ML19281A891
| ML19281A891 | |
| Person / Time | |
|---|---|
| Site: | Dresden, Nine Mile Point, Oyster Creek |
| Issue date: | 02/02/1979 |
| From: | Catton I CALIFORNIA, UNIV. OF, LOS ANGELES, CA |
| To: | Bates A Advisory Committee on Reactor Safeguards |
| References | |
| ACRS-CT-1088, NUDOCS 7904160099 | |
| Download: ML19281A891 (4) | |
Text
e.
grACTOR S ArFC'l*CS 65. NRC Boelter Hall 2567 FEB 121979 university of California (q@
Los Angeles, California 90024 D lll2 ll2 b 4r5r3 February 2, 1979 hal i l t
1 i To:
Dr. A. Bates From:
Ivan Catton
(
Subject:
Mark I Owners Group Meeting, 22-24 January 1979, San Jose, California
[b 4 J[p d f Copy To: Professor M. S. Plesset General Observations. The Mark I program has been underway for over three years and there are still surprises. The use of orificing for scaling and validity of the results at small scale is in question as wel1 as LLL
~
tests showing that loads derived from 3D tests are greater than loads derived from 2D tests. Chugging loads seem to be small as long as the operator activates the ADS soon enough.
It was not clear that all plants require that he do so. The coupling between the S/RV quencher and the suppression pool ullage is not understood and as a result the stuck open S/RV or ATWS cannot be completely addressed.
Scheduling. The Mark I program is nearing completion and closure is planned to occur at the end of May. Between now and the end of May three meetings between the NRC staff and the owners group will take place (the first is planned for 15 February, the second for 8 March, and the third and final for 15-16 May). At the final meeting, the draft NRC position will be presented.
It is anticipated that an NRC position letter will go to the utilities on 31 May.
Following closure of the program, the various utilities will be asked to make plant unique studies and to schedule and complete whatever plant modifications are necessary. NRC would like this to be accomplished before December 1980.
Some utilities with two plants feel they will not be able to complete the required modifications before 1981.
~
I 7904.16COf
They will be given an opportunity to argue for an extension and if their arguments are not satisfactory, will be asked to shutdown.
Orificing. There is still a great deal of ignorance about the effects of orificing to achieve similitude with small scale experimental facilities.
Recent EPRI results at 1/11.7 scale show a 44% increase in upload pressure when the oridice is moved from the downcomer to the vent. The impact load is increased by downcomer orifices because the pool surface remains flatter than in their absence. A great deal of discussion about how one should orifice followed hearing about the EPRI experiments.
EPRI recomended use of a split orifice (50% at vent and 50%, at the downcomer) as a compromise.
GE will try to justify that the split orifice pool surface behavior data is applicable to all plants.
Three Dimensional Effects. The LLL 1/5 scale tests show that upload pressure for the 90 sector (3D) were about 15% higher than for the 71/2o sector (2D). The GE 1/4 scale results are for a 7 1/2o sector and even though already higher than the LLL 71/2 sector results, the question of why a 15% penalty should not be imposed was raised. No explanation for the differences between the GE and LLL experimental results was obtained at the meeting. The NRC staff wants an explanation before they will consider not penalizing 2D results. Vent header deflectors may be needed if the penalty is imposed and some plants (Browns Ferry) may have difficult redesign and i
modification problems.
Short Downcomers. Owners group members would like to shorten the down-comers to decrease the submergence and to obtain relief from the high air clearing loads. The problem becomes one of the steaming phase and the effects of themal stratification. The NORCO full scale (22 1/2 sector) experiments were run with subinergences as low as l'6" to demonstrate that 2
2 tMs was a viable solution to their problem. The assumption is made that the operator activates the ADS within ten minutes.
Even though significant stratification has taken place, steam bubble collapse outside the downcomers has not yet occurred. As long as this is the case, the violent chugging and large lateral loads on the downcomers will not occur because water flowing out of the break will be subcooled. How quickly the operator activates the ADS will be a plant unique question. The NORCO test report will be available in a month or two.
Fluid Structural Interaction. The forcing function needed to carry out a structural analysis was shown to be uncoupled from the structure.
This simplifies the problem a great deal. Rigid wall data can be used and a series of difficult experiments will most likely not be necessary. A report to be released in the near future will contain the details of this study. The rigid wall pressure distributions will be supplied to the utilities for their plant unique analysis. The rigid wall pressure distri-butions will be for both the air clearing phase and the steam chugging phase. The chugging phase information will include frequency data.
Safety Relief Valve T-Quenchers. The S/RV T-quencher program is sumarized as follows:
a) develop T-quencher geometry, b) carry out full scale test at Monticello (single and multiple value actuation) to obtain baseline data for modeling, c) carry out 1/4 scale experiments at NUS facility to produce para-metric variations for model confirmation d) develop a predictive model (QBUBBS)
The model is supposed to contain all elements of quenc!;er behavior from air clearing to full steam flow. Details of model will be presented at the 3
...s
, IS~ February review meeting. Certain aspects.seem weak but with proper reliance on Monticello data and 1/4 scale data for trends may be adequate for load determination. Again, proper assessment of the QBUBBS model must wait for its documentation.
A separate aspect of the quencher performance is its behavior at the high temperatures resulting from a stuck open S/RV or from ATWS.
Both KWU and CNEN tests show that the quencher operates as long as one or two degrees of subcooling are maintained. Thernal stratification characteris-tics of the pool become important and have not yet been properly assessed.
The amount of subcooling will be a function of the ullage pressure which, in turn, is dependent on the RHR induced circulation and buoyant plume behavior above the quencher. The Monticello tests were only for durations of ten minutes or so and probably not long enough to determine the quencher to ullage pressure relationship.
4
.