ML20005B918
| ML20005B918 | |
| Person / Time | |
|---|---|
| Site: | Comanche Peak  |
| Issue date: | 09/01/1981 |
| From: | Brink B CITIZENS FOR FAIR UTILITY REGULATION |
| To: | NRC OFFICE OF THE EXECUTIVE LEGAL DIRECTOR (OELD) |
| References | |
| NUDOCS 8109160118 | |
| Download: ML20005B918 (27) | |
Text
.
September 1, 1981 UNITED STATES OF AMERICA Eunm coluel>o. sos.m KUCLEAR REGUI.ATORY CC IISS10!.
BEFORE THE ATO.IIC SAFETY AND LICENSII:G ECA3D IN the .latter of TEXAS UTILITIES GENERATING Docket Nos. 50 h45
.COI.1FANY , ET AL. 50-446 (Comanene Peak Steam Electric (Applicatior. for Station, Units 1 and 2) Operating License)
CRUR'S FIRST AND SECOND SET OF
.SUFFLEIfENTAL ANSWERS TO NRC STAFF'S FIRST AND SECOND SET OF INTERROGATORIES Citizens for Fair Utility Regulation (CFUR) files this its partial supplemental answers to NRC Staff's 1st and 2nd set of Interrogatories, pursuant te Board Order.
CFUR files answers for the following interrogatories:
Cl-3, Second supplemental answer C 9 '3, 4 and 6, second C3-2, Second supplemec.'al answer C3-3, Second supplemental answer 53 @;
C3-4, Second supplemental answer C3-5, Second supplemental answer -
%se
"'T W 9198tf ,, C.
-1 C3-10, First supplemental answer L- %
Otriceef;3j Dc:1.eting ry 2 C3-11, Second supplemental answer g Emach C3-12, First supplemental answer fi m C3-19 through 22, First supplemental answers o) \ -
@/ '
C4-10,First supplemental answer /f C4-12, First supplemental answer j G l 5 19 4 Sf C4-15, First supplemental answer =D f Ci
/
CFUR will file additional ans.lers to additional inte forthwith.
>L fbf 8109160118 810901 3500 PDR ADOCK 05000445 ggl g PDR
c- ,.
. Cl-3 2nd Sucole entarv Ansvar A) "Agreenent of Settlenent ('Agreenent') dated Decenber 16, 1977 between Westinghouse Electric Corcoration and Owners of Conanche Peak.
Fron nage 13 of Agreenent: "'Testinghouse will orovide to Owners free of charge the effort required for the orecaration and defense of the Conanche Peak FSAR sections within the Westinghouse scope of resconsibility in.accordance with the requirenents of Regulatory Guide 1.70, Revision 2,
' Standard Fomat and Content 'of Safety Analysis Reparts for Nuclear Power Plants - IMR Edition', dated October 16, 1975. ('Rev. 2')"
Fron page H-6 of Agreenent: " Vendor, free of charge to Purchaser, shall implenent the provisions of ... Rev. 2 for those cortions of the CSAR for Conanche Peak which are within Vendor's scope of supply as identified in Revision 6 of the FSAR Scope Split dated Novenber 23, 1977. ...
Vendor's portion of the Rev. 2 changes are in Chapters 1,3,4,5,6,7,9, 11,12,15, and 16 of the FSAR."
B) CrUR asserts that the Applicants relied on Westinghouse to originate the following portions of the CPSES/FSAR:
Sections 1.5,1.6, Aopendix J AN, 3.6N.1, 3.6N.2, 3.7N, 3.9N.1, 3.9N.2, 3.9N.3, 3.9N.4, 3.9N.5, 3.10N, 3.11N, 4.1, 6.2, 6.3, 4.4, 4.5, h.6, Appendix 4A, 5.1, 5.2.1, 5.2.2, 5.2.3, 5.2.4, 5.3, 5.4.1, 5.4.2.1, 5.h.2.2, 5.4.3, 5.4.4, 5.4.6, 5.4.7. 5.4.10, 5.4.11, 5.h.12, 5.h.13, 5.4.14, 6.1N, 6.3, 7.1, 7.2, 7.3, 7.4, 7.5, 7.6, 7.7, 0.1.1, 9.1.2, 9.1.4, 9.3.4, 11.1, 11.3, 15.1.1, 15.1. 2, 15.1. 3, 15.1.h . 15 1. 5, 15. 2.1, 15. 2. 2, 15. 2. 3, 15. 2.4, 15. 2. 5,
- 15. 2. 6, 15. 2. 7, 15. 2. 8, 15. 3.1, 15. 3. 3, 15. 3.4, 15.4.1, 15.4. 2, 15.h . 3, 15.4.4, 15.4.6, 15.4.7. 15.4.8. 15.5.1, 15.5.2, 15.6.1, 15.6.2, 15.6.3, 15.6.5, 15.7.1, 15.7.2, 15.7.3, 15.7.4, 15.8i 15.8.1, 15.8.2, 15.8.3, 15.8.4, 15.8.5, 15.8.6, 15.8.7, Appendix 15A.
C) CFUR asserts that the Applicants relied on Westinghouse to provide input, review and connent on the following portions of the CPSES/FSAR: ,
Sections 1. 2. 2, 1. 3.1, 1. 3. 2, 1.h , 1.7, 3.1, 3. 2. 3. 5, 3. 8. 5. 2. 5, 6. 2.1,
- 6. 2. 5, 8. 3.1, 9.h , 11. 2, 11. 6, 12.1. 2, 12. 2.1, 12. 2. 2, 12.4, 16. 2, Appendix 17A D)' CRTR asserts that the Acolicants relied on Festinghouse to review the following nort$ons of the CPSES/FSAR:
Sections 1.1, Accendix 1 AB, 3.73.1, 3.78.2, 3.73.3, 3.73.h , 3.7BA, 3.9N.6,
- 3. 93.1, 3. 92.2, 3.93.3, 3.93.4,- 3.98. 5, 3.93.6, 3.103, 3.119, 5.h .5, 5.h .9, 6.1,6.2.2,6.2.3,6.2.4,6.4,6.5,8.1,8.2,8.3.2,8.3.3,91.3,9.2.5,
- 9. 3. 3, 9. 5.1, 9. 5. 2, 9. 5. 3, 0. 5.h , 9. 3. 5, 9. 5. 6, 9. 5. 7, 9. 5. 8. 10. 3, 10.h .1,
Cl-3 (cont) 10.4. 2, 10.4. 3, 10.h .h . 10.4. 5, 10.h . 6, 10.u. 7, 10,h , g , lo,h ,9, 3 0,g ,79, 10.h.11,10.4.12,10.h.13, 3 0.4.15,10.h.16, Accendix 11 A,12.3,1,
- 12. 3. 2, 12. 3. 3, 12. 3.h , 14 . 2.1, 1h . 2. 2, 14 . 2. 3, lh . 24 , 1h . 2. 5, 14. 2. 6,
- 14. 2. 7, 14. 2. 8. 14. 2. 9. 14. 2.10, .h 2.11, 1h . 2.12, 15. 7. 5, 17.1, 17. 2, 10.4.14 l
I M
CFDR's Supplemen.tal Answers to NRC Staff Interrogatories, cont.
