ML19323B668
| ML19323B668 | |
| Person / Time | |
|---|---|
| Issue date: | 03/25/1980 |
| From: | Advisory Committee on Reactor Safeguards |
| To: | Advisory Committee on Reactor Safeguards |
| References | |
| ACRS-1715, NUDOCS 8005140049 | |
| Download: ML19323B668 (21) | |
Text
- '
.e MAR 2 51980
,[.g
'T NUTES OF THE ACRS SUBCOMMITTEE ON 8 0 051400*f g ABILITY AND PROBABILISTIC ASSESSMENT
@, RE
.,3 WASHINGTON, DC g
s s -
j,i FEBRUARY 6, 1980
.' f'i
.gQ C.
p\\
S" g7 The ACRS Subcomittee on Reliability and Probabilistic Assessment met with representatives of the NRC Staff on February 6, 198L te, discuss the following:
The development of a response to Congressman Udall's July 27, 1979 1.
letter concerning consistency of actual component failure experience with that projected in WASH-1410, and the probabilities of occurrence of the September 24, 1977 Davis-Besse and the March 20, 1978 Rancho Seco events using WASH-1400 methodology.
The deveSpment of a repsonse to Comissioner Gilinsky's December 18, 2.
1979 letter concerning nuclear plant risks versus risks from other electricity generating methods.
Development of quantitative safety goals for nuclear power plants.
3.
22, 1980 A notice of the meeting appeared in the Fedetal Register on January (Attschment A). A copy of the detailed presentation schedule is attached (Attachment B).
A list of attendees at the Subcomittee meeting is attached (Attachment C).
A list of documents provided to the Subcomittee is attached There were no written or oral public statements from members of (Attachment D).
The entire meeting was open to members of the public.
the public.
MEETING WITH THE NRC STAFF (OPEN SESSION) 1.0 Subcommittee Chairman's Opening Remarks Dr. Okrent, Subcommittee Chairman, introduced the members of the Subcommittee and noted the purpose of the meeting. He pointed out that the meeting was being conducted in accordance with the provisions of the Federal Advisory Committee Act and the Government in the Sunshine Act EM that.Mr. Gary Quittschreiber was the He stated that no requests for oral Designated Federal Employee for the meeting.
statements nor written statements from members of the public had been received with regard to the meeting.
t THIS DOCUMENT CONTAINS P00R QUAUTY PAGES rr -
1 Fabruary 6, 1980 2-RFA NRC Staff Comments 3egarding Consistency of Actual Component Failure _
2.0 trperience with that Projected in WASH-1400 W. Vesely, NRC Staff, concluded that new component failure rate data and WA d
point estimates for component failure rates are consistent, but that error sprea s In addition, there for active components should be a factor of 10 instead of 3.
He is a plant-to-plant variability of 10 to 30 above the generic failure rate.
indicated that the NRC Staff is reevaluating WASH-1400 with the new data and it appears that core melt could be about a factor of 3 greater than shown in UAsh 1400.
F. Rowsome added that he believed the frequency of core damage events for the industry is probably higher than suggested in WASH-1400 but that compensatory conservatisms would reduce the magnitude of releases and consequences.
ACRS Fellows Comments Regarding Consistency of the Actual Component Failure 3.0 Experience with that Projected in WASH-1400 E. Abbott, ACRS Fellow, discussed the difficulty one has in comparing component He noted that the rates are frequently failure rates from different sources.
based on different things, e.g., NPRDS may report a pump packing leak as a failure, when in fact the pump could still deliver proper flow if called on to He noted that most LERs are a result of entering the limiting condi-operate.
tions for operation (LCO) in the technical specifications; therefore, if aquip-Another ment fails when the LCO is not applicable, the failure is not reported.
problem with trying to determine failure rates from LER reports is that they D. Okrent suggested that the present report only failures and not successes.
NPRDS rulemaking procedure could be modified to provide better failure rate data collection if the Committee wished to provide comments in this regard.
