ML19290E266
| ML19290E266 | |
| Person / Time | |
|---|---|
| Issue date: | 02/01/1980 |
| From: | Marsh L Office of Nuclear Reactor Regulation |
| To: | Check P Office of Nuclear Reactor Regulation |
| References | |
| NUDOCS 8003100067 | |
| Download: ML19290E266 (15) | |
Text
-
MEETING
SUMMARY
DISTRIBUTION
- H. Denton L. Gifford, GE, Bethesda
'O. Eisenhut S. Sandoz, GE, San Jose R. Tedesco G. Sherwood, GE, San Jose B. Grimes A. Levine, GE, San Jose L. Shao W. Hodges P. Check R. Frahm D. Crutchfield R. Woods A. Schwencer L. Marsh D. Ziemann G. Lainas V. Noonan G. Knighton T. Ippolito
- 0. Brinkman K. Kniel Z. Rosztaczy D. Fieno T. Novak
- R. Silver
- S. Nowicki
- J.
Shea
- P. O' Conner
- T. Wambach H. Thornburg K. Seyfrit S. Bryan E. Jordan
- 0RB 3-PMs
- C. Berlinger
- M. Aycock OELD OI&E (3) 050 (3)
R. Fraley, ACRS (16)
TERA Docket Fil NRC POR Local POR NRR Reading File RS Reading File Receptionist, Bethesda Attendees
- Denotes person to receive a copy of slides 8003100 O6$
c gaa arog
-[ "5 d
UNITED STATES
- ,"w i j NUCLEAR REGULATORY COMMISSION
... E WASHINGTON, 3. C. ll0555
't
[
a....
TIE 3. '980 MEMORANDUM FOR: Paul S. Check, Chief, Reactor Safety Branch Division of Operating Reactors FROM:
Ledyard S. Marsh, Reactor Safety Branch Division of Operating Reactors THRU:
Vincent W. Panciera, Section Leader, Reactor Safety Branch, Division of Operating Reactors
SUBJECT:
SUMMARY
OF MEETING WITH GENERAL ELECTRIC CCl1PANY REGARDING STEAM EFFECTS ON BWR CORE SPRAY DISTRIBUTION Representatives of the NRC staff and the General Electric Company met on November 15, 1979, in Bethesda, Maryland, to discuss GE's plans in the subjec* area and responses to several staff questions. A summary of the meet.ng, an attendance list, the staff's questions, GE's cate-gorization of the staff questions, and copies of the presentation slides are attached.
L '* L Ledyard B. Marsh Reactor Safety Branch Division of Operating Reactors
Enclosures:
As stated cc: See attached list
ATTENDANCE LIST, NOVEMBER 15, 1979 CORE SPRAY DISTRIBUTION MEETING V. Panciera NRC/ DOR L. B. Marsh NRC/ DOR W. Hodges NRC/ DSS R. Frahm NRC/ DSS R. Woods NRC/I&E J. Alai GE G. Dix GE L. Rodriguez GE
ATTACHMENT 1 Summary of Meetine with GE to discuss Steam effects on Core Soray Distribution The meeting was requested by GE to discuss their progress to date, and
'cu give their responses to several staff questions generated during the review of GE's August,1979 report " Core Spray Design Methodology Tests."
GE briefly reviewed their testing and analyses in their efforts to in-vestigate and ecdel the effects of steam on SWR core spray. Single and multinozzle tests in a full scale mockup of a SWR-6 upper plenum in an air environment were conducted at the Valicitos, California facility,
.%re recently, multinozzle tests in a full scale mockup of a 30* sector of a BWR-6 upper plenum in various steam environments were conducted.
These tests have been performed by GE in developing models and techniques to predict core spray distribution in typical post LOCA environments.
This phase of the overall effort to ensure adequate BWR post LOCA fuel spray cooling has been called by GE the " methodology" phase. Once the methodology is developed and accepted, each BWR licensee must ensure acceptable fuel cooling by applying the methodology to their particular plant design. This phase of the effort will be the final step in assuring adequate core spray in steam environments, and has been called the "appli-cation" phase.
The staff's questions from its review of the GE report " Core Spray Design Methodology Tests" which had been informally given to GE previously, were tentatively grouped (by GE) into two categories (corresponding to the two phases): those pertaining to the methodology being used by GE in developing
2 models and techniques to predict SWR core spray in a steam environment and those pertaining to the application of the methodology to individual reactors in assuring acceptable fuel performance in LOCA situations.