C 3 -2 In addition to CFUR's first supplemental answer, CFUR responds:
In NUREG-0737, " Clarification of TMI Action Plan Requirements,"
Task I .C .1, the NRC staff found what they cbscribe as " recurring deficiencies" in the guidelines being developed for the evaluation and development of procedures for transients and accidents. The staff comments:
"Specifically, the staff has found a lack of justification for the approach used in developing diagnostic guidance for the op-erator and in procedural development. It has also been fcund that although the guidelines take implicit credit for oreration of many systems or components, they do no' address the availa-bility of these systems under expected plant conditions nor do they address corrective or alternative actions that should be performed to mitigate the event should these systems or compon- '
ents f ail . . . . " (P . I .C .1-2) .
"The analyses conducted to date for guideline and procedure development contain insufficient information to assesr the extent to which multiple failures are considered. NUREG-0578 concluded that the single-failure citerion was not considered appropriate for guideline development and called for the consiceration of mul-tiple failures and operator errors. Therefore, the analyses that support guideline and procedure development should consi-der the occurrences of multiple and consequential failures. In General, the sequence of events for the transients failures such that, if the failures were unmitigated, conditions of inade-quate core cooling would result.
" Examples of multiple failure events include: Multiple tube ruptures in a single steam generator, and tube rupture in more than one steam, generators failure of main and auxiliary feed-water; failure of high-pressure reactor coolant makeup system; an anticipated transient without scram (ATWS) event following l a loss of offsite power, stuck open relief valve or safety /re-lief valve , or lospor main feedwater; and operator errors of I ommission or commission.
"The analyses should be carried out far enou to assure that all relevant thermal / hydraulic /gh into the neutronic event -
phenomeria are identified... Failures and operator errors during the long-term cooldown period should also be addressed.
"The analyses should support development of guidelines that de-fine a logical transition from the emergency procedures into
.the inadequate core cooling procedure including the use of in-strumentation to identify inadequate core cooling conditions.
Rationale of this transition 7hould be discussed. Aditional in -
formation... includes:
1.A detailed description of the methodology used . . .
- 2. Associated control function diagrams, sequence-of-event ciagrams, or cthe rs , if used; 3.... bases for mulitple and consequential failure considerationc;
- 4. Supporting analysis, including a description of any computer codes used; and 5.... description (s) of the applicablity of any ceneric results to
CFUR's Supplemental Answers to Staff Interrogatories, cont.
1 4
C3;^ cont.
plant specific applications....
"Pending staff approval of the revised analysis and guide-lines, the staff will continue the pilot monitoring of emergency i procedures described in Task Action Plan item I .C 8 (NUREG-0660.) l i
For PWR's this will involve review of the loss of coolant steam-generator-tube rupure, loss of main feedwater, and inadequate l core cooling procedures.'"
The nodified computer codes must be tested to obtain a measure of absolute accuracy of the calculated values with respect to some recognized standard such as the small-break 16CA sequences tested with Semiscale and/or LCET. The margins of safety (e.g., if operator had waited an additional length of time, the allowable cladding temperature would be exceeded) and are required to be determined.
From 10 CFR 50.34 (a)(4) and (b)(4): "A final analysis...with the objective of assessing the risk to public health and safety resulting from operation of the facility and includ-ing determination c' (i) the margins of safety during nor-mal operations and transient conditions anticipated during the life of the facility and (ii) the adeo.uacy of structures systems, and components provided for the pervention of ac-cidents. . .and taking into account any pertinent information developed since the submittal of the preliminary safety analy-sis report."
From 10 CFR 50.34 (b) and (b)(2): "The final scfety analysis report shall include information that prece:.t; a safety analysis...of the facility as a whole, and shall include...
evaluation...to show thm safety functions will be accomplished."
Failure to modify computer codes to account for the parameters cited in NUREG-0737 will lead to adoption of codes that will not realistically predict plant behavior. In addition, a number of other tasks in UUREG-0737 and the Applicant's compliance to them in NUREG-0797 are relevent to this contention.
Part A I.A.l.1. Shif t Technical advisor - potential for operator error I.B.l.2. Independent Safety Engineering Group -- operator error.
I.A.2.1. Immediate upgrading of training - operator error.
II.B.l. Reactor Coolant System vent.s - noncondensible gases II .:B .2. Dec[%h Review Systems used in Postaccident Operations -
Hydrogen and noncondensible gas . levels.
I .D .1 Control room design features - operator errcr, misleadint indications.
II.L.3 Direct Indicaticn of Relief and Safety Valve Positions -
Misleadir:g indications.
II.E.1.1 Auxilliary feeduater System Evaluatin - maintencnce errors, misleading indications II.E.3.1. Emergency Power - Pressurizer Heaters - Maintenence errors
-S-
CFUR's Supplemental Answors to Staff Interrogatories, cont.
C 3-2, cont.
II.F.1. Accident Monitoring Instrumentation - operator error plus misleading indications II.F.2 Instrumentation, Inadequate Core C ooling - operator error and misleading indications II.K.1.10 Procedures, Operability status of mainknance items -
operator and/or maintenance errors SMALL BREAK II.K.2.13 LOCA W/O Auxilliary Peedwater--analysis of Small Break LOCA 's II.K.2.6 Reactor Coolant Pump Seal Damage -- limits validity of small break LOCA model II.K.2.17 Potential for voiding in RCS during Transients -
Small Heak LOCA II.K.3.1 Installation of Automatic PORV Isolation System - small break LOCA II.K.3 2 Overall Safety Effect of PORV Isolation System - small break LOCA II.K.3 5 Automatic G rip of RCP during LOCA - maintenance error small heak LOCA II.K.3.30 -Revised sma11 break LOCA methods to show compliance with 10 CFR 50 Appendix K IIK. 3.31. Plant Specific Calulations for CPSES -small break LOCA 's II,X.3.46 Natural Circulation in Depressurization of RFU during small break LOCA 's.
e N
CFUR's Supplemental Answers to NRC Staff Interrogatories, cont:
C)-3 The computer codes utilized in the accident sequence analyses contained in the CFSES/PSAR, Chapter 15, do not in general, sufficiently account for hydrogen generation in the primary coolant loop, operator errors, maintenance errors, multiple faiures, consequential failure, equipment failures of a secon-dary nature, misleading indications from instrumentation and central systems, non-condensable gases and small break 10CA's.
As a res ult, these analyses, in general, are deficient to the point of providing in,, accurate answers.
C 3-4 This response supersedes CFUR's supplemental response to inter- .
rogatory C3-4:
The computer codes addressed in Contention 3 are the following:
LOFTRAN, SATAN, LEOPARD, FACTRAN, COCO, TACT, THINC , TWINKLE, LOCTA, WREFLOOD, TURTLE, WFLASH.
CFUR contends that the following computer codes ured in Section 15 of the CPSES/?SAR, in general, produce inaccurate answers for the following reasons:
1.) There are insufficient allowances for operator and/or mainten-nance error.
2.) The single-failure criteria interpretation used inthese codes are tco restrictive in that they do not sufficiently allow for analysis of multiple failures, consequential failure , and equip-ment failures of a secondary nature.
- 3. The codes do nct sufficiently predict the consequences of a small break LOCA.
4.) There are insufficient allowances for the formation of hy-drogen and uncondensible gases in the. primary coolant loop.
5.) The computer codes do not sufficiently allow for the ramifi-cations form the intrumentation and central system, particolarly with regard to operator error.
CS-5 See UUREG-0737, item L.C .1. Also see response to interrogatory C 3-2, part A . These items comprise " pertinent information" under 10 CFR 50 34 (b) ' ': ) . Failure to include them in analysis of accident sequences will lead to use of computer codes providing inaccurate answers, ac iney pertain to parameters relevant to operations at CFSES.
C3-10 List of relevant parameters including but not limited to:
- 1. Overator errors. Errors of ommission and commission based on insufficient or inaccurate information and/cr failure to. proper-ly interpret data to titigate accident and transient consecuences at the proper point.
CFUR's Supplemndal Answers to NRC Staff Interrogatories, cont.
C 3-10 , cent.
2.) Maintenance errors. Failure to maintain equipment which could impact on safety functions in the primary coolant loop, the second-ary loop and the engineered safety features.
3.) Hydrocen formation. Creation of hydrogen in the primary cool-ant loop.
4.) Sincle Failure criteria-internretation. Two or more independ-antly occuring single failures that,if left onmitigated, would result in conditions of inadequate core cooling.
(b.) Conc'quential failure--tnose failures that occur as a result of one or more single failures.
(c.) Equipment failures of a secondary nature--failures of equipment to perform unspecified yet assumed functions in an accident sequence.
5.) Misleadinc indications. Failures of reactor instrumentaion systems to provide accurate and reliable operational data as well as the absence of direct indications.
6.) Noncondensable cases. Creation and/or release of noncondensable gases that are contained in the primary coolant loop.
7.) SMall-break 10CA consecuences. Enclosure 1, pages 9 through 12 of " Report of CFUR's Position on Each Contenticn", April 10, 1980, documents CFUR 's basis as to the following:
"The freqt.ency of the TMI accident is much greater than one in a million and would be categorized as a credible accident of either the infrequent variety or the limiting fault variety.
The two accident sequences analyzed in the CPSES/FSAR which most closely rewemblec +he start of theTMI sequence are the loss ofhormal feedwater flow sequence and/or the feedwater system pipe break sequence.
Those sequences utilize the LOFTRAN, FACTRAN, and THINC ccm-puter codes. Yet.none of these codes contain the capability of determining the amount of hydrogen generated during the accident sequer.ce...."
8.) TMI-2 accident descrintOn. '
The accident involved three different elements:
(a.) Maintenance Error: A number of maintenance activities were intertdned with the TMI-2 accioent. The actions of the foreman and two auxillery operators in an attempt to unclog a resin plug in a pipe leading from a con-densate polisher iniitated the sequence of events.
Prior maintenance activities also were contributing factors...the feedwater valves which were improperly lef t closed are ohious contributors. . .In addition. . .
sluggish response on the part of maintenance to cure problems with instrutentation in the control room also contributed, especially when it is recognized that maintenace tags obscured the view of some of the instruments which were operating normally.
,(b.) Equipment Failure: Although the pilot-operated relief valve performed norc. ally when it opened, it fail- d tc close properly which wil be referred to as its secondary function.
(c.) Operator Error: Besides not being trained for this accident sequence, the naterial available to operators had apparent-ly convinced them that the coquence of events which cccured
~g-
- CFUR's Supplementary Answars to NRC Staff Interrogatories, cont.
C3-10, cent.
at TMI-2 were improbable. The number of errors are lengthy and complex, and would take long to enumerate...certain contributors to these errors was the fact that the FSAR 4 was deficient but never challenged:
" Based on our training, it was impossible...if you look' back through everybody's training and theFSAR and safety analysis and thebuilding construction, you will not see a paragraph that-pro jects thattype of transient. (It) is so particularly foreign and unblievable that it has absolutely no significance. That's why nobody did anything about for two days. (NRC Soecial Inouiry Group, Three I4ile Island-A Report to the Commissioners and to the Public , I4itchell Rogovin, Director, January 19S0,
- p. 43.)
CFUR 's Partial Substantive Objections to Applicants ' Statement of Objections, July 23, 1960, fudher documents that NUREG-0578 mandates that "further analyses of small LOCA 's are needed" and that "mora 1d a different kind of analysis of accident analyses is nee' According to " Clarification of TMI Action Plan Req' 3", NUREG-0737, Task I .C .1, p.IC .1-2, "NUREG -oS78 conc u that tne single-failure criteria was not considered appropriate for guideline development (guidance for the evalua-tion and development of procedures for transcients and accidents) and called for the consideration of multiple failures and operator errors. Therefore the analyses that support guideline and procedure development should consider the occurences of multiple and consequential failures. In general, the sequence of events for the transcients and accidents and inadequa.te core cooling analysis should postulate multiple failures such that if the failures were unmitigated, ... inadequate core cooling would result...the analyses should.be carried out far enough into the event to assure that all relevant thermal / hydraulic //ntheok_
phenomena are identified... failures and operator errors during t' 3 long term cooldown preiod should also be addressed. The aimlyses should support development of guidelines that define a logical transition from the emergency procedures into the inadequate core cooling procedure including the use of instrumen-tation to identify inadequate core cooling conditions."
It should be note that in "CFUR 's 4th Set of Interrogatodes to Applicants" that in answer to interrogator 12, the Applicants do nn3 take exception to items I.C.l., II.K.3.30, and II.K.3 5 According to 10 CFh 50.34 (b) (4) , " A final analysis and evaluation of the design and performance of structures systems and components with the objective stated in paragraph (a) (4) ~regarding asses-ment of risk to public health and safety and ac.equacy of structures, etc., provided of accident for the p]revention consequences of this of accidents section and the and tacing mitigation into account any pertinent information developed since the submittal of the preliminary safety analysis report."
In order to comply with 10CFR 50.34 (a)(4) and(b)(4) , it will be necessary t, perform the analysis with " pertinent information"
-q.
CFUR's Supplementary Answers to NRC Staff Interrogatories, conte C 3-10. c ont .
regarding the parameters reflecting the sequence of events at TMI.
C 3-11
~
Operators errors will lead to inadequate responses to transient and accident conditions due to failure to properly interp t data or to rtid inadequate data which may be misicading. Mai nance errors will lead to failures in systems and components which ould impact on safety funcitons, inhibiting or eliminating their response to transcient and/or accident conditions.
Hydrogen generation would lead to potentially explosive mixtures which could wreck containment components or could . inhibit core cooling by blockage of coolant intakes or pump' '
1 =
Sigle failure citeria interpretation in too narr w a sense does not prevent development of muldple failure and consequential failure hazards which X'ould disable o/ seriously . restrain systems and components that impact on safety functions. Equipment failures of a secondary nature may undermine the consistency and effectiveness of primary function responses.
Misleading indications could induce operator error or could affect automatic operaticas of systems impacting on safety functions.
Noncondensable gases could lead to coolant flow olocxage, pump cavitations and creation of flux tilt conditions which may lead to partial core melt.
According to UUREG-0737, Task I .C .1, p.I .C .1-2, UCREG -0578 concluded that the single-failure criterion did not appropriately accennt for multiple failures and operator errors, nor "id it acccunt for consequential failure. As a result, new sequer.ces of events should postulate multiple failures, consequential failures and operator
- errors such that failure to mitigate would create conditions of l inadequate core cooling. Accident sequences should be carried out' l sufficiently far enough to identify relevant phenomena...ar.d should i consider use of instrumentation to identify inadequate core cool-l ing conditions.
Analyses of small break LOCA 's are required under TASK II .K.3.30,
- p. II .K.3.30-1 ci NUREG-0737 Both I .C .1 and II .K.3.30 are re-
. quired of applicants for operating license, including thore for-l CPSES. The Applicants e acknowledged no exceptions to those i
requirements in their a 3rs to CFUR 's 4th set of interrogatories.
I C 3 -12 -
l " Realistically predict plant behavior" means fo' recast the action
! or reaction of CPSES with a high level of confidence. The accident i
sequence analyses supplied for the Applicant at the tire this contention was writtn . cro . clearly inadequate to provice a prcter l
' basis for plant design and for the development of cperatcr trcin-ing programs and orera tint procr 'iures. The ap72icant fcilcd t:
analyse mas than the initial minutes of a transient, wheras such analyses should have covered a time period until a stable system had i bean assured.
l
-q O
CFUR's Supplementary Answers to NRC Staff, cont.
C 3-12, c ont The Applicant has failed to take into account maintenance errors as they affect both safety-gade and non-safety grade materials .
The applicant has failed to account for operatcr errors that compound errors. The Applicant has failed to consider multiple failures and consequential failures in their accident analyses.
In Adition, the applicant has failed to account for problems with non-condensible gases in the primary coolant loops. The Applicant has also failed to compensate for misleading indications for the instrumentation and c system.
CFUR contends that those changes proscribed in NUREG-0737, items I .C .1, II.K.3.30 and II.K.3 5 constitute " pertinent information" undar 10 CFR 50.34 (b) (4). The nuclear Regulatory Commission has pro-posed for licencing requirements rules on May 13, 1981,which in-corporata .. tem I .C .1 recommendations for " Analyses of small-break LOCAs and of transients and accidents that involve postulated multiple failures, consequential failures and operator errors which if unmitigated, could lead to inadequate core cooling". CFUR' contends the new NRC requirements are an attempt to recognise these analyses as "
10 CFR 50.34Cb)(4) . pertinent information" within the scope of Thisprof$edrulealsoacknowledgesthe"pertiner.ce"ofcarrying an analycle suffidently into the event to " identify all signifi-cant phenomena" and " address possible failures and operator errors during the 1cng-term cooling phase." Also, by proscribing analyses that "suppor t development of guidelines that define a logical tran-sition from the emergency procedures into the inadequate core cool-ing precedures including the use of instrumentation to identify inadequate core cooling conditiors." If failures of instrumenta-tion systems lead to creation and usage of misleading indications of core and primary coolant loop conditions, analysis of accident sequences with inadequate core cooling conditions are undermined sufficiently to prohibit realistic response to transient con-ditions. Consideration of the effects of misleading indications is crucial to the development of effective and " pertinent" accident analyses. Moreover, development of analyseu under item I.C .1, would be inhibited by a failure to accaunt for hydrogen formation in the primary coolant loop. Since such formations were shown to have blocked coolnnt flow at T:J.I, procedures to identify in e adequate core cooling conditions and to mitigate their consequences would be deficient and would fail to meet the criteria of 50.34 (b) (4) in providing a afficient analysis of " structures, systens, and components provided for the prevention o7 accidents and the mitigation of the consequences of accidents."
Development of accident analyses under item I .C .1, would be inhib-ited by a failure to account for noncondensible gases in the primary coolant loop. Since such formations were shown to have blocked coolant flow at T:iI, procedures to identify inadequate core cooling conditions and to mitigate their consequences would be de-ficient and would fail to meet the criteria c: 50 34 (b) (4).
Maintenance errors are " pertinent information" as they create po-tential for "pcstu]ated multiple failure , consequential failures, and operator errors, which if up.aitiga ccd , could lead to
, -it -
CFUR's Supplement ti Answers to NRC Staff, cont.
C 3-12, cont /
inadequate core cooling." On May 12,1981, in a letter to the NRC , the advisory committee on Reactor Safeguards argued that "a supporting infrastructure of procedures, information and trained personnel is as impcrtant to safe operation of nuclear power plants as capable licensed- operators." The ACRS said, it "has become increasingly aware of a lack of requirements for such support systems." Areas outlined by ACRS as its key concerns are: Availability of knowl~edgeable maintenance personnel who are familiar with plant-specific hardware... qualifications, such as training for service personnel responsible for instrumentation, electrical distribution, fluid system testine omputer software, and water treatment functions; criteria for cec _Jions, or the need for repair modification or replacement of equipment with observed in-service deficiencies; planning for procedural and record-keeping;ractices for maintenance and other service activities, work, controls and communications practices for service activities that may adversely affect public safety as a raslut of errors;. completeness and accessibility of maintenance and servicing information for power plant equipment; and conditions for using contract maintenance and service per-sonnel in place of trained licensee employees for supporting ser-vice functions.
Although CFUR does not considr the ACRS problea areas as wholly inclusive, we do consider these areas to be " pertinent infor-mation" under 10 CFR 50.34 (b) (4). CFUR contends that the use of " pertinent information" as defined in our basis, will lead to development of accident analyses and utilized computer codes that have the potential to " realistically predict plant be- ,
havior."
In Addition, CPSES is the first Westinghouse plant to atter.pt to operate without a boron injection tank. On a normal Mectinghouse PWR, the highest head source of water available is provided by the CVCS charging pumps and therefore , it is the most responsive safety injection system during high-pressure transients or ac-cidents.If the boran injection tank is deleted, the planned safety margin would be compromised.
The Applicants position (" Summary of Meeting on CP Design Change and Res May 26,ponses 1931) istothat RSB"the Questions" by S.B.
steam line breakEurwell is theofonly the NRC designStaff, basis accident (Chapter 15) for which credit is taken for the bcran in-jection tank." CFUR contends that the true value of a high break
' system is during a high pressure accident which may arise as the result of a small break or partia$ fuel blockage. Example:
While the toran injection tank may not be taken credit for in any accident sequence other than the large-break LCCA, the operator is left without the margin of safety provided by the high-hecd safety injection system on ' occasions when non-autonatic actions are called for during a hi6h-pressure' accident.
C 3 -19 Yes.
