P-REM
- 1.1E-4 1 -14.7 5400 -11.4 4200 1
- '2 5.4E-5 - 5 ~. 0 4000
- . -3.3 2800 3 1.7E-4 -13.7 3400 -8.4 2300 4 1.2E-4 -4.0 1800 -0.4 700 5 1.7E-4 -86.5 2E4 -82.6 2E4
- 6 2.6E-4 -86.3 1E4 -82.4 l
1E4 t u, r
x 4
~
SUMMARY
OF VALUE-lMPACT ANALYSIS I
l (NO INTERDICTION) r
[ DELTA NET BEN $ NET BEN $
l CORE OFFSITE PER TOTAL PER ALT hELT ($XE-6) P-REM ($XE-6) P-REM
[ 1 1.1E-4 +20.4 270 +23.7 210 2 5.4E-5 +11.5 200 +13.2 150 l 3 1.7E-4 +41.1 170 +46.4 120
- i. 4 1.2E-4 +32.2 90 +35.8 40 5 1.7E-4 -32.5 940 -28.5 900 l 8 2.8E-4 -3.4 620 +3.5 570 l I
m b .
i DESIGN CRITERIA 0 POINT BEACH IS BASE PLANT 0 SYSTEM INITIATION PRIOR TO S/G DRYUP (I HR.)
O COCL TO LOW PRESSURE RHR ENTRY CONDITIONS IN 3 HRS FOLLOWING INITIATION O LIMIT SERVICE WATER IEMPERATURE RISE TO 1QOF 0 MINIMUM DEPENDENCE ON OR INTERFACE WITH, EXISTING FACILITIES l
i l
l
/
1 0 04 p. stew 4-.uness---I, '
j
- 4. .. . . . .
'. = = >
G - i...
x u. .m.I
.p.
9'd
,e g - .
f.M c
I,
, ~
?
p in. ,
g g , O h
.I es 33' p}
o-1 : - .
e :o{o{ d H ij J $' .s I
b
((
I g i EE cH : EH( y g 5
o-i( It 3E E*
Y E ll
=c == ..
El CH : 04( h I T L ;
I I I a
l l ' TV ..
1 1,a
- II lr &
I I i
I I .
I I > .
ln
. lag - gg-
- 8. .
' ^
_ha.4 m
DIFFERENCES FROM GAR PROPOSAL 0 PRESSURE / FLOW CONTROL AT FLASH IANK 0 FLASH IANK BOTTOMS COOLER ON PUMP SUCTION O
DIESEL GENERATOR AND AUXILIARIES LOCATED IN FLASH TANK BUNKER 0 CONDENSER AND BOTTOMS COOLER COOLED IN PARALLFi 0 VENT LINE RETURNED TO RC DRAIN TANK 0 ADDED PRESSURIZER SPRAY LINE MM
r / .
4 ., .
i x 3E l uv s==mw/ yg iG 1
(o -
-
- M *
>j>
i..r ,,- ~,, L ' O ., ' ' ' - .
<,:=>::.. : . . . , , 1 _"
O .,
vW Y
+ "-
t- .
e , l.a I
. l .'g r; d
s 3 i se M(
1.I Mi M :
El*
c j I
i I
I ,
2__
,. i 5 :;f -
I" 32 ii I
- a [ 1 1,
g . .
J i a 1- .
l.g VI
- s .
Eg g .
I I ETALDATED 005T COMPARISCII
($ in Thousands)
BASE WITH PRIMARY RICR PRESSURE BASE RHR BLOWDOWN RRR 23,173 32,983 39,988 36,421 Direct cost ,
19,206 27,879 33,390 30,412 Indirect cost
. TOTAL 42,379 60,862 73,378 66,833 contingency 10,595 15,215 18,345 16,708 Owners Cost 4,238 6,086 7,338 6,683 Escalation & AFUDC 8,580 12,094 16,582 13.281 TOTAL 65,792 94,257 113,643 103,505
\
V s
b i k