ML20031F166
| ML20031F166 | |
| Person / Time | |
|---|---|
| Site: | Sequoyah  |
| Issue date: | 08/17/1981 |
| From: | Strehlow R STREHLOW, R.A. |
| To: | |
| Shared Package | |
| ML20031F167 | List: |
| References | |
| NUDOCS 8110190291 | |
| Download: ML20031F166 (5) | |
Text
- .
' ' s,d ',-
Attachment
'^"I
' E P4: 55 Evaluation of 4
.-g QUARTERLY PROGRESS-REPORT #3
- .: SEC.
Research Program on Hydrogen Combustion'and Control Dated June 16, 1981 TVA Sequoyah Nuclear Plant and the July 23, 1981 Progress Report Meeting Prepared for Mr. James Milhoan, P.E.
Office of Policy Evaluation Nuclear Regulatory Commission Washington, D.C.
20555 Prepared by Roger A.
Strehlow Consultant August 17, 1981 i
B I IdI962R LXA
I 1
SCOPE OF THIS REPORT This report will discuss all aspects of the third quarterly progress report and the presentations at the progress report meeting of July 23, 1981.
I RESEARCH TOPICS Completed Research There are four research studies that appear to be completed and satisfactory for evaluation by the st'aff.
These are:
- 1. Halon research (Atlantic Research Corporation) 2.
Spark Ignition Electromagnetic Interference (Keiser)
- 3. Combustion Tests (Fenwal)
- 4. Catalytic Combustor Tests (Acurex)'
Research That is Underway
- 1. Igniter development (EPRI Whiteshell)
This is an appropriate test of some alternate thermal type igniters in various H -steam-air mixtures.
It should pro-2 vide useful information when it is completed.
- 2. Lower Flammability Limit and Extent of Reaction (White-Shell)
This is a good test and should provide useful information.
After discussion sith the investigators I have no problem with the use of spark ignition in any of these tests.
This is because it was explained to me that a strong spark will indeed ignite these weak mixtures and they are interested not in the ignition problem but in flame propagation away from the source and the extent of the contents that are involved in combustion.
The m
.I a
2 spark just makes it easier for them to instrument the. experiment.
This is because glow plugs exhibit a large variation in the delay from their activation to ignition.
- 3. Laminar Spherical Deflagrations (Whiteshell)
As I said at the meeting I can see no real need for these tests.
There is sufficient information in the literature such that one could predict the pressure time traces from these experiments - at least for Air-Hydrogen mixtures.
Also, the addition of steam could be ~~l led vwole ' as the addition of carbon dioxide and there is data in the literature on the burning velocity of Air-Hydrogen-carbon dioxide mixtures.
- 4. Effect of turbulence and structures on Spherical defla-grations (%hiteshell)
As I said at the meeting the results of these tests will stand by themselves and not be extrapolated to any other _ con-figuration, Thus the information gathered in this series of tests will tell you nothing about, for example, the effect of an ice condenser section on flame acceleration.
- 5. Sphere and connected pipe (Whiteshell)
My comments here are identical to those in item 4 above.
I can see no useful extrapolations-of this information.
- 6. Hydrogen control studies (Acurex/ Factory Mutual)
I feel that this set of tests, both the small scale preliminary tests at Factory Mutual and the large scale ones at Acurex, are very worthwhile.
Furthermore the test procedures and test matrix are well thought out.
These tests should supply
3 9
useful information relative to the eficacy of fogging a hydrogen-A air mixture to prevent combustion.
- 7. Distribution / mixing (EPRI-Hanford)
I am not sure of the necessity of this ' test since it i
only models the lower compartment.
The real,que'stions concern the behavior of the flame processes in the ice,pondenser.
- 8. Igniter Endurance (TVA-Singleton)
This is a well thought out set of tests.
Not only will
/'
it check the igniter endurance in the presence of hydrogen but
'it will also show how good a hydrogen scavenger the plugs are.
OVERALL COMMENTS I still feel that the overall questioh of flame acceleration in an ice condenser configuration has not been addressed and I v
am still of the opinion that the configuration of the ice con-denser plants is potentially very dangerous if a continuous hydrogen burn were to occur (a'la Three Mile Island) I am.very strongly in favor of the glow plug concept because they cause ignition at low hydrogen concentrations and the weak flames thus produced could never accelerate to produce damaging over-pressures or detonations.
This means that the flame / acceleration questions is probably moot.
Nevertheless, I want to make it clear to everyone that, in my opinion, the proposed turbulence, obstacle and pipe experiments (items #4 and 5 above) will answer no questions concerning the behavior of<the flame when it passes through an ice condenser.
4 4
As I said at the meeting, one would have to. construct a section of the ice condenser using full scale baskets and supports in the vertical position to determine how it would react to either an upward.or downward propagating flame.
Even through this wculd be a very expensive single test it is the only way that I know that the question of flame-ice condenser interaction-can be answered.
M t
t e