ML20198H671
| ML20198H671 | |
| Person / Time | |
|---|---|
| Issue date: | 08/19/1985 |
| From: | NRC |
| To: | |
| Shared Package | |
| ML20198H658 | List: |
| References | |
| FOIA-85-769, RTR-NUREG-0956, RTR-NUREG-956 PR-85-113, NUDOCS 8601310074 | |
| Download: ML20198H671 (1) | |
Text
_ _ _ _ - _ . - - - - - - - __ - - - - - - -
c.
V s
?
/'""%'
f' -
UNITED STATES 2-s l NUCLEAR REGULATORY COMMISSION
%..'.'../, Office of Public Affairs Washington, D.C. 20566 No.85-113 Tel. 301/492-7715 FOR IMMEDIATE RELEASE (Honday, August 19,1985) -
NOTE TO EDITORS:
The Nuclear Regulatory Comission staff has issued a draft report describing the "...NRC staff and contractor efforts to reassess and updain t'ne ,
agency's nuclear analytical power procedures for estimating accident source terms for plants.
A source term is defined as the quantity, timing and characteristics of the release of radioactive material to the environment following is attached a core for yourmeltinfomation.
accident." A copy of the report's " Executive Sumary" The staff is inviting written coments on the draft report. They should be addressed to the Secretary D. C. of20555, the Comission, Nuclear Regulatory i
I (- Comission, Branch.
Washington, Attention: Docketing and Service They should be received by October 7 this year. The draft report is
- titled " Reassessment of the Technical Bases for Estimating Source Terms" and is identified as NUREG-0956. Single copies are available, free of charge, on written request to the Division of Technical Information and Document Cor, trol, Nuclear Regulatory Comission, Washington, D. C. 20555.
i 8601310074 051223 PDR FOIA g SHOLLYB5-769 PDR 19
- (
.' mama mme )R-Miu.Ndr
~~
COMMITTEE TO BRIDGE THF GAP. (53 FR. 3 /9F])
1637 BUTLER AVENUE s203 y' .[d'- '
LOS ANGELES. CALIFORNIA 90025 j( ( (213) 478 0829
'85 StP -3 All:49 August 30,1985 Attn: Docketing r, Service Branch U Secretary of the Conrnission D&g'"g*-[' o Nuclear Regulatory Consnission Washi..gton, D.C. 20555 RE: NUREG-0956, request for extension of public comment period
Dear Sir:
The NRC has issued NUREG-0956 as a draf t report for public conrnent . According to NRC press release No.85-113, the public conrnent period closes on October 7,1985, or about 60 days af ter the release date.
I am writing to request an extension of the public consnent period from 60 days to 120 days. The reasons for my request are as follows:
- 1) We did not receive the draf t report until some weeks into
(. the 60 day comment period and we know of other interested parties who have not yet received it at all.
- 2) Due to the summer season, many interested individuals are on vacation. In particular, many universities will be out of session until the middle or end of September. Uni ve rsi ty
< faculty and others will only then receive the report and begin to review it.
- 3) NUREG-0956 is of substantial length and, more important, it may be of considerable regulatory significance. Therefore an adequate review may take at least 120 days.
Considering that the comment period for proposed rules (which take up a single page or two in the Federal Register) is typically 60 days, a 120 day comment period for NUREG-0956 (which is perhaps 150 pages long) seems only reasonable.
Sincerely, n . 0 N ][, . C b dab , ' Y b f , ,f Steven Aftergood 3
Q^
0956 C a.,voJu PDR Executive 01 rector I
a --
1 d00lEt W WGEN i M/M hWs mensen gygg YN-OW Cc'KED :
- r. L5haC THE UNIVERSITY Of ALABAMA
'85 SEP 19 A1M{se of Arts and sciences Department of Biology GFFICE 2: EECAtiA
- Correspondence Address -
00CKEi;NG & SEPVICi WNN Louis 'G. WILLIA"3, Ph. D.,
12h6 Northwood Lake Northport, AL 35k76
. September 16, 1985 (205) 339-1535 Docketing and Service Branch Secretary of the Cocznission U. S. Nuclear Regulatory Commission WASHINGTON, D. C. 20555 Firstly: Did the NRC send a copy of NURErA)956 to the library or l to anyone at the University of Alabama? I am unable to find a copy for I examination and cocznant. I'm sure that I need to cocx:ent on this doc.usent.
Secondly: Please do extend the time limitations to 180 days as beginning issuance, August 7, 1985.
Enclosed is a photocopy of my suggested letter to SCIENCE NEPIS based
(, on the photocopy of their treatment of National Waste Policy Act. At this I date my letter was not M published. It could appear, in part,later. j Is the area served by TVA going to be allowed a separate treatment '
from states and other regions by the NRC7 -
Do agreement states get speheial " Exceptions (species)?" Can the differences be spelled cut for us in Alabama? Could be-called a refererendum?
t Only the State Administration and the legislature made this possible, but neither the governor nor the legislature of Alabama really know what they have O{9 done because the special interests have carefully crechestrated their own n version of what is good for Alabama. The media has given No coverage of
,{@
m what the # agreements
- now mean, and the interpretation of the National Q '
Waste Policy Act is very unclear on safety factors.
W G
T.
We are all concerned about releases of ionizing materials, doing a Core Melt down (NURErA)956), but much that is in the US EPA h0 CFR Part 191,
% allows too many known loopholes for leakage especially after 10,000 years.
DQ Apparently the DOE and the NRC has not looked at the extensive data on *
. leakages at Savannah River, Oak. Ridge and Washington State, for very short-
' q'$% % term possible leaks that could occur under the new set of standards. Using Q any radiondelide like radium as a re3Attve standei is prepoosterous.
D Many of my niceegraphed handouts have been sent over the years to the
-Q,w U f
NRC that deal with scane of these problems in the real world. My vocal
,5 D participation at various public hearings dealing with high-level ionizing ,
.l materials are very to the current disseussions.
When I can actually see the document and attempt,$ to siphon out the Q jargon to lay terminology I will hope this 1 give firstover[ view.
o m :: ; ;;O916 d'#4 M PDR NUREG Louis G. WTT.TM 7
/O 0956 C PDR r
swuiULUGY 205/348 5960 POST OFFICE BOX 1927 UNIVERSITY. ALABAMA 354%
W#
@ BUbf &-f(SG l O fo'&
. .s- u h0 S'-D0 $
SET
....w... 30ETE,rES
=
Safe Energy Communication Council 1609 Connecticut Ave, NW. Suite 201 Z 8@ I9 AfD:31 Wcshington. D.C. 20009 i
)
(202)483-8491 17 Septemb h~ .%
k*t&SESERVIG t.rt: Its .
Docketing & Service Branch sRANCH Secretary of the Commission US Nuclear Regulatory Commission Washingtone DC 20555 To Whom It May Concern:
It has come to our attention that the comment period for NUREG-0956,
" Reassessment of the Technical Basis for Estimating Source Term," is applicable 90 days after 7 August, its date of issuance.
We request that the comment period be extended from 90 to 180 days.
The length of the document, its highly technical nature, and the extensive impact NUREG-0956 could have on NRC regulations are major factors involved which warrant an extension. -
(' In addition, the release of NUREG-0956 in August precludes a timely response by many individuals on vacation during that month, and many others on the West Coast have no access to Federal Register notices which are not available there for 2-3 weeks after publication. -
We believe the magnitude and the number of reasons cited above justify extending the NUREG-0956 comment period from 90 to 180 days and strongly urge your immediate approval of the extension.
l Sine rely, Se WV nYW et De
~
~
l Director l
C CC20'"?- 850917 PDR NUREC 0956 C PDR f)b
. .!. ". ^', :.? <
3? 5 '
$U.l U,.y $.f:c.j',ll4 i ' .J. - *
( i i D 1
l cenew fw mene.new me.ource.
- cnticas us.e Energy Prorect
- Enwronmental Action
- Enwronmental Acton Foundation
- Enwronmentat
- pw.cy insatute o Fnenes of the Earin e Joes en Enwgy
- Media Access Project
- organiang Media Protect
- PutWac Media Center
- S,erra ~
cha e soier Loney Tesocommun.cationa me ren one Action Centw a .The Fund for Secure Energy
- Un.on of Concerned Scientiste e+
~ IJ k*
Q.
- "*[pu$%
0 f.n .
_4 =
Pu--bicC t=z- en _hU$ ="
Corgess Watch . Crecal Mass Energy Project . Hearth Research Group . Utgaton Group e Tax &% .'c up '
Septanber :.7,1985 I9 A10:30 Docketing & Service Branch y m,;, .
Secretary of the Ocnmission Qf[I'N 4 sti,.3, U.S. Nuclear Regulatory Ccmnission INANCH Washington, D.C. 20555
Dear Sir / Madam:
We are writing to urge that you extend the omvent period set for NUREG-0965,
" Reassessment of the 'Ibchnical Basis for Estimating Source Term."
'1he notice published in the August 7,1985 Federal Register (Vol. 50, Ib.
152, p. 31937) noted that public coments were due by October 7,1985. 'Ihat provides just over 60 days to omment on a docunent that is both lengthy and technically ocuplex. By ocnparison, on other proposed rules which were shorter or more straight-forward, longer atmnent periods have been allowed.
Given the role NUREG-0956 is likely to play in influencing the shape of
( NRC regulations in the future, it is ne ssary that the Ormission make provi-sions for securing the broadest public input. One key elenent in accomplish 2ng this goal would be providing a ocmnent period that is long enough to both allow interested persons learn of the availability of NUREG-0956 as well as to read, digest, and formulate scme tinughts on the docunent.
Accordingly, we would urge that you consider extending the 60 day connent period to at least 120 days and perhaps 180 days. We hope that you give our request serious consideration.
S' ly, Ken Ibssok i Critical Mass Project l of Public Citi e5e?:200:3-B50917
. PDR NUREG 0956 C PDR l -~
KEB
.( -
,.J.. ., P '
( ,
Crecar Mass Energy Prosect . 215 Pennsytvaw Avenue S E. . Washrgon. D C. 20003 . (202) 546-4790 8
e et'
o M RO' Q p _ p - (fg((
om @
CHORICv d MISSISSIPPI POWER & LIGHT COMPANY
)) Helping Build Mississippi E d M ana5 P. O. 8 0 X 16 40. J A C K S O N. MIS SIS SIP PI 39215-1840 -
C" n i -
October 7, 1985 NUCLEAA UCEN 4 SAFETY DEfdA'WENT Secretary of the Commission Nuclear Regulatory Commission ,r -
Washington, D. C. 20555 6. .f Attention: Docketing and Service Branch
Dear Sir:
SUBJECT:
Grand Gulf Nuclear Station Units 1 and 2 Docket Nos. 50-416 and 50-417 License No. NPF-29 File: 0260/11500 Comments to NUREG-0956 AECM-85/0325 The purpose of this letter is to transmit coceents made by Mississippi Power & Light Company (MP&L) to the draft of NUREG-0956 " Reassessment of the
, Technical Bases For Estimating Source Terms".
MP&L feels that NUREG-0956 represents a major advance in the NRC's severe accident assessment technology. In regards to the " science and engineering" aspects of NUREG-0956, the source term phenomenological models and computer codes which incorporate these models represent a great i improvement over those used in the Reactor Safety Study (RSS). These codes can be used to improve source term estimates for severe accidents. The MAAP
% code, developed by the Nuclear industry during the IDCOR program, contains
'C similar models and provides an alternative means of assessing severe accident R source terms. Technical differences between the IDCOR and NRC severe ace', dent Q models have been identified. It is appropriate that these differences be
'O N resolved prior to the regulatory application of any source tern code.
l?. d i s[i MP&L's overall comments regarding NUREG-0956 are presented below
['s, .' and detailed comments are included In attachments 1 and 2.
,. J. o The BMI-2104 source term analysis methods are more accurate and more
+( . realistic than those used in the Reactor Safety Study (RSS).
' y o Other equally accurate and realistic source term analysis methods of
..,O . equal validity are available, principally, the MAAP code.
.K :
qi g N' o Preliminary risk analysis results on the Surry Plant indicate that the overall severe accident risk is lower than predicted in the RSS.
'; The consequences were higher in some sequences but lower in most.
M Thus, existing regulatory policies are conservative. There is therefore no immediate need to apply these new methods in the k4 g regulatory arena.
g\ P b^h.,851007
- 0556 C peg J13AECM850100201 - 1 Member Middle South Utilities System . _ g
. - - . .. a . 39t l]U s
s '
- AECM-85/0325 Page 2 o When, and if, regulatory practices regarding severe accident risk ard re-evaluated using improved analytical methods, the industry should be icvited to participate in the development of the groundrules for this regulatory practices re-syalestion. Thers ground rules should account for accident probabilities and operator intervention. And, the new severe accident Eagulatory practices should be balanced with increased requirements on issues which are found to be of greater risk than assumed in the past and relaxed requirements where the risk is found to be smaller.
MP&L appreciates the opportunity to comment on NUSEG-0956. Please call if any points or comments need further discussion or clarification.
Yours truly.
pdf , e I-L. F. Dale Director TWS/GWS/SHH:vog Attachment cc: Mr. J. B. Richard (w/a)
Mr. O. D. Kingsley, Jr. (w/a)
([.
Mr. R. B . McGehee (w/a)
Mr. N. S. Reynolds (w/a)
- Mr. H. L. Thomas (w/o)
Mr. R. C. Butcher (w/a) dr. James M. Taylor, Director (w/a)
Office of Inspection & Enforcement U. S. Nuclear Regulatory Commission Washington, D. C. 20555 Dr. J. Nelson Grace, Regional Administrator (w/a)
U. S. Nuclear Regulatory Commission Region II 101 Marietta St., N. W., Suite 2900 Atlanta, Georgia 30323 Sb J13AECM850100201 - 2
i i
k' Attachment 1 ,
I MP&L COMMENTS ON NUREG-0956 CONCLUSIONS AND RECOMMENDATIONS Conc 19sion 1. The BMI-2104 suite of computer codes represents a major advance of technology and can be used to replace the Reactor Safety (RSS) methods.
MP&L Comment. We are in agreement, but the comparable IDCOR-developed MAAP code may also be used to replace the RSS methods.
Conclusion 2. Principal omissions and oversimplifications in the RSS methods have been corrected in the new source term codes. Fission product chemistry, retention in the reactor system, and mechanistic aerosol behavior are now accounted for, at least in an approximate manner.
MP&L Comment. We are in agreement, but the same can be said of the MAAP code.
Conclusion 3. Remaining areas of uncertainty have been identified in the new source term analytical procedures and indicate areas of research that should be pursued. Uncertainties persist in some of the areas where major advances have already been made.
MP&L Comment. We are in agreement. We feel that the resolution of technical .
(.' differences between the NRC and IDCOR should be given priority not only because these differences are significant but also
)
i because their resolution will provide the nuclear industry with !
an alternative (and perhaps easier to implement) means of assessing severe accident consequences via the MAAP code.
Conclusion 4. The new analytical procedures have been extensively reviewed, including a review by a special study group of the American Physical Society, and all phases of the source term reassessment effort have been documented.
MP&L Comment. The review process to which the BMI-2104 codes have been subjected is extensive, as documented in NUREG-0956.
Conclusion S. The analytical procedure is complex and involves several scientific disciplines. Successful application of the analytical procedure requires a thorough understanding of the problem to be solved, including the plant characteristics, the accident sequence description, and the purpose of the analysis. A quality assurance procedure is also required.
MP&L Comment. We are in agreement. Because of the number of disciplines required to perform severe accident analyses, it is important to simplify the process wherever possible. One area of possible improvement for the BMI-2104 codes package is a reduction in the number of codes involved by incorporating them into fewer I (preferably a single) larger code (s). This would reduce the occurrence of modeling inconsistencies, calculational duplication, and input / output errors.
J13AECM850100201 - 4
k Conclusion 6. New source terms have been calculated for selected accident sequences for five reference plants that represent major reactor and containment types in operation in the US. These selected sequences have provided a sufficient test of the l capabilities of the computer codes.
MP&L Comment. Although much testing of the BMI-2104 code package has been performed, it is likely that as more sequences and plant types are evaluated additional code revision will be required.
Conclusion 7. For most accident sequences, the largest single factor affecting source terms is containment behavior. A delay of several hours in containment failure will reduce source terms significantly.
MP&L Comment. For most plants, the results reported in NUREG-0956 demonstrate the importance of the containment behavior during severe accidents. This appears to be correct for GGNS despite the assessment that the " source terms at Grand Gulf are more af fected by suppression pool bypass than by containment failures" as stated in Section B.4.4 of the report, since the CPWG results, as per Appendix C.2, show that suppression pool bypass is predicted to occur late (during core / concrete interactions) and thus is expected to have little impact on offsite consequences. Suppression pool bypass and suppression
(;. pool decontamination factors probably still remain as important areas for future investigation for Mark III plants. With respect to the effect of containment failure delays, it is probably more correct to state that the larger the difference between the time of core melting and containment failure, the smaller the source term. Containment failure at or near the time of core melting results in the largest source terms.
Conclusion 8. Source terms were found to depend strongly on plant design and construction details, thus making development of useful generic source terms difficult. ,
MP&L Comment. We are in agreement.
Conclusion 9. New source terms for many accident sequences were found to be lower than those in the RSS, but some were larger. The reductions were found mainly because containment integrity was maintained and natural processes reduced airborne concentrations of fission products. The larger source terms resulted from early containment failure, which is still predicted in some cases, and the improved description of ex-vessel processes, which leads to larger release estimates.
Therefore, generalizations are inappropriate.
MP&L Comment. The key point is that overall risk is lower than those in the RSS. Thus, existing regulatory policies are conservative and k
there is no immediate need to revise these policies.
J13AECM850100201 - 5
I 1
- I k
Conclusion 10. A comparative risk appraisal for the Surry plant using the i RSS accident frequencies, source terms based on BMI-2104 l results, and a preliminary re-evaluation of the behavior of the containment shows a reduction in estimated risk compared with the RSS. The reduction results about equally from new source terms and new evaluations of containment behavior..