-l;L-
CFUR's Supplemental Answers to NRC Staff, cont.
C3-20.
We intend to challenge the accuracy of the following codes in the FSAR:
LOFTRAN, COCO, TACT, FACTRAN, TUINKLE, LOCTA, THINC , TURTLE, SATAN, WREFLOOD, LEOPARD, WFLASH.
C 3 -21 In general, these computer codes are not able to accept the following parame.ters:
1.) Operator errors: No evidence of allowance for errors of ommission or commission.
2.) Maintenance errors: No evidence of uilowance for errors in maintenance of primary, secondary and engineered safety features.
3.) Single failure criteda: Too strict an interpretation applied to the criteria leading to exclusion of multiple failures, consequential failures and equipment failures of a secondary nature.
4.) Small break LOCA: Inability to predict consequences accurately.
5.) Hydrogen formation: Inability to analyze hydrogen generation in terms of adequacy of core cooling.
6.) Noncondensible gases: No evidence of allowance for nonconden-sible gases in the primary coolant loop.
7.) Misleading indications: Inability to analyze effects of mis-leading indicators from instrumentation systems.
C3-22 The Applicants have not presented sufficient or satisfactory evidence that any accident sequence in Section 15 of the FSAR has i adequately compensated for these parameters. In fact, in the case of many of the stated parameters there is no clear evidence that the applicant has addressed the parameters in any substantite fashion. The Applicants have consistently chosen a represer.ta-l tive accident to analyze, ignoring accidents they consider less credible or probable. They have used a narrow set of parameters to be examined.Mostly those involved physical conditions within the core, primary coolant loop, pressurizer, etc. Yet, there is no clear evidence of a systematic attempt to incorporate in-plications of operator or maintenance error or the operational i
status of structures, systems and components designed to main-tain the proscribed conditions. By restricting their analyses to, a simple single failure transient or accident, they fail to concurrent failures, consequential failures or failures of equip-ment to re turn to pre-event status, thus threatening the ersdi-bility of an analysis based on narrow sets of parameters and credible failures.
The Applicants have not provided an analysis of a small break LCCA that embodies enough of the paraveters to permit rcalistic pre dic tiore: of CPSE3 cehavior.
CFUR's Supplemental Answers to NRC, cont.
C 3-22, cont.
NUREG-0737 includes a series of items (listed in response to Interrogatory C3-2) that address issues involving our parameters.
Since these items arose after TMI-2, it is not likely that they were addressed sufficiently in general, to permit incorporation-of these items in pre-TMI-2 computer codes and corresponding ac-cident sequences. Comparing these items with the .'oplicante responses to them in NUREG-0797, the Safety Evaluai.on Report for CPSES, a number of points are made.
First, critical assumptions regarding conditions and operational status of structures, systems and components at ~0SES were tade.
If those assumptions are undermined by the items of NUREG-0737, the potential for axistence and significence of our parameters is increased.
Second, many of the items in NUREG-0737 have been dealt with in an insuf#icient manner, as recorded in NUREG-0797 Among those in dispute are I.A.1.1, I.B.l.2, I.C.I., II.B.2, and II.F.1. A complete list is found in response to C3-2, part A. ,
Third, the Applicants have not presented sufficient evidence to show that the computer codes utilized in the CPSES/PSAR account for the items listed from NUREG-0737 or correspondingly for the parameters previously stated that may arise and show significnat rcsults of the failure to compensate for thoce in Part A.
CFUR contends that failures to provide timely resolution of the problems posed by the items of Part A leads to the dreation of the stated parameters, although they are not the only source of '
deviation. At minimum, the computer codes must be adjusted to provide for a comprehensive analysis of Part A items and their corresponding parameters to permit " realistic" predictions of
- plant behavior.
l l
9
CQ-3 Intsrrogstorv Do you oss rt in Contcntion 9 thit as a result of oonration of Conanche Peak, there will be " radioactive releases"? If so, state the basis for such assertion. Do you contend that such " radioactive releases" w311 result fron nornal coeration or on3y as a result of an accident? State the basis fi" your oosition in this regard.
Oricinal Answer a) yes. FSAR Sections 12 and 15, Contention h basis and the series of NUREG-0521 reoorts, " Radioactive Releases from Nuclear Power Plants (Year)".
b) Yes. Sane as (a)
Fotion To Connel Although the first part of CFUR's answer to this Interrocatorv centains sone of the information which the Staff seeks in Interrogatory C9-3, CFUR's response does net answer the second question posed in that interrogatory. By nerely answering "Yes" t o that quesilon, CFUd does i not state, as is requested, whether the radioactive releases with which it is concerned will result from nornal coeration of CPSES or only as a result of an accident. Accordingly, CFUR should be directed to answer that question.
Sucolenentary Answer (5/22/81) Sone radioactive releases will result fron normal operation, and it cannot be ruled out that sons will result from transients, accidents er incidents.
Telechone Contact 8/18/81 NRC Staff acknowicdges that Interrogatory has been properly suoplem/n.ed. Staff refuses .:diitional oral voluntary disclosure.
Voluntary Mritten Disclosure The object of this contention is to insure that planned bsteh releases of radioactive gases will be acconolished during neteorological conditions which nininize radiation exposures. During normal operation and anticipated operational occurrences, sone radioactive gas will be accunulated which will, on o ccassion, be released in a planned nanner to the atmosohere. D:tring some transient, accident and/or, incident sequences a somewhat larger amount of radioactive gas nay te successfully accumulated (not released to the atnoschere). It is possible that sone of these radioactive gases will also be released to the atmosphere in a planned nanner.
10CFRI20.1(c) states that the applicant should, in addition to c'o mplying with the requirenents set forth, nake every reasonable effort to naintain radiation exoosures as low as reasonably achievable.
Action taken by the Acolicant to nake nlanned hatch releases during neteorological conditions which nininize radiation exoosures (in addition to conolying with regulations sticulating cernissable levels of radiation, .fF-j
- _ _ . _ - _ - - _ _ _ _ _ - _ _ _ l
C9-3 (cont.)
radio ctivity in cfflutnts, d2 sign crit rie, cnd liniting conditions for coeration) co.olies with the requirenents of 10C:3J20.1(c).
9
[
f i
I f
I
EU.
Intorrognorv Stata tha bacio for your assartion that thors will be "cffcets of radioactive releases on the general public other than at the exclusion boundary".
Oricinal Answer The EEIR reports,10CFR Part 20 and 10CFR Part 50 including references cited.
Motion To Connel ... Neither the Staff, the Licensing Eoard nor the parties should be required to speculate as to the cortions of "the BEIR reoorts, 10 CrR Part 20 and IC CrR Part 50 including references cited" which contain the basis for C UR's assertion that there will be " effects of radioactive releases on the general oublic other than at the exclusion boundary".
Sueolenentary Answer (5/22/81) There is no reason that CWR can think of for assuming *. hat there will be effects of radioactive releases on the general public only at the excitision boundary. CrUR does not know wh-t the Staff is asking.