E. Epler noted that of the dozens of traumatic events which have occurred'at nuclear power plants, he could not find any where component failure was the major contributor.
S l
V~
~
~" ~
~n-p
_,,MQ
g/
e Fabruary 6, 1980 RFA NRC Staff Comments Regarding the Probabilities of Occurrence:ofithe September i
4.0 24, 1977 Davis-Besse and the March 20, 1978 Rancho Seco Events F. Rowsome, NRC Staff, discussed the sequence of events that occurred at Kancho Seco on March 20, 1978 and at Davis-Besse on September 24, 1977 (Attachments I-5).
He pointed out that the probability one assigns to a historical event is entirely dependent on how broad a class of events one takes to represent the event. One can get a probability of any number between one and zero depending on how narrowly one WASH-1400 draws the class of events for which the probabilities are defined.
instrumenta-did not provide for the Rancho Seco event where loss of non nucient WASH-1400 did predict the loss of feedwater transient, but tion power was lost.
the probability was not appropriate to B&W plants because of design differences.
Rowsome indicated that it is exceedingly unlikely that the application of WASH-1400 techniques would have unfolded the precise details of the sequence of the He noted that in the Rancho Seco event the auxiliary feedwater specific events.
system was s. tarted when the steam generator level indication just happened to If the indication had not accidently drifted down there would have drift down.
beea no signal for cooling water and no instruments telli.g them they needed cooling water and a possible core melt could have occurred. Epler suggested this:should be looked at in more detail.
It was the opinion of F. Rowsome that using the WASH-1400 methodology to make absolute predictions on bottom line risk is the least trustworthy application of the WASH-1400 techniques. He suggested the tools be used to draw qualitative inferences on the strengths and weaknesses of systems.as an independent way of finding errors in design, operating procedures, maintenance techniques, etc.
F. Rowsome noted that system reliability is a " tricky measure of risk" because accident scenarios differ from plant to plant, such that it is not a uniform measure that can be applied to relate to risk. He suggested that PAS not attempt to provide the Subcommittee with a probability for the Rancho Seco and Davis-Besse events since orthing comparable to these events were covered in WASH-1400.
He said be could arrive at any number between 10-2 and 10-4 for these evects depen-ding on the breadth of the classification of events.
l
_ ~ -.. -
.~
u
-=
l
.i
- Febru ry 6, 1980 RFA ACRS Fellows Comments Regarding the Probabilities of Occurrence of the 5.0 September 24, 1977 Davis-Besse and the March 20, 1978 Rancho Seco Events W. Kastenberg discussed the ACRS Fellows c>nclusions regarding the probabilities of the Rancho Seco, Davis-Besse, and Three Mile Island-2 events using WASH-1400 methodology. He provided the following as best estimates for these events (Attachments 6-8).
J Davis-Besse 1.2x10-3/B&W reactor year i
)
Rancho Seco I.2x10-4/B&W reactor year 1.5x10-4/B&W reactor year TMI-2 S. Ditto felt that the probability of 8.6x10-3 per year that the ACRS Fellows used for loss of NNI-Y in determining the probability of the Rancho Seco event was too small.
W. Kastenberg discussed the WASH-1400 methodology for determining operator errors, when called upon in an emergency, as a function of time (Attachment 9).
His calculations assumed the operator had correct information and procedures.
people were reading more into the WASH-1400 human error rate calcula-Rowsome felt tions than what had been intended.
NRC Staff Discussion of Recent Findings on Coal / Nuclear Risk Comparisons 6.0 R. Gotchy, NRC Staf f, discussed some of the conclusions reached in the CONAES, 10).
SAI, Technicron, and NUB'IG-0232 comarisons of coal with nuclear (Attachment He noted that some of these studies have not factored in the latest known infor-mation uncovered during the GESMO hearings but that in the end it really doesn't l
make much difference. He felt that most of the studies being done simply take the results from previous studies.
R. Gotchy indicated that he has briefly reviewed the SAI coal / nuclear study and that there is no bottom line since it's primarily a matrix for interfacing dif-ferent models. He said the SAI work simply quotes work that others have done.
He noted an error in the dose conversion factor in ICRP for lead-210, which is a daughter of radon, coming from mining and milling. He indicated this as a dominant source of population exposure for uranium mining,and milling operations.
He indicated that the previous calculations were a factor of five to six higher
.w-o.e o.
=
e
]
. Februnry 6, 1980 RPA than they should have been due to an error in the consideration of critical organs.
H. Lewis noted that he was not an advocate of comparing coal and nuclear risks but that if it is to be done one should consider the low-probability /high conse-f L. Lave noted that arguments quence event of the coal cycle CO2 greenhouse effect.
would lead to warming effects of the earth which may can be made that the CO2 increase the productivity of the land. Gotchy added that it may also delay the next ice age and for these unknown e f fects, it is best not to consider it.
Gotchy indicated that there is a large ongoing campaign to discredit the health Lave noted that there are some good scientists who are interpret-effects of SO.
2 health effects as unproven, small, and even zero. Saunders compared ing the S02 the controversy with cigarette smoking, where you can get experts supporting both views.
ACRS Fellows Findings on Comparisons of Nuclear Plant Risks Versus Other 7.0 Methods of Generating Electricity D. Johnson, ACRS Fellow, summarized the report prepared by the ACRS Fellows com-paring risks associated with generating electrical power by coal, oil, hydro, and nuclear. In summary, he concluded that even though the various fuel cycles have components that are not quantifisd at present, that there appears to be a general consensus on the ranges and numbers of health effects and basic conclusions.
It was pointed out that even though there are several studies comparing risks of the different technologies that they are not necessarily independent.
8.0 Development of Quantitative Safety Goals for Nuclear Power Plants It was the general feeling of the ACRS consultants at the meeting that there is Lave no single quantitative safety goal number that can be found acceptable.
suggested desegregating the events into various consequences and probabilities and looking at ranges of estimates with emphasis on verifiability. He also sug-gested an alternative was to not set absolute number goals but rather use compari-sons among other electrical generating technologies and let people decide whether they desire electricity on that basis.
P. Slovik suggested that risk benefit Members analysis or decision analysis might be used f.o decide the safety goals.
of the Subcommittee indicated a full blown decision analysis taking several months to perform would not replace the need for specific quantitative safety goals.
Fabrusry 6, 1980 RFA 9.0 Problems Identified as Possible Safety Issues The following problems were identified as possible safety issues:
J. Ebersole discussed the possible use of quantitative criteria as a tool 1.
He postulated a failure of the Browns Ferry Plant's steam for solving problems.
line to the high pressure feedwater in the auxiliary building, which houses all the supportive and emergency equipment, just upstream of the DC driven external Ebersole postulated a pipe failure in the vicinity valve outside containment.
of the external valve which would disable the valve and the safety equipment in the A single failure of the valve inside containment along with the above room.
sequence would lead to a lack of core cooling.
C. Michelson postulated the possibility of spilling coffee in the operating 2.
console and suggested that one should look at the effects.
J. Ebersole noted that light bulbs in the control room could fall out 3.
during a seismic event and cause short circuits in open cabinets.
J. Ebersole expressed a concern that PORVs are depended on for feed and 4.
bleed and may, in f act, be gagged shut.
10.0 Concluding Remarks D. Okrent suggested that the most fruitful path for the Subcommittee to follow 1
f in developing safety goals would be for someone to write down some " proposed Okrent noted approaches" and then for the Subcommittee to review and comment.
that he intended to have something ready for the Committee to look at by early Summer 1980.
i The meeting was adjourned at 5:50 pm on February 6, 1980.
For additional details, a complete transcript of the meeting is available in the Nuclear Regulatory Commission Public Document Room, 1717 H Street, N.W., Washington, D. C. 20555, or from International Verbatim Reporters, Inc., 499 South Capitol Street, S. W., Washington, D.C. 20002.
