Subject: Report of
of reactor design with them. They were all very cooperative and anxious to aid
us in every way possible.
The main points in these discussions from my point of view are listed
The control and safety rods at
inches/minute, enabling a skilled operator to bring the reactor to full power
of 25 KW in about three minutes. Safety controls are not associated with the
rod removal rate but with the more realistic concept of power rise rate. At
This is necessary since the temperature of the reactor, the power region
in which you are working, and the flow rate of the cooling water all effect the
rate at which the power rises and not the control rod alone. This point was
illustrated in the region of 1 KW power for a cold reactor. In this case the
rod removal rate was about 3.5 inches/minute because of the practically constant
temperature of the reactor (water flowing at a rate such that a very small
differential in temperature would dissipate the 1 KW) and therefore practically
all of the control was in the rod. At higher powers the temperature increases more
rapidly and the rod must be removed much faster to allow the same power increase
rate. The condition for constant power rise is
where [delta]k is increase in k due to rod removal, T the temperature of reactor
soup, [sigma] the temperature coefficient and To the mean, cooling water temperature.
At 1 KW in the
remains about constant. Therefore,
slowly (3.5 inches/min). At higher powers
making it necessary to increase
the control rods, and both stated that rapid rod removal rate was desirable for
some experiments in which short time exposures were needed.
from a safety point of view nothing would be gained by removing the rods as slow
to five minutes. At a speed of 1"/min. the
least 20 minutes to reach full parer (removing both rods). Since the reactor
will be started and stopped several times a day, considerable unnecessary time
would be spent each day in bringing the reactor to full power. This will result
in increased operation costs.
The facts are as follows:
The power can be controlled automatically from about 50 milliwatts up
to 25 KW within a fluctuation of less than 1%. Control is better at low levels
than at high. The control rods are B10 rods worth about 80 grams each and can
be positioned and read to 0.018". The position to this accuracy is indicated
on the control panel.
The shim rods (two) are of 1/32" Cd 3" x 18" and are worth about 27 grams
only as an extra safety rod which could be used to compensate for various
was used for automatic control.) and is one of the points that should be cleared
up by additional correspondence as it may affect our design.
The reactor is allowed to rise in power with a period of five seconds.
Any faster power rise causes the rods to be automatically released.
method and the remaining gas is allowed to pass up the stack after a "hold-up"
of about one week. The system is shown schematically below with approximate
Dilution air, flowing at the rate of about 100 cc/minute, enters the
3/4" tube inside the 2" tube at A and is called cooled by cooling coils (inlet water
3.5°C) to about 13°C after first flushing the air space in the reactor. It has
been found advisable to use a baffel of about 2" to 3" in diameter to direct the
air over the volume without directly striking the "soup" surface. The cooling
coils condense the water vapor and take out some of the entrained materials.
The air then passes through a filter 3" in diameter and 17" long
containing stainless steel wool. No check has yet been made at
determine what has been collected in this.
This trap is connected through a 1/2" orifice to a rotor blower
capable of developing a pressure difference of about 8" of H2O.
At this high pressure point a small amount of the gas is bled off to
the stack to prevent diffusion in the position opposite direction. There were differences
of opinion regarding the advisability of placing the bleeder at this point.
From this point the gas goes to a pair of recombiners containing
several inches of platinized Al spheres. Each recombiner is 5" x 5" x 8". Near the
base (1/2") of these is a thermocouple and two others above, each at a higher
position. The temperature at the lower couple is about 470°C (at 25KW) and
somewhat lower for each of the others. A record is kept of these temperatures
and the criteria used is to replace the catalyst when the lower couple temperature
becomes loss than the middle. To date, after 4700 KW°hr of operation, no change
has been noted. It is not necessary to preheat the catalyst since the system
actually operates better when the catalyst is cold.
The H2O leaves the recombiners as steam and enters a condenser about 2"
by 11" containing about 20 feet of stainless steel tubing (about 3/4") through
which the cooling water (same as in reactor) is passing at 0.2 gal/min. This
condensed water flows by gravity back to the reactor through the inner 1/4"
tube (C). The fission products and dilution gases then pass through a pipe, the
dimensions of which are being sent to us, some 1000 feet long to the exhaust stack.
A valve and trap is provided at this point point D to drain off any necessary H2O to make
room for the addition of nitric acid.
About 120 to 150 cc/min of gas is thus exhaused at 25KW. About one week
is required for this to reach the stack. There is no indication of any activity
after reaching this point. A blower operated by a 3/4 hp motor exhausts the gas
by use of a venturi tube up the stack.
Tho dilution used at
operating at 25 KW. It would be well below the safe level at 10 KW.
All men were enthusiastic about the system which has been operated for
4700 KW°hr without trouble. It is recommended that this system be placed in our
unit either in the envelope or against the super-structure inside wall. Due to
the small size of the unit this would be feasible, enabling the eventual
on this is being prepared.
When the reactor is to be filled it will be necessary to first fill the
reactor with pure H2O and obtain a zero reading for the chambers. Then
This process is repeated with the necessary re-concentration of removed soup
until the reactor becomes critical.
It was emphasized by Dds.
(> 4) placed at different positions so that the true 1/C curve can be bracketed
(See Figure below.) and thereby determined.
It was suggested that B4C be obtained from
paraffin drained off. The remaining mixture has about the consistency of fresh
concrete and can be placed in a form for cooling. The form used for the curtain
process will have to be developed in our case since the boron is to be placed
against the concrete slabs forming the door to the thermal column.
The cooling coils consist of three 20 ft sections of 1/4" I.D. stainless
steel (347) having 1/32" wall thickness. The total flow through these tubes is
3.2 gal/min with a pressure of 60#/in² required to obtain this flow.
The water is cooled to 19°C at the inlet and at 25 KW the water exit
reaches a temperature of 71°C. At
85°C the control rods are dropped.
It is very important to avoid cracks in the graphite about the reactor.
This is due to the presence of fast neutrons. The standard graphite blocks are
held together by the use of Al pins. Detail plans for this stacking is to be
The exposure ports at
The rectangular ones had a 1/8" shoulder to prevent the radiation from having a
direct path to the outside of the reactor shield. The
lined with steel and plugged with Bi, Fe, graphite, Pb all sandwiched to produced
sufficient shielding. These sections were easily removed.
The "glory" hole was circular and had concentric cylinders with a
slight taper. The cylinders offered a variety of different sized exposure
ports as well as an excellent shield.
The details of exposure ports on the
are being reconsidered in view of the above new information.
The question was raised regarding the advisability of using the uranyl
sulphate. It was suggested that we contact
(1) In the thermal column Bi should be used instead of Pb. An increase in the flux
of a factor of ten is obtained by its use. This should be studied in some
(2) The bubbling effect has been measured but can be neglected in comparison with
temperature coefficient. They do not ever consider it at
(3) The temperature coefficient is 0.72 gm/degrees at low temperatures and 0.95 gm/°C
(4) Total uranium content is 970 gram, with about 100 gm excess when the reactor
(5) Lower power level for automatic control is 50 milliwatts with fluctuations of
less than 1%. No information was available on frequency of these.
(6) The neutron source is Ra - Be and has a strength of 200 millicurries. It is
placed about 1 foot from the reactor sphere,. If Sb - Be source is used it will
be necessary to use Ra - Be to start after long shