ML19289A056
| ML19289A056 | |
| Person / Time | |
|---|---|
| Site: | Trojan File:Portland General Electric icon.png |
| Issue date: | 12/20/1978 |
| From: | Office of Nuclear Reactor Regulation |
| To: | |
| Shared Package | |
| ML19289A054 | List: |
| References | |
| NUDOCS 7901040196 | |
| Download: ML19289A056 (9) | |
Text
.us esc
'e' UNIT E D e,t Af t s NUCLE /.R RCCut/JCRY 00 / MS01Cf.'
e
,e
- v. Assmiss o c. nsts i
I SArtTY EvtLUAT:c:4 CY THE 0 FICE OF MUCLE'e EACTP: ?EG'JLAT CN S'.'DPORT ! h" 000E0 FOR t00!FICATIC" Ci LICENSE PELATEC TO [aT!C? DI WE*TINCWSE r,cc3 pystgAt;ey.: rt ntrW'r ti on Uestingneuse was infot.:.ed on " Arch 21, 1978 by one Of their licensees tt,at an error hac eeen discovere1 in their ECCf. Es al act!0n "Mel.
This errer sas c; 'en to octn :ne ole'wn arc heatur codes.
. e sti'i: 'ca '
ceter.inen ',y enal yses tha t the f uel r:0 hec
- Mlance equotira in the LCCT! I!.'. S ' T a'l VI cc:es was in crrte en1 tnat the LCC/. e nlyse:
ora"iousiv suNiitted by tveir customers were incerrect and creaictea pe?L cl y te-eratures PCT'd which were toc lau. "estin: house cr.cernined tut only half of tee volueetric heat aaneration due to r;tal-va te r re3 tien was useo in calculating tne ci:ddin, tenocratur-5.
'tr rf"f i tcied sa t e +v ocM +.i n n e r,i e.tod e,i nc e p r el i mi n a ry n e.t ut t e *,
I h n e,
?a indict.te, that sv :o olt.n s would "ot aaet tre 2200'F linit of 10 CFi bu.46 at the calculated oaxinun overall reakinn f actor licit.
<.e s t i nc.
hcuse retitica t'ieir costaners anc. MC on Mar:h 23,197J s.hile the utilities notified f3C thr0 ugh the regional Of fices of !nsnc: tion anc Enfor lient.
Preiptly uten notifica 1on Dy..estinchouse, the UDC staf f Assessed tne imniate safety sienificance of in1s infor-'ation.
'lo notec certain points tNat iv.icat?c no irredato action ras recui*ed to assure safe c; ereti;n o' the pl a.nts.
First, nest plants operata.it a peakin-fcctor si nificantly belnu the naxi"un neuin; factor use(1 for safety C
c:lculctions.
By mRing safety ccmtations at f acters niphar then acto:1 onerctin; levele. the facility has a wice rance of flexibility, without the need for nour to hour r2coopJtations of core status. The difference bet.<een t** ac tual rect.ing f actors and the naxie'.t" calculct?d redinc factors, for mest plants, would of f set the penalty resulting fro 1 the correction of,tna error.
Second, for most roa:ters tnere are 7901040t%
2 a number of very plant-specific para.*eters which bear upon aspects of the ECCS performance calculations. Utilities ao not generally take crecit for these plant-specific paraneters preferring to pr0 vide a simpler conoutation which conservatively disregards these incivicuelly small credits. Third, the error in the Westinghouse comp;tations relates to the zirconium-water reaction heat source. This is an aspect of Appendix K, which is generally recogni:ed to be very conservative.
New exeerimental data indicate that the methocs recuired by Appendix K ape' eciably over estimate the heat source. Thus, unile the error in f act entails a ceviation from a specific recuirenent of Aprendix K, it coes not entail a natter of inmediate safety significance.
Westin; house continued to evaluate the impact of the error on previous plant saecific LOCA analyses and perforned scoping cal:ulations, sensitivity studies and some plant-scecific reanalyses.
In accition, Westinghouse investigateo several moaifications to tae previcss'.y approved methods which if accreved by tre NRC staff would offset some of the inneciate inoact of the error on Technical Specification limits and on the plants operating flexibility.
On March 29, 1978, Westinghouse and several of their customers met with ncabers of the NEC staf f in Bethesda. Westin1nouse cescribed in detail the origin of tne error, explainea now it af fectec tne LOCA analy>v>,
and how the error had been corrected and characteri:cd its af fect on current plant specific analyses.
In order to avoid reduction in the overall peaking f actor (Fg),luation model modifications wnich wouldWestinghouse three proposed ECCS-LOCA eva contribute a compensating reduction of PCT. They were characteri:ec as follows:
1.
Revised FLECHT 15 x 15 Heat Transfer Correlation This new reflood heat transfer correlation which had been recently developed and submitted by Westinghouse in Reference (1) was proposed as a replacene.it for the carrently approved FLECHT correlation. To determine the benefit, the proposed correlation was incorporatec into the LOCTA IV heatup code and was found to result in improved heat transfer curing the reflooo portion of the LOCA.
3 2.
Revised Zircalov Emissivity Based un recent EPRI data (Reference 2), Westinghouse croposed to modify the presently approved equation for Zircaloy cladcint emissivity to a constant value of 0.9.
The higher emissivity (previously below 0.8) provides increased radiative heat transfer from the hot fuel pin during the steam cooling period of reflood.
3.
Post-CHF Heat Transfer Westinghouse proposed to replace their present post-CHF transition boiling heat trantfer correlation with the Dcugall-Rohsenow film boiling correlation (Reference 3) which they stated was includeo in Appendix K to 10 CFR Part 50 as an acceptable post-CHF correlation.
These three model nodifications were classified as generic, applicable to all plant analyses.
Subsecuently, as oiscussec below, these changes were rejected by the tlRC staf f as providing generic benefit. However, a portion of the creoit proposed by Westinghouse was approved by the PRC staf' for certain specific plants, which nad previoed specific calculations with the new 15 x 15 correlation. During the period tiarch 29 to April it,1973, Westinghour.e nrovided us with additional sentitivity Analysn Anr. niant specific analysis in which they evaluated the effects of some changes to plant-specific inputs in the LOCA analyses. These were as follows:
1.
Assumed Plant Power Level A reduction of the plant power level assumed'in the SATAH VI blowdown analyses from 102*, of the Engineered Safeguards Design Power (ESDR) level to 1020 of rated power was pronosed.
Previously, analyses had been perforned at approximately 4.5% over the rated power. This change was worth aproximately 0.01 in Fg, and is refered to as $ESDR in Table 1.
2.
COC0 Code Input A modification to the COCO code input (Reference 3) to more realistically model the painted containment walls was proposed.
Since the paint on containnent walls provides additional resistance to heat loss into the walls, the COCO code calculates an increase in containment back pressure, which results in a
4 benefit to the calculated peak claccing temperature of 0 to 40'F, during the reflooding transient.
The meenituce of the benefit is depercent on the type of plant and the heat transfer prcperties of the paint, and results in up to 0.03 denefit in F, and is Q
referred to as AFCP in Table 1.
3.
Initial Fuel Pellet Temoerature A modification of the initial fuel pellet tomoerature from the design basis to the actual as-built pellet temperatures was
- proposed, in the present LOCA calculation:, Westinghouse nas assumed margins in the initial pellet temperature. The margin available is plant-specific anc ranges from 23'F to SS'F.
Use of the actual pellet temoerature rather than the assumed value results in a reduction in pellet temperature (stored energy) at the end of blowdown, as calculated by the SAT A'! C0de, of Soprox-inately 1/3 of the initial cellet te cerature margin. Uesting-house has provided sensitivity analyses s.hi:h indi: ate that a 37'F reouction in fuel ;ellet temperature at end of blowdown is worth apercximately 0.1 in FQ. This is referred to as AFPT in Table 1.
4.
A::umulator Water Volum: Con:iceration Westinghouse has evaluated the effect on ECCS perfomance of reducing the accumulator water volume, and has deter nned that for those plants for unich the downcomer is refilled before the accunulators are emptied, there is a benefit in PCT.
The sensitivity studies have indicated that this benefit in FQ is pl ant-speci fi c. This is referred to as a FACV in Table 1.
5.
Steam Generator Tube Plue:ina Consideration In previous analyses Westinghouse has assumed values of steam generator tube plugging whien were. greater tnan the cctual plant-specific degree of plugging.
Sensitivity analyses submitted in Reference 4 were used to evaluate the benefit available by realistically representing the plant-specific data. For the plants affected, the benefit in PCT ranged from 7 to 66'F which was conservatively worth from 0.007 to 0.'66 in F. This is O
referred to asafsG in Table 1.
Discussion Pnd Evaluation The infor 4 tion proviced by Westinghouse was separated into two categ0rics; the generic evaluation model nocifications and tre plant scecific sensitivity stucies and reanalytes. The UC staf f reviewed the reaking factor lirits proposec by Westinghouse to verify their conservatism.
The netal-water reaction heat generation error in the Westinghcuse ECCS evalut.tien model was evaluated by us to leternine an appecpriate interim pe nal ty.
Westin;Scuse provided t,<o pr*11minary separate c'fects calcula-tions which incicated that a maximum penalty of fron 0.14 to 0.17 was appropriate to cot";ensate for the mocel error. The staf f,censervatively roundea inis penalty up to 0.20.(Reference 5)
Westingnouse also oroposed several cortensating generic channes in their evaluatien model to of fset any necessary reductions in peaking f actor cue to the error. These changes were assesscc by us as follows:(Reference 5) 1.
flo credit would be given at this tire for the changes in the post-CHF heat transfer ccrrelation anc new Zircaloy emissivity data.
2.
Fartial credit (700) woulu be given at this time for Uie sie of the new 15 x 15 FLECHT correlation only for 01cnts unich had proviced a specific calculation demonstrating that such credit wcs a;propriate.
Based on this review we develooed recommended interi") peakina factor limits for all the coerating plants and ceciaed 'inat any othir plant-specific interin factors (benefits) not related to the generic review should be considered separately.
In addition, the staf f revierted plcnt-specific recnolyses for DC Cock Unit !!ois. I and 2. Zion Unit Nos. I cnd 2 and Turkey Point Unit tio. 3 which had corrected the error in r:etal-water rer tion.
In these analyses the Dougall-Ronsenow and Zircaloy enissivity crevits were not consicered, while the new 15 x 15 FLECHT correlation was included. We concluded that these reanalyses could serve as a basis for conservatively determini,1g interim peaking factor limits for these plants.
For nost of the operating plants our generic review resulted in a lower allowable peaking f actor than Westinghouse had preposed. However, in one case, Westinohouse had crocosed more limiting peaking factors in order to prevent clad temperatures at the rupture noce from exceeci7 2200*F. We concluded tnat it would be properly conservative to use the minimum of these values.
Based on plant specific sensitivity studies, perfored by riestinghouse, the licensees have submittec renests for interim plant-specific benefits.
We reviested these sensitivity studies and recommendec that aporo-priate credits be acceptec.
The results of these analyses are shown in TaDie 1.
We informed each licensee by telephone on April 3,1978, that they should administratively reduce the plant's peaking f actor limit from the limit containec in the Technical Specifications to the interim peaking factor limit contained in the right hand column of Table 1.
In those cases where the limit in Table 1 is 2.32, this represents no change from the Technical Specifications limit. The peaking factor limit of 2.32 is generally supported and aporoved for Westinghouse reactors enploying constant axial effset control operating procecures (P.eference 6).
For the reactors having an interin peeking f actor linit of 2.31, we requested no further justificat1on of the limit. This is cecause tne generic analysis supporting the limit of 2.32 approacnes the linit only at beginning of the first cycle.
Since the affectec reactors have operated past this point, it is clear that the maximum attainable reeking f actor will be less than 2.32.
While this margin has not been quantifiec, we are convinced it is substantially greater than the 0.01 for which we are recuiring no additional Justification from tne piants with an interim limit of 2.31.
For the reactors with an interim limit less than 2.31 we recuested that the licensee furnish administratively imposed t,ocedures to replace Techn; cal Specifications either:
1.
To provice a plant specific constant axial offset control analysis of 18 cases of load following which would ensure that the interim limit would not be exceeded in nomal operation of the powen plant, or, at its option, if sucn analysis were unobtainable, inappropriate or inwf ficient, 2.
To institute procedures for axial power distribution monitoring of the interim limit using a system designed for this purpose.
If such systems do not exist manual procedures could be used as indicated in our Standard Technical Specifications 3/4 2.6 and ancillary Specifications.
7 We requested the licensees to confirm by letter that they have adopted tne above interim LOCA analyses, interim peaking f acter limits and aaministrative procedures by April 10, 1978, if their reactors were operating, and by April 17, 1978, if the reactors were not operating.
Conclusion We conclude that when final revised calculations for the facility are submitted using the revised and corrected model, they will demonstrate that with the peaking factors set forth herein, operation will conform to the criteria of 10 CFR 50.46(b). Such revised calculations fully conforming to 10 CFR 50.46 are to be provided for tne facility as soon as possible.
As discussed herein, the peaking f actor limits specified in the particular Orders or Exemptions issued for the affected f acilities, with operating surveillance requirements, as applicable, specified in Orders or Exemptions for particular plants, will assure that the ECCS will conform to the nerfor-mance recuirements of 10 CFR 50.46(b). Accordingly, limits on calet'ated peak clad temperature, maximum cladding oxidation, maximum hydrogen genera-tion, coolable geometry and long term cooling provide reasonable assurance that the public health and safety will not be endangered.
Oate: December 20, 1978
e Deferences 1.
R. S. Deucell, W. M. Rohsenow, " Film Boiling en tre Insite of Verti:ai Tu0es witn Urwara Flow of the Fluid at Low Qualities",
MIT Retort 9073-26, Secterber 1963.
2.
EPRI Report NP-525, "High Temperature Properties of 2ircaloy.
Oxygen Alloy", March 197 7.
3.
WCAP-9220, "',:estingbcuse ECCS Evaluation Model, February 1975 Yersion", February 1973.
4.
UCAP-Ec85
" Perturbation Technique For Calculatine ECCS Cooling Perf or: nance", February 1977.
5.
Memerancun: 0,os: toc:y to Eiseniwt and '.oss, "Metsi-Water Peaction Heat Generation Error in Westinghouse ECCS Evaluation t'odel Computer Progran." April 7,1978.
f.
T. Mori tt, et al., "Pcuer Distridution Control and Loa 1 Follo.sien Proteocres, " WC' P-6355 (P r0erietiryl and UC AP-8003 ('ior,-P roprie ta ry ),
September 1974 8
9
9 T1.BtE I KT Fo aFg afar 0 aFILEDIT'I ISE aF p alpy aF g afgcy In LIMIT I, Mitt aFL50R c
5 Q
2 PCT Fq Analysts
- F 0[D 2 Ioop Ft. Beach 1 20?$
2.32
.16
.2 2.32 2. 2 81 2.32 2.26
.01 029 Pt. ecach 2 2025 2.32
.16
.2 2.28 2.32 2.28
,03
.f46 2.32 Ginna 1912 2.32
.26
.2 2.32 2.32 2.32 2,32 053 Srwaunce 2112 2.25
.03
.2
.05 2.13 2.25 2.13 01
.02
- 2. l f, 1valrie Island 1/2 2181 2.32
.ol
.2 05 2.16 2.26 2.18 01 02
,03 2.24(a) 3 t emp North Anna 21HI 2.32
.02
.2 2.14
- 2. 12 2.14 2.14 Beaver Valley 2011 2.32
.15
.2 2.21 2.32 2.21
.0 36 2.31 farley 19 J1 2.32
.24
.2 2.32
- 2. 12 2.32
.01
.005 2.32 e
5 ery 1 2111 1.t15 02
.2
.06 1.13
- 1. st s 3.11 03
.025
.023 1.81 Soery 2 2 11 1.85 02
.2 06 1.13 1.HI 1.13
.03
.025
.023 1.81 Iveter Point 3 2019* 1.90 14 0
.03 2.01 2.v6 2.01
.cro 2.03 luency Point C 2195 2.05 00
.2 05 1.90 1.91 1.90
.01 1.91 4 s emp e
Indian reint 2 20fE 2.32
.31
.2 2.23 2.23 2.23
.01 2.24 Indian Folnt 3 2125 2.32
.01
.2
.06 2.25 2.19 2.19
.01 03 2.23 Iso 3an 1915 2.32
.26
.2 2.32 2.32
- 2. 32
.01 031 2.32 Sa tt* 1 2115 2.32 06
.2 2.10 2.32 2.18
.01
.024 2.21 lion 1/2 Flc)** 2.01 0
.03
- 2. 0-1 2.01 2.0c(,}
(col I Z l f,1
- 1.90
.03 0
.03 1.99 1.91 1.90 l
1.90 C4*i 2 2190* 2.10 01 0
0 2.11 2.11 0
0 1
0 0
2.11 t
Fg
- 0: edit in Fq for PCI serlin to 2 in"F limit.
F :02 - Metal Wter Reaction penalty on I.
r q
for Improvements to 15sl5 f tEOti Correlation.
F,g((gg-Credit in Fq fp(g
- Staf f estimated Iq Isased em 2200"F PCT limit.
Ist
- Westinghouse putosed in lased on stored energy sensitivity studies.
- thnotes reanalysis at Fq oIJ value error <nere(ted.
3
- th notes reanalyses at fg old value, ce ror (ovve(ted, an.mslator Vol. Change of 100 f t, a((umulator pressure of f 50 p la
(*) Te.cse limit s are appilt, lite assuming II(ensee evidif f es muesulator (ovnlitions as altropriate. I f. wit. l'a alt te Islan.s 112 F =2.21. lita 1/2 59=l.9 l
q
.