ML20027A879
| ML20027A879 | |
| Person / Time | |
|---|---|
| Issue date: | 09/20/1982 |
| From: | Catton I Advisory Committee on Reactor Safeguards |
| To: | Igne I Advisory Committee on Reactor Safeguards |
| References | |
| REF-GTECI-A-49, REF-GTECI-RV, TASK-A-49, TASK-OR ACRS-CT-1504, NUDOCS 8208290015 | |
| Download: ML20027A879 (2) | |
Text
__
C7- /fd V
?OA?'
RECE,,.
MEMORANDUM f
U R C.
20 Sept 1982 SEP 2 71982 PM 8,9,1011Slih3,4,5 6 i
T0:
I. Igne i
f USNRC - ACRS Mail Stop H-1016 U).
bM FROM:
Ivan Catton EPRI/SAI Experimental Studies of Downcomer Mixing for PTS SUBJ:
Assessment, Sunnyvale, 27 August 1982 EPRI is conducting a series of experiments at SAI Sunnyvale to The experimental investigate cold leg and downcomer SI mixing processes.
apparatus is a 20 foot channel that is 30 inches high by 6 inches wide (a vertical slice through the cold leg) connected to a 10 foot high x The apparatus is full scale in 8 foot wide box that is 10" in thickness.
A test is the vertical direction but very distorted in the horizontal.
The run by filling the system with hot water and injecting cold water.
injector is a 3 inch diameter pipe located 16 feet from the box that represents the downcomer.
Injection is from the top of the channel.
The injection is to simulate HPI. The experiment we observed The involved injecting 64*F water, dyed red, into stagnant 145 F water.A factor 80 F temperature difference yielded a density variation of 1.7%.
of ten higher is needed to reach values expected under the more severe PTS incident.
Even though the scaling is highly skewed, the basic phenomena are The simulated HPI coolant mixed with the what one would expect in a LWR.
hotter fluid near the injector. The mixing process is one of jet mixing The mixed SI coolant and stagnant as costulated by CE in their model.
fluid, being more dense, flows along the bottom of the channel representing the cold leg with almost no further mixing. The flow in a PWR cold leg will be even more stable as it will have a much larger density difference.
The cold mixture enters the downcomer and falls because of the negative Mixing takes place as the colder fluid mixes with the warm buoyancy.
Hot fluid that is displaced flows into the top of water in the downcomer.
the channel. The temperatures remain steady until the hot fluid is Once this occurs there is a general cooling pushed out of the downcomer.Even though one cannot use the test results directly, of the whole system.
being able to observe the various processes was a valuable experience.
The fluid mechanics and heat transfer appear to be governed by jet mixing and entrainment at the SI inlet and by a negatively buoyant plume in the downcomer. These processes are now being modeled by Theofanus [Purdue University], Levy [S. Levy, Inc.], and Rothe [Creare).
The key to successful modeling will be determining the proper entrainment The work at Purdue, EPRI and correlations for the two mixing processes.In the meantime, reasonable estimates Creare should yield these parameters.
)
are possible based on our knowledge of similar mixing problems.
8208290015 u
Catton to Igne 20 Sept 1982 The two primary parameters, aside from geometric considerations, are SI momentum flux and SI to primary system density differences. None of the experiments conducted to date have matched these parameters.
Typically an attempt is made to match Froude number.
If hot and cold water are used, then the velocity, or momentum flux, must be much lower than desired. This decreases the SI jet mixing and will lead to conservative results. One can then conclude that the mixing observed by Creare and by SAI will be less than expected. This must be balanced by the density differences being too small which will decrease the mixing.
Presently, it is my best estimate that the initial mixing will be dominated by momentum processes.
To obtain more realistic density differences salted water will be used as an injectant. Presently when this is done, both EPRI and Creare cool the injectant relative to the primary fluid and monitor temperature. This method is not appropriate as mixing of cold salted water with fresh water involves two different processes. One must measure salt concentration if salt is what makes one part of the fluid more dense than another.
In my view, a great deal of progress is being made on the thermal-hydraulic side of the PTS question. The learning curve has been steep and reasonable estimates can now be made.
Sound experimentally based predictions of primary system temperature in the downcomer will be possible within the year.
.....