The staff's questions and their categori:ation are attached.
GE was prepared to discuss only their responses to the staff's methodology questions, and stated that the application questions would be addressed during that phase of the overall effort. It was GE's contention that by acceptably answering the staff's methodology questions and by presenting corroborating data, the GE methodology for predicting BWR core spray in a steam environment would be substantiated.
The staff stated that if all methodology questions were acceptably answered, then the general methodology would be accepted. However, the staff stated that the separation of the questions into two groups might be oversimplifying the contents of the questfons. For example, certain questions may appear to be " application" type questions, but their response may bring to light questions concerning the validity of the overall methodology.
GE responded that the methodology itself should be justifiable, independent of specific application questions, but if methodology concerns arose during the responses to application questions, those concerns would be answered.
GE then presented their responses to the staff's methodology questions.
The staff noted that GE's methodology report and their responses to the staff's questions were given in terms of flow density (gpm/ bundle) at
t 3
certain core radii, but no data was given for the bundle predicted flow.
rates. Since acceptance of the methodology depends on how well the technique predicts individual bundle flowrates, the staff requested that questions pertaining to predicted or actual bundle flowrate be answered with bundle by bundle data, not with flow densities.
Additionally, the staff requested GE to document the information presented on the slides shown at the meeting, as well as the accompanying explanation.
G9 requested, and the staff agreed to fonnally ask both the methodology questions being discussed at the meeting and the application questions which would be addressed at a later date. These questions (Attachment 2) had been previously given to GE informally. The staff concluded that the following questions, however, had been acceptably addressed at the meeting and would not be formally asked and did not require any further response:
la, Sa, 7, 9, and 11.
GE requested, and the staff agreed to send GE a letter, when all methodology questions have been acceptably answered, stating our acceptance of their techniques. The staff stated that acceptance of the GE methodology depends on satisfactorily responding to the methodology questions referred to above, as well as any other methodology type questions which may arise in our review of GE's responses,
t ATTACHMEtiT 2 ADDITI0tlAL It4FCRt'ATI0tl REQUESTS (1 ) With regard to the ccre spray distribution data taken at Lynn and presented in this report:
a) Explain how the measured individual bundle (or partial bundle) flow rates were used to calculate the (average) bundle flow vs. core radius data presented in your report.
Include a table and/or figure showing which bundles (or portions of bundles) were used to determine the spray flows presented in your report for core radii of 27", 33", 39",
45", 51", 57", 63", 69", 75" and 81" respectively.
b) Explain how the data presented (average spray flow vs. radius) can be intarpreted by the reader to allow deternination of the conserva-tively bounding minimum spray flow that was actually measured in the Lynn facility for any individual bundle at each given radius ( 1 ").
27 If the data as presented cannot be used for this purpose (i.e., if you cannot state, for example, that "in all cases the minimum was within (x%) of the average") then you must present new figures for each case used in your confirmation of the GE-core-spray-distribution methodology, showing minimum bundle core spray flow measured at each radius (for radii 1 ").
27 c) Indicate the uncertainty that should be applied to the measured minimum bundle flow rates due to bundle flow measurenent errors, calibration errors, total sparger flow uncertainty, etc.
t
-2 (2) With regard to the GE spray distribution methodology and its confimation by the Lynn tests for 218" I. D. BWR/6 plants:
a) State the spray flows for one CS system and for two CS system operatic, : hat GE believes would be the minimum present in any bundle following any size LOCA under all conditions when LOCA analyses take credit for core spray cooling.
b) Describe in detail all steps in the derivation of the minimum values presented in 2-a.
For example, are the flows presented the mean values of "GE methodology superposition" calculations less 2-sigma?
c) Describe how differences between the predicted post-LOCA conditions and test or calculational conditions assumed or measured in 2-b are accounted for, including but not necessarily limited to:
- 1) post-LOCA pressures outside the range where HSF (and/or Lynn) tests were perfomed (i.e. you used 29.5 psia, but credit is taken for spray cooling up to C150 psia);
- 2) spray temperatures outside the range tested, or a combination of post-LOCA pressure and spray temperature different from any combination tested.
(3) a) Compare the minimum measured flows (#1-b less the uncertainty stated in #1-c) to the minimum flows stated in !2-a, and compare both to the SWR /S - 218" minimum design spray flow of 3.25 gpm per bundle.
- t b) If any of the values presented above are less than 3.25 gpm, exclain how you will justify acceptability of the spraycooling coefficients assumed in the 218" BWR/6 ECCS
" Appendix K" analyses, which are based on FLECHT tests with 3.25 gpm supplied to the electrically heated single bundle, provide a senedule for any new tests (FLECHT, etc.) you may propose.
(4) With regard to application of the GE superposition methodology to plants sizes other than 218" (where Lynn results are not available for direct comparison, i.e., 238" BWR/S or 251" BWR/4, etc.) and considering the differences between the Lynn results discussed above and the " super-position methodology" results (including measurement uncertainty sub-tracted from the Lynn results), quantitatively discuss how much re-duction will be applied to the minimum channel flow calculated for such plants by applying the GE superposition methodology.
(5) On the morning of July 5,1979 an NRC staff member (R. Woods) witnessed two representative tests at the Lynn facility. He noted and comented on the fact that certain low-flow bundle locations did not appear to benefit significantly from operation of two spargers as opposed to one sparger (the spray flow to those locations remained essentially the same for a one and for a two sparger steam test). Since justification for continued operation of BWRs pending completion of final testing (which includes Lynn testing) partially relies on the minimum-bundle-spray-flow margin believed to be present due to 2 sparger operation, Woods' observation would appear to be a significant result of the Lynn testing. However, your report does not specifically mention these results and does not present data in sufficient detail to verify or refute the observation.
a) Accordingly, provide core flow maps (in the famat of Figures C-1 through C-37) for all spray distribution tests perfomed to date. Also provide an index which indicates the groups of tests (i.e., two or more) that have comparable steam flow, spray flow, and temperature conditions and which involve one sparger and then both spargers operating together (indicate the series run the morning of July 5,1979). provide a test matrix for the addi-tional figures containing the information on page 25 of your report. The only such series now in your report are Figures C-1 C-3 and C-5 (for upper, lower, and both
- pargers at 29.5 psia, no core steam flow, and 145"F), and Figures C-2, C-4, C-6 which apparently represent a repeat of those tests. For position x=6, y-6, both of those series show less flow for two sparger operation than for either sparger operating alone, b) Quantitatively comment on the apparent failure of the super-position principle for one and two sparger operation.
. (S) With respect to the steam flow and flow paths used during the Lynn tests, you state that the maximum steam flow that can be provided to the fuel bundles without "significant" droplet di-version is around 20,000 !/hr. We note, however, that such' a steam flow rate corresponds to a decay heat removal (due to vaporization) of only about 2% of full core power. Credit is taken for core spray cooling under conditions when decay heat is at least 5% and/or significant depressurization (flashing) is still occurring. We are therefore concerned that the steam flow regime investigated at Lynn is not representative of postulated post-LOCA conditions. Accordingly, please justify the acceptability of the steam flow and steam flow paths in the Lynn facility regarding ability of that facility to carservatively represent predicted post-LOCA conditions, including but not necessarily limited to the following:
a) Qualitatively explain why the " balanced steam flow" condition (equal steam flows through " fuel" bundle and upper plenum / steam separators) represents the worst case for methodology confirmation, b) Quantitatively, for 29.5 psi pressure and 145'F spray flow (or similar conditions where you obtained extensive data) provide a plot of ni.nimum spray flow measured in any channel 64>27") as a function of core steam flow. On the same plot, indicate any data points you obtained with different steam flow splits (i.e.,
not " balanced" between fuel and separators). Indicate the range of core steam flow expected during times when credit is taken for core spray cooling, including all sizes and locations of postulated LOCAs.
. c) Explain the basis of the statement on pg.15 that "The facility design with no vapor injection into the bypass region is judged acceptable."
d) Justify the conservatism of the steam flow distribution selected for the tests, since other parallel channel steam flow rates (not necessarily even in the same direction) re-sult in the same AP and may yield a lower minimum single channel spray flow.
(7) State the 1.) cation of the " steam vent" (pg. 21) and describe how flow out the steam vent was controlled and/or measured.