@e G
- tme e e h We Ng 6 w eneee e
e v
Fedcral Register / Vol. 45, No.15 / Tuesday, January 22. 1980 / Notices 187
~
,,.ncrators (OTSG) and other features of Rancho Seco events using WASil.100 cnd any subsequent documents will be Babcock a'nd Wilcox designed nuclear methodology.
available for in<,pection and copying at
- 3. Itisk comparison of nuclear plants the Public Document Room.171711 plants.
- /legulatory Actirifics. htarch 5.1900, with other methods of electricity Street. N.W., Washington D.C. 20555.
Washington, DC.1he Subcommittee will generation.
Copics of the environmental report are D
ig review regulatory guides and revisions
- 4. Quantitative safety goals for also being provided to the State no existing regulatory guides; also, it nuclear power plants.
Planning Coordinator, Office of the pay dist.uss pertment activities which Further information regarding topics Governor. 2320 Capito! Avenue, sficct the current licensing process and/
to be discussed. whether the meeting Cheyenne, Wyoming 82002.
has been cancelled or rescheduled, the 3.The scoping process willinclude a ci!!
or reactor operation.
Chairman s ruling on requests for the meeting to be held in the Lodge Room of ACRS Full Committee hieetings opportimHv to present oral statements the Elks 207 E. Alain Street. Riverton, at allotted therefor can be frhrvary 10-12.15WO Wyoming, on February 13,1980 at 7:00 u. ; a prepaid telephone call to p.m.This meetmg wdl provide for a 7
A. *NRC Bulletins and Orders resulting otm.
from the Three Mile Island. Umt. Nuclear the cognizant Designated Federal briefing of interested parties concerning X)
L power Plant Accident.
Employee, hir. Gary Quittschreiber, the proposed action and alternatives D.
- ACRS Annual Report on the NRC (telephone 202/634-3207) between 8:15 and opportunity for comment on the Safety Research Program, a.m. and 5.00 p m.. EST.
scope of the proposed statement.The C.
- Proposed crite -ia for Mark I Dynamic Dated: January 16,1980.
participation of the public and all Containment.
D.
- Proposed operation of the Three Mile John C.llo3 e E
1 was Island. Unit 1. Nuclear Power Plant-Advisory Committee Afanagement Officer.
invited. Copies of this notice will be E.
- Proposed modification of NRC Criteria p uma m,d ie a es el mailed to all affected federal, state and fur Siting Nuclear Facihties, F. ' Proposed ACRS report on nuclear 8TG Coot MMW IoCal ageBCies, and other' interested will power plant component f ailure rates and persons. Written comments concerning S
abil tic assessment of nuclear plant
[ Docket No. 40-4492]
the scope of the proposed statement will be accepted until February 29,1980 G. *Recent operating occurrences at Federal-American Partners Uranium
- 4. After the environmental report has naclear facihtics Mill, Gas Hills Mining District, Fremont been analyzed. a draft environmental March 6-3,1960 County, Wyo.; Availability of impact statement will be prepared.The Environmental Repo t and Intent T DEIS is expected to be available to the
~
Agenda to be announced.
Prepare a Draf t Environmental Impact public for review and comment in June.
y APri/10-12,19s0.
Statement Concerning Renewalof a W80 Source Material License pmposd achn, Agenda to be announced.
8 Deted. january 17,1930.
AGENCY:U.S. Nuclear Regulatory DEIS. or scoping meeting and any lal'n C. Ilo>te, Co:nmission (NRC).
written comments should be directed to Advisory Committee Afanagement Officer.
ACTION: Notice of availability of D. hf. Gillen, U.S. Nuclear Regulatory tra twe so-iss ra,d i-ri-am e <s.*l environmental report and intent to Commission, Division of Waste sem coot nmm prrpare a draft environmental impact hianagemt nt,483-SS. Washington, D.C.
lly
'I.
statement and to hold a scopmg 20555, phone (301) 427-4103.
m pting.
