ML19259B512
| ML19259B512 | |
| Person / Time | |
|---|---|
| Site: | Dresden  |
| Issue date: | 02/01/1979 |
| From: | Ziemann D Office of Nuclear Reactor Regulation |
| To: | Reed C COMMONWEALTH EDISON CO. |
| References | |
| NUDOCS 7903010342 | |
| Download: ML19259B512 (8) | |
Text
~;F a*fc<<9, b
UNITED STATES NUCLEAR REGULATORY COMMisslON h
9 W,iSHING TON, D. C. 20555
, 9 :,.., 7.,}
^
i *wn -/ /
February 1,1979
~.....-
Docket No. 50-10 Mr. Cordell Reed Assistant Vice President Commonwealth Edison Company Post Office Box 767 Chicago, Illinois 60690
Dear Mr. Reed:
We are reviewing the t., ore Spray Distribution Evaluation submittec in support of Dresden Unit No.1 operation by your letter dated December 23, 1978. To continue our review, the additional information identified in is required.
Please provide the requested information by May 1,1979.
Sincerely,
..'s.
~[
n.~
s..
,. v..~..a
~,
Dennis L. Ziemann, Ch'ief Operating Reactors Branch =2 Division of Operating Reactors
Enclosure:
Request for Additional Information cc w/ enclosure:
See next page 7903010342_
p
- . ":-de'l Reed February 1,1979
- w/erciosura:
- 'r. d hr. U. P.CWe
- shan, Lincoln & Beale Ccunselors at Law One First National Plaza, 42nd Floor Chicago, Illinois 60603 "r. B. 3. Stephenson Olant Superintendent Dresden Nuclear Power Station P.;ral Rcute Al "erris, Illinois 60450
'J. S. ' uclear P.egulatory Co nission Jirry L. Barker
?. O. Ecx 706
- rris, Illinois 60450 In an '. Sekuler
-ss4 stint Attorney Ger.eral ir. iron ental Centrol Division
'IO '<!. Randol:h Street Sui:e 2315 Chica;;, :llinois 60601
:-ris Public Library f:2 'iderty Street "erris, Illinois 60451
h-
-u
-~.4-:,..
/-
~
1
,A Ms-
^ _ -
=----__........_-w
.0-.-_,-
r.4-. -.*
.e
. ~,
=-e
_..s
-~..-r:_-
1 g
- ]
ENCLOSURE ADDITIONAL IfiFOR'% TION RE0VIREMENTS tie D2-2cl 64 December 1978 I.
Justification of t'ethodology's Applicability to D-1 The NRC staff has provided tentative approval of the GE ceneric methods that are utt'.ized in NED3-24164, but the approval is subject to verifi-4 cation of the methods utilizing the 30* steam testing facility at Lynr.,
Massachusetts. The Lynn facility is a full scale mockup of a 30' sector of the upper plenum of a 218" I. D. BWR/6 reactor. There are significant differences between the Dresden i design and a BWR/S desien, including:
- 1) D-1 uses nozzles which produce a relatively narrow, hollow core spray pattern whereas the BWR/6 nozzles produce a solid core spray pattern;
- 2) the D-1 reactor core is much smaller in diameter; 3) the D-1 nczzles
^:
are approximately 2-1/2 feet above the core whereas BWR/6 nozzles are approximately one foot or less above the core; and 4) tne D-1 nozzles are aimed downward 60* below horizontal whereas the SWR /6 noz:les are aimed approximately horizontally.
In summary, the D-1 gecmetry has some of the characteristics of the Big Rock Foint ring spray syste:- (where g'
GE felt the Lynn facility might not be applicable) and some of the features of the BWR/6 design (where GE believes the Lynn facility is rest
~
applicable).
In view of these facts, we believe the Lynn results will provide somewhat less assurance that the methods are applicable to D-1 as compared to BWR/6's and similar design plants. Accordingly, pl eas e:
I.l.
Explain qualitatively how you intend to justify applicability
~
of the methodology to D-1.
e I.2.
Explain quantitatively how much penalty (reduction in minimum spray per channel) you propose to apply to the D-1 spray e
distribution results to account for uncertainties in the methodology as applied to D-l.
II. Selection of Simulator Nozzle and Single dozzle Effects The GE generic methodology utilizes " simulator" noza.
you have
+
done for D-1.
However, the generic method (approved su st to con-firmation at Lynn) will select simulator nozzles with the prime selection criterion being that each individuc1 simulator nozzle must produce, i
in air, a spray pattern closely matching the spray pattern produced by the actual nozzle in steam.
