ML20010J584
| ML20010J584 | |
| Person / Time | |
|---|---|
| Issue date: | 08/21/1981 |
| From: | Catton I Advisory Committee on Reactor Safeguards |
| To: | Boehert P Advisory Committee on Reactor Safeguards |
| References | |
| ACRS-CT-1362, NUDOCS 8110050472 | |
| Download: ML20010J584 (1) | |
Text
.gg YO g k p )i.os*Ancet.ts: scnoot.or rucia 21 August 1981 AND APPLIED SCIENCE hf
( T Ob V 2
/T0:
Paul Boehnert g-
- s. (.r i
g ' d- %!
PDR 9)M id l
FROM: Ivan Catton 4
d' Ci 4
,A
SUBJECT:
The Rensselear Polytechnic Insti reon L of '-- BWR Simulation The RPI facility is a small scale simulation of a BWR for study of the reflood phase of a LOCA using freon as a coolant. The vessel is pyrex and contains three simulated fuel bundles. Each bundle is an annular channel properly orificed at top and bottom. The inner cylinder is a heater and the outer a pyrex pipe. Its design pressure of 30 psig is equivalent to a BWR pressure of 300 psig. The lower plenum is distorted and the jet pump simulated very simplistically. Vapor generation occuring in the lower plenum during a BWR LOCA is modeled by predetermined freon vapor flow into the lower plenum.
,A typical test is conducted by flowing freon into the system through sprays on top of the coil and freon vapor into the lower plenum.
Various flow rates of both liquid, vapor and liquid subcooling were investigated.
The results obtained are interesting. The spray flow immediately bypasses the heated channels and begins t, fill up the lower plenum. The vapor flow from the lower plenum reaches the upper plenum through the heatei channels (fuel bundles). Coolant does not enter channels from above.
The lower plenum fills until it blocks off the jet pump. Vapor must now all flow through tha simulated fuel bundles.
Coolant is entrained with the vapor and the resulting two phase aixture has a high enough pressure drop that coolant is forced through the jet pump and out the break.
Even through there is no reflooding in the usual sense, the two phase mixture does a good job of cooling because it has rather high liquid content.
Breakdown evera ally occurs at the top of the heated channel and coolant enters. Under some circumstances breakdown of the hot channel never occurced.
Visual observation of the various thermal hydraulic processes is very valuable..
One can obtain a qualitative understanding ve y quickly, a
Freon can be a useful. tool in scaling certain thermal hydraulic pro-cesses'because they allow a test to be conducted at a much lower oressure than if water is used.
If condensatica as well as evaporation is important then results based on freon can be misleadina.
Freon's low thermal con-ductivity decreases condensation rates almost to zero relative to water.
As a result the BWR simulation with freon does not include subcooled spray condensing lower plenum stream. The vapor flow rate to the lower plenuu
(
is based on calculations made by GE. The vapor must flow out of the lower plenum and can only flow up through the heated channels following sealing off the jet pump. With a predetermined vapor supply rate to the lower plenum and very little condensation, the RPI results may oe atypical.
Further CCFL at the channel upper orifices could be influenced by subcooling if condensation can occur.
The thermal hydraulic processes observed in the RPI facility are not the same as those described by the General Electric evaluation model.
l Modeling a BWR with a freon system, however, may not properly scale the various thermal hydraulic processes and lead to the observed differences.
On the other hand, the observed phenomena a~re sufficiently different from the GE Efi predictions that an explanation should be requested.
In my opinion, i.his is not a safety question even if the results are e.orrect. The results should be oatt,of considerations that might lead to changes in the BWR EM codes.
n110050472 8108L PDR ACRS 4 CT-1362 P D,R.
4