MP&L Comment. As stated above, the impact of a reduced risk estimate is that existing policies are conservative.
Conclusion 11. For the other plants, further analyses need to be made before any conclusions can be drawn about changes in estimated risk.
The fact that source terms for some accident sequences are not lower that those in the RSS suggest that significant reductions in estimated risk may not ba found in all cases.
MP&L Comment. Conclusions regarding other plants are pending the analysia of these plants.
Conclusion 12. Research programs that address the remaining major areas of uncertainry in the source term technology are currently in place and being pursued by the NRC.
MP&L Comment. We agree that this is appropriate.
( Conclusion 13. A major conclusion of the APS study group confirms the NRC staff position that source term research must be continued in order to complete the regulatory actions being considered.
MP&L Comment. We are in agreement, but again, existing regulatory policies are conservative.
e Se J13AECM850100201 - 6
)- ( Attachment 2
) -
MP&L COMMENTS ON NUREG-0956 RECOMMENDATIONS Reconnnendation 1. The new source term analytical methods should be used to re-evaluate regulatory practices that are based on the RSS methods. Insights from new analyses should be applied to
. reconsider the use of TID-14844 assumptions. Improvements are so significant that utilization of the new methods is warranted while additional confirmatory research is being completed.
l M*SL Comunent. Although the new source term analytical methods represent
- significant improvements over previous methods NUREG-0956 has shown that existing severe accident regulatory policies are conservative. Thus, the immediate application of these methods is not warranted. In the long-term, a realistic treatment of the source term issue is the preferred method of establishing regulatory practices. Since source term reductions are not expected for all postulated severe accidents for all plants, this re-evaluation of regulatory practices should be balanced
increased requirements on issues which are found to be of
~ greater risk than assumed in the past, and relaxed requirements where the risk is found to be smaller. Since l{
l the overall risks are lower, the overall requirements should be less restrictive. The industry should be invited to participate in the development of the groundrules for this regulatory practices re-evaluation.
These ground rules should account for accident probabilities and operator intervention.
Recommendation 2. A particular version of the new codes called the Source Term Code Package will be maintained as a reference code and is the recommended analytical tool for NRC analyses of accidents severe enough to result in complete core melting. Additional technical insights can be obtained for all accident conditions with the NRC's detailed mechanistic codes and their experimental data bases.
MP&L Conssent. This is a excellent approach for maintaining analytical consistency.
Reconsnendation 3. The Source Term Code Package was designed to provide best-estimate results (i.e. , without intentional bias) .
In any regulatory application, careful consideration must be given to the purpose of the evaluation, to the desired margins, and to the uncertainty levels. Close coupling between the research effort an the regulatory effort will l be required in assessing uncertainties and evaluating
( technical issues.
MP&L Comment. See comments on Recommendation 1.
J13AECM850100201 - 7
,, ,.a.s .
., e v.M ,, 2;r.n. -
J'E 000tEl StRitER
[f'?W .
m m fh-YM MN p%
- y. .,
oveecnogss)
H 3.::ta: hor 15, 1 2 5 <
as. p.c0 ,
i, 0.\3:
- , ,D'% Docketing and f.crvica Ercnch Secretary of the Cc :-ission U.S. *:uclear Re;ulatory Con-istien D @ Jp $0 29 ,
Tashin; ton, D.C. 20355 Cfr:- .
00:h'{t[yk BRk,y?.i':r f Mi DeFr Sirs:
I an ' riti g h erress ny cones = eurrounding tha recent RCs.n :f ;
1.'.T.23-OF56 fcr - uFli: connent. It i: n- c-inisn tiri ::V ^; e -f fr -raa:' * ;
regulatien: tha- des 17:ith the nuclerr indurtr/ ch: .1d te ;ivan a ;rert dati :f thou;ht, tnd a nari- r. of utlic in ut. :uld rsnini you thtt ne are E.s , es:1c' Then you are : e= t: rr:tect fren the otential nona n of ; corly -= ;ad n :1:ar i programs. Fr.cn it c:ces to an icsue as cericus a: ths analytical procaluras ;
ursd to esti ats ;;rcible nuclear accid 2nts, then it ssens : Einfully (:ricu: t.' .zt . ,
( the ute. st cautien and tho :;htP '- ust be used.
s -
The 90 day peri:d which you have offered fer public co=e".t and input is dangerously inadequata. Ey home is in a state that has thr2e very treuti?s ce ccmercial generating reactors in the 600 or larger ne;a ratt cara. city. "a al::
hava a sneller re et:r that has also been both cestly, and ha:ardan: ': T.? - r-r:unding connunity. All are in .cotential er.rthp ahe re; ions,riith three in e'sp~c-ially h'sh rick areas on tha Califo m ia coast. I met inrict that all int.2 rested parties, both e. vert and laypersons, te given ,cre ti e to revie-- this ver/- critical -- l new report. Excessive haste in this area could be not only unnice, but petentially i disastrous.
If you are tr'M/ng to restere public confidence in an 1.dustry t'.at hns been i i plagued by decertien, nis ent, and enemens costs, t!.en I stren;17 su;;ert you do as nuch as reesible to solicit the public's input and refrain fr= c1: sed-( de r ncetings. I a. confident that the desir:d result, refer .%ciser -Irr*- r. d
. increa ed ublic tr:,st, will be enhanced. = " "" n -
]L - jf l p/ .. . '
A- '
~-
0956 C
, 4,.' ' ; 9..* i . 4 4 ~ i !
l 0 Y
y)( 4k;,F . .. .
' I.?.19., k. -
H ,
i:
..: , Er .trcc_4 re_-17 on thir --='.".er W.11 to e,- recirtad. .
Li" *:017s Larry 1, .rd:r 1 5 9 2 . !!rt.r d c y Trez..:, Cl.
?3701 J
t v
e J
e 4
+
I a !
no i
I E
l i
k' i j !
1 1
- . . . )
1
?'. .-- , -
-.,1 : , - , _ , , _ . _ , --: 1 _; :- , . . _ _, _ , _ _ -
7- ,
i h.0 AKA5 mihMKR q
@DRDED BULL M(SC t Ml h~". O.lUAGG-6956) f September 14. 1985 BScgrt p~
USMC
, 'f D o c t. e t i n g and Service Bronch Secretary of the commassten U.S. Nuclear Regulatory Commassion gg $3316 p3,.43 Washtngton, DC 00555 5rtgg y .
D3CKE Tag'h ffg ,
Dear Sar/Hs.:
4RAhCN Re: comment Period for NUREG-0956 Accordang to the Federol Register notsce on the availobility or NUREG-0956. *Roossessment or the Technical Basis for Estsmoting Scuree Terms,. ror comment (50 Fed. Reg. 31937, August 7. 1985). comments must be received by October 7, 1995.
This as cr.ly o 60-day comment persed for o rathee lenstby, technical dscunent. It should be noted that the NUFE0 : ntoans many re r ere ces os well wnich woule MS/e t: be tM:e:uskly esentered for sutstantive c mmentory. Teese rc:ts, os well os the press o' other cusiness. Pende- it uni s k el> that
- wcul-have time to submit substontive comments before the October 7 deadline, Other Crcspective 20mmenters are D*Cbocir Focat9 stEslOr ;
situations. 3 r. s e e s , cc.stdertng the sweepang regulotc'> Ohcrges
( ,. whsch are predicated on source term reossessmer.t. it would seem proper tc allow a much longer comment perzod on teis document, Sixty days is the length er the :cament perice rer son > r:uesne rulesorings which de net involve such substantive and technical issues, See, e.9.. 50 Fed. Reg. 07006 (Modiracotien or GCC 4 Requirements for Protection agosnst Dynomse Errects or Postuloted Pipe Purtures, whsch had a 60 co; correct perice).
I would therefore request that the corment peri:s r:r NUREG-8956 he ewtonded to 180 days.
E1ncerely, g . .-
Euson L. Htoet chse ctercens for Responsible Energy, Inc. g f
8075 Munson Rd.
Mentor, CH 44060 LOle) 055-3158
- i cc: Ccngressmon Dennis Ectort ' ,i-
\'
t rly r A A
$$ ^/ s g ei t JAl $
"** """M")R Wise Nd/4 g O,.
s o TERR.
I. LEWlS M0 E(cd954) m . a 19149 f DOWTCD D)cketir.g and Service Branch ' ' ;W
' secretary of the Goatiesien '
1 wraa '85 SEP 16 M1:28 Washingt5n D.C. 20555 ~
) cr_,,:. as :; .
e j
Sir: 00~ .U M T S G ,i The connent period for the NtFG 0956 is totally unfbYThis rasasessee nt took ths NRC 6 years to put together aft'er the TRI accident. Now the public is asked to get its e xtnents together in days.
Even if t.he Dublic has the renourcee .Which it doesn't, itjstill unftir to ask for this speedy a consent period.
I respectfully request an extension of tre cornant peri:>.1 on NtREG 0956.
Very truly yours,
/ w
/ - *< \
'i M. l. LEWIS
- 6504 ORADFORD TERR.
PHEA, PA.1914e n
lIb17j.cc /
/ Oi [k & 'I 1 .'
tf jg-) l ' '
k2.' lun C . -u/ct. 'l,,,
o ?f 1 I
2L? cm1 .
o ..u,:y.m a&cG1f a~, &
-udw km!:6g l
.r .
~
I 5
L _ - - - - -
L .
J F
wg "'"l*L"'R Mrse Nok (NUREG.69ffo NTS m*m
,, LTITON
'85 SEP 16 Mt :28 us East mm street .
$[3N[ O.71r BANCH i .- September 12, 1985 j
jC**"*" Secretary of the Commission atent , ce wre Docketing and Service Branch 5"* 'a^r U.S. fluelear Regulatory Consnission U $," Washicgton, D.C. 20555 kghard loits,on l'" b'* "
hacen irtold
Dear Sir er Madam:
Alan Pots 6v toma s.urnan Wry S*"
I am writin8 as director of a 2500-member citizens' group on 1.ong Island. We are outraged by the highly inadequate period
,,,,,,c,,,,,,,,,, of the de Reissin has alktted br pMic comnt updhg Nor 2 wdn XUREG-0956, " Reassessment of the Technical Basis for Esticating of can i Source Term."
lanw.I h6" This very technical document may exert a significant influence K5" C'"" upon our lives here on Long Island. Your agency will use this q ud cieueu v uoni uk document to make crucial decisions about emergency planning, a r
rui tnd sharenarn o ponenes controversial and important issue in the Shoreham Nuclear Plant 7'"ad' d 'he 8*"h Frwedi d u 1icensing disput e.
croup #* e, wh Fork mmceow. tano Aie rn.mn in spite of our keen interest in this issue, we only today g,$"p' received your notice of the public comment period which was LI Feends of cleerwetet published Wednesday, August 7th, in Federal Register Vol. 50, gngrymd No. 152. We have not yet received a copy of NUREG-0956, and, Nonh Fork naturally, h&Ve not had time to read, analyze it, and inform our l>*'ma"**'d C =d arcebership about its contents.
North brk Orgwner.tv c' hiear Erposute
- EINnY se, tners, When the Commission has considered other critical nuclear safety Nee,r enam ccm, issues, the public has beets given many months to offer comment.
Peeach uas.'
Tor example, nearly nine months was allotted to the public for
["""J.,17,%,al"" cc:maent on the Decom:cissioning Rulemaking. 90 days is simply too Satease5ound short a period to adequately review and comment uport a lengthy, 5*
s nref a"P ^"""*","'
so, ehe r.opi technic.il, and significant document .
I
- s. .. .nh.o scat wakm 9.m.i ues The Commission aust knew as well as we that the public is prone yl$'fy,'j,, , to a basic mistrust of federal bureaucracies. Often changes in
- ~o.,,,, o important regulations seem to happen mysteriously, without
- E"~4' 4 ,,
adequate public notification or input. If more time is not allowed for public comment in this case, the public will have yet more reason to lack confidence in your egency's procedures.
i
-2 AV-e .s .
a **
l .
L,4- a yy ,t a H. .
We respectfully request that you extend the period for public comment regarding NUREG-0956 from 90 days to 180 days. That length of time should allow this citizens' group adequate time to conument and so contribute meaningfully to the Connirsion's analysis.
Sincerely, Nora Bredes Executive Coordinstor cc. Rep. Thomas Downey Rep. Robert Mrazek C
O e e 1
I 26
s o,?
October 2, 1985 f Mr. Denwood F. Ross, Deputy Director Office of Nuclear Re6ulatory Research U.S. Nuclear Regulatory Commission Washin6 ton, D.C. 20355
Dear Mr. Ross,
The following paragraphs are written in response to your request for comments on NUREG- 0956, "Draf t Report for Comment", " Reassessment of the Technical Bases for Estimating Source Terms". I express my personal opinions as a private citizen with thirty-five years experience in nuclear technolc6Y-In gene ral, I found the report to be un excellent summary of the state of the art. It's shortcomings relate to its extremely narrow focus on scientific issues derived from poorly thou6h t out accident senarios. My specific comments follow:
The fuel meltin6 temperature given on pa$e 3-6 as 2277 C is inconsistent with Taole 3 1 and with any aata that I am aware of. The melting temperatures for Zr ana UO given in 2
Taole 3 1 are correct.
I do not believe that the MARCH code as descrioed on pages 3-6 and 3-7 is realistic. The zirconium claadin6 of fuel
( roda will lose virtually all its etrength aoove 900 C and it will oxidize and crumole rapidly above about 1200 C. This loss of strength and integrity will allow the fuel which is cracked into 0.25 cm. on a side rectangular fragments to crumole into a rubble bed. TMI-2 has a rub' ole bed that prooably was procucea at least partly by this process. The rubble bed will be supported by the s tubs of the fuel rods below the bed. See Nuclear Science and Engineering, Vol. 4, 1958, pp 180-199 for tne configuration (
of the core after the ESR-1 celtdown.
t With respect to the MARCH codes Puel deoris beds which clump onto the lower support structure at temperatures aoove 1200 C will be61n to transfer heat to the reactor vessel at substantiel rates by radiati.on. Tnie transfer will cool and ,
s top the progress of the debris mase until all the water is '
gone from the botto= cf the vessel. The aspect of a decris mass sitting in the bottom of a reactor vessel ana heatin6 the whole vessel to a white heat to cause it to fail by loss of strength (not by meltin6) is awsome. If the NRC considors this to be a real (if remote) possibility, then the NRC shoula require that reactor cavities be designed to collect water 2 rom croken pipes or other sources to surrounc to vessel with sufficient water to keep it cool in auen a ma ior accident. Reactor cavities a re flooded during refueling.
(
~
GG f*WiWW gw 3 2 PP-I
(
With reference to Table 3.1, page 3-18; The UO2 f0P light water reactors is sintered in a steam-hydrogen atmosphere at 1700 to 1800 C. This atmosphere results in an oxygen potential in the fuel of about -300 kJ/ mol. This is equivalent to the oxygen potential of the steam-hydrogen atmosphere in the reactor vessel at TMI-2 which is known to have suppressed the formation of molecular iocine (12 ). During irradiation the zirconium cladding reduces the oxy 6en potential of the fuel s till further. It follows that molecular iodine (I2 ) should not be listed as a " fission product" in Table 3 1. See also Nuclear Technology, Vol. 40, Oct. 1973, pp 297-305 Fi Es. 1 and 4.
The footnote to Taole 3.1 is incorrect and inconsistent with the data in the table and the statement on page 3-6. Tne lowest meltin6 composition in the UO 2-ZrO2 %s i at acout 2500 C.
syste The last paragraph on page 3-33 re of the "certain circumstances" under which elemental
" quires clarification lodine may be present". Universal understandin5 of the conditions constituting these "certain circ ~umstances" is essential to control iodine genera tion in accidents. Figure 4 or the above reference describes the oxygen potential necessary to produce elemental iodine as bein6 more positive than about -200 kJ/mol.
That is more than enough oxygen to burn hydrogen.
( In my opinion, NUREG-0956 places too auch attention
, on the scientific aspects of hypthetical accident sequences and on computer codes based on such hypothetical sequences. It is not clear that the sequences are realistic.
In =y opinion, section 7 of NURE3-0956 shoula pay some attention to future research on safety-rslated measures to prevent accidents and to insure
- terminated sequences" (the term used on page 3-19 ). These measures m16ht include educational pro 6 rams for NRC and utility staffs relative to actions to be taken under reactor accident conditions. The measures might include s tudies on safety-related systec design chan6es. For example: Perhaps the containment atmosphere could oe kept low In oxygen and high in nitrogen and hydrogen to suppress iocine forzation. Pe rhaps the containment could be pressure relievea te prevent rupture by bleeding gas to a gas fractionation and purification systec. j Sincerely,
. 47 Walston Chuob 6d&
j450 MacArthur Drive Murrysv111e, PA 15c66
(
sh
i
! . ~ f bV
/
s' Oak Ridge (615) 576-3342 Institute
{~ - Associated Post Office Box 117 for Energy Universities Oak Ridge. Tennessee 37830 Analysis October 15, 1985 Dr. Nel Silberberg Of fice of Nuclear Regulatory Research Division of Accident Evaluation Fuel Systems Research Branch U.S. Nuclear Regulatory Commission Washington, D.C. 20555 Subj ect: COMMENTS ON NUREG-0956 Dear Mel The following comments are submitted to provide focus for our discussion of November 6. They are not intended as a formal set of comments.
- 1. Scope / title of the report. The focus of the report is on the NRC set of computer codes for predicting source term; it would be appropriate for the title to reflect that. Actually, knowledge of the source term is much broader than the BM1-2104 codes (as presented in NUREG-0956), out NRC has been thus far unable to capture such breadth in a computer program (and perhaps never will).
- 2. Need to integrate containment performance into the suite of codes. It is apparent that the containment working group failure criteria were not factored into the calculated results of Chapter 4. This lack of inte-gration clearly detracts from your claim of "best-effort" results.
- 3. BWR source terms. The relative lack of attention to BWR ohenomena alluded to in the APS report is still present in NUREG-0956. What do you plan to do about it?
4 Risk analysis. The risk analysis for Surry in Chapter 6 and Appendix D is erroneous, in my opinion, and quantitative results should be deleted f rom the report.
The deficiencies of the risk analysis include
- a. lack of consideration of operator actions D. use of a 10-year old set of sequence probabilities
- c. inadeouate peer review of the BM1-2104 V sequence that dominates
( the risk
~ s ., -
%u- ,,(", # / W '[G A f7
I Dr. Mel Silberberg 2 October 15, 1985 C.
- d. inadequate consideration of containment phenomena as developed by your two working groups.
The only quantitative statement that you can justify would be "The BMI-2104 3
suite of codes as applied to Surry with current input data give an order of magnitude reduction in risk as compared to that of the RSS." Speculating about the relative risk of BWRs is unwarranted until you have done a better job of integtating external findings (i.e. SASA program, IDCOR logic) into your program or actually improved your BWR codes.
- 5. Bypass sequences. Since the V sequence and perhaps other bypass sequences may dominate the risk in PWRs with large dry containments, have you planned a review and upgrading of the Battelle computation of V?
- 6. Operator actions. While not strictly needed for bounding source term estimates, consideration of operator actions is built into risk event trees. Ilow do you plan to consider operator actions?
- 7. "Best-estimate" results. Before you can claim best estimate results, it is necessary to improve the computation with parameters you know about L , such as those cited above.
- 8. Uncertainty analysis. It seems to me that uncertainty analysis should focus on the effect of specific parameters on risk, rather than source term. A given parameter might be uncertain by orders of magnitude, but if it can't increase predicted fatalities by 1 (or whatever threshold you prefer), it isn't worth pursuing. PRA coupled to uncertainty analysis should be used to prioritize research in the source term program, considering also research being funA ^d by others.
sincerely, I\
wh Irving Spiewak IS:aon cc: Chris Ryder, NRC
(
$ Y <W T
- f. .
( ., .
Washington Public Power Supply System 3000GeorgeWashingtonWay P.O. Box 968 Richland, Washington 99352-0968 (509)372-5000
{
October 7, 1985 Mr. Denwood F. Ross, Deputy Director Office of Nuclear Regulatory Research U. S. Nuclear Regulatory Comission Washington, D. C. 20555
Dear Mr. Ross:
Subject:
Review and Coment on NUREG-0956 The Washington Public Power Supply System has reviewed NUREG-0956
" Reassessment of the Technical Bases For Estimating Source Terms".
We are encouraged by the results of this reassessment in conjunction with the
(, industry's IDCOR effort, and recomend that the NRC continue efforts to identify appropriate areas to apply these results. The reassessment of source terms has the ability to save ratepayers millions of dollars with no reduction in the level of protection afforded public health and safety.
Specific coments from our reviews are provided in the attachment.
Very truly yours, l
l 1
G. C. Sorensen, Manager i Regulatory Programs j l
I l
i b
i 2pp l
I. .
I ,
ATTACHMENT h
- 1) As noted throughout the document and particularly in Chapter 8, Conclusion 8, " Source terms were found to depend strongly on plant design and construction details, thus making development of useful Generic Source terms dif ficult." Examples of such plant specific features that were considered in NUREG-0956 would be the use of a certain kind of concrete, the shape of the reactor cavity and the routing of ECCS piping in PWRs. Other plant specific features not mentioned in the NUREG, but which could impact the source term results would be such things as the effect that the containment sump area geometry has on turbulence, flow rates and deposition rates; the surface area available for deposition is likely to be much higher than assumed due to the presence of cable trays, cable, piping, instrument lines, various kinds of equipment and supports, etc.
This conclusion should hardly be considered surprising and certainly should not be construed as an impediment to usage of the new methodology in the regulatory arena. The Commission noted in the supplementary infonnation accompanying its recently published " Policy Statement on Severe Reactor Accidents" that this effor.t is just one part of a larger program which will, by design, account for plant specific contributions to risk. Plant specific features which contribute to the difficulty apparent in developing Generic Source terms need to be addressed primarily through such efforts as the QUEST studies, etc., which are ongoing.
{
The logical course in addressing this issue would be to use the approach
- followed by the industry in the IDCOR program.
- 2) Improved understanding of equipment failure rates and event initiators could shed significant light on risk assessments, since risk is defined as frequency multiplied by consequences. Both the Reactor Safety Study
. (WASH-1400) and NUREG-0956 concentrate on consequences, which is appropriate given the nature of the regulatory climate. However, in the application of such research one must look at both sides of the equation.
Cost benefit analyses which begin with severe consequences and an assigned probability of occurrence set at an artificially high value, not reflective of reality, does not serve the goal of providing a more coherent technical and scientific basis for regulation.
- 3) The logical place for NRC to begin testing this new methodology would be in the preparation of new and/or revised value/ impact assessments for use in prioritization of safety issues in NRUEG-0933. This would provide not only a comparison of the effect of the differing methodologies in a regulatory environment but would also allow an alternate assessment of the utilization of staff resources. Revised safety issue prioritizations of HIGH, MEDIUM, LOW, and DROP, based on new and improved analytical methods would contribute to improved agency management of scarce resources, minimize the impact to industry from overreaction by NRC to issues of negligible safety importance, and focus NRC and industry
( efforts on those issues which are truly significant in terms of public health and safety.
/
.W
~
~~ '
f /
Q l- f q-Nuclear Information and Resource Service 1616 P Street, N.W., Suite 160, Washington, D.C. 20036 (202) 328-0002 l
September 16, 1985 Docketing & Service Branch Secretary of the Commission U.S. Nuclear Regulatory Commission Washington, D.C. 20555
Dear Sir or Madam:
I am writing regarding the public comment period set for NUREG-0956, " Reassessment of the Technical Basis for Estimating Source Term." It is my understanding that the comment period is 90 days from the date of issuance, August 7th. Such a period of (h; time is inadequate.
I am hereb /reque ting an extension to this comment period from 90 days t 180 da s for the following reasons. First, NUREG-0956 was issue vacation.
in Jt summer month of August when many people were on e6ond, the sheer length of this document merits a longer than usual comment period. NRC routinely publishes proposed rules, much shorter that this NUREG, for which the public may be given several months in which to respond. For example, NRC extended the comment period for the Decommissioning Rulemaking from May to November, 1985, thus allowing a comment period of nearly 9 months.
Equally important is the highly technical nature of the document, necessitating far more work than otherwise, and requiring potential commenters to seek out qualified persons to analyse the technical information presented. Fourth, it usually requires ,
several weeks for prospective commenters to become aware and request copies of such documents, thereby decreasing the time available to them to review and comment. For example Federal Register notices often are not available on the West Coast for two to three weeks after publication. Lastly, and perhaps most
((
important, given the major role NUREG-0956 may play in the chang-ing of NRC's regulations, it seems only fair that NRC allow an adequate comment period.
_g ^<77 Arm y A rr 6
m-
- x. -- .
^L' k- -
1; ..
Y, In sum, there are many reasons for NRC to extend the comment period for NUREG-0956, and few, if any, reasons not to. I hope the Commission will respond favorably to this request in an i effort to seek as much review and public input into its analyses of source term as is possible.
Sincerely, ,
y .
./.
lkty/ W* " '
Eugene Rosolie Research Analyst cc: NRC Commissioners
- f. D.F. Ross, Jr., NRR V
t i
B
(
6b
p- .
'Q Q
~
, pggposED RULE -
.- HARVARD UNIVERSITY g[ff_jk D3CKETC;.
( DEPARTMENT OF PHYSICS V3hSC JEFFERSON PHYSICAL LABORATORY CAM BRIDGE. M ASS ACHUSETTS o2138 U.S. Nuclear Regulatory Commission
'85 S919 P4:09 l l
1717 H Street - :
Washington, D.C. N. .a-00s ..: iau
_ ,f.a i5:.
Ehl.NCH 11 September 1985 I
Dear Sirs,
I have received a draft copy of your report NUREG-0956 with a request for comment. You request comments within a time that is difficult for me, as it is the start of a new term, so my comments must be brief.
At a later time, I may be able to go into more detail.
Yours sincerely, tb Nf -
Richard Wilson Mallinckrodt Professor of Physics cc: APS. study group members E^"ert 3 ndm'h k
- # 7%' 62%
I l 2 n'-
I AcknowW by eeftf...S..E.P... 2 31985
5 ,
I
(' Consents on NUREG-0956 Draf t, dated July 1985, received September 1985 by Richard Wilson.
n page 5-4 to p.
members of the APS study group also vistied KFK Karlsruhe, Battelle Columbus Labs, Idaho Labs.
Conclusion 1. The first part of the sentence is supported by the APS study V.D., p. S109, 2nd colurn 2nd paragraph and V.E.2. But the 2nd part of the conclusion is not supported by the APS study group report. Attention should be paid to APS conclusions V.E.4, V.E.5, V.E.6 V.E.7, V.E.8 If it means a temporary replacement it's alright because the BMI-2104 suite is better than the RSS set; but not as good as it should be.
Conclusion 4. This is too sweeping. Extensive review does not meon sat iuate review. Note 7th line page S8 of APS study. Until the work is properly published it cannot be fully reviewed. The statenent that " documentation in its present form would not be suitable for (peer reviewed) publication" should be considered as a comment upon the documentation and not as a comment on the need for peer review. A sad state of affairs.
( Conclusion 7 is a very important one. It is important to note as in the APS study (V.E.3) that this is not dependent on details of the computer codes.
If it did, it would not, at the present time, be believable.
Recommendation 1. Again, an unwise emphasis on analytic methods and not on cantadement strength.
Recommendation 2. The APS study group indeed commented on the need for an integrated package. 12 months ago an adequate one did not exist. "One can place little reliance on elaborate systems unless each appropriate system has been thoroughly tested"(v.D. summary) "The suite of codes does not yet meet both of these requirements simultaneously" (V.E.2 and V.E.5)
Recommendation 3 It is unclear that the Source Term package gives best-estimate re s ults.
Page A-7 I'm glad you finally found a drawing for hurry *-
Page C.1-16 line 6 from bottom. Though " residual" the uncertainties are great.
'Ihe amount of La release is important.
Page 7-9. paragraph 5 is too optimistic. It is unclear to ne that the data has anything to do with the CORCON and VANESA prograns and vice versa. The data
[sfelf has not properly been published so that it is hard to tell. f)
CITIZENS RCRINST NUCLEAR 9@WER l- y gy 220 S. STATE STREET el202 m m PR-wh Marce c"*
- 7 ,y :
EG (NUNG-Q9ffo] RsWc September 17,1985
'85 SEP 23 Pl2:07 Docketing & Service Branch Secretary of the Commission crrici c: Sgt,; ,:.
U.S. Nuclear Regulatory Commission 000F.ETmG & SEf(VI
Washington D.C. 20555 BRANCH Dear Sir or Madams -
I am writing regarding the public comment period set for i NUREG-0956, " Reassessment of the Technical Basis for Estimating i Source Term." I am requesting an extension to this comment period from 90 days to 180, thereby extending the comment time to February 7, 1986.
I amke this request based on the following reasons. For theos of us not based in Washington it often takes several week sto become awareof, request and receive such documents. Also, given the highly technical nature of this document a careful study of the document may take some time especially for thoes members of the
( public who work in other fields.
Given the major impact NUREG- 0956 may playit seem only fair that the NRC allow an adequate time for public comment.
Sincea rely, Y/$
Paul D. E neke 9
y
[ "
g ,
7 my s"U J e- q
.6
-v ACM W -H(y f c%
{! 2hesosass sat (N00.6G-OWh
,i< :
i k -
22'.'[
STATE OF lLLINOIS DEPARTMENT OF NUCLEAR SAFETY '85 SP 27 A1928 1035 OUTER PARK DRIVE SPRINGFIELD 62704
' 'j g
(217) 546-8100 BRAT:CM TERRY R. LASH o.aenoa September 25, 1985 9
i Secretary of the Comission b U.S. Nuclear Regulatory Comission Washington, D.C. 20555 Attn: Docketing and Service Branch Re:
Coments to Draft Report, " Reassessment of the Technical Bases for Estimating Source Terms", NEUREG-0956 affecting 10 CFR\Part 100 s
Dear Sir:
(.:
The Illinois Department of Nuclear Safety intends to submit c nts to the Nuclear Regulatory Comission concerning the dr t titi of the Technical Bases for Estimating Sourc erms". Ho ver, staff " Reassessment preparing the coments will be unable to meet the c ent deadline of ober 7,1985.
Therefore, the Department respectfully r quests an extensio to November 7, 1985, for the purpose of submitting come ts.
Your consideration of this request is greatly appreciated.
incerely,
,4
, ht \ l-
Te ry R. Lash Director TRL:sp 9
f*e rf ff 9 LN 6 Jf4 m s t r 7- II
~ '
"" "'QR-HlSe Nc+icg,
. 040k&-0%)
Coalition for Nuclear Power Postponement
_ y 2612 East Robino Drive, Wilmington, DE 19808- g Telephone (302) 999-7380 "M September 21, 1985
'85 SEP 25 gio32
, Docketing and Service Branch ..
j Secretary of the Commission U.S. Nuclear Regulatory Commission kCIIkh ;'
BIAN .H J
) [
Washington. D.C. 20555 j Dear Sir; I.hereby request in.the names of all people who are trying to rush comments to you to comply with what ,1 consider, is inadequate time for comment, an extention of the present comment period3 to 180 Days or more, in the matter of NUREG- 0956-
" Reassessment of the Technical Basis for Estimating Source Term."
L Consider that this Document represents countless man-hours by NRC Staff Employees, striving to comply themselves with the time constraints put on Staff by the Commissioners, to give them the wherewithall they needed to bring Utilities.beleagured with the costly and cumbersome Backfits mandated after the TM1 Debacle in 1979, out from the Reg-ulatory yoke by cancelling and severely altering the safety-related measures required 3 by simpl; altering the way that emissions from an accident are calculated and assessed.
True to form, but undarstandably, after years .af Proposals, Public Hearings, etc. , Staff
(,
produced the document v e have before us for our final consideration prior to its becoming law in the Spring of 1986.
'Ihe" fast track" status that is afforded this Document in the form of a 90 Day' comment period is appalling in light of the awesome v eight that comments will be given before final approval. But it seems that another rationale vill pervall- No Comments / No Opp-osition!
While years rould truly be needed to study the Document before making comments, that of course.v ould be absurd to ask. But because many of those who wish to make comments have follov ed the process on Source Term long enough not to need background on the subject, a modest increase in the time allowed will makt it possible to elicit well
. thought out public comment, before v e embark on a course that will not only effect the future cost of nuclear pov er in this countrj, but more importantly, will effect the basic safety of millicns of us who live next to operating nuclear plants.
Because this Document vill also affect those vho are yet to live near the prototype
" Standardized Plants". this comment period vill prove to be a momentous occasion, overshadov ed only by the revelation that generic problems with existing nuclear plants will make their continued operation neither technologically nor economically feasible.
1
{
J,jy y , fd d / > _ l o
7 i '[ ,
l2 W - .
SEP 2 51985
-rurc................
II
+ t
- l J Coalition for Nuclear Power Postponement 2812 East Robino Drive, Wilmington, DE 19808
.k Telephone (302) 999-7380 L
U.S. Nuclear Regulatory Commission September 21, 1985 Page Two Until that date arrives I hope that you will allow extra time to those who ponder their comments on NUREG 0956, so that their reservations with the Proposals will be duly documented, v eighed and incorporated in the final Regulation stemming from these proceedings.
I tnist that my request of you will find your favor.
Respectfully Submitted ;
,-)
r s.9/ 1 (;l:
ILLINOIS INSTITUTE OF TECHNOLOGY Lewis College of Science and Letters l Department of Mathematics September 23, 1985 1 Mr. D. F. Ross, Jr. US Nuclear Regulatory Commission l Washington, DC 20555 h
Dear Mr. Ross:
i
- 1. Please provide me with the materials regarding Draft NUREG-0956 (a ;
copy with exhibits). l
- 2. In view of numerous deficient and fraudulent leak rate tests, such .
as the Zion tests, and in view of the present deficient testing ,
methodology which allows to " determine" any desirable value of the
( leak rcte from any set of the data, the source term calculations are of extreme importance to the nuclear safety.
It is impossible to provide concrete comments in the few remaining days (till October 7,1985). It appears that the issuance of Draf t during the vacation time was very unfortunate because it excludes the academia from '
reviewing the subject Draft.
I request that the extension of the comment period to (at least) 90 working days be immediately granted. It is a common practice of your office to extend the legal limits for considering even such urgent matters as emergency relief petitions (petitions on LaSalle and Zion, for example). To deny the academia an opportunity to contribute to the subject matter by setting a deadline for comments in early October is I
very unfortunate and must be corrected. ;
I Sincerely, IO l I Z. Reytblatt, Assoc. Prof.
c, ce 3 gl4;fp@Fi c
~ '
u
.L f==. 851816 IIT Center Chicago, Illinois 60616 U14 soe-nut l IA
Coalition for Sr.cm%g:.cgJ;m=c,:r=;.'!=ina: ! ==,gry:p f/n S'u .
j Responsible "5! ili& pnp J"JI"a WPJ#'duln Jue"czian* '
EU E'M"I*'50" N # E N"EE EM'8 M " "
'^$cEI!EiEI"c! I'IsiSII*E!tI5IriefEEE$!vE"'
4
. u Education co i== = ^ =a "= a=
315 W. RIVIERA DRIVE
- TEMPE, AZ 85282 * (602)952179f 16 P1 :06 l October 5, 1985 !
QF % / i k ;:, I u00XEI'NG & SEi<v:r.; i BRANCH Docketing & Service Branch Secretary of the Commission g we m gl U.S. Nuclear Regulatory Commission t' -- O, P0',d Washington, DC 20555 ERQRCMD .RUI.E3 -
(h $~
Dear Sir or Madam:
I am writing in regards to the public comment period that has been set for NUREG-0956, " Reassessment of the Technical Basis for Estimating Source Term." A comment period of 90 days, beginning August 7th, the date of issuance, is an unfair and inadequate time constraint to grassroots organizations such as CREE. I am asking for an extension to this comment period for the following reasons. Many of our organization members were on vacation during August and early September. Since our office was unstaffed for that period, we did not receive information about this document until late September. We have requested a copy of the Draft NUREG-0956, (' but due to large volumes of requests for NRC documents from other intervenors across the nation, I am sure there will be a 3 to 4 weeks lag time,at least before our group even sees this document. 3 I understand this document is quite lengthy and technical in nature, therefore making the remaining time for study and comment insufficient. To refuse an extension of the comment period would be contrary to the reasoning for having said comment period, i.e. to seek informed review and public comment. I hope that you will view this request favorably and extend the comment period to at least 180 days. I am sure there are many more groups, besides CREE, who need tais time extention, and who will therefore provide you with informed input. Sincerely,
% AL B rbara S. Bush -
Executive Director Coalition for Responsible Energy Education cc: NRC Commissioners ( D. F. Ross , Jr. , NRR Arizona Congressional: U.S. House of Represenatives/ Senate q 4; , g t e + % e
, r -e- g, A p.
~
o
""""L"PRdx)de 6s04 dor'o"aE8c tie 23, .e ea,1.
PHILA., PA.19149 e gg49g (. si ~\ Docketing and Service Branch Secretary of the Commission
.]L, p .. - . ,
- p. ;
IENRC 6 5- " Washington, D.C. 20555 N
Dear Mr. Secretary:
/g 8
- The October 7th deadline for comments on NtREG 0956 is totally unfair ad I believe that the Oct 7th deadline was manufactured to limit evidence reaching the Comiision in a timely manner. I enclose an artic1 about a recent studty and rebuttal by the Pa Health Department. The Study shewed that there were excess health effects in the Three Mile Area that appear very related to the TMI accident. Tha rebuttal of the ya Health Derartment that said the study was flawed used inflated and absolutely wnong manufactured numbers to take the original study to task.
There have been health effects from the TMI accident. These health effects are high4r than predicted by the NRC. There were relaases apparently higher than originally assumed by the NRC and there was proof that ( these releases were higher than originally assumed by the NRC. In light of the recent evidened concerning TMI releases and health effects. I again request an extansion of the emAant perbd on thie NLEEG 0956 Raassessment of the Technical Bais for Estimating Source 'Ivras. Ah o nicase tell me of your decision because I don't get a Federal Register
. and have to take off work to see it at the Library.
very truly yours,
.q .
( % 2.N l M. l. LEWIS 6504 BRADFORD TERR. PHILA., PA.19149 l 4 1 # # f -
,ai" l- -(_
,,g-2 7P-s _ _ _ . . _ _ . . _ _ . _ _ _ _ _ . . . . . . . . - . _ . . . _ . . . . _ _ _ . . ._ _
a
hc,.ews.,.~t.*.hhag . Patdot-ibw HARRISBURG, PA., OCTOBER 6,1985 State's TMI study cloudedi
- by survey ' method doubts By Frank Lynch pI 1 fundatt {Intriot Neaus 4 The state's recently released , FT -- r--h
study cf health effects of the 1979
,D.1
=
M ,* ; Three Mile Island accident may have been flawed by expanding y
'~-- D;1L '
the survey areas beyond the pre- d , M ] ' " "5 - = ' ' o.,rvl . W --- b _ scribed five and 10 mile zones. W. ,. . , ' , According to 1980 census fig-utes. the state Department of
,,,,,,,,j
~ .f.
.erv .""""*.*.... . .- _m. _ - ~j He2tthincluded 28.610 people who i L' )l, j r-t -e.
- ~
*"6'I? d*1""** "'"'
11 ( irther than five miles from + f- u.w.. .- 5 ~
- th) cndonderry Twp. plant in thel
- f. [ -
.d7R#'"?.' c. .... , 1 populati:n listed for those who . '
'. . . - < J rse . c.p.M,. a.r .a j,,, * - '
i live within five miles.... !l M ;14/.P.:. a a ** ' v' I - "" % dr.?~- Another.'l22.000 people .who. t 4"^ :*
.e.. .
c.a., w w N live farther than 10' miles from the ); 4.t=4.... '-*- W' plant were included in the po ulav Q-"'/ .. . ,,,,' il:n cf those. living,p"wijhl ' 10 - miles.
-'~
1 d5
-. e - -
, -$: v
- l. .... .;.,.,,-
'_-1 THE RESULT, accUding to ep-demiologists and statisticians con e i
,w q1 . * $ ? N (w__dif - -, _7
'M -- -
acted ' by the. Sunday *.PatriotJ.,j ._3 _-- _a - 1ews, is that if there actually V d - - -- e_ .
'-'---.7 A
~
vere adverse health effects such '.% , f '. O is increased cancer cases among e. w. _'%.- .W hose living close to the plant, the
'igures would be diluted by ex.
, "[ ,
/ -
#'"I'se'e ns'i!x*e"a#
stra
$g"e' thing SM Areas incl'uded in 5-mile population study i"ui7t.5e'n'A'yEv'a'nfaTale^ tin"v"et ,. Areas included in 10-mile populatiori study tity professor of statistics. "1 think icu would substantially dilute (as- cess cancer rate [in the five mile' ago. concluded that no adverse the tota:s listed by the Health to get even a zone), and not excess rate beyond health effects had been found so partment,14 percent.of the p sumed 'ew cancer rates)I miles away." the five mile zone." he said. "The far in pt-ople who live around TMI. lation figured in the five-Dr. George Hutchison. Harvard larger population would dilute the site of the nation's worst commer- statistics live outside that 2 arefessor of epidemiology, con- overali cancer rate." cial nuclest accident on March 28, while 42 percent of those saidi
- urred. ; 1979.
"Let's suppose there is an ex. THE STUDY, released a month Cornparing census figures with See STATE'S-Page A1D
,('
/P
!. y &wn
- ~
l'g tQ.e # Nd l20h
. Machwd r __ _ ._ _ _ . . . _ . _ . . - . _ .
w 0e***" e sutt P R - A, M dW
...' ( A00.gG.0954) 5 Bechtel Power Corporation Engmeers-Constructors
( Fifty Bea;e Street San Francisco. Cahfornia 1 Mad Address: P O Box 3965. San Francisco. CA 94119 October 10, 1985 ow g ; Secretary of the Comission i U.S. Nuclear Regulatory Comission Washington
,85 C " c]0:21 .
0.C. 20555 l e,. . 1 Attention: Docketing and Service Branch 5 &:cd. c.
Subject:
? >. 7 Comments on NUREG-0956 ;
" Reassessment of the Technical Bases for Estitaatirig Source Terms" Gentlemen:
Comments were requested on NUREG-0956 in the August 7,1985 Federal Register. Sechtel interest in the use of realistic tource terms for regulatory and for design purposes has been reflected in a number of public statements by our people, our active support of the 10COR program and our leadership in the AIF's new special action group on regulatory application of new source term information. Since the " source term" is'the only sourc.e of public risk from a nuclear plant accident, source terms are a fundamental building block for a C,- large body of regulations and have a large impact on the public perception and public acceptance of nuclear power. We are pleased that NUREG-0956 basically supports our view that the source terms used for regulatory purposes in the past have been signifit:antly overestimated. Perhaps even more important is recognition of the fact that the nature of the " source term" is quite different from the original concep~t used as a basis for many of the existing regulatory guides -- the cloud of elemental fodice gas concept led to many design features and regulations that. do nothing to reduce any restdual risk due to hydrogen explosion, steam explosion, etc. As stated in Conclusion 10, the most thoroughly evaluated plant, Surry, showed an overall reduction in risk compared with the Reactor Safety Study which in turn showed significantly reduced risks compared with previous risk concepts. A major part of this reduction is due to the recognition that most of the fodine will combine with cesium to form a salt rather than remaining as elemental fodine and recognition of other natural process which would retain radionuclides in the reactor coolant system and in the plant structures. This is discussed in connection with Conclusion 2. This is recognized in Recommendation One on the same page, XXIII, which says
" Improvements are so significant that utilization of new methods is warranted while additional, confirmatory research is being completed." Another aspect is noted on page C.1-6 -- steam explosions were found not to be a source of significant loads. We believe that enough information is available now to justify using lower source terms than have been used in the past and that the NRC should make the necessary judgments now. We believe that NUREG-0356
( confirms the original reassessment done over six years ago -- the assessment that led to the conclusion that the source terms originally used for regulatory purposes were unnecessarily restrictive.
" _q-a @ @ gr =
OCT 161985 Acknowiced by L. J. . '"""" [ k Y f'f'-
l l
. , Bechtel Power Corporation
(.. l l Secretary of the Comission October 10, 1985 ) U.S. Nuclear Regulatory Commission Page Two l ! We believe that the almost complete reliance on computer codes is not ( warranted. While the title of the document is The Reassessment of the Technical Basis for Estimating Source Terms, it does not utilize much of the experimental information developed from laboratories or the factual information developed from actual accidents and incidents. The discussion in the last paragraph on page 4-3 highlights the fact that many very low frequency events were evaluated only to provide insight into the source term ! phenomena and not because they are significant contributor s to risk. Page 4-28, first whole paragraph, states that the calculations in this chapter I are not meant to characterize the expected or most likely containment behavior, rather they represent simply a study of the potential challenges in a parametric study of the source term. Parametric studies are of limited value in trying to determine public risks and should not be used for regulatory processes esrecially if it is known that the cases identified are not likely to happen. ae believe that this purely mathematical approach is (. very confusing and misleading. For example, on page 3-6, it is stated that for Surry an early containment failure scenario was analyzed even though the peak pressure was well below the estimated failure pressure. In the discussion relating to Recommendation 2 on page 8-7, it is stated that the
" Source Term Code Package" was reviewed only in the context of severe accidents that completely melt the core. Again at the bottom of page C.1-7 it is stated that the steam spike induced failure of the Zion containment is an event of very low probability contrasted with the high likelihood events which came. out of the Reactor Safety Study methodology.
Page C.1-15 (and the bottom of page 14) is another example. It is stated that "Rather extreme assumptions had to be utilized to produce loads of sufficient magnitude to challenge a large dry containment" including dispersal of 100 percent of the core and 100 percent oxidation of all cladding. There is too much emphasis on these improbable sequences and too little discussion of more probable sequences that should be considered in the plant design and procedures. The sensitivity of the conclusions to containment failure or bypass of suppression pools (for pressure suppression containments) raises questions about the types of reviews performed to date. This would appear to be a question of civil engineering rather than a question of basic physics. IDCOR devoted a significant part of their effort to the containment question. They concluded that tne containment capability was much greater than previously estimated and that even if the containment failed the fission product release to the environment would be much less than estimated in past studies. { 7b
Bechtel Power Corporation G Secretary of the Comission October 10, 1985 U.S. Nuclear Regulatory Comission Page Three The American Physical Society group recomends further research but it does not state that research must be done before regulatory actions can be taken as is stated in Conclusion 13. In fact, the American Physical Society report essentially states the opposite - the APS specifically stated that they included no considerations of regulatory actions and did not connect the two at all. More detailed coments a"? contained in the attachment. We appreciate the opportunity to comment eri this important subject and would be pleased to meet with NRC personnel to discuss our views in more detail. Sincerely,
'& f)h M. Levenson
( ML:bbe Attachment m 7c
1 e COMMENTS ON NUREG 0956 !
, h Page XXII Item E Conclusion Six, states: "These sequences have provided a i
sufficient test of the capabilities of the computer codes", but there is no
' indication that the codes have been benchmarked against accidents, incidents or experimental work. An analytical analysis that has not been benchmarked has only limited credibility.
Page 2-8, Item 2.4 - 2. The attenuation for elemental fodine is also greater i than assumed in the Reactor Safety Study since this study did not recognize that iodine reacts with material such as steel, light metals and concrete.
)
Page 2-10, the references to the German risk study are only to the original 1 old German risk study. The newer, revised study which shows much lower source ! terms has not been referenced and should be used instead of the older report. Page 3-6, large full paragraph, center of the page. The statement is made "An early containment failure scenario was then analyzed, even though the predicted peak pressure was well below the estimated failure pressure". This is an example of a parametric analysis being supplied an input that the actual analysis says will not happen' This is very unfortunate because the subsequent analysis and risks associated with it have not been properly identified as being parametric rather than the result of the predictive l analysis. Page 3-9, Paragraph in the center of the page. The entire core, the support (, steel and molten steel from the bottom head of the vessel are all assumed to fall coherently from the vessel upon vessel failure. This appears to be in conflict with all experience in the physical world in such phenomena. It may be easier to model it this way for the computer but doing so ignores the physical factors that yield incoherency. Page 3-9,' bottom paragrah. The discussion that the calculation of containment loads and the calculation of container performance with the March 2 code "is not considered reliable", is extremely important. Why is the code used for a calculation if it is not considered reliable? This is especially true when the same paragraph says that predicted failure of the containment is controlled by the criteria input by the user rather than by the scientific analysis. Page 3-29, Item 3 at the bottom of the page. The early containment failure is an assumed event for parametric reasons and should not be included in risk studies that people will assume represent " calculated" cases. It also should not be used for regulatory decisions. Page 3-32, Item 7, top of the page. This discusses resuspension. There are no references to support the assumption that a major fraction of the resuspended material is an aerosol. Material that has coalesed, plated out, settled out, etc., might be ejected by violent forces but there does not appear to be a mechanism for making any significant fraction of such material a stable aerosol. It is most likely to be " chunks" or " flakes", that fall like rocks. i 7d
' Comments on NUREG-0956, continued h
Page 3-39, Discussion of in-vessel fission product release from the fuel and aerosol generation uses the assumption that 100 percent of anything vaporized becomes a stable aerosol. I believe that in the past the government has spent hundreds of millions of dollars attempting to generate stable aerosols in connection with biological and radiological warfare programs. None of these programs had any success -- only the computer seems capable of generating large quantities of high density stable aerosols. Page 3-41, in containment failure modes, why is the extensive experimental program on reinforced pretensioned/ post-tensioned concrete-structures that was performed for the gas cooled reactor programs not referenced? This represents a major source of data on failure modes of concrete pressure vessels. The experimental evidence is that they do not fail catastrophically -- they crack, vent and close up. Page 3-41. The summary states that the codes can be run successfully, can be transferred and have been peer reviewed and therefore by implication they are correct -- but there is no statement of whether the answers given are accurate and no statement about benchmarking them against the physical world. Page 4-3. The first full pargraph at the top notes that the signficance of this accident sequen:e lies entirely in its challenge to the codes since it is (~ generally agreed that its risks signficance is very low. Such analyses should \- not be included in a report whose objective is to analyze risk and make regulations. Page 4-3. The last paragraph refers to a number of sequences that were assumed to progress to a severe core melt. Here again the basic idea of letting the codes help one predict when there is or when there is not a risk is completely bypassed when one interjects such assumptions. Page 4-4, under TMLBC sequence, last item. Such technical documents should differentiate between "could", meaning "not impossible", and what is really possible. (40,000 747 airliners "could" crash in one week but it is not possible -- that many 747s do not exist. ) The general public translates could to will and while this document is not written for the general public, the intervenors and certain members of-Congress will make this a public document. Page 4-10, first full paragraph at the top, sequences including containment failure have been selected to provide a test of the codes. This philosophy has been commented on before. Page 4-14, Figure 4.2. Logically it seems difficult to believe that the temperature of a peripheral node at the middle of the core will rise so much faster than the central node at the top of the core. The peripheral nodes at the center of the core have a heat sink in the form of the reactor core barrel. Page 4-28, first and second full paragraphs. It is not clear how one can come ( to the conclusion that the analysis is specific to sequences and to plants when some of the analysis being done are parametric analysis and not actual modeling. 2 1
Coments on NUREG-0956, continued ( Page 4-35, last full paragraph, puff releases at containment failure are not likely to be largely aerosols. Material which has settled out or plated out or coagulated may be dislodged or ejected but what mechanism exists to disperse it finely enough to generate an aerosol? Page 4-36, Figure 4.14. It is not clear how the quantity of material leaked to the environment can be almost an order of magnitude greater than what has become airborne. This applies to the time between between 200 and 350 minutes
- maybe this is just difficulty in interpretation but it isn't clear.
Page 4-44, top of the page. It is not clear how one can assume that the time of release for everything can be taken as the start of core melting -- even if the containment building airlock doors were wide open. There is a significant time between start of melting and when anything could be released. Secondly, the input by the user of values such as failure pressure or failure temperature, independent of the design, makes this a parametric analysis and not a true risk analysis. Page 5-5, first paragraph at the top referring to the APS review. While the APS review group might have been broad based and independent, they appear to have only analyzed the NRC funded code work. Therefore it cannot be stated that their study accurately reflects the status of all the source term (.' analytical work and support research. The programs of other countries and the programs of organizations, such as EPRI, do not appear to have been thoroughly
, reviewed by the APS group.
Page 7-9, second paragraph under Containment Pressure Loads. Some of the hydrogen behavior work done at Sandia is done in shock tubes and care must be taken in assuming such data is relevant to actual containment building condi tions . Sandia people involved with this program have stated that the work was not directly relevant but was done only to get a fundamental understanding of the hydrogen ignition process. Page 8-7, Recommendation 2. Since the computer codes that comprised the code package were reviewed only in the context of severe accidents that completely melt the ccre, no recomendation is made regarding analysis of lesser severity. Since the complete core melting does not necessarily arise from the i code itself but from either input assumptions as to when and how failure occurs or by parametric inputs, it is not clear what this means. Page B-8. The arbitrary assignment of values to verbal descriptors is entirely arbitrary. The range from unlikely to certain has been given a factor of 10 in two cases and 100 in 2 other cases. It would be no more l arbitrary to differentiate certain from unlikely by at least four orders of magnitude. Page B-9, last paragraph. It is not at all clear why the opinions of a
" reactor safety comunity" have any special validity on a subject that is
.(- technical and somewhat narrow in scope. These opinions might be more meaningful if the opinions of the people who are structural / mechanical experts were separated from the group of people who are analytical / code / mathematician / computer experts (and not necessarily experts in questions of structural mechanics).
e Comments on NUREG-0956, continued h Page B-15. The discussion under Surry, first paragraph, raises a very interesting question. The implication is that the building spray system reduces safety because it increases the probability of a hydrogen burn? -- such a burn being one of the few residual risks left. Therefore, one has to ask -- is the spray system a safety item or does it increase the risk? I l Page B-15, under Surry, the assumption that a meltdown is coherent is not technically supported. Every operating reactor has a significant range of , power densities, has a range of cooling factors, has a range of heat sink ' availability, etc. To assume it is coherent is the same assumption that for some years prevented the boiling water reactor from being developed. The argument made by many of the reactor safety comunity at the time of the original safety reviews of the BWR was that all of the steam bubbles might collapse coherently and the reactivity added would be enough to blow up the plant. In fact, page C.1-5 identifies that core melting incoherencies is widely recognized. Page B-17, second paragraph, under Sequoia Discussion. It is hard to understand the assumption of a 50 percent probability of containment building failure. This seems to conflict with all the discussions on containment integri ty. k-Page C.1-5, last paragraph. Why were the standard problems not structured to be best estimates rather than perhaps high by several orders of magnitude? Page C.1-6, under Probablistic Aspects, the conversion from analyses that are parametric to a statement that this represents the uncertainty range is questionable. The parametric range is absolutely arbitrary and does not necessarily reflect any technical (ncertainty. Page C.1-7, Results for SP 1. Since it does not appear that particle size or reaction rates have been taken into account, it isn't clear how the results can have " negligible uncertainty". The role of the passive heat capacity of the thousands of tons of concrete, steel and other materials must certainly play a part -- a part that varies depending upon rate of heat release. Page C.2-3, the third paragraph. This identifies that the capability pressure has been assumed to be the catastrophic failure pressure and that 1 percent tendon strain in a prestressed concrete containment has been assumed to cause catastrophic failure. The actual experimental work done on the HTGR tests, etc., would indicate that this is very far from the real physical world. Not only is there no evidence that catastrophic failure occurs for reinforced concrete structures at this pressure -- there is in fact no evidence that reinforced concrete structures of this type can be catastrophically failed at any pressure. All the evidence indicates that at some strain, concrete cracks, relieves the pressure buildup, and then closes up under the tensions ( of the steel . 4
m NttEBER P. O. Box 14000. Juno BE ACH. FL 33408 l , 290PO$fD RULE M ! (NU2EG-0937o) % l o' E_- A L:/ FLORIDA POWER & LIGHT COMPANY
. l Y OCT 2 2 toss f 383 W%p'/.3
.c 1
Mr. Samuel J. Chilk, Secretary - U. S. Nuclear Regulatory Commission ' OCT 241935 f2' ' Washington, D.C. 20555 c, s a g Attention: Docketing and Service Branch 4 ca &
Dear Mr. Chilk:
The following comments are submitted in response to the issuance of l NUREG-0956, " Reassessment of the Technical Bases for Estimating Source Terms"(50 FR 31937, August 7,1985): l l These comments do not address the detailed technical aspects of NUREG-0756. Florido Power & Light Company is a participant in the Industry Degraded Core Rulemaking (IDCOR) program and { regards IDCOR as o proper forum for performing a rigorous technical review of the NUREG. At this point our comments are confined to the prospect of using source term research results to support changes in regulatory requirements. We believe that the nuclear community has enough information to identify potential changes to'the regulations and that the Commission should set int'o motion the process by which such changes con be codified. We think this should be a high priority effort and urge the NRC staff to stort developing positions on regulations that con be improved. Very truly yours, p y' ,r,ds,Jr.
. . Will' Group V' President Nuclear Energy
. JWW/ MAS / cob Q.b((b b* 5
( l - f '- MASI/001/1 OCT 2 51985 se knowie+yn; : m . , , ,, , , ,, ,,,,,,, 9 PEOPLE, . SERVING PEOPLE a
DON IKK."CiN
,f .
,mPR-M m Neut w O10l%-c956) 9 October 24, 1985
{'d
~
CONNENTS OF OHIO CITIZENS FOR RESPONSIBLE ENERGY, INC. ON NUREG-0736,
- REASSESSMENT OF THE TECHNICAL BASES FOR ESTIMATING SVURpE
+ TERMS * ', ' I ' * '/ / 10 The NRC has recently released NUREG-0956, entitled
~.p s ,
" Reassessment or the Technical Bases ror Estimating Soure 94 -
4> Terms,' ror public comment. Ohio Citi: ens for Responsibi e 4.3. ,. sr 2 's Energy. Inc. ('OCRE') hereby submits its comments, pursuon ko, ,g the notice in 50 Fed. Re9. 31937, (August 7,
/ .."
.a' 1985).
OCRE believes that NUREC-0956 correctly presents the state or technology on severe occident source term evoluotions i.e., highly uncertain. For example. NUREG-0956 contains many statements revealing the immaturity o'r this technology. It is frequently emphasized that there ore large uncertainties ossociated with source term calculations (pp. 4 4-46, 8 8-2). Areas in which uncertainties persist are notural circulation in the reactor vessel, core melt progression and hydrogen generation, in-vessel fission product release from fuel and aerosol generation, retention and revoporirotion or fission products in the reactor coolont system, rission product release and aerosol generation
~
from core-concrete interaction, scrubbing erriciency or 1 suppression pools and ice comportments, and contoinment pressure 3 '. loods and roilure modes. P. 3-38. These are moJor oreos encompasssng almost oilstacets g or severe occident progression. The roct that uncertaintie's' remain in so many creos reveals that this technology gs mo [s.suitople .... ror application to the NRC's regulatory purposes at Q'4.3-d M A % c-9. D t his-k m e tth',= W ( . In addition, source terms are round to plant-specific and ..
< -n ! Q.&s - ~~'
OCT 31 sB5 IO 4 f'F- Acknowrewd by ca rd . . . . . . . . . . . . . . . .
- ~ - .
y
.1 . '. .
' accid 3nt sequonco Spocific (p. 4-1), os woll os strongly l t '.
dependent on details or plant design and construction (p. 8-4). l 1 It is also cautioned thot the onolytical procedures are complex
. [
\ ' and involve several scientific disciplines, and they must be subject to a quality assurance program because 'there is o signiricant chance or making mistakes.' NUREG-0956 at e-2, 8-3. The NUREG olso correctly reports the rindings or the Americon Physical Society ('APS*), with which OCRE ogrees. APS noted the uncertointies ossociated with source term predictions and specifically expressed concern on the lack or experimental validation or computational models and the insurricient evoluotion or occident sequences and phenomeno which could result in increased source terms. APS concluded that '(i)t is impossible te make the sweeping generalization that the colculated source term for any occident sequence involving ony redctor plant would always be o small fraction or
, the fission product inventory at reactor shutdown
- and 'it is not possible to derive roctors by which the source term For all radionuclides and all reactors con be changed from the values reported in the Reactor Sorety Study.' APS, Radionuclide Release from Severe Accidents at Nuclear Power Plants,
. Conclusions and Recommendations and Executive summary.
No doubt in recognition of these uncertainties and criticisms, the Foreword to NUREG-0956 states: The next step orter analysis or comments con NUREG-09563 will be to complete o study of the opplications or these models to the risk oppraisal or rive reference plants. That work, which goes beyond the science and engineering of NUREG-0956, will, when
- published, also be issued for brood comment and review. gnly then will we be able to use this new technology in a review of regulatory practice.
( NUREG-0956 ot xiii, emphosis added. Mb m
't. .
, Howevor, in epito of theco cov20ts, Recommendation l' (p.'8- .
E e
- 6) states that the new methodologies should be utilized to
( reevoluote regulotory practices before the completion of r j-
; research which would remove these substantial uncertainties.
l This is o total nonsequitur, and it reveals that the source term l l i I
'{ issue really involves o political decision which has already j l
lf been mode, i
, The fait occompli nature of this issue (re9ardless of what l I l further research may show) is demonstroted by various NRC policy i documents and statements. See, e .'g . , the Re9ulatory Agendo, wherein proposed changes to emergency planning and reactor l$ '
siting regulations are on the books and owoit only on official ' i
'f go-cheod. NUREG-0936. Vol. 3. No. 4. pp. 79. 109. The Policy l
,$ ond Planning Guidance portion or the NRC's Annual Report hos ;
,; emphasized the source term issue and its regulatory implications
. since 1982. The NRC 5torf hos even 'put out a pleo recently for t
- 1) industry representatives to tell NRC what regulations they wont t
] _ changed in response to source term research ' Inside NRC. Nov.
- 10. 1984. P. 14.
s It would behoove the NRC to reco11 that, under the Administrative Procedure Act, changes to regulations must be supported by the rulemoksng record. 5 USC 553, 706. Nothing in NUREG-0956 supports any relaxation in the NRC's regulations,
; given the uncertainties and lock of experimental Volidotion of this technology. Regulatory changes mode before the completion of confirmotory research would clearly be held orbitrary and copricious by a reviewing court. Since the NRC certainly wants to conform its practices to the low, the NRC should agree with 'k Mc L
__-_--__- -- - ._ - _ . ~
, , -- -___--..,........_; ....._._m.
,s . , e . .. -t . .
.e $ .' *0 dR,E t h a t no regulatory changes should be raade bef ore tho e . , .
}
- CORP 1Gtion Of ConfirDatory r@500rch, t ,
Respectruity submitted. 3 I
=
I Suson L. Hiatt OCRE Representative 8275 Munson Rd. Mentor, OH 44060 8-(016)255-3158 I s' l i 4 I .h 6 s
'I D
4 i*P
-l5
=
h 10d
r ff C6agpon for m;'J;/statma;m;;;cawegn,;; mra,ra:a .
? Res Toasible "#E!diMJ M'!J" Jai 'JHn' '#MWJ#Mie"Jann' '
Ene w , N E"E N W E"og;EUGEE 2"n;;g;pjgg;cy;,;ggug;g;;gg g;gpu S' N e Education ====*-"u-== , f RIVIERA. DRIVE 315&w.
- TEMPE, AZ 85282 * (602) 968 g ag pguuggR ;
N 4 October 27,41985
. ,n i PRW h 3- 4 g
Docketing & Service Branch LNU -
-i .~
Secretary of the Commission $' 0g OgG "h .5ca [f U.S. Nuclear Regulatory Commission C 9c , Washington, D.C. 20555 S # s * 'idJ
Dear Sir or Madane:
fpei
. m@ \
On October 5th, I wrote to you to request an extension of ninety days beyond the already alloted comment period on NUREG-0956, based on the time delays involved in securing the draft proposal, the lengthy and technical nature of that proposal, and the need for detailed and informed review and public comment. Such an extension would enable interested citizens and intervenors nationally the opportunity for such review and comment, something not possible under the present time limitations. I am writing again concerning additional difficulties which have been placed on respondents such as ourselves. On October 26th, I received a response to my request for a copy of (, Draft NUREG-0956. The response stated that I had to mail our group's request to a different address, and that I needed to enclose $10 to receive a copy. We mailed to the exact address published in the Federal Register (see enclosure). Moreover, the Register made no mention of the $10 fee. Effectively, we have lost another month before we can have the oppor-tunity to review this important material. In fact, it will be impos-sible under the existing time limitation, as set forth in the Register (supra) for us to review and comment on the draft proposal. Obvious 1v, any other potential respondent likewise relying on the information published in the Federal Register will experience the same problems. Considering these extenuating circumstances and the technical nature of the material, it is only fair that the original ninety day comment period be extended. In light of these hardships affacting our organization and other possible respondents who have relied on the Federal Register, to refuse said extension would be contrary to the reasoning for having the comment period: to seek informed public comment. Sincerely, g g ( Barbara S. Bush, Executive Director Coalition for Responsible Energy Education cc: NRC Commissioners /Ross/Az. Cong. delegation Enclosure $2 3 - , , A c %; L . v>~( i7 2 PP-
. F:dct:1 Regist:r / Vcl. So. No.152 / Wednesday. August 7.1985 / Noticss 31937 l
"aw'ar'd'cne. tim 2 minigtenta f:r the gne end cndinfSeptember 619M. in the Chritmen. hr. W. Reed inhnson. Ctry J.
year period beginning September 6.1gy following a mounts: Edles.
- -~ . Daltd. Augast 1.19a5
- w ee l wu. e ne.a. i"'"",,', C. Jean $boenatet.
g Secerrory todeAppealBoard ca.ww im c , as , _ p . =c ,.,a.ae ,com.,. t,. n , w e. Y, , , s eco PR Doc 85-15r28 *, fed &+85. 3 45 am) a cave, i.e.:a ra cesrme. o o e a. ta m s a. vs , o se a orwr. _' s2no owwo ecoc rho-et-e p.,e e , - m.m s v.wr t.o. sm n.e. m wo = .e or,wc e. ..ocen. . ss. oe,. I i em im .
- em r. 49. v oa
. w v. n = +s ,, , as an ssuance of NUREG-0954; c, me, t chaa to.-p w..y ga.u ur msw.a u . mov .e-. *to. sew w wi s'** g " Reassessment of the Technical 8 asis t.,r a4 som a c.,m.=,
en. wete.gaa m . u-~3' o%oo--e e see,c.me see.r e-.mo .--s.n ,e __. s.w t., 44 saar a n c. _ --,e is, ' io no for Ea.tirnating Source Tetrns,. vw <=. -o-.- w.m i v.y w.. c+ . a.w.a u _ o , a w.
,c= io ma .aa me L..v., wo. i v. s .a pue.6 sm The Nucleat Regulatcry Commission o, L. < am c.,. a A i
. e ~w i. ,. coc.v. o,.o n.an .g.ao.=,n
- e. .v ,. t; no has issued a report. NUREC-0956. (DrafI
- e. u a c. aa e. ,o%.a e, awa si.o. - for Comment]" Reassessment of the
*' Technical Bases for Estimating Source aum s. u o . m.e w . _ o m"m er we i . , e r e, au- a ar soo Terms. The report was written by the sua u. t sm c,wc.j. o..w,- -e a , e u.w u.m m. . ,l is om NRC eta (f.
se re en ow. .n- . e-. w -, -. -. j DrafI NUREC-0956 describes the
.pi s,=. . v,e is, ma c .r. sas effort by the NRC staff and their cgon .% ; t,. _
_) contraetors to reassess and update the
>= t c .r.w smi a.e. vt._ - ww. . a v. po, e : a too
_.jcn. m. r.e ena==5r.e o...a w-,
.. < , e em -- . __ - ;.' analytical procedures used to estimate em severe accident source terms for nuclest noi, t.o. s . c.a w ...a ,,,v .__.qwie u m w.5=
t.u. 6.i, s . .*mnme. ei mei. mawe n a voo aan -r. weo*.es.. l- t aan power plants. A source term is defined a w -. s . = L.p.see a.reavn ,o, w w=*v .m . ,eq .aw ..m.m as the Nuantity, timinI, and f.e (so. S. .i c.,w p.yan we D = pa.se a.8. m or.wm2 8 6.J.itreig eat 7 tow is - . - . . . _ . _ . i sisist. characteristics of the release of
, radioa6ve material to the environmeat
. following a core melt accident. Source terms are used in many areas of the was made for field These awards pre made to supplement funding availhbilitfand state support the funding of tra:rurig activities of held programs. nations "8f*d,{P oc, s e3r port, the Y
prrgrams, nation il and state support units. bar ass iciations. law schools and pr cedures, the results, the review units. bar associa tions. law schools and other interest *d parties to supplement process, insights on risk, future research. other interested 1 arties, the funding of training activities l Request fori roposals: Training '"d C "CI"8I "8-These furids w.ll be awarded on a Public comments are being solicited non recurring bms under the authont) Develepment and Delivery of Pub. L. 06-411 and section memorandurr . April 3.1985). on Draft NUREC-0956. Comments stE(a)(1 gB) and section too6'al!3) of should be sent to the Secretary of the Re Legal Lervices Corporation the Legal Services Corporation Act of intends these grants to increa.e and nunini n. uclear RWaton 19~4 as amender.
'"
- I improve the quality oflegal sernces to Attention: Docketing and' Service There will be no refunding rights for the poor throgh the training of the these one time g ants. Branch. (by October 7.1985 )
Individuals oho will provide those legal Notice is issued pursuant to section services, Draft WRE-66 is avadaWe for 100"(f) of the Letal Services Corporetion John C M')*,' inspection and copying at the NRC
. ended, with a request Public Document Room 1717 H Street.
Act of 19.ts4a as afd recommendations for comtr.e h"* &* &l* d' 'C'5 Washmston. DC. Copies may be withm a pened f thirty (30) t ap from im Doc aF sus Ned a-6 45. a 45 ami obtained by callmsgo21275_:oco or the date of pub ication of this notice. swuc coca -4:o46 m wntmg the Supenntendent or The grant a ard will not become Decuments. U.S. Govemment Printing
~~-
Office. P.O. Box 37082. Washmgton. DC , effcctis e and s ant funds will not be distnbuted pnst to the expiration of the NUCLEAR REGULATORY 20013-7982. thirt). day per o d. , COMMISSION (5 tLS C. ss2 (ell D ATt A;l comments must be received b) Dated this nda de> of August 1985 the legal Serdcas Corporttion wtthin 30 (Dock et No. W20.Ot.Al For the Nuclear Regulatory Commsvon days from the date of publication on this Denwood F.Ross lr, Boston Edi non Co. (Pilgrim Nuclear
"*"*** Depurp Director. Of te of Nuclear Regularo.7 Power Stat \on): Assignment et Atomic FOR PURTNrR ;NFO tuATION CONTACT Sately and JCensIng Appeal Board S'I'#II A Cynthia B. He ne. Nationat Trairun8 [FR Doc. 45-1B*:- Ned 84-85, a.45 am)
Coordinator, t.egal Services Notice is hereby given that, in s wi,a cootrseu i.a
, Corporation. Office of Field Senices. accordance with the authority conferred 733 Fifteenth Ltreet NW Washington by to CTR : 787(a). the Cha!rman of the .
D C.20005. . Atomic Saft ty and ucensir.g Appeal R AILROAD RETIREMENT BOAR.D SUPPd WENTABY INFCRMATlo*c Grants Panel has a" signed the followi3 panel are awarded on a competitive basis. and members to serse as the Atomic Safety Agency Forms Submitted for CMS ( purst. ant to the Legal Services and Licenang Appeal Board for this Review Corporanon's onnouncement of operating lijense amendment suiilabilit,5 of 'und . Announcement of proceedmg- Alan S. Rosenthal. , ActNcy: Railroad Retirement Board M eWcAmid f
4U4 LE 12d. O g[ CNea:sm,; s.~ , sistans. sanvANTs or rMs suuAcut.ATs McAar or [my eso west nu Avsmus s ,g u onnes. usewooaw . .. 2 ]
- Y~g.
}
October $ % k. -
;, . , h. ,1-)
Docketing ervice Branch
- Secretary of the Commission /, , -
U.S. Nucimr Regulatory Commission vashington, D.C. 20555 ( Dear Sir or Madans I am writing regarding the public comment period set far NURm-0956,
" Reassessment of the Technical Basis for Estinating Source Tara."
It is sy understanding that the comment period is 90 days from the date of issuance, August 7th. Such a period of time is inadequate. I sa hereby requesting an extension to this comment period from 90 days to 180 days for the following reasons, yirst, NUREG -0956 was issued in the summer month of August when many people were on vacation. Second, the sheer length of this document merits a longer than usual comment period. NRC routinely publishes proposed f' rules, auch shorter than'.this NUREG, for which the public may be given several months in which to respond. For example, NRC extended the comment period for the Decommissioning Rulesaking from May to November,1985, thus allowing a comment period of nearly 9 months. (* Equally important is the highly technical nature oftthe document. I-necessitating far nors work than otherwise, and requiring potential ' commenters to seek out qualified persons to analyse the technical infarnation presented. Fourth, it usually requires several weeks for prospective commenters to become aware and request copies of such documents, thereby decreasing the time availabli to them to review and comment. For exmaple, Federal Register notices often - are not available on the West Coast for two to three weeks after l' publication. IAstly, and perhaps most important, given the major role ' NUREC-0956 any play in the changing of NRC's regulations, it seems only fair that NRC allow an adequate comment period. In sua, there are sant reasons for NRC to extend the comment period , for NUREG-0956, and few, if any, reasons not to. I hope the . Commission edll respond favorably to this request in an effort ' to seek as ~agsk review and public input into its analyses of source ' tern as impible. Sincerely, . lOnkmy er Al Sister Barbara Bacci, IHM - ' DIM Responsible Investment Committee ( _ g ' Li.f 0 [ ,07 L L p. 3t
r 1
'e<
M l
.ec e n e en - m** - . .
onFOSft' ~M ,T Cl-
'__t f ;
(ftl0826-C9&r.) l f amam LONG ISLAND LIGHTlNG COMPANY I i snOREH AM NUCLE AR POWEm ST AtlON *
- O BOM 628
- W ADING river. Ntw TOMK 11792 NOV 0 71985 ;
4 Mr. Samuel J. Chilk, Secretary U.S. Nuclear Regulatory Commission Washington, D.C. 20555 66, , B
Subject:
Request for Public Comment NUREG-0956,
" Reassessment of the Technical Base for Estimating Source Terms"
Dear Mr. Chilk,
on Wednesday, August 7, 1985, the NRC published the subject Roquest for Comment in The Federal Register. The Long Island Lighting Company wishes to provide the following comments and cbservations on the subject. The BMI-2104 suite of computer codes provides certain specific advances in state of the art in simulating the phenomenology of core melt scenarios. Major advances have been made in the area of treatment of chemical forms of iodine and other fission products. For the five reference plants studied, however, none are similar to Shoreham's BWR Mark II containment. In addition, the NRC staff has concluded the following regarding the source term reassessment:
" Source terms were found to depend strongly on plant design and construction details, thus making development of useful generic source terms difficult (pg xxii) . "
It should be noted that due to the strong influence of the core-concrete interaction on Mark I source terms, that extrapolation of Mark I results to Mark II design is not valid. This is especially true for Mark II designs such as Shoreham that have downcomer vents directly below the reactor vessel, thereby providing a drainage of corium to the suppression pool which limits the core-concrete interaction. Therefore, in associating Mark II source terms to a Mark I design, no account is taken for the quenching of a molten core by the suppression pool water, thereby retaining most fission products and reducing the potential for containment over-pressure failure. p 3 y'tj d T 1 rr- n.- w wf
. .. i. -
Il
e e . V t Section 4.9 (Aerosol Retention in Containment Pressure Suppression Pools) does not reference the extensive EPRI test program underway at Battelle-Columbus. Validation by experimentation was a big concern of the APS report on source terms. Table 4.13 entitled " Plume characteristics and release fractions for Surry and Peach Bottom Sequences" provides detailed descriptions of sequences assigned to WASH 1400 release categories. In reviewing this table it would be helpful to know, if possible, the expected frequency of release on an absolute or 1 relative basis. Table 5.2 provides a comparison of IDCOR and BMI-2104 source terms. However, the corresponding text (Section 5.3, Pages 5-6) does not adequately describe the reasons large discrepancies exist between IDCOR and BMI results. For example, why is there an order of magnitude difference in the time of release for the Sequoyah station blackout sequence? In view of the advantages noted above for the Mark II containment, we believe a source term reassessment for the Mark II. design would show diminished and more realistic values. ( LILCO wishes to thank the Commission for this opportunity to express our views on the important issue of the source term and NUREG-0956.
..c .
~_ .
~ .p. r 2 . .
John D. Leonard, Jr. #- Vice President, Nuclear Operations
'-DF/ja
( flb
B ., Spxcific Comments ,
, 1. Page 4-32, top cf page. It is stated th:t thD siza of the conteinm:nt opsning l- (in the range 2.1 to 11 sq. inch;s) dass not effcct tha p;;ak containment pras-sure in the THLB' sequence for Zion. Failure to isolate (28 sq. inches) does affect peak pressure.
4
.( ! would take issue with the BH1 analysis which went only to 15 hours. The IDCOR analysis for this sequence predicts ultimate containment failure at 32 hours.
{, Would such failure be precluded by containment openings in the range 3.1 to 1 11 sq. inches?
- 2. Page 4-37, first para. under 4.10. It is stated that the " amounts and timing of releases of radioactive material to the environment" were determined. The timing of releases in the BMI-2104 methodology was more assumed than " determined.
The analyses of the containment working groups was not factored into the BMI sequences. In effect, BH1 arbitrarily selected containment failure pressures and openings for illustrative purposes. The writeup should more accurately describe this part of the process.
- 3. Page 4-40, Table 4.13. The 3-asterisk footnote should also be attached to the AB- and THLB'- sequences listed in the table and the footnote reworded to state that the BHi-2104 releases are to ground and not to an atmospheric plume.
- 4. Page 5-3, last para. It is stated that the APS group consisted of "12 scientists" (true), "most from the academic community" (true), "two from non-nuclear corporations" (not true) and "one from a national laboratory" (not true).
The composition of the group might be characterized as from universities, 7; retired from national labs, 3; from a private non-nuclear corporation, I and from a non-nuclear government agency, 1. { 5. Page 6-1, bottom. I question the desirability of reporting a " comparative risk appraisal" without a re-evaluation of the frequency of various accidents sequences.
- 6. Page 6-5 and 6-6. Figure 6.1 and 6.2. These figures should be deleted for a number of reasons. The " comparative risk appraisal" on which they are based does not include the re-evaluation of frequency (see comment 5). They display 6000 early fatilities for WASH-1400 information where WASH-1400 itself displays
. only 3500, without explanation (see attached figure from WASH-1400.) The analysis on which they are based (Appendix D) has a number of flaws (comments 10.1 and 10.2 below).
d . M1All Oak Ridge Associated Post Office Bos 117 # b-Universities Oak Ridge. Tennessee 3 830 Silve r Sp r ia 9 Y& y c=-4, gy[ n n ( } m pp 7 II
e (
- 7. Pages 7-8 and 7-9. 1.ast paragraph of 7.1.5 and last paragraph of 7.1.7, These paragraphs are not related well to each other and to the rest of the report.
The " direct heating" phenomenon should be evaluated in a coherent way in the NRC program, with the goal of defining if it is a significant challenge, and , if so, it needs quantification and modeling. The reference to Section 7.1.1 makes no sense.
- 8. Page 8.5 Conclusion 10. As stated in Comment 6, I believe it's premature to publish quantitative estimates of fatality reduction.
- 9. Page 8.7 Recommendation 3. The Source Term Code Package is said to be designed to provide "best-estimate" results. While there may be no intentional bias, NRC contractors have tended to select " conservative" input parameters that give answers on the high side of the likely range of variation, Tor example, BHi's choice of containment failure pressures tend to be on the low side. Some discus-sion of this issue would be desirable somewhere in the report. I don't believe it is equally likely that BHI-2100 underestimates releases as opposed to over-estimating releases for the sequences covered.
- 10. Appendix 0 comments.
As a preamble to understanding the Surry risk, I prepared Table 1. This table shows that 77% of the " risk allocation" comes from the V sequence (bins 11 and
- 12) and 12.2% comes from the early containment failure mode (bin 5). The table assumes that risk is proportional to iodine release.
10.1 - The V sequence. The BMI-2104 calculations of the V sequence has had little or no exposure to the peer review process. The release fractions reported are high relative to those reported by others for the V sequence, i.e. Stone
& Webster for Surry and IDCOR for Zion (Zion is not really comparable.)
Some of the BH1-2104 results are counter-intuitive. For example, the secondary building removes about 75% of the radioactivity in the AB-p sequence but only 20% of the radioactivity in dry V. Why? NRC should subject the BH1-2104 V sequence to peer review prior to its use in a risk analysis. This risk estimate is very 10.2 - Containment failure prior to core melt (Bin 5). high for a number of reasons.
- a. Thelargert.0CAfailuresconstituteonlyabout10%ofthisbin,yetAB-$
was used to estimate release, eliminating the effectiveness of primary system trappingofaerosols,Characteristicforsmall-breakLOCAs. THL8-fould have been a better choice for calculating release.
- b. The containment arbitrarily was assumed to fail to atmosphere, the worst case, when several other modes were available, ,
- c. No consideration was given to avoidance of early containment failure via operator-initiated venting.
- d. Based on the above, I would expect that the risk of bin 5 is overestimated by at least a factor of 10.
( IZh
8 l ( 10.3 - leolation failure, , sequences (bins 6 and 7). Having no risk contribution i from/sequencesseemscounter-intuitive. NRC should examine these more care- l fully, there being a broad spectrum of failure-to-isolate modes combined i with all avellable accident sequences. l 10.4 - Page 0-1. l Typo under large LOCA, Category 4. Should be 1 x 10-II . (; l l O ( IA c
* .. e
- 0 e. .
I i i 3 8 *l'Il 8 8 3 36. 173 -
, e i i s iis i e i , i n ii i s e a a #as
~
. ~
I .
~
~
,e W
Itr$ , ;
= : _
1 2 10"O ,e
~
e L A . Q_ p e w m
~
- f. -
de
, m N _
AvWege FJfve W gg th
~
f-
- l. m I I I'' '
I g I t t tttIf f f f I f IfI 19 M M 10 1 10 2 100 . Earty Fatelltle. X
- FIGURE 5-3 Probability Distribution for Early Fatalities per Reactor Year 1
- Note: Acoroximate uncertainties are estimated to be reeresented by
! actors of 1/4 and 4 on conwouence magnitudes and by f actors
( of 1/5 and 5 on probabilities.
. It d
1 i fa.J8s. / NUR GCs - o 9 s'4 S'sk A /la cale a n f.. . . _ .
. . Rosk Allocals., RoskA//au?,,,
. DidlasaFvd .n Clu C.Is. X
. . Be n... 1%gua,cy
- 84. I d t o" I 7. T x /U" ' 0.307$ % /O'# O. 4
/ .
- 2. /
2 2. . V < i o-7 7. f . f o
-b /. F
~
3 3. I K lo
~
S. / d /O' O.0/52 O 2-M .3. / ( /O'7 f, o < # 0' 3 . C. / ff'
~ ** ' 2 . 7-5 2 7 e to ' 5. 9 x io / . / o . 53 -
$ 4. 6 x to 2. 7 s / O $,O/79
~
53/o'3 0 034o 7 4. t < 10 '
~' /, 35 /* $
f 2.'7s 10 S'. o x t o' 3
. 5, / s 6. 0 9 6. G < to ' 7. 1 s / o'3
(:. 6 9 a ro'3 d. b~/ ?b~ d6
/0 _ 7.To16' .
~
2, r, 2. _ . _ _ .. . . _ . . . . . _ A 9 . ?-
. _ . . . // .. . 3 s / 0 .'.. . . 9..F F ./o'I .- .
Y 7 ' d' (. /o /
~
/2 /x/0 4. / 4_ /. .
o O
/3 3 x / o .
.,0. 12.7 5 O* l l} 9, 5'x /6' /. T x t o" # .. .
~
/C .2 s to'5 o o
/00 Sum Y. 4 x 10 '
~
f{r.2 s O ( I2e
-- g- , , ;3 _ r; -- ---
k -} - . _ _ _.h_ - - _ _ _ _ _ _ _ _ - - - - - DROFDSED RULE
~
h, Mb Y '.l CNUREG-s99) ! (. L'SIh[ ARKANSAS POWER & LIGHT COMPANY ag3 i POST OFFICE BOX 551 UTTLE ROCK. ARKANSAS 72203 (501)3714XD f October 31, 1985 IS A10:23 g ,. . 001: : '-r.':, fi .. : .
~ 'ffjyi ..::
OCAN1d8503 Secretary of the Commission U. S. Nuclear Regulatory Commission Washington, D.C. 20555 ATTN: Docketing and Service Branch
SUBJECT:
Public Comment on Draft NUREG-0956
" Reassessment of the Technical Bases for Estimating Source Terms 3, Gentlemen: -
Attached are our comments on Draft NUREG-0956, with comment expiration date of November 7, 1985 (50 Fed. Reg. 175 9/10/85). (, Very truly yours, r 1 .
. Ted Enos Licensing, Manager JTE/MGB/sg Attachment
( piv vi c er-13 MEMeEA MsOOLE SOUTM UTILITIES SYSTEM
e
,, 1 Attachm@nt to DCAN108503
- Pags 1 ef 4 i ATTACHMENT
( COMMENTS ON DRAFT NUREG-0956 l General This effort seems to have been generally well conceived and directed and the report has been well written and organized in a manner that enhances the degree to which it is informative. The inclusion of Chapter 2 and the contribution it makes to an accurate perspective on this issue is particularly commendable. The emphasis on a mechanistic rather than a deterministic approach is extremely important and appropriately pervades the report. However, there are some departures from this. Comments on these departures are included in the more specific comments below. The report seems to have a general preoccupation with uncertainties and occasionally confuses the concepts of uncertainty and sensitivity. Such an overemphasis on uncertainties can have at least two adverse effects. First, it can generate naive perspectives that view the uncertainties as expressed in NUREG-0956 as completely undermining the knowledge gained by this major advance in technology, i.e., they fail to recognize uncertainty as a - i mathematical expression rather than an expression of doubt. Second, it can* generate a lot of expensive research with limited value by attempting to - reduce or define uncertainties with little impact or for which qualitative i knowledge is adequate. When specific uncertainties have significant impacts on conclusions and are amenable to further reduction and/or definition then (. further efforts are warranted. A good example of this type of uncertainty is that for which additional research is described in Section 7.1.8, Containment Failure Modes. In general, this effort and report describing it make a significant and
. positive contribution to the science and engineering of Nuclear Power Plant Risk Analysis.
Specific
- 1. We are pleased to see the recognition in Conclusion 8 of the importance of plant-specific features. We believe this recognition fosters a mechanistic approach to risk analysis and discourages generalizations that frequently overstate problem applicabilities and mischaracterize problem solutions for some.
- 2. Recommendation 1 is particularly important and appropriate.
Unrealistic a'rumptions frequently cut two ways, both enveloping variations it. input conditions and creating new unnecessary problems and risks. It is important to avail ourselves of more realistic assumptions as they are developed and made available. However, care should be taken to avoid iterative or incremental regulatory policy ( 13 b
' , ', . . i
. 4ttachment to SCAN 108503
- Page 2 of 4 changes such as occurred on the ECCS issue in the early 1970s. A march
(._ of succeeding interim or revised final regulatory policies wastes resources, generates confusion, degrades credibility and reduces the likelihood of high quality, deliberately developed, well integrated improvements. Such a situation should be carefully avoided.
- 3. On Page 3-19, second paragraph, mitigating circumstances are listed for the potential underprediction of in-vessel releases for some sequences involving a node of molten fuel slumping out the core region and subsequently residing in the vessel for an hour or more. The first mitigating circumstance listed is a limited surface area and the last
; is lowered temperatures d e to quenching. It appears possible that the limited surface area may be enlarged during quenching due to resolidification and potential fracture mechanisms during rapid quenching, i.e., the creation of rubble rather than molten mass. There may be good reasons why this is unimportant, but the list of mitigating circumstances as listed seems to beg this question.
- 4. On Page 3-21, a statement is made that, among Brownian movement, gravity, turbulence, flow fields with shear, and inertia in flow fields with curvature the first three forms of relative motion are dominant for agglomeration considerations and are the only three modeled in -
i, TRAP-MELT. It would appear that,'for flow through the reactor vessel , i and steam generator, flow fields with shear would be significant and, - for flow through reactor coolant pumps and piping and ECCS piping, o inertia in flow fields with curvature would be significant for r agglomeration considerations. Again, there may be good reasons why the ( first three are dominant but such reasons are not clear from the discussion given, a
- 5. As mentioned in the general comments, there are some departures from the emphasis on a mechanistic approach. It is recognized that deterministic choices of mechanisms are occasionally mandated.
However, the frequent mention of steam explosions throughout the report, e.g., Page 3-40, seems to introduce a phenomenon that is not mandated nor warranted given a failure to identify evidence for mechanisms to create such a phenomenon during any defined sequence. Another departure from the mechanistic approach is illustrated by the statement on Page 4-3 that operator intervention to-correct faulted plant conditions has not been included in the postulated accident scenarios. It appears that this deterministic omission was necessary to achieve plant states relevant to providing insight into source term phenomena from severe core melt sequences. That is understandable. However, this fact should be highlighted early in the report to discourage qualitative misapprehensions of *.he potential for such actual severe core melt sequences. In particular, as efforts progress from source terms to risk analyses, f ailure to adequately consider operator action will be a very serious shortcoming.
- 6. Consistent with the importance of the mechanistic approach is the importance of careful attention to the attainability of configurations
( inherent in the sequences used. An illustration of the importar.ce of this careful attention is the discussion of the results for SP-A on I3c
e
,..*\
. Attachment to SCAN 198503 i . Page 3 of 4 Pages C.1-14 and C.1-15 in which the attainability of both high coolant system pressures and temperatures high enough to cause massive fuel
(. failures is brought into question. Preoccupation with examining all mechanisms for release can very easily create situations in which the 3 plant configuration required to support the sequence being examined involves mutually exclusive conditions. The existence of these situations can frequently be very difficult to detect because of the subtlety of the mutual exclusiveness. We believe additional attention in this area is warranted, though identification of additional specific examples of inadequate attention in this area is beyond the scope of our review. Rigor in this area is basic to the efficiency and value of a mechanistic approach.
- 7. The relevance of Appendix 8 is very difficult to ascertain. Its
, application is obscure and such work based on subjective opinions of experts seems to add an unnecessary aura of subjectivity to the report thereby damaging its credibility. We suggest that it either be deleted or an expanded discussion of how it was used in other portions of the effort (especially the risk insights of Appendix D) needs to be included.
- 8. Discussion in Appendix C of SP-1 and SP-2 raises a question regarding 3, consistency of assumptions and the degree to which there are mechanistic bases for assumptions. The discussions of SP-1 assumes an -
initial containment pressure of 60 psia due to primary system boil-off.- The impression given is that this is an assumed rather than a calculated value. The discussion of SP-2 assumes an initial
- h. containment pressure of 30 psi. This is explicitly stated to be an assumption. The reasons for the difference is not obvious from the material presented. For SP-2 no mechanism for the assumed initial containment pressure is stated. In addition, for SP-2 the units " psi" are used intermittently. Where incremental pressure changes are involved, the units " psi" may be appropriate, but the usage in the fourth and fifth lines of the last paragraph of Page C.1-8 is clearly inappropriate.~ It appears that the units should be " psia." Similar unit problems appear in the discussion of SP-3. The basis for and apparent discrepancy between the initial containment pressure assumptions in SP-1 and SP-2 deserves further attention.
1
- 9. The report includes values called " warning times" in tables 2.1 and 0.3. These values are left undefined except for an ambiguous one sentence description on page 2-5. Their value to the report seems quite limited. We suggest that they either be deleted or a better explanation be provided as to what they are, how they were determined l and what they are used for.
l i i ( l l l
r-E
- g. -
....i
. Attachment to SCAN 198503 l..t Page 4 of 4 Typographical Errors - NUREG-0956
.(
Page Para Line I " experimented" should be " experimental" 3-17 3 8 3-37 3 2 " core-concreate" should be " core-concrete" l 3-38 Item 5 " general" should be " generation" C.1-11 2 2 "quaatity" should be " quantity"
'3
( 9 k 13e
i ,
, I f-l g UNIVERSITY OF CALIFORNIA, BERKELEY BERKELEY
- DAVIS
- IRVINE
- LOS ANCELES
- RIVER $tDE
- SAN DIECO
- SAN FRANCISCO SANTA BARBARA
- SANTA CRUE f l DEPARTMENT or CHEMICAL ENGINEERINC l BERKELEY, CALIFORNIA 94720 '
1 November 21, 1985 t Mel Silberberg Branch Chief, Fuel Systems Research Division of Accident Evaluation Office of Nuclear Regulatory Research Washington, D.C. 20555
Dear Mel:
Thank you for bringing us up to date with some of the progress on source term estimation since the APS Study Group completed its report. Thank you also for the opportunity to comment on the draft report NUREG-0956,
" Reassessment of the Technical Bases for Estimating Source Terms".
Overall, I found the draft of NUREG-0956 to be a reasonably accurate (- account of progress made with the following important exceptions. In my view, this draf t overestimates the level of documentation of the codes used in BMI-2104, the quality of the peer review process for BMI-2104, and the degree of validation (comparison with meaningful experiments) of the various codes. I also feel that this draf t is unbalanced with respect to the very limited attention given to the experimental parts of NRC research on the source term as opposed to its coverage of code development and code output. I also think the sensitivity studies (QUEST) deserve considerably expanded coverage. I elaborate on these general topics below. One of the greatest frustrations during the APS study was our repeated inability to get clear and appropriately thorough descriptions of the bases of the codes used. This is reflected in our conclusion V.E.S'. and recommendation VIII.E.3. I personally feel the idea that everything must be published in peer review journals may have been overstated. What I believe is that as much as possible of NRC sponsored research be sub-mitted for journal publication to gain addition review by a larger reader-ship and to convince the technical community as a whole (not just the narrow source term community) that source term estimation is indeed soundly based. Not all of NRC's sponsored research is appropriate for journal publication. Nevertheless, these aspects must be written up as NRC reports with the same level of clarity, precision, and thoroughness demanded of journal published articles. I do not believe that ORNL/TM-8842, although considerably improved from the draf t version available to the APS Study Group during its deliberations, meets these criteria. Would a reasonably well informed scientist or engineer, but one not steeped in source term ( research, be able to gain sufficient insight into what was done in order to be convinced that the approach was soundly conceived and soundly executed? w- *2,hcA y .u t- W W r- N
3 8 Nel Silberberg November 21, 1985 I think not. I also think that poor documentation probably has impeded NRC progress. As one example, take the question of the decay heat generation rate which you mentioned had been treated differently in the MARCH and CORCON codes. Surely, since decay heat is the driving force for the accident its method of calculation should be clearly identified. _ I did find (page 80) mention that MARCH 2 uses ANSI /ANS-5.2-1979 but there was no indication in Chapter 5 on CORCON as to what it used for decay heat generation. While the numerical changes may or may not be important, I got the impression that the BCL people didn't know what the SANDIA people were using. Such inconsistencies, especially when found late in the study, do not inspire confidence in reviewers. Without fully documented accounts of what was done, critical peer review becomes enormously difficult if not impossible. Of course, the technical Expert Peer Review panel of 14 scientists during the period when much of the source term analytical procedure was being developed was necessary. I regard this exercise more as the use of consultants to provide additional perspectives and guidance to the study, but it was not peer review of the final product. The transcripts of the meetings show that the members often arrived with inadequate time (or motivation) to review material sent them. Moreover, as with the presentatior.s to the APS Study Group, these meetings were dominated by
- h. presentations of code output. The discussion of the engineering, physics and chemistry underlying the codes was not dealt with adequately. More-over, it was a disappointment that when some of the same questions on the underlying science or engineering asked by the Technical Expert Peer Review Committee were subsequently asked by the APS Study Group 3 Elear, convincing, thorough responses had not been formulated by the NRC con-tractors.
The question of code validation has barely been scratched. NUREG-0956 requires a clear definition of what is meant by validation. To me, this ultimately means comparison of code output with experiment. Both well controlled, limited phenomena, small scale experiments and realistic, adequately instrumented, large scale experiments are necessary to insure that all important phenomena are modeled with sufficient accuracy. See APS conclusions IV.C.10.e, f, g, V.E.7, VIII.C.f and recommendation VIII.E.1, with emphasis on the word " integrated". The slowness with which data from large scale experiments are being released is quite distressing. In many sequences, it is calculated by TRAP-MELT that 85% of the iodine and cesium are retained within the primary system and this is one of the main reasons for lowering of the source terms of these elements compared to WASH 1400 (especially for by-pass sequences). Yet we know virtually nothing about the experiments at LACE and MARVIKEN and the experiments at PBF have not been completely analyzed from the point of view of fission product transport and deposition. Is it any wonder then that ORNL/TM-8842 concludes (page 362) that "the quality ; assurance level for the TRAP-MELT code is not presently veryhigh" and ( (page 372) " TRAP-MELT is lacking in formal validation". I feel CORCON J l l i
/Tb i
i l.
- e. . .
o i Mel Silberberg November 21, 1985 1 is at a similarly low level of experimental validation. NAUA in virtue of experiments at ORNL/NSPP and DEMONA is probably one of the better validated codes, although questions of natural ton-vection and thermal stratification still bother me. I Since our understanding and predictive capability is now and will continue to be incomplete, sensitivity studies must be an essential part of source term estimation. This is APS recommenda-tion VIII.E.2; see also the summary for VI.D. The idea is stated explicitly for fission product removal by suppression pools in the last two sentences of conclusion VIII.B.7; our ability to predict suppression pool removal is related to the assurance that aerosol sizes produced earlier in the accident are indeed in ranges where the decontamination factors are large rather than at the minimum of the curve. Similarly, fission product releases computed by VANESSA are sensitively dependent on CORCON's predictions of melt
- temperature. It would help to interpret the more important sensi-tivities from the mechanistic point of view, i.e. releases from the melt are sensitive to temperature because of the exponential dependence of vapor pressure on temperature. Releases also may be very sensitive to chemistry (chlorides in the concrete) .
(; I also have the following specific comments: XIX Conclusion 2 - add qualifiers "A number of important" omissions and. . .now " partially" accounted for. XIX Conclusion 3 - add qualifier "Some" remaining areas. XIX Conclusion 4 - This conclusion needs to be considerably toned down. XX Add experimental results as an area of improvement, esp. suppression pools, DEMONA. XXI Add relocation oE_ deposited fission products, especially where molten as an area of uncertainty. This can be important for revaporization og alteration of sequence. Also interpret natural circulation to include thermal stratification in containment. XXII Conclusion 7 - Change last sentence to "A delay of several hours in containment failure, following the cessation of fission product release to the contain-ment,is computed to reduce source terms significantly. " 3-6 Point out in last paragraph how sensitive computed ( source terms are to these user selected inputs. Iilc;
b . ( Mel Silberberg November 21, 1985 l 3-10 P 3 - natural circulation above "and through" the core. 3-19 'IPl - I believe this paragraph underestimates the potential importance of the time of release of volatiles from the core because this will feed into predictions of aerosol size and hence deposition. 3-19 If the silver is returned or retained in the core region, what influence does this have on aerosol generation? Are volatiles refluxed in code calculation? Will molten silver dissolve or alloy with zircalloy? 3-21 last -IP . . .but " assuming the low gas velocities computed by MARCH rather than much higher natural circulation velocities", generally, the mos t important. . . 3-22 TRAP-MELT does not consider nucleation of new aerosols which and might be important if flow rates are higher and plumbing is 3-35 hotter due to natural convection. ( 3-36 first & - I believe inadequate attention has been given to the possibility of a sustained release from the core following s failure of the reactor pressure vessel. See APS IV.C.10.i. I agree that we have a good understanding of many phenomena related to aerosols. But this does not mean we can predict everything with the needed accuracy because of the complex cher.istry, complex geometry, unknown thermal hydraulics, etc. The impression given here is misleading. It is like saying we know the laws of classical mechanics and therefore can predict the motion of turbulent fluids. Point out that this is especially true because of the sensitive dependence of deposition rates on aerosol size. 3-39 Natural convection will also influence the timing of the release and the deposition rates because of changes in composition, temperature and residence times. 3-41 Weaken summary in view of discussion above. 4-18 Is the revaporization from the upper plenum and plumbing following reactor vessel failure, especially with some of the core remaining and with natural circulation) p[ ( IT/
q e . . .
'- ,e i
Mel Silberberg November 21, 1985 l 4-37 add qualifier...SPARC is high enough " assuming the size of aerosols reaching the suppression pool are sufficiently large, that the computed" releases to the... I hope the above consnents are decipherable and of use to you. Sincerely, h Simon L. Goren Professor SLG:ncm A A/S C 4 l Ne a
v . ~/ WILLIAM R. STRATTON PRESIDENT ( STRATTON AND ASSOclATES,lNC. 2 AcOMA LANE
- LOS ALAMOS. N.M.a7544. (sos) s72 370<,
November 20, 1985 Mr. Denwood F. Ross, Deputy Director Office of Nuclear Regulatory Research U. S. Nuclear Regulatory Commission Washington, D. C. 20555
Dear Mr. Ross:
I enclose herewith my comments on the document, " Reassessment of the Technical Bases for Estimating Source Terms," NUREG-0956. These comments are divided into two sections; fi rs t , those that I think are of major or general o concern, and, paragraph. second, page by page comments organized by chapter, page, and The U.S. members of the (now dissolved) ANS Special Committee on Source Terms have been advised of the extension of the comment period and have been urged to submit their personal comments. ( It is my opinion that the document is not satisfactory as it now exists, and I presume to offer an idea for its revision and/or the writing of NUREG-ll50. If the intent is to create a document representing the point of view of (and acceptance by) the technical community of specialists in severe accident analyses (as opposed to a docu: ent representing only the NRC point of view), the technical community must be asked to share in the preparation of such an assessment of the state of knowledge. IDCOR and EPRI comqto mind at once as organizations that could provide such assistance. The assistance could be very In theearly long review of draf ta of chapters, or even in the writing of chapters. run, a document so prepared would have a much quicker and wider acceptance, and a better understanding by all parties. Sincerely yours, 3
/ . !-- ,
.. -7. .,1 William R. Stratton WRS/bj Enci k
Jg c o Lv dde n . wr tf nF~ 852074
$ Tf'-
4
, - e $j .- %
f (- ..t i REVIEW OF NUREG-0956 { Points of Major Concern
- 1. The title of the document is " Reassessment of the Technical Bases for Estimating Source Tems." A review, then, can ask how well the document satisfies the reassessment function, and how closely it adheres to the goal as defined by the title.
- 2. The current evaluation of severe accidents and consequent source terms is a multi-organization, world-wide ef fort. The diversity of effort provides a great deal of assurance that significant physical and chemical phenomena will not be missed. The review process during technical meetings, for example, is no different from review of reports of a new scientific discovery. The criticisms are very of ten sharp, and result in an improved, more technically defensible position. The technical work produced by IDCOR SWEC, EPRI, ORNL, NYPA, the FRG, French, etc. , as well as BMI, has been reviewed in this manner and should be a part of the " Reassessment of the Technical Bases for Estimating Source Tems."
This is not the case. These several studies have not been considered, and NUREG-0956 can be regarded only as a statement of NRC's position relative to the evaluation of source tems. It is not an assessment of the knowledge of the technical community. The title should be changed, or the reassessment should include results of the entire technical community.
- 3. The stated intent is to produce "best estimate" evaluations of source terms.
(- The evaluations of source terms described in BMI-2104 (on which NUREG-0956 is based) are indeed better than the estimates in WASH-1400 or the arbitrary conditions in TID-14844, but cannot be described as "best estimates." For example, the containment failures incorporated into the BMI work are described as a parametric study and "are not meant to characterize the expected, or most likely, containment behavior." (See page 4-28, third paragraph.) However, relative to containment behavior, a best estimate is attainable if the results of the containment working groups are f actored into the BMI analyses. Is this step now underway?
- 4. The matter of uncertainties is emphasized, but the subject is confused throughout the document. An uncertainty of a factor of 10 up or down is very important for a source tem that is, for example. 0.05 of core inventory, but it is unimportant if the reference value is 0.0005 times core inventory of the element in question. The QUEST study is referred to as an uncertainty study, but is admitted to be a sensitivity study on pages 3-29 and 6-4. Both uncertainty and sensitivity studies are important, but the difference must be kept clear.
- 5. The American Nuclear Society is a professional society. It is not a part of an industrial group, as stated on page x1x, first paragraph.
k
7.. 1
./-... . J ,
- Pcge 2 :
Review of NUREG-0956 I. .
- 6. A comparison of BMI-2104 calculated source terms to those calculate 1 by other groups (IDCOR, SWEC, EPRI, NYPA, ORNL) shows a disturbing and consistent difference: The BMI results are higher, sometimes by significant amounts.
Given the serious responsibilities that the NRC has, these dif ferences must be explained and understood. A major effort must be mounted to obtain agreement on physical and chemical assumptions, mathematical approximations, computational formulations, etc. Clearly, the same physics, chemistry, and mathematics should give the same result independent of the computer or programmer. This matter is of the utmost importance. Design of plant, safety features, formulation of rules, regulations, safety guides, etc., depend on resolution of the source term matter. The BMI code suite cannot be accepted until resolution of results is obtained.
- 7. There is no indication of when or at what point the program, experimental and analytical, planned for the future will come to a satisfactory conclusion.
Some measure of what source term or tenns is good enough must be found. A deminimis criterion or criteria are badly needed, both for consequences and for probability.
- 8. The history discussed in Chapter 2 is incomplete. The first study of importance was WASH-3, in 1950. The ACRS first proposed the assumptions in TID-14844 in 1960, and the investigations of the past half-dozen years were stimulated by the accident at Three Mile Island.
- 9. The document overemphasizes the dangers of generalization of source term results. A great deal is possible; e.g., many PWR accident sequences in large, dry containments are functionally similar and can be so compared. Furthermore, accidents can be grouped as to initiating event (e.g., large break, small break, etc.). If the worst case (e.g. SyB) is acceptable, all other S sequences in 2
which something works will be less severe. The whole problem is difficult enough without making it worse. .
- 10. The risk appraisal presented in Chapter 6 and Appendix D is of questionable
, value. It is not necessary for a " Reassessment of the Technical Bases for Estimating Source Terms." The appraisal uses source term results from BMI-2104, which are admitted to be a parametric study of containment integrity. Hence, the appraisal has an air of unreality about it. If a risk appraisal is to be performed, a closer comparison to WASH-1400 would be helpful. To be precise, a direct comparison to accident categories in WASH-1400 (Tables V 3-14. V 3-16, and Table V 2-1) should be made. For example, have PWR-1 and BWR-1 really disappeared? Further, the large body of modern calculations, with a small number of exceptions, shows no early fatalities. Therefore, this measure of the severity of an accident is best dropped and person-rem, or some related variable, should be adopted aa a supplement to (and ultimate replacement of) WASH-1400. C'
f:i , e
~'
Pcg3 3 Review of NUREC-0956 l (-
- 11. Finally, the implication is 'given throughout that the BMI code suite
[{. is complete, correct, adequately reviewed, and accepted. It is, indeed, a very -large' step forward from WASH-1400 TID-14944, and safety guildes 1.3 and 1.4, but is not regarded as complete, correct, asequately reviewed, or accepted. The very large differences between NRC and the technical connunity must be resolved, t F f, I L G 1 k
1 b )- '~
\
= COMMENTS ON NUREC-09h t
( (The format follored is to note the page and paragraph for each chapter, followed by comments) f Executive Stannary: I 1. xvii,1; The intent to use improved analytical methods in WASH-1400 and to reassess the assumptions in TID-14844 is laudable, and should be encouraged. 2.xvil.2; An additional, independent validation study and uncertainty analysis ! abould be performed. The existing " uncertainty" analyses is better described as a sensitivity study. ' 3.xv11.3; The absence of BWR Mk-II in the NRC effort is notable. This ' has been completed outside the NRC funded program. j .
- 4. xvii,4; The peer review by the panel of 14 scientists was never completed for the final draft EMI-2104.
5.xix,1; The American Nuclear Gociety is a professional society; it is not [ a part of an" industry group." i 6.xix,2; Because the newer evaluations (including others than NRC-EMI) - show no early fatalities for essentially all sequences, a better 1 measure of consequence is person-ren with some estimate of distribution of dose. , , L A reference to the 1986 publication should be given. 7.xix.5 (conclusion 2); The matter of containment strength and non-importance i of steam spike and steam explosion should be mentioned. I 8.xix,6 (conclusion 4); The implication is given that the code suite is ( complete and the review is complete and satisfactory. This is not the case. I t L _
s al
, Pcgh 2
( 9.xx, Table ES. I. #6; The " mechanistic" treatment of core-concrete interaction may be questionable. It has not been subjected to pser review except cursorily by the APS. 10.xxi, Table ES.2; An indication of relative importance to calculated source terms is necessary. An uncertainty may be large, but unimportant because the reference value is very small. An additional uncertainty is the time of leak or fail'ure of containment. This is not mentioned. 11.xxil, conclusion 7; The correct description of the fission product chemistry is of equal or greater importance than the correct ,,- evaluation of containment strength. 12.xx11, conclusion 8; Much generality h possible, e.g. many PWRa in large dry containments are functionally similar and will {' respond similarly to accidents. Some few sequences are design dependent, but complications need not be invented. 13.xx11, conclusion 9; The new evaluations of source terms that are much , smaller than those in WASH-1400 dominate, overwhelmingly, those very few that are comparable. The iodine factor is essentially - eliminated; early containment failure is unlikely. The conclusion is incorrect.
- 14. xxii, conclus. ion 10; This is a couplete pus.zle because the containment response is part of source term evaluation.
15.xxiii, conclusion 11; This conclusion is unnecessarily negative. 16.xxiii, d (Continuing Basearch); A major point missed is the need for an overall, peer revie4 of BMI-2104 and the assumptions and boundary conditions that are imposed upon the source term calculations.
f ,
'7v &1 4
Pcg2 3 * ( .
.17.xxiii, Recommendation 1; Correct; even with inadequacies and errors, this would be a major step forward.
18.xxiii, Recosamendations 2 and 3;- The implication is given that the NRC
" Source Ter a Code Package" is complete, correct , and undisputed.
This is not true and far from correct.
- 19. xxiv. References; Reference should be given to the ANS Source Term Study, the IDCOR program, the several SWEC reports, the work sponsored by the PASNY, etc.
Chapter 1: 1.1-1,2; .
" Setting priorities" implies the existence of a deminimis criterion
- (or criteria). This is largely ignored but is very important.
2.1-2,5 : The most important of the Source Tenn evaluations assume,_o,that no active mechanical or electrical equipment operates. Hence, operability of active compotents is irrelevant unless needed for public health and safety. This is not the casa. 3.1-3,5 ; The NRC studies relating to containment capability (Contairunent loads and Containment Performance Working Groups) should be issued for review and undergo a peer review by a designated group of international experts. 4.1-3,6; ; Croups other than BMI and IDCOR are involved. SWEC, ORNL, EPRI. NYPA, ANS, foreign groups should be acknowledged.The effort is international. 5.1-4,2; BMI-2104 is not the most recent quantification. The process is world-wide and continuous. B)tI-2104 has been criticized as incomplete, inaccurate, and wrong in places. (* 6.1-4,4; Iho peer review af E!!I-2104 was not comprehenstve. is not complete. It was and Some parts have not been reviewed at all. I E
- f.
- Page 4 Chapter 2
! 1. 2 -1. 1. 2 . 3; The brief history is incomplete. The first estimates of release of radioactive materials in an accident is found in WASH-3 (1950). This document set the pattern. The Geneva Conference in 1955 and the Windscale accident in 1957 are important for their influence. . Prior to the publication of TID-14844 in 1962, the ACRS issued a letter in 1960 recommending essentially the same asstanptions. The stimulus for NUREG-0772 was external to the NRC staff. 2.2-1.3.; The real shortcomings in WASH-1400 were not recognized until af ter the accident at TM1-2. The early criticisms were in the
'( probability estimates and did not affect significantly the estimates of risk. The errors in consequences, however, are very -
significant. 3.2-4.1; The iodine was required to behave much like a noble gas in its airborne and leakage behavior. The later writing of Safety Guides should be mentioned. The significant change was that, given the initiator (e.g. a pipe rupture), the iodine and noble gases would be "immediately" dispersed throughout the ' containment. 4.2-5.4; Mention of plutonium is a red herring. Pu02 is extremely refractory, dense, and to make and keep it airborne would be most unlikely. 5.2-7.1; "Some amount of volatile species of iodine" begs for a quantification. ( l
v
. nygg Page 5 l(-
6.2-7,3; "The Reactor Safety Study (with all its weaknesses)" is an unnecessary and incorrect chop. The RSS was and is an extraordinary study.
- 7. The world-wide investigation of source terms derived from the THI-2 accident, the Kemeny Commission investigations, and the letters in 1980 from three national laboratories and one industry laboratory that pointed out where NRC rules, regulations, and guides were in error relative to iodice behavior. The R&D effort did not originate in the NRC.
8.2-10,6; The phrase, " cover other initiating events as an envelope." is a very useful generalization. The same philosophy was used in the ANS Source Term Study. NUREC-0956 ignores this possibility , l for generalization. 9.2-11,5; The inclusion of " accident sequences of very low frequency" begs the questions: how low and how low is low enough to be ignored.
- 10. The NRC-Sandia " uncertainty" study may be, in fact, a sensitivity e
s t udy. The NRC needs additional advice in this regard. 11.2-14.2; The review of BMI-2104 was not thorough and is not complete. Some parts have not been reviewed at all.
- i Chapter 3:
- 1. 3 - 15 . 1; Host fission products are not radioactive or were of short half-life and have decayed to a stable scace.
2.3-15,5; The uncertainty in ORICEN is unduly pessimistic. The users of ORICEN should be consulted and quoted.
^
3.3-22.2; A deposition velocity of 0 for CaI is surely an error that must (, be corrected. i l I i l L
2v BL Pcg3 6
~
4.3-22.4; Experiments on re's uspension of deposited aerosols have t been performed by Fauske and Assoc. These should be mentioned. 5.3-23,7; An independent peer review of the Vanessa code has not been started. This, and related physical phenomena, should be complaced before NUREC-0956 is reviseu. 6.3-29.4; The importance of an uncertainty (a factor of 100 is mentioned) depends on the base casa estirate. This is not quoted.' 7.3-30. Table 3.2; Coding symbols are quoted. These are useless to the reader. The 6th and 7th items in the table are, essentially, nonsense statements. 8.3-32, 6 and 7; If containment sprays are operating, the source term is very smal.'.. 9.3-32.11.12. & 13; If the suppression pool is effective, the source term is extremely small and a large uncertainty is unimportant. ' 10.3-33.4; The iodine was created as a noble gas; chemistry was ignored. , 11.3-33.6; "Some elemental iodine" begs for quantification. .
- 12. 3 -35 .3; The face of tellurium atoms when they decay to nascent iodine is ignored. It would be chemically fixed in place.
- 13. 3 -35 .4; Vanessa has not been reviewed. It may or may not repres ent "a major advance in the accuracy of source term analyses."
The statement is defensive. 14.3-38. Table 3.5; A major uncertaint.t not mentioned is the set of assumptions made by the user of the code.
.,~
f, _ 15.3-40.3; Deposition on ice should be no different than deposition on other surf aces, depending on temperature and wetness. Chapter 4: 1.4-1,1; A stamary of all calculations should be included to allow the reader to judge the results and conclusions.
- 2. 4-1,2 ; A large amount of generality g possible and should be noted; i.e. for source term considerations, there is no difference between 10~3 to **
a f actor of M and net. 3.4-1,3; The Mark II containment u different and must be analyzed by NRC-BHl. This has been done by SWEC with surprising results. 4.4-5. V Sequence; The effectiveness of the safeguards building was not included in the calculation. The " submerged V-sequence" variation appears to be ignored. - In general, it appears to be the case that the retention capability of buildings exterior to the containment is ignored. 5.4-10,2; The antainment failure mode was " selected in the March code." Apparen ly, t.he risk analyses in this document (0956) are , bas.ed upon arbitrary assumptions relative to containment failure, ignoring the work of the two containment working groups. 6.4-10,4; The implication is that all 1.5 metric cons are radioactive. This is not the case and should be clarified. The 10 billion curta number is unimportant because of very rapid radioactive decay. 7.4-22. Table 4.10 and 4-23. Table 4.11; The release numbers quoted in the tables are to the containment volume, not outside containment. I (' This in not clear from text or table. Curies should also be tabulated. l 1 tl, I
Page 8 ( 8.4-32.4 4 5; Calculations other than those by BMI are acknesledged here. The very large number that have been completed by others is ignored. This is a major f ailing of the analyses in this doctament . 9.4-35, 3 44; The containment failures are postulated. This should be stated. 10.4-40. Table 4.13; The containment failure modes and times are arbitrary. This should be stated. 11.4-45.3; The release fraction of 0.5% of I needs more justification before it can be accepted. A review is necessary. 12.4-45.5; The BMI-2104 results discussed derive from arbitrary assumptions
- r. . containisent failure and neglect of auxiliary buildings.
( 13.4-46.1; The QUT.ST study is admitted, elsewhere, to be a sensitivity acudy. Therefore, it should not be used to estimate uncertainties " . in risk. Chapter 5 . 1.5-3; Major factors identified by the peer review group included better
- accounting of water and the effectiveness of auxiliary buildings.
It is not clear that either of chase recommendations was followed adequately. 2.5-6.2; The NRC uncertainty study was a sensitivit' n.3y, as admitted elsewhere in this doctament. An indepenkn un reainty study is needed. 3.5-8.7; Results for all investigations (at aoout July 1984) are presented in the ANS Source Term Study. It is not clear that this enormous body of information is used at all. The neglect is a major f ailing of this analysis , i
OvPA
~
k . Page 9 Chapter 6: 1.6-1.4; Probabilities are those of WASH-1400; thus, apparently stema explosion phenomena suse be included in the analysis. This is completely incorrect and ignores the general conclusion that steam explosions of magnitude sufficient to cause containment to fail can be ignored.
- 2.6-2,3; The analyses for the CCDF plots apparently use the BMI-2104 calculations in which the containment failure modes are admitted to be paramet.ric. Thus, no credence can be given to these analyses.
3.6-2,5; Because of low source terms, the use of early fatalities is no longer a good measure of an accident, and the use of latent , ( fatalities is questionable. Some other measure of dose must be used.
- 4.6-4.6; The QUEST study is admitted to be a sensitivity study.
5.6-7,3; Retention of fission products in the reactor' building has been , calculated by IDCOR. These and other studies should be a part of this analysis and that planned for NUREG-1150. Chapter 7: . 1. The experimental program that is described is large, and apparently covers the matters of major interest. However, the relative importance is not obvious, nor is any evaluation of fered as to what knowledge is sufficient for regulatory purposes. I.e., what release fraction for each isotope
~
is low enough that it can be ignored. This implies that a deminimis criterion or criteria must be developed. This should be addressed in conjunction with an experimental program to keep it within reasonable bounds.Page 8
%z Pcg2 10
( From another point of view, it is apparent that !.'RC-BMI source term results are consistently dif ferent f rom thos e of ot he r inves tigators . In order to evaluate the necessity of experiments, agreement as to the state of knowledge must be reached. A large and intensive effort to resolve differences must be started and continued until agreement is reached. Chapter 8: 1.8-1,1; The peer review of BMI-2104 is not complete as stated. The complete suite and some parts have not been discussed.
- 2. 8-1, ' Conclusion 1; The BMI-2104 suite of codes is a significant advance over methods used in WASH-1400, but is scarcely complete or a best estimate. Review and improvement is needed.
3.8;-1, qonclusion 2; A major deficiency in past work was the neglect of (- deposition in auxiliary buildings. This has not been corrected adequately in BMI-2104. ' 4.8-1 Conclusion 3; The relative importance of uncertainties is ignored. . Many are minor and will have only a small ef fect on source terms when fully resolved. 5.8-2 Conclusion 4; Again, a complete review (and implied acceptance and agreement) is claimed. This is not true. The " uncertainty study" is admitted to be a sensitivity study.* 6.8-2. Conclusion 5; The statement is confused; the " purpose of the analysis" invariably is to obtain the best, most realistic estimate of problem at hand. 7.8-3 Conclusion 7; For many sequences, chemistry is of equal or greater importance. The high solubility of cesium and iodine salts can limit ( releases independent of containment int e:Irit y. Further, even a leaky containment can plug with aerosols , but not with gases. l 1
~ ' ~ Q&
~'
Page 11 (
- 8. 8-4, Conclusion 9; The conclusion is contradicted by -the discussion, where it is admitted that analyses are incomplete. The conclusion does not note that the overwhelming number of analyses. show large reductions from WASH-1400.
9.8-5, Conclusion 10; This conclusion is irresponsible, because ,the containment failure modes used in the BMI-2104 analyses are admitted to be an arbitrary parametric study. See page 4-28, paragraph 2. The risk values, therefore, are without merit. 10.8-5, conclusion 11; The conclusion is contradicted by the discussion. 11.8-6, Recommendation 1; This is correct; the new methods should be used while improvements are continuing. 0 12.8-7, Recommendation 2; The " Source Term Code Package" has not been , reviewed, is believed to be incomplete, and may have errors. 4 13.8-7, Recommendation 3; The Source Term Code Package cannot give a -
"best-estimate" source term estimate if the many qualifications , earlier in this document are to be believed. Review is necessary.
- Appendix B:
1.5-15,2,3,& 4; The assumption is made that operation of sprays and fan coolers in the containment will lead to conditions that allow a massive hydrogen explosion to occur that. will cause failure of containment. This is nonsense, is contradicted by discussions elsewhere in 0956, and contains hidden assumptions about the postulated accident, metal-water reactions, hydrogen ignition.
. and strength of containment.
k
W
- Pcg3 12
( - Appendix C: 1.C.1-3.1; The group of " expert analysts" apparently did not have any containment design and construction experts. The work should be reviewed by experts in design and construction. 2.C.1-4, Standard Problems; The choice of challenges is complete enough to provide a basis for analysis. 3.C.1-7.2; The steam-spike-induced failure is judged to be "of very low probability." Given the analyses, why not zero probability? 4.C.1-8,1; Failure is estimated at 16 hours for the Zion containment. Other analysts predict much longer times, suggesting that review of the working group kould be advisable. 5.C.1-9,1; Again, why not a zero probability? (* 6.C.1-10.1 and' C.1-11,1; The matter of batteries forI 1,ykdvs
.a ;_ m: in the ice -
condenser containments is mentioned. Can this be regarded as a passive safeguard? Clastly, it is noe in BMI-2104.
- 7. C.1-ll,2 and C.1-12,3; The postulated early failure for Mk-I and Mk-ll containments should be analyzed further for leak rate and deposition in the surrounding reactor building.
8.C.1-14,2; The short review of the postulated ." direct heating" sequence shows very clearly that extreme assumptions are required, and that depressurization might occur prior to vessel failure or even fuel melting. A more intensive study is needed. 9.C.2-3. Executive Summarf of the Containment Performance Working Group (CPWC); Generally, this group has completed a very useful study. The results should be reviewed by a group of individuals who are u__
depf. Pcg2 13 i l (.- l active in the design and construction of containment buildings. p Further interaction can modify estimates made for the most 1 i likely failure and reasonable extremes. l These data must then be ied into the BMI analyses to product better l estimates of source terms. e
. e ..g o .}}