Telechone Contact 8/18/81 C7UR unable to obtain any further explanation af interrogatory. Staff insists that CFUR should provide page numbers of cocuments.as ordered by Board.
2nd Sucolenentary Answer 10CFR120.105 and 20.106 prescribe pernissable levels of radiation in unrestricted areas and radioactivi.ty in effluents to unrestricted areas. These requirenents apoly to all unrestricted areas - not just at the exclusion boudary. With respect to design objectives and liniting conditions for operation,10CrRI50 Aopendix I, Sec. III states that account shall be taken of the cunulative effect of all sources and oathways within the plant contributing to the particular type of effluent considered and 'that ectination of exoosure shall be nade with respect to such cotential land and water usage and food oathways as could actually exist during the tem of olant coeration.
In addition,100 R150 Accendix I states that the characteristics ,
attributed to a hypothetical receptor for the puroose of estinating internal dose connittnent shall take into account reasonable deviations of individual habits fron the average. These requirenents obviously apply to an evalustion ot the effects of radioactive releases on the l general public other than at the exclusion boundary (in addition to persons at the exclusion boundary).
Reginatory Guide 1.111 supplies nethods for estinating atnospheric transport and disoersion of gaseous affluents in routine releases fron light-water-cooled reactors and Regulatory Guide 1.145 supolies atnospherie l dispersion nodels for ootential accident consequence assess.ents at nuclear olants. Ueither rostrict an evaluation of the effect on the general oublic
'b
- 99 4 (cent) ,
. to just ths exclusion boundary.
Chanter IV of the BEIR I reoort Section II.B "Atnoscheric Discersion and Renoval Processes",Section II.D "Iand-based Nuclear Facilities",
Section IV.C "Other Sources of Radiocontamination",Section IV.D l
" Redistribution of Radionue11 des",Section IV.E " Radiation Effects In Soil", Sectien IV.F "Radionuclide Entry into Plants and Radiation Effects", and Section V.C "Aninal Produ.:ts as Sourc.es of Hunan Exoosure" document sufficiently that effects of radioactive releases on the general public occur other than at the the exclusion boundary.
Appandix VI " Calculation of Reactor Accident Consequences" of
'r: ASH-1400, Appendix II " Pollutant Pathways" of "Public Health Risks of Thermal Power Plants", UCLA-ENG-7242 by Starr and Greenfield, and
" Radioactive Releases From Nuclear Installations", Vol. 2, pp.17-152, by Clarke and MacDonald all document effects of radioactive releases ,
on the general public other than at the exclusion boundary.
I All of these documents (some of which CrUR does not consider conservative) substantiate that there will be " effects of radioactive releases on the general public other than at the exclusion boundary".
CFUR is not aware of any oroposed basis that there will be effects of radioactive releases on the general public only at the exclusion ;
boundary.
l .
i w . . . .. .
. C9-6 Interrocatorv Idsntify the "various transoort' nochanisna" referred to in the contention.
Original Answer See C9-5. More conolete answer provided with direct testinony.
Motion To Connel ...In these interrogatories the Staff is seeking an identification of the "various transoort nechanisns" referred to.. . -
CFUR's answers are totally unresponsive. ...
Sucolenenta: r Ansver (3/22/81) Those transport nechanisns detailed in "AIRDCS-EPA:
A Computerized Methodology for Istinating Environnental Concentrations and Dose to Man from Airborne Release of Radionuclides", Oak Ridge Nat'l Laboratory, TN, Decenber,1979.
Telechone Contact 8/18/81 Staff insists supplenentary Cs not specific enough.
Requests page numbers so when they obtain the document they will know where to look.
2nd Sunclenentary Answer Document nunber is P380-147838.
Modes of exposure include (1)irnersion in air containing radionuelides, (2)exuosure to ground surfaces contaninated by deposited radionuclides, (3')innersion in contaninated water, (h) inhalation of radionuclides in air, and '5) ingestion of food produced in the area. Atnospherie and terrestial transport nodels are included on pages 8 thru 30.
Methods of calculating radiation doses and intake rates by persons are included on pages 36 thru 54 Terrestial transport input paraneters are included on pages 81 thru 100.
The balance of the report contains the introduction, a section on how to use the code and listings of the code and sanple runs which may also .be helpful.
l 1
- lL D' -
C 4-10 1st Supplem2ntary Answer, CFUR to NRC Staff:
Those of the Lewis C ommittee. See attached copy of "CFUR'c auport on Each C ontention" dated 4-10-80, regarding C ontention 4.
C 4-12 ,
1st Supplementary Answer, CFUR to NRC Staff:
Report of the President's Commission on the Accident at Three Mile Island; those of the German Risk Study Summary, issued August 9, 1979 by the Federal Ministry of Research and Technology in West Germany; NUREG-0642, "A Review of NRC Regulatory Processes and Functions", p.p. 6-2, 8-3, 8-2, and 7-8 letter, C ouncil on Environ-mental Quality to John Aherne, March 20, 1980. See attached copy of CFUR's Report on Each Contention" dated 4-10-90, regarding Contention 4.
C4-15 1st Supplementary Answer, CFUR to NRC Staff:
Most probably. A hydrogen explosion could conceivably sever both normal and emergency core cooling as well as cause the integrity of the con-tainment structure to be violated if the hydrogen were to explode.
This limiting catastrophic event was not adequately addressed in WASH-1400. Yet, it is clear from the TMI accident that the possibility of sufficient accumulation of hydrogen in either the reactor and/or the containment for an explosion to occur is gretter tha( heretofone imagined. As established in the position statement for this contention, THI's accident occured with less than 500 reactor-years of coffreial op,eration. The TMI-2 accident is a credible accident. It is also a l known fact that a hydrogen explosion occured at TMI "At about 9h hours into the accident, the hydrogen in the reactor containment '
l building ignited....", IEEE Spectrum, The Technical Blow-by Blow,
- p. 42 Depending on the requirements concerning hydrogen venting, two l
positions are possible: 1.) Hydrogen venting of the primary coolant will be installed. Much larger quantities of hydrogen than that ex-perienced at TMI-2 will occur in the ovtt of a partial meltdown similar to TMI-2. Enough oxygen is present in the containment building to sup-port combustion. The hydrogen recombiners installed at CFSE3 would not be able to remove the hydrogen before a spart from either operator action or fror automatically controlled equipment ignited the oxygen.
-Is%
L. ,
C 4-15, c ont .
Therefore, this accident sequence should be evaluated. 2.) Hydrogen venting of the primary coolant system will not be installed. In the event of a partial meltdown, such as occured at TMI-2, non-condensible gases in the primary coolant loop present a problem in that restric. tion er blockage of pr'. mary coolant flow may occur. Such a sequence could lead to full meltdown with all the attendant hazards associated with cuch occurances. Steam and/or hydrogen explosions then present a serious hazard in that overpressurization of the containment may occur.