(8) On page 34 Figure V-5, centerline flows at Lynn vs HSF are compared. However, other parameters such as nozzle spread (transverse to flow direction), azimuthal asymetry, etc may also be significant and may be different at Lynn comapred to the HSF.
Provide data to show that the other parameters were not signi-ficantly different.
(9) Explain the effects of running the calibration tests (pg 25) at a spray temperature different from the spray distribution tests (185'F vs 145'F resp.).
(10) Provide the vertical axis' scale for Fig. VIII-3 pg. 52.
't
. (11) Justify conservatism of the 30' sector minimum spray flow result's compared to the full 360' geometry in view of Figure IV-2, which shows considerably more flow to some bundles for a 27' sector compared to a full 360* geometry (also see question f(12)-9) below).
(12) Answers to severai questions in our Feb. 3,1978 letter to GE were deferred until Lynn results became available. These de-ferred questions are listed below with their original number, and should be answered now that Lynn results are available.
- 2) (1/19/78) Provide copies of the references cited by Dr. Sandoz at the 1/19/78 meeting regarding size of the steam condensing agion sur-rounding a nozzle. Describe why GE believes that this data is appro-priate for acclication to a BWR spray system (e.g., that the geccetry, spray flow rates, subcooling, and steam pressures are similar in the referenced tests and in BWR's following a postulated LOCA). Please include pictures of typical BWR single nozzle spray patterns in steam.
5)
(1/19/78) We have heard several presentations regarding test programs to be accomplished at the Lynn, Massachusetts full scale 30 -sector 0
steam test facility. Each presentation has emphasized investigation of either core spray (CS) distribution or counter-current-flow-limiting (CCFL) phenomena. In reality, the two are closely coupled.
Please provide a written description regarding how the facility will be utilized to investigate the closely coupled relationship of CS and CCFL phenomena.
6)
(1/19/78) Quantify the expected effects of the smaller amount of steam condensation that is excected to occur in the " hydrodynamic" region. Why does GE expect that this condensation will not invalidate the " separability" hypothesis? (The January 19 meeting disclosed that approximately 25:; of the total concensation is expected in this region.)
7)
(1/19/78) What air updraft velocities will be utilized in future Vallecitos air-water full scale tests to simulate steam velocities in the post-LOCA environment? Justify the conservatism of the simulation, including magnitude and direction of the air flow with respect to pre-dicted steam magnitude and direction following a LOCA.
~
. 9) (12/15/77) Provide dJcumentation regarding why GE believes steam and water flow patterns in the Lynn 30' test facility will adequately represent the flow patterns that might be present in a full 360'.2 actor upper plenum following a LOCA. Include dis-cussion of tests both with and without the " pie-shaped baffle" in place. (Answer to this question should be included in or coordinated with your answer to question 11 above.)
- 11) (12/15/77) P ovide the "CCFL delay vs. zero spray coefficient" tradeoff results (discussed in slides SCR-5 through SCR-8 (12/15/77)) for the sizes and t pes of, jet-pump SWR plants whose results were not presented j
at the 12/15/77 meeting, and for the second most limiting break location for "LPCI-Modified" BWR's.
- 12) (Previous numoer 1-C, 9/2/77) The proposed tests do not include possible effects due to the different steam qualities that might be present under various conditions. Water droplets entrained in the steam may change the interaction of the steam and the spray cone. Describe how GE plans to quantify such possible effects experimentally and/or analytically.
- 17) (Previous number 11,9/2/77) Justify your assumption that one-half of the " Appendix K" quoted core spray heat transfer coefficients can be used when spray flow to a bundle is below minimum design flow. You should provide results of experimental spray heat transfer coefficient measurements taken at lower spray flows. Also. you should quantitatively demonstrate that actual penetration of the assumed (lower) flow into the bundle is consistent with your CCFL data and correlations, under all conditions p edicted by your ECCS calculations where this assumption of lower heat transfer coefficients is made.
ATTACHMENT 3 Categorization of Staff Questions Methodology Aeolication lal x
b x
0 x
2a x
b x
C x
3a x
b x
4 x
Sal x
b x
6a x
b x
x c
x d
X 71 x
8 x
1 9
x 10 x
11 x
12-2 x
6 x
7 x
9 x
x 11 x
12 x
17 x
I - These questions were satisfactorily answered at the November 15, 1979 meeting.