^*
Advisory Committee on Reactor Safeguards, Subcommittee on
SUMMARY
- 1. Description of the Proposed day of f anusry,1980.
Reliability and Probabilistic Action-Federal-American Partners For the Nuclear Regulatory Commission.
-21.
20 Assessment; Meeting (FAP) has operated a uranium millin the Ross A.Scarano,
- ted The ACRS Subcommittee on Gt s liills hiining District of Wyoming Chief Uraniamaccoveryl.icensingsconch, since 1959 under NRC Source hiaterial Division of Waste Afanagement.
Reliability and Probabilistic Assese.nent License No. SUA-067. FAP is currently p gw,_,,,g,,,,,,,,,,,,,
will hold an open meeting on Fel.uary 6.1980. in Room ING. 171718 '.A., NW.,
applying to renew this Source h1aterial License and to obtain approval for a mill Wa shington. DC 20355.
ill The agenda for subject meeting shall expansion and use of a new tailings 21, be as follows: Wednesday, February 6, disposal system. The mill is located in
[ Docket Nos. 50-275 OL,50-323 OL]
1980. 8:30 a.m..mtil the conclusion of Fremont County on the Gas Ifills route i
port b ;siness, approximately 60 kilometers (50 miles)
Pacific Gas & Electric Co. (Diablo The Subcommittee will meet in east of Riverton. Wyoming.FAP,s Canyon Nuclear Power Plant, Units 1 l
executive r.ession with its consultants proposed plans call for the expansion of and 2); Change of Place of Argument and fellows to explore and exchange the mill process capacity from 000 W
opinions regarding the topics being metric tons of ore (950 short tons) per January 10.1980.
Itc discussed.The Subcommittee will also day to 2.680 metric tons of ore (2.950 The location of the oral argument in hear some brief presentations and hold short tons) per day and disposal of this cause scheduled for 9:30 a.m.,
discussions with representatives of the tailings into a mined out pit Wednestlay, January 23,1980, has been ibe NRC Staff.The following topics will be approximately one mile from the tuili.
changed irnm the U.S. Tax Court to the 9e
- 2. Pursuant to the National United States District Court. Courtroom discussed:
- 1. Consistency of actual component Eavironmental Policy Act of 1969 and No.12.19th Floor. Federal Building and failure experience with that projected in the regulations of the Commatoa in 10 Courthouse. 450 Colden Cate Avenue, WASil.1400.
CFR Part 51, FAP has filed aa b"
"#I8C#' #T.##I" a.
its
- 2. Probabilities of the Sep. 24.1977 environmental report in suppoit of their it is so ordered.
l Davis Desse and the Alarch 20,1978 applications. The environmental report t
&N~
' t-1ENIATIVE PRESENTAT104 SQlEDUIE
' ()
AmS SuneaumEF ETING RELIABILI1Y AND PROBABILISI1C ASSESSMENT FEBRUARY 6, 1980 1717 H Street, NW, WASHINGIW, DC R0041046 WEDNESMY, FEBRUARY 6,1980 Approximate Presentation Time Time 8:30 - 8:40 10 min D. Okrent - ACRS Openug Statement, introductory remarks 8:40 - 9:05 15 min W. Vesely - NRC/ PAS FEC.Statt conclusions regarding consistency of actual component failure experience witn that projected in WASH-1400 1
9:05 - 9:35 15 min E. Abbott - ACRS Fellow Fmdmgs concernmg tlE consistency of actual couponent failt.re experience with thatprojectedinWAFd-1400 9:35-10:15 Open Executive Session to discuss Subcormmittee conclusions on caponent failure rates 10:15-10:25 Coffee Break 10:25-10:50 15 min F. Rowsome - NRC/ PAS NRCStattconclusionsrefrdingthe7roba-bilities of occurrence o the Septem:er 24, 1977 Davis-Besse and the March 20, 1978 Rancho Seco events predicted on the basis of WASH-1400 failure rates and methodology.