The amount of hydrogen vlich escapes to the containment building in the event that restriction or blockage of the primary coolant flow does not take place is directly proportional measure to the size of the f break in the primary flow. A break equal t,a or less than the TMI-2 {
break is not assured. l The hydrogen explosion that took place at TMI-2 caused a pressure spke of 28 psi. Had.more hydrogen excaped due to a larger break, due to more rapid formation or if the hydrogen had been ignited at a later ,
time, it is reasonable to ac;ume that the precsure spike would have been greater. CPSES has hydrogen recombiners installed in the con.-
tainment building which can be operated remotely from outside the con-tainment. But the design, paramenters used for the containment hydrogen monitoring system to be operational do not require this status until 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> after the accident. (CPSES/FSAR , p.6.2-83) As noted above, .
theT.I-2 hydrogen exploded at about 9 hours into the accident. THis .
I system would have no value in this circumstance. The CPSES recombiners j are designed to limit hydrogen concentration to or below four volume l percent based on the release model indicated in Regulatory Guide 1.7 i dated March 10, 1971. (CPSES/FSAR , p. 6.2-81 and 1A(B) -3) . This [
guide has been revised at least twice (S,ept. ,1976 and Nov. ,1978) ."
Even then, an exception to the guide is taken concerning assumptions ;
for the analysis of hydrogen production and accumulation in the con- ;
tainment based on the " maximum credible accident" (CPSES/FSAR , p. 6.2- :
i l 104, Figs. 6.2, 5A , 8 and 9) . According to the NRC model and assum '
ing total mixture, the volume percent of hydrogen would exceed 4%
after 25 days for the release rate of R.G.1.7 But what would happen in an accident sequence similar to TI-2, a credible accident? When l wo11d the hydrogen recombiners be turned on? Ey what procedure and ,
l according to what indication? At what rate nf would hydrogen be !
t i formed? At what rate would hydrogen leak into the containment? Mhat j
-lb- -
C4-15, c ent.
events may take place to ignite localized hydrogen? In the event of a small hydrogen explosion (20 psi) would the recombiners continue to -Zunction or would they turn into a source of oxygen to support successive explosions? .
CERTIFICATE I declare (or certify, verify or state) under penalty of per-jury that the preceding answers to NRC Stafft interrogatories are true and correct to the best of my knowledge. .
Executed on this SmL day of September,1981.
Betty kbink Respectfully submitted, s
s BettyB/ ink CFUR 7600 Anglin Drive Fort Worth, Texas 76119 l
l l
i l
l t
I t
f
- 1 1-
'QyQGMS (I & (7m ka< J. M ~~W/'^W'
. Position: At the tima of the TMI-2 accidsnt, such an cccurrence was thought to be " incredible." In fact, after the accident, the NRC staff found that:
"The accident at Three Mile Island Unit 2 it.volved a sequence of successive failures (i.e. small-break loss-of-coolant accident and failure of energency core cooling system) more severe than those postulated for the design basis of the plant. Therefore, we conclude that the accident at Three Mile Island was a Class 9 event." Matter of Public Sernee F.lectric and Gas Co. (Salem Nuclear Generating Station.
Unit 1), Docket 50-272, "NRC staff rescense to question no. 4 of ths Atc=ie Safety and Licensing Board" at 3 (e=phasis added).
C.:ITA does not agree with this categorization. Instead, CFUR contends that the 7MI-2 accident sequence was mis-categorized in the first place - mcre in line with the finding of the President's Cc=ission on the Accident at Three Mile Island when the found that: - -
"...the probability of occurrence of an accident like
- hat at Three Mile Island was high enough, based on WASH 1400, that since there had been nere than 600 reactor years of nuclear power plant operation in the United States, such an accident should have been exceeted durine that osried." Report of the Presicent's Cc=ission on the Accident at Three Mile Island 32 (1979) (anchasis added).
Clearly, if one accepts this premise, an accident even mere serious than Three Mile Island is credible in light of the fact that the TMI-2 accident occurred in less than 500 reacter years of operation. The problem is how to identify those sequences which fall in the credible category. WASH-1400 supplies information upon which to base the sequence. FIR-3 invoives a s=all LOCA vith an equivalent diameter of about 1/2 to 2 inches combined with failure of the contaiment spray injection system followed by containment failure due to overpressure. The median probability assigned this accident sequence by WASH-lh00 is 2 x 10
-6 ~
per reactor-year. However, the WASH-1400 soort was criticised in .a reevaluation by H. W. Lewis' Risk Assessment Grono initiated by the NRC. The Lewis Group concluded that WASH-lh00 failed to enchasize sufficiently the uncertainties involved in the calculation of probability and that the bounds of error on the estimates of accident sequence proba-bilities were greatly understated. In light of the criticisms of the WASE-1400 study made by the Lewis Cc=ittee, the Nuclear Regulatory Comission reexa:.ined its views regarding the MSH-lh00 study and made the following statement:
"The Comission accepts these findings (of the Lewis Comittee) and takes the following action Accedent Probabilities: The Com.ission accepts the Review Group Report's conclusion that absolud.e values of the risks presented by WASH-1400 should not be used uncritically either in tie regulatory process or for public policy purposes and has taken and will continue to take steps to assuro that any such use in the past will be corrected as appropriate. Irr particular, in light of the Review Group conclusions on accident probabilities, the Cc=ission does not regard as reliable the Reactor Safety Study's nu=orical eninate of the overall risk of reactor accident.
qW M& Y __ -__ _. . __ --
V-D In cdditien, the ACRS has cddressed this'issuoi
"...tha centainmant dasign pressure is based on the assumption that core melting will be =aintained and that no fuel =elting will occur.
The containment does not include provisions to cope with a molten core or the heat, hydrogen, and other asoects of an accident in which the whole ceremelts."
' "The single failure criterion and other failure control design bases should be modified as necessary to encourage consideration of pro-gressive, cemnon cause, and nultiple failures arising frem a single initiating event."
"Except for a few limited cases considered during the east few years, the staff has been unwilling to investigate potentially sirnigicant safety natters if they were not identified as pa-t of the ' design basis'. It's consideration of the ramifications of accidents invol-ving degraded safety features perfomance or other circunstances leading to accident consequences beyond those covered by the ' design basis' was toerestrictive, causing both industry and the regulatory staff to be inadequately prepared for anticipated accident circun-stances. There has been a salutary change in the NRC Staff views of such matters since the the TMI-2 accident that seems responsive to the need. Future organizational arrangements should assure that this interest will be sustained."
" Accidents beyond the current design bases should be considered in deciding on the future approach a to ... design, and to emergency measures."
, "...the SER consists pri .irily of repetitive ' boiler clate' which often tends to obscure and provide little amolification of safety issues. The result is that the SER has been=e a document of little value to those pecple responsible for safety reviews of nuclear facilities." h"JREG-06A2, "A Review of NRC Regulatory Processes and Functions' , p. p. 6-2, 8-3, 8 2, and 7-8.
No such salutory change in NRC Staff views is in evidence in this proceeding and ' boiler plate' analysis appears to satisfy the Staff. But the health and safety of the public in the vicinity of CPSES requires something more than this approach.
The Council on Envirerzental Que.11ty states:
"The past failure to discus 4 th g consequences of the full range of potential accidents and theirkundemines the basic purpose of the National Enviren= ental Policy Act te infor= the public and other agencies fully of the potential consecuences of Federal prcposals and to previde a basis for infomed decisions..; We do not believe the Co=nission's prior legal justification for severly limiting the discussion of nuclear accidents and their consequences in EIS's is any longer sustainable, assuming it ever was."
1.etter, CEQ to John Ahearne, March 20, 1980.
The following regulation is cited for justificatio..
"If...the information relevant to adverse i= pacts is inportant to the decision and the means to obtain it are not known (e. g. , the means for obtaining it are beyond the state of the art), tho agency shall weigh the need for .ie action against the risk and soverity of poss-ible adverso i=cacts vers the action to proceed in the face of the uncertainty. If the agenc'/ prococds, it shall includo a worst case analysis and an indication of the pretability c1r irarobability of it's occurrence." 40 CFR, part 1502,22(b) (1979).
M3
r . -
.. 4-5
, Th3 N S chnirntn, John Ahcarns, in answer to the CZQ, said a staff rece= mend-ation to abandon tce "old AEC policy" and to discuss serious accidents in enviror. . ental i= pact statements is now before the Co:=.ission and would be given prempt consideratien. ( Inside N?f, March 24, 1980, p. 5). An 574 staff paper (Secy 80-131) advocates, as an interim NRC policy, censideration of core nelt events in envire==cntal impact statenents and safety reviews.
( Inside h?I, April 7,1980).
In any event, CRTR contends that a PiR-3 accident is a credible a eident and the consequences of st.oh an accident should be calculated for CPSES..Even in the event that the . probability of the accident cannot be proved, enough ur-certainity exists that the accident should be evaluated to neet the "conser-vative requir-ments" of 10 CFR , Part 50. '
In addition, CFU?. contends that an accident sequence based on site-specific initiating events should be analyzed and the consequerees determined. The CPSES area is noted for the unusually high frequency ard intensity of tornadoes.
An accident sequence whereby every designated non-safety function is assumed to be demolished abruptly while both reactors are operating at full load should satisfy this purpose. A ce=bination of tornado-induced missles which initiate additional turbine-generator missles which destrey piping, condense s, and every other so-called. nonsafety ite could bc
_ considered as the initiating event.
I
- \
\
i l
l l
l k-h - - - _ - - _ - - - _
% # 9]
'E' .
c
- With resoset to th9 conpentnt parts cf ths Study, the Cemission ext ets ths Staff to nnko usa of them as appropriato, that 13. where the data base is adequate and analytical tachniques pernit. Taking due acecunt of the reservations exeressed in the Review Group Report and in its presentation to the Cc:=nission, the Cc=issier. suoports the extended use of probabilistic risk assessnent in regulatory decisio=aking."
(NFC Statement on Risk Assessnent and The Reastor Safety Study deport (WASH-lh00) In Light of The Fisk Assessment Review Group Report, January 18, 1979.)
Recently, Mr. Lewis noted that WASE-lh00 had it least identified the relative incertance of various accident types:
"For example, WASE-lh00 concluded that transients, small LCCA and hunan errors are i=portant centributors to overall risk. yet their study is not adequately reflected in the priorites of either-the research or regulatory grovos. These three items - transients, small loss-of-coolant accidents and human errors - were the central features of theThree Mile Island accident." ( H.W. Lewis, "The Safety of Fission Reactors", Scientific American (March 1980), p. 64)
This conlusion is shared in "The Geman Risk Study Su:-mary" issued August of 79 by the Federal Mir.istry of Research and Technology in West Gemany. The study concludes that 72 nereent of all hypothetical core-nelt accidents are caused by sna11 reactor pipe breaks. For this kind of accident, about two-thirds of the risk is in human failure and the remainder in equipment failure.
One reason why human failures create so much risk is that most postulated accidents ce=e from small reactor leaks such as occurred at TMI, not from large ,
pipe breaks. Large pipe breaks, which e=oty a lot of reactor water in a lhrry, have to be handled pre =otly and autcmatically, nostly without operator intervention. This is not so with small pipe breaks.
In light of the above, it is auparent that the probability of a small LCCA is much larger than' that used in the MSH-lh00 study. In like manner, naintsaance arror, operator error and /or equipment malfunction could contribute to the orobability of failure of the containment sorny injection systen. Error bou.ds determined fren the Lewis Connittee working papers ecuoled with use of hunan errer rates experienced under stress and the use of a 95% confiderce level vill establish this probability. Contain=ent failure by overurt sure is described as follows:
"According to an NFC source, contaiments were expected to withstand even core melt- until the nid-1960's when the idea becane'too exnensive' to consider. . . .The limit is now 50 usi but 'with margins' - it can
. withstand 100 psi. The IMI pressure spike went up to 28 psi. But other kinds of accidents - a stean exolocien, for instance - could cause pressure to exceed 100 psi, particuJ arly in a core nelt react- .
ion with concrete in which carcen dioxide, stean and hydregen nay be liberated." Inside NEC, *lalune 2. No. 7- April 7,1980, p. 7.
(
CERTIFICATE OF SERVICE I certify that a copy of the foregoing document has been forwarded to all parties of record this M/ h_,19El,byde-posit in the United States Mail.
Ad-dnistrative Judge Marshall E. Miller Mrs. Juanita Ellis U.S. Nuclear Regulatory Comission President, CASE Atomic Safety and Licensing Board Panel 1426 South Folk Street Washington, D. C. 20555 Danas, TX 75224 Dr. Forrest J. Remick, Member Atomic Safety and Licensing Board 305 E. Hanilton Avenue State College, PA 16801' Dr. Richard Cole, Member David J. Preiswer, Esq.
Atomic Safety and Licensing Board Assistant Attorney General U. S. Nuclear Regulaton Comission Environmental Protection Division Washington, D. C. 20555 P. O. Box 12548, Capitol Station Austin, TX 78711 Marjorie Ulman Rothschild, Esq.
Office of Executive legal Director Jeffrey L. Hart, Esq.
U. S. Nuclear Regulatory Co= mission 4021 Prescott Avenue Washington, D.C. 20555 Danas, TX 75219 Nicholas S. Reynolds, Esq. Arch C. McColl III, Esq.
Debevoise & Libernan 701 Connerce Street
, 1200 17th Street, N.W. Suite 302 Washington, D.C. 20036 Dallas, TX 75202 Docketing and Service Section Atomic Safety and Licensing Board Panel Office of the Secretary U. S. Nuclear Regulatory Comission .
l U.S. Nuclear. Regulatory Comission Washington, D. C. 20555 Washingten, D. C. 20553
,- @ R Atomic Safety and Licensing Appeal Panel 4 g U.S. Nuclear Regulaton Comission
/Y Washington, D.C. 20555
.;f 200. %
\
A U v=> m O w 3 38f' . -
- u. ' ~ . . ,,,, ~2]
$W G/
Q. l --
Yetty ink or i l .t > j CFUR 7600 Anglin Drive Fort 'clorth, Tenac 7511c i
!