ML20207A272
| ML20207A272 | |
| Person / Time | |
|---|---|
| Issue date: | 08/13/1985 |
| From: | Kelber C NRC OFFICE OF NUCLEAR REGULATORY RESEARCH (RES) |
| To: | Telford J NRC OFFICE OF NUCLEAR REGULATORY RESEARCH (RES) |
| Shared Package | |
| ML20207A169 | List:
|
| References | |
| FOIA-86-678 NUDOCS 8611100161 | |
| Download: ML20207A272 (3) | |
Text
% UNITED STATES j
['
g
., . 'f gg
. E NUCLEAR REGULATORY COMMISSION WASHINGTON, D. C. 20555 os c
- x. . . . '
%-Wade 78 August 13, 1985 MEMORANDUM FOR: John Telford, DAE FROM: Charles Kelber, RES
SUBJECT:
ESTIMATES FOR REVIEW GROUP I have analysed the problen set forth in your meno to distribution dated August 9, 1985, as follows:
- 1. The noiten core nass, ranging fron 20 to 50 tonnes, falls into water of depth 100 tc 200 cm., in the form of a right circular cylinder of diameter ranging fron 15 to 450 cm. The acceleration, g, of the cylinder is 1000 cr/s2
- 2. The energy density of the reiten core mass is 1.2 108 J/ gram.
- 3. Mixing progresses fron the base of the cylindrical plug towards the top, ar.c proceeds until the plug has penetrated the entire depth of the water, whereupon ar explosior takes placc.
- 4. The plug is nixed with the water to length L= 0.07 (9/11) 2 d , where 2 d=gt2 and 9 is the difference of liquid densities, 11 their sum, and the coefficient of 0.07 is cited by D. L. Youngs . Of course, if L exceeds the length of the plug, the entire plug is mixed.
- 5. The energy content is converted into kinetic energy, K.E., by the conversion ratio of 0.15, characteristic of a confined explosion. This factor is based on the admittedly tentative analysis of Krein, et al.
D. L. Youngs, " Numerical Simulation of Turbulent Mixing by R-T Instability,"
Physica 12 0 (1984) 32-44 (North-Holland, Amsterdam) 8611100161 861030 PDR FOIA 1 CURRAN 86-678 PDR
/G/ .
l RESULTS I
For 50 tonne mass falling as a plug of diameter 450 cm., I estimate a mixed mass of 36.6 tennes, yielding a K.E. of 6.6 GJ. Even with a reduction by a factor of four from " SIMMER" effects, there is adequate anergy to blow a nissile through the top head.
For a 50 tonne mass falling as a plug of diameter 15 cm., on the other hand, the amount of mass mixed is 40 kg. and the K.E. is about 7.2 MJ. Moreover, the thin structure of the stream indicates a lack of effectiveness of confinement, so that a conversion efficiency less than 0.15 might be expected. If the fall were modeled as a rumber of thin streams, more mixing might be expected, and greater yield, but I don't know how to model the interaction.
Other sets of parameters fall between these limits.
CAVEATS Youngt, work applies to initially plane surfaces. The correlation is based on histly limitec data. The theory clearly predicts a degree of mixing prcportional to acceleration, so that accelerations caused by pressures from rearby explcsions, or from high pressure ejection will cause more mixing to take place. Although Youngs irdicates that a 3-D computer simulation is prcbably infeasible even using a CRAY conputer, he does present 2 D simulations of breakup of jets. The penetration distance is arbitrarily defined and may not correspond to the pre-nixing stage in a clear-cut way.
The flow through the complex structure of bottom grid plates, flow corrector plates, core support plates etc, may affect the flow characteristics of the mass of molten core significantly. (There apparently is about 15 tonnes of steel in tnese structures. The PLUGH code should adequately model the fic+ 6 through the holes in the steel. If the heat content of the core is as advertised one probably should expect that whatever comes through hasn't lost much comentum te what has been frozen on the ablating walls.)
The effects of gas bubbles retained in the melt have been neglected, but they may dorainate explosivity, though not mixing except through the bulk density.
2 i
l Conclusions
- 1. Experiments with transparent simulant fluids with a range of density ratics near 10 should be used to derive a correlation similar to Youngs's. The fluids could be transparent to light, microwave, or x-radiation.
- 2. The correlatier. should be extended to include the case of a thin jet subjected to an impinging pressure pulse, and to the case of an immersed sphere subjected to such a pulse. In this way one could seek a model of mixing induced by a nearby explosion, as suggested by Anderson. An extension to a l
number of parallel, thin jets is desirable.
l l 3. The resulting correlation should be used in a 1- or 2-D hydrocode simulation (SIMMER) tc determine the need for further tests, and indicate their nature.
1 .-
hhNO,'
.t G Charles Kelber cc: R. E. liinogue D. F. Ross
- 0. E. Bassett 3
-_. _ _ - _ . . __ . . _ _ _ _ _