Your stated criterion in the last para-graph of page 6-1 is not consistent with the above criterion.
In fa ct, the criterion stated on page 6-1 could have the effect of selecting the simulator nozzle which would produce minimum interaction phenomena, whereas the desired result is to produce in air whatever interaction h
y t
e
'+
N y
}. " -
".,, )"
g 9
e p
, y.
- l s.,
j
2 Aenonena actually would occur in steam. Accordingly, please prcvide the following:
11.1.
Qualitatively discuss your selection criteria for the simulator nozzle (i.e., elaborate on the brief pg. 6-1 discussion, stating all factors which were considered and why).
II.2.
Provide figures showing normalized flow density vs. core radius, for all simulator nozzles considered, for several representative flow rates (#or comparison to Figures 4-5, 4-6 and 4-7 wni:5 show the D-1 actual nozzle distribution).
II.3.
Describe in detail how the individual nozz'e data is cnaracterizec for use in the superposition calculational codes. More detail is required than the individual nozzles' centerline flow dist-ibu-f on, which is what you have shown in your Figures 4.5, 4.6 and 4.7 involving single nozzles' spray dutributions.
(A graphical example would be helpful).
II.".
Provide data suitable to quantify the azimuthal asyretry (i.e.
unequal flows at " mirror image" or otherwise equivalent ge: metrical positions) of:
- 1) the actual D-l spray nozzle; and 2) the simulater nozzles censidered.
Two exanples of acceptable ways to do this would be with figures such as 4-6 but showing and comparia; da a taken with the nozzles rotated about their o.en axes (i.e. screaec a fraction of a turn on the connecting nipple) to three or more different positions for each nozzle so tested, or with data taken from the individual nozzles counted vertically with ceasuremer.s across several different diametrical paths.
The da ta shoulo be presented at several representative flow rates for each nozzle.
III.
Sensitivity of Soray Distribution to Various Parameters (Multiple Nozzle Effects)
The siumlator nozzle selected will in all probability be found to best represent actual nozzle spray patterns in steam at only one pressure (or relatively narrow pressure range) and one flow rate (or relatively narrow range of flow rates). Sensitivity of the overall spray distribution (from many nozzles) to these and other factors must be determined. Accordingly, please provide the following quar.titative estimates of the ef fect on spray distributic.
(particularly on the minimum spray flow to any channel) of:
III.l. System pressure.
Figure 9-2 is deficient as it does not include effects due to atypicality of tha simulator nozzle at different pressures.
111.2 Flow ra te.
Figure 9-3 is deficient as it does not include atypicality of the simulator nozzle at different flow ra tes.
III.3 A'zimuthal position. A conservative representation of variation between similar azimuthal locations must Se considered, for example by subtraction of the worst local deviation from your " final" distribution. Use of radial average values, which is essentially what you have done, is not acceptable. Measurements of the type presented in Figures 7-7 and 7-8 ( only forthe flow, pressure, and the temperature under consideration) might be acceptable to determine the worst local deviation, but such measurement must include flow measurement across diametrical paths that terminate half-way between adjacent nozzles, as opoosed to diametrical paths that terminate at nozziss as shown on Figure 7-5.
Inclusion of measurements along diameters not terminating en a nozzle could be crucial, especially for the outermost fuel channels.
III.4 Spray tempera ture. This may be a smaller effect (oer your text) than the other effects mentioned in this section, but it must be taken into account.
Figure
,-l is deficient in that it does not take into account atypicality of the simulator nozzle at various temperatures.
III.5 p.andom rotation of individual nozzles and groups of adjacent nozzles about their own axis, Single nozzle asym e:ry effects are discussed in 11.4.
III.6 Snall errors in the exact aiming of individual nozzles.
6.1 You state an aiming tolerance of 12.
Please describe how you measured the aiming point of each of the 60 nozzles no.,-
on the sparger in the Dresden 1 reactor, and describe how you duplicated that aiming pattern in the mockup sparger used for testing. Justify that the sum of all errors involved in this process results the stated tolerance (12)in nozzles on the test sparger within of the nozzles on the actual sparger in the reactor.
6.2 Provide data to quantify the effect un minirum channel flow cf variation (from nominal) of +2' and -2* in aiming angle of individual nozzles, and adjacent groups of nozzles, at seve. a1 fl ow rates.
4
-4 6.3 If pg. 9-2, 7th line fron bottom should read "re-air.ing" (not reaming), explain the magnitude of the difference in spray pattern caused by the re-aiming in the context of your responses to questions III.6.1 and III.6.2.
(If
" reaming" is correct, disregard III.6.3 but see V.5).
IV.
