Memorandum from Pelle Isberg to The Reactor Group
Typescript
3 pp.
May 9, 1953
MurNBsamplingsystem050953


[page 1]

Reac. Memos

Raleigh May 9, 1953

To: The reactor group.
From: Pelle Isberg.

It has been stated that the sampling system should be able to do the following things.

The basic components of a system capable of doing these things are:

Arrangement.
2 and 3 will be hot from gaseous fission products, and should therefore be located in
the concrete box on top of the reactor shield. As 4 must be directly accessible for the
reading of the manometers, it has to be put outside the reactor. Note, however, that C)
cannot be performed if the airfilter has this position between the manometer and the
reactor.

Materials and some details.
1 and 2 should be made of stainless steel and electric level indicators provided. The
arrangement of the insulators for these indicators seems to be a major problem because
of the space they need.

A satisfactory airfilter has to be found.

The requirements on the valves that are to handle the soup are as follows. They must be
fastclosing stainless steel valves without gaskets and without spaces from which the soup
cannot be easily washed away. If suitable valves cannot be found, I suggest that a
valveless system be developed. Figure 1 shows one principle for such an arrangement.
Filling : D open. Suck gently on E. When the level reaches F, cut off the pump and let
air in slowly through E. Emptying : D closed. Apply a low pressure on E. Note that all
tubes between G and the trap must be short and wide to prevent pressure gradients.


[page 2]

Emptying of the reactor.
A fast and safe system for the em/ptying of the reactor can be made along the same lines
as figure 1 and is scetched in figure 2. All the soup can be brought up into three or
more containers with boron between them. tThe whole system can be placed in the concrete
box.

Note on the off-gas system.
The behaviour of the air volume on top of the soup and in the recombiner is essential
for the sampling system. Before the latter can be finally designed, one has to know
how fast air can be pumped out of this volume and how the pressure in there will vary
under these and other conditions.

Pelle Isberg