10:50-11:15 15 min W. Kastenberg - ACRS Fellow Fmdmgs cw enung the probabilities of the Septem:er 'z4,1977 Davis-Besse and the March 20, 1978 Rancho Seco events using WASE-1400 methodology.
j 11:15-12:00 Open Executive Session to discuss the probabilities of the Davis-Besse and Rancho Seco events 12:00-12:30 15 win R. Gotchy - MtC/DSE and W. Rhyne - SAI Discuss any expected changes to le R 32 conclusions as a result of the p eliminary findings of the latest coal / nuclear risk report fitvHnga.
l 12:30-1:30 Break for Lunch l
L Q1 Jl *1
+
2:
Tentative Schedule February 6,1980 f
g Approxiznate Presentation Time Time 1:30 - 2:30 10 min each W. Kastenberg/D. Johnson /J. Griesmeyer ACRS Fellows Findings concerning the caparisons of nuclear plant risks vs other methods of generating electricity 2:30 - 3:50 Open Executive Session to discuss 91Wttee's conclusions on tuclear plant vs other electricity generating risks 3:50 - 4:00 Coffee Break 4:00 - 5:00 15 min each R. Wilson /L. Lave /P. Slavik - # ES Consultants r m ts on the development 01 ge ntitative safety goals for nuclear power plants 5:00 - 6:00 General discussion on quantitative safety goals 6:00 Adjournment l
l O
g t-
~ - - -
ATTENDEES LIST ACRS Members Miscellaneous D. Okrent, Chairman A. R. DuCharme, Sandia Labs J. Ebersole A. S. Heller, B&W W. Kerr R. Leyse, EPRI J. C. Mark D. H. Risher, ~W C. P. Siess M. A. Linn, TV M. Bender S. R. Blazo, Bechtel H. Lewis D. Walker, OPS S. Lawroski W. K. Brunot J. Dann, McGraw-Hill ACRS Staff G. R. Quittschreiber, DFE*
ACRS Consultants S. Saunders S. Ditto E. P. Epler N. D. Singpurwalla W. C. Lipinski P. Slovic R.' Wilson L. Lave C. Michelson ACRS Fellows W. Kastenberg D. Johnson E. Abbott J. M. Griesmeyer NRC Staff W. Vesely R. Bernero R. Rowsome
?. Manning P. F. Riehm R. Gotche Designated Federal Employee
- l l7 Li
DOCUMENTS PROVIDED 'f0 THE SUBCOMMIT1EE FOR THIS IEETING 1.
Viewgraphs shown at the meeting are provided as Attachments 1-10.
A complete set of all handouts are provided in the meeting transcript and in the ACRS Office file for this meeting.
2.
ACRS Fsilows Report, " Analysis of Feedwater Transient Sequences in B&W Nuclear Steam Supply Systems," dated February 1980.
3.
Nuclear Regulatory Commission Staff Report, " Evaluation of Davis-Besse and Rancho Seco Feedwater Transients on September 24, 1977 and March 28, 1978 using WASH-1400 Data" 4.
Memorandum from F. Rowsome to R. Fraley, "ACRS Query on Material Relevant to Udall Letter: Davis-Besse and Rancho Seco Transients," dated February 1980.
ATTACHMENT D s
,-r-.
-e s.___,
ALTERNATE QUESTIONS o
DID WASH-1400 CONSIDER OR PREDICT ACCIDENTS OF THIS TYPE?
o COULD WASH-1400 METHODS HAVE ALERTED ANALYSTS TO THE POSSIBILITY OF SUCH ACCIDENTS IF THE METHODS HAD BEEN APPLIED TO THE AFFECTED PLANTS?
a o
WHAT IMPROVEMENTS IN WASH-1400 METHODS OR DATA ARE NEEDED TO PROPERLY CONSIDER SUCH SEQUENCES q
IN RISK ASSESSMENT?
o CAN WASil-1400 M5Til0DS SERVE A USEFUL FUNCTION IN ANALYZING ACTUAL EXPERIENCES?
b
~
v
[
t DID WASH-1400 CONSIDER OR PREDICI TMI, DB, OR RS INCIDENTS?
I.
ACCIDENT CLASSES INVOLVING TRANSIENT-CAUSED STUCK-0 PEN PRESSURIZER RELIEF VALVES ARE CONSIDERED o
THE CONTINGENCIES AND FREQUENCY IN WASH-1400 IS INAPPROPRIATE TO B&W PLANTS II.
RS o
ACCIDENT CLASSES INVOLVING COMMON MODE TRANSIENT INITIATION AND DEGRADED RELIABILITY IN RESPONSE SYSTEMS ARE CONSIDERED
- o. THE COMMON MODE DEPENDENCE ON NON-SAFETY GRADE INSTRUMENTATION POWER SUPPLIES OF MAIN AND AUXILIARY FEEDWATER SYSTEMS IS NOT PRESENT AT m
SURRY
COULD WASil-1400 METHODS HAVE ALERTED ANALYSTS TO THE POSSIBILITY OF SUCH ACCIDENTS IF PERFORMED ON THE AFFECTED PLANS?
1.
FREQUENT EXPOSURE TO TRANSIENT-INDUCED LOCA COULD HAVE BEEN PREDICTED 11.
RS COMMON-MODE FAILURE OF MAIN FEEDWATER, INSTRUMENTS, AND AUT0 START OF AUXILIARY FEEDWATER COULD HAVE BEEN PREDICTED l
O
(
i J-
WHAT IMPROVEMENTS IN WASH-1400 METHODS AND TO PROPERLY TREAT SUCH SEQUENCES IN RISK ASSE BETTER METHODS TO PREDICT OPERATOR BEHAVIOR o
0F COMMISSION AS WELL AS OMISSION)
NOTE THAT BETTER QUALITATIVE SEQUENCE PRE ACHIEVED BY MERELY ATTACHING TO RISK AS QUALITATIVE REVIEW 0F POSSIBLE OPERATOR B QUANTITATIVE DATA IS CRITICAL ONLY TO QUANT PREDICTION SYSTEMATIZE SEARCH FOR COMMON CAUSE FAILURES o
TREAT PARTIAL OR BRIEF FAILURES o
-L.
-J
i
~
'I CAN WASH-1400 METHODS SERVE A USEFUL FUNCTION IN EVALUATING OCCURRENCES?
FAA & NASA USE FAULT TREE ANALYSIS Tills WAY o
KEMENY AND ROG0 VIN USED EVENT TREE ANALYSIS o
TO ORGANIZE THE "WilAT IF" EXERCISE:
ALTERNATE SEQUENCES
.~.
RISK ASSESSMENT SUGGESTS THAT A MESSAGE OF THE o
RANCHO SECO MAY HAVE BEEN MISSED:
SUSCEPTIBILITY TO COMMON-MODE MAIN FEED TRIP, OPERATOR CONFUSION, AND AFWS AUT0 START FAILURE i
?
L Lq-g...
-- g
% L s~e, c
IRVIS-BESSE P
FREQUENCY OF FEEDWATERIMM5tc#T 3/ YEAR 7
P PROBABIUTYPORVLIFTS 0.5/De%ND p
(C1AGGEDORNOTGAGGED)
I 1-P PROBABILITYOFIRIP 1/ DEMAND g
(1-P 8 1) g PROBABIU TY PORY FAILS OPEN 3X10-2pegno
/
Po s
1-P PROBABILITYOFHPISAcTuATION 1/ DEMAND u
P' PROBABILITYOPERATORSDEFEATHPIS 0.027/DEMND u
a 1-P,,
PROBABIUTY OPERATORS BLOCK 0.999/DewND o
POW WITHIN REQUIRED IIME
(>20 MIN)
~ 1.2 x 10-3/YR 1
_J
R#lCHOSECO 8.6x10-3/ YEAR P
FREQUENCYOFIRANSIENT 7
(LDSS OF NNI+RfD 1-P PROBABIU TY PORV DoES NoT LIFT 0.5/ DEMAND p
(GAGGEDORNOT) 1-P PROBABILITYOFREACTORIRIP 1/ DEMAND g
(1-P )
- 1 g
1-P,
PROBABIU TY CODE SAFETY VALVE OPENS 1/DBWO p
(1-P,)
- 1 g
1-P,
PROEABILITY CODE SAFETY VALVE SHUTS
.l/ DEMAND g
(1-P,): 1 o
Pu' PROBABILITYOPERATORIHROTTLESHPIS 0.027/Dewe
~ 1.2 x 10-4
/
1 b
i l
~
', b,,
THEE Mll.E ISl#0 2
P FREQUENCY OF FEEIMATER 3/ YEAR 7
P PROBABILITY POW l_IFTS 0.5/DD%ND J
p (GAGGEDORNOTGAGGED) l-P PROBABILITYOFIRIP
}/ DEMAND g
(1-P ) # 1 g
P, PROBABILITY POW FAILS OPEN 3 x 10-2 engno
/g o
1-P PROBABILITYOFHPISACTUATION 1/D0%ND u
Pu' PROBABILITY OPERATORS DEFEAT HPIS 0.027/ DEMAND N
P,,
PROBABILITYOPERATORSPAILTOBLOCK 0.125/ DEMAND POW WITHIN REQUIRED IIME ( 15 MIN) 4
+1.5 x 10 /YR.
i j.
.. r :
OPEPATORFAILURE P
0.9 5 MINLRES AFTER I.ARGE LOCA 01.
30MINUTESAFTERLARGELOCA 0.01 SEVERAL HOURS LATER 0.2 - 0.3 INHIGHSTRESS(AVE)
IE, RS, TMI:
3 TERATORS THROTTLING HPIS FOR.1/2 HR.
(,3)3 = 0.m 7 P ' = (P)N
=
u IE, BLOCK POW IN IIME (1 20 MIN) 1MI,=l-PN = l-(0.1)3 = 0.999 g,
TMI, Ib NOT BLOCK POW IN TIME (*15 MIN)
( 5)3 = 0.125 NP,, = pN
=
Q
/
m
- 7'
0o b e N }.'t DOSE COMITMENTS PER GWy(e) FOR THE URAN!UM FUEL CYCLE OCCUPATIONAL ar q aAJ Pjg(!C NUREG EE5MO CONAE5 &
NUREG BEngu CMAE5 &
CONAES & Cusrur OPERATION 0332 COSPUP 0332 C0$ PUP Corrected for RN*
MINING 88 250 250 140 630 600 00 MILLING 88 120 00 40-1750 120 120 20
{
1 9
9 10 CONVER$10N 0.9 1
ENRICHMENT 09 0.7 0.7 0.3 0.2 0.2 3
FUEL FASRICATION 12 12 12 0.6 0.6 0.6 REACTOR OPERATIONS 560 570 1240 130-160 77 76 REPROCESSIM (U.PU) 28 25 25 400-510 340 360 TRANSPORTATION.
6 4
4 13 0.5 0.2 IRA 40. Fut( STORAGE.
WASTt M87.
t TOTAL:
W/0 REPROC.:
760 960 1600 330-2100 840 800 190 W/ REPROC. :
700 900 1600 790-2600 1200 100tP*
550 i
eCerrected for typographical error in ICRP-2; used as dose conversion factor for Pb-210 (from decay of redon) in feed.
- cApparent typographical error in CONAES Table 9.5; correctly adds to 1200 (same as SESMO).
NUREG-0332 was an update of GESMO, correcting for Rn-222 doses from Pb-210. and using a 100 year Environmental Dese NOTE:
Commitment (EDC) rather than the 40 year EDC used '. GESMO and in the recent Table S-3 (10 CFR 51) hearings.
"I
-