Relation of Tested Parameters to predicted LOCA Condition Since you have not submitted an ECCS-LOCA analysis utilizing a model that the NRC staff has found to be wholly in conformance to the re-quirements of " Appendix K", it is not possible to draw final conclu-siens regarding acceptability of the pressures, temperatures, and flow rates selected for your tests. Nevertheless, earlier analyses raise cuestions regarding your selection of the set of " primary evaluation conditions" which are derived from your D3A.
In particular D-1 has a very " flat" break spettrum with a considerable rangr of break sizes having PCTs very close to the D3A. Alsc, a large range of break sizes results in core uncovery, i.e., credit is taken for core spray cooling in the analyses.
Therefore, any degracation of core spray distributionover a large range of break si es could increase PCT above 2200*F which would cause yourcor. sideration of " primary evaluation conditions"alone to be unacceptabie. Accordingly, please provide the following:
IV.1 For each of several specific break sizes covering the break size range where core uncovery is predicted, provide a separate Table which shows the length of time during which the CS system will experience conditions in various pre-specified ranges of core pressures, spray flows, and spray temperatures.
The ranges of core pressures, spray flows, and spray temperatures should be. selected in coor-dination with your responses to item III so that you can also specify in the Tables the expected mir imum channel spray flow rate for each of the time periods.
This can be attempted by utilizing combinations of the " sensitivity studies" proviced in response to item III, i.e., conservatively estimate how to combine the effects of varying each of the several parameters considered individually in item III, roulting in an estimate of the minimum flow to any channel for each particular set of conditions listed in the several tables.
V.
Miscellaneous Additional Information Needed V.l.
How do you propose to justify the acceptability of the soray cooling coefficients assumed in your LOCA analyses with tne knoYledge
.s of the 0.8 gpm your tests indicate for peripheral channels?
Specify any low flow (FLECHT or other) tests you will reference, and compare c; war levels in the test bundles so referenced to power levels expected for your peripheral bundles.
V.2.
Many of the sarameters discussed in Section III above would be particularly important for peripheral bundles. That is, slight atypicalities in core angle, azimuthal variation, or aiming angle could severely reduce flow to bundles that are only margionally within the predicted spray pattern.
For example a variation between +2' and -2* from nominal in aiming angle would move the position of the s; ray core edge by more than 2", which would likely leave some bundles with zero flow. Justify conservatism of your stated peripheral bundle minimum of 0.8 gpm in view of the many uncertainties discussed in Section III.
~
V.3.
State and justify the penalty (reduction) you propose to apply to the minimum channel spray flow rate that you predict for pressures of 155 psia to account for uncertainties in extrapolation of your lower pressure data (45 psia maximum, or 75 psia maximum including " vertical nozzle" data).
V.4.
How do you propose to justify acceptability of assuming spray cooling credit during the long term Post Incident (PI) period? Your analyses of test data conclude that the core will not be covered when only PI pump flow (1600 gem) is available.
If you propose to utilize one or more CS pumps, ycur answer must specify the time available before the contain-ment reaches an unacceptably high level and/or must specify system: (and their capacities) which could remove water from containment.
V.S.
Your report does not discuss nozzle reaming except on page 9-2.
Please explain, and justify (the validity of comcaring data befora and after reaming.
If you meant "re-aiming," disregard V.5 but see III.6.3)
V.6.
The flow measurement system is not visible in the photograph (Figure 7-9). Does it sit on the floor beneath the header?
Provide a detailed description of the orientation of the sparger, collectors, supporting equipment, etc. and justify that splashing or aleshing from the floor or other surfaces (present in the test but not in the reactor) would not affect.
the spray distribution mea turements.
Y.7.
With respect to your 2-to-3 second collecting period:
7.1 Justify your use of such a short collecting period. What uncertainty do you propose to include in your minimum channel flow rate to account for inaccuracies introduced by the short collecting period?
7.2 Describe how you initiate data collection to avoid effects due to the starting transient of the test rig (i.e. before full flow is reached and the full, steady state spray distribution is achieved). Repeatability of the results by itself gives no assurance that startup effects are absent; you could be erroneously measuring t;ie same startup effect with each repeat test.
7.3 Explain why the " central plume" causes unsteady spray density in the center of the core.
It is possi'cle that the unsteady flow in that region is caused by the short measuring period and/or the starting transient.
V.8.
Provide details of all differences between the full-size sparger mockup, including piping leading to that sparger, and the actual sparger and supply piping on the 0-1 reactor.
You state on page 7-1 that " insignificant" differences exist, but this is not adequate information to allow us to indeperdently reach the same conclusion.
e S