ML20127M235
| ML20127M235 | |
| Person / Time | |
|---|---|
| Issue date: | 02/22/1985 |
| From: | Catton I Advisory Committee on Reactor Safeguards |
| To: | Boehnert P Advisory Committee on Reactor Safeguards |
| References | |
| REF-GTECI-A-03, REF-GTECI-A-04, REF-GTECI-A-05, REF-GTECI-A-47, REF-GTECI-SG, REF-GTECI-SY, TASK-A-03, TASK-A-04, TASK-A-05, TASK-A-3, TASK-A-4, TASK-A-47, TASK-A-5, TASK-OR ACRS-CT-1801, NUDOCS 8505230216 | |
| Download: ML20127M235 (3) | |
Text
.
ZF-
~
13DR 06\\3% $
1
_ gy
..
- l-
, --o =y;
- ' ]
..s t e
,,e,~e-
, e,
"..
- 9. *t*
4-4g, 1/
February 22, 1985' TO: paul Boehnert FROM: Ivan Catton f
-s SUBJECTtThe AEOD Transient and SECY 82-256 My understanding of the AEOD concerns that lead to SECY 82-296 I find myself in were based on AEOD reports describing the transient.
agreement with their perceptions of its seriousness having been at TMI l
j shortly after the acxcident and listening to discussions of whether or to flood the steam lines. I believe this was the genesis of Carl not Michelson's concern as well. The AEOD rteports clearly define the transient as do some of the A-47 reports. One can argue that the concern L
is without basis by showing it has a very low probability of occuring.
Of course the AND-1 incident may make this difficult. This particular issue seenms to be one that ought to be considered by those responsible for Generic issues A-3, A-4, and A-5 as well as A-47. One can also argue that our calculational tools are sharp enough that we don't need experimental verification. I would make such arguments for some of the j
issues driving the MIST program but have difficulty with this one.
i Calculations of steam generator behavior under normal operating conditions is difficult. Under many accident conditions we must consider steam generator. heat transfer is not too important. For example the i
LBLOCA primary system tempera,ture and pregsure time histories are not preatly influenced by a poor steam generator'model. Slower transients iL nE are impacted significantly as certain LDFT experiments have shown.
- i. QS The cuestion nere is whether or not we can calculate blowdown of a full g
steam generator with sufficient accuracy to yield temperatures adequate o
for thermal stress analysis and determination of steam generator tube thermal stress. I do not accept Dr. Sheron's statement that ruptured
- ,o-W i
((-@. steam generator tubes are a given. Those who deal in pRA plac l
on whether or not they fail and somehow the number needs to have a basis.
Calculations have been made for full steam generator blowdown.
a For l
EEb I believe there are many caveats associated with the results.
example, how do separators and driers behave when flooded?
8' l
At the second TAG meeting Dr. Harold Sullivan noted that steam line/SGTR' tests were needed because heat transfer coefficients under i d suen circumstances were not known.
I agree with his observations.
0-There are a number of circumstances one must consider.
The steam This case generator may overfill because the MFW is not tripped.
0D OL leads to the bottom portion of the steam generator being subcooled Here the stratification is well defined
- g eA about 50 deg. K or so.
y 90 provicing the, codes can track an interface within the liquid phase t
G
- which I don' t think they can.
Some minor changes could alleviate In this case professor Theofanus is probably correct this-problem.
- l%
N Further as long as the steam generator is cooling the primary system.
N
[
my preliminary estimates of water hammer show that it probably will 3
l not occur.
f A MFW driven overfill will lead to a stratified steam generator.
8 Heat transfer to a stratified fluid -is difficult to estimate unless it 1
~.z.
.g. _.
P
- (j The heat
.is. boiling and one knows the local value of the subcooling.
transfer from the primary system may depend on the amount of I
stratification.
I know of no analysis or experiment that addresses 4
[
.this.
The primary system pressure, as I noted above, is very sensitive to. heat transfer to the seconda, y side.
If such analysis or other r
I substantive information exists to show that this is not a concern, 4
would like to see it.
A more complex thermal hydraulic problem arises if the AFW is the cause of overfill.
At the top of the SG the AFW will run down a j
few tubes near the injectors.
The tube sheets will cause a certain This will occur amount of spreading and mixing with upwelling steam.
at each tube sheet.
If,however, the initiating event is a turbine trip, the SG pressure is usually raised 100 psi or so and.there will be some collapse of voids.
The AFW will rundown the nearby tubes until l
reaching a tube sheet or saturated water.
One could carry on from here describing the various mixing, or non mixing, processes that might l
-occur.
It is my view, however, that we do not know what the conditions are when and if we must deal uith a SLB.
h The AFW may penetrate all the way to the bottom forming a lake
[
i l
of cold water while displacing the hot fluid.
Under these circumstances one would not expect a water hammer to cause a SLB.
On the other hand, i
simple calculations indicate that the AFW may be saturated by the time it reaches the first tube sheet.
For the latter, we must concern ourselves with what the temperature of the water is when it spills over the inner shall of the steam generator.
The AFW is at 70 deg F or so i
and is mixing with water at 400 deg F or so.
If 100 deg F subcooling water hammer will occur and a full steam generator with
- results, varying amounts of subcooling will blowdown.
To place probability numbers on SGTR, one must address the The AFW driven multitude of possible circumstances described above.
overfill is the more probable for a B & W reactor and also the most 7
difficult to analy:e. Before one can decide whether or not Dr. Sheron, Jones or Prof Theofanus are correct in assuming that the experiment Mr.
planned to address the question is adequate and SECY-82-296 is a clear statement from those doing the calculations is satisfied, needed.
Dr. Sheron's aopeal to A-47 is inappropriate.
A-47 addresses instrumentation and control.
Although the question of multiple SGTR failure has arisen from A-47 considerations, the thermal stresses leading to their failure is a thermal hydraulic question requiring use l
of the advanced codes. The question remains as to whether or not the calculations can be made with tolerable uncertainty.
Those responsible for A-47 do not have the computing horsepower nor the proper talent to do the job.
i one must Addressing the physical processes more directly, consider multiple choking at the tube sheets as well as within the The blowdown initiates from a state that is separators and driers.
iphomogeneous in phase within the two phase region and stratified The effective stagnation state for the l
within the subcooled regions.
i creakflow calculations is not very well defined.
L
. -. ~
gy.
pp y.
.. g.-:.
sq
~
To summarize, I believe the. steam generator overfill induced i
combined LOCA described.in SECY 82-296 is well defined.
If there are reasons or arguments countering my above discussions, I am unaware of them.
I don' t believe the AEOD transient can be written off out of hand.
Mr. Jones recitation of a number of meetings and menos seemed to be more of a description of how compromise was arrived at than technical.
They may well be sound but I would need to see them before accepting them.
professor Theofanus' comment that "we have done enough blowdown calculations" may be true but, it my mind, it is premature.
Dr. Sneron's shifting the question to A-47 is not warranted.
I The modifications to MIST needed to perform what I believe the AEDD transient to be are minimal.
Some modifications of the geometry of the upper SG internals would be needed.
Additionally one could blowdown to the atmosph'ere.
Before pursuing such a course, however, I would like to hear from the TRAC code developers.
They should be l
asked to do calculations for the planned test and for the AEOD I
The overfill case should be that resulting from MFW and AFW.
l transient.
A discussion of the results and their confidence in them would settle least put out where we can discuss it.
A part of the the issue or at discussion should include how well they know fluid subcooling and flow at the SL entrance ( water hammer potential) and SG material temperatures (SGTR potential).
f I
I l
{
l l
9 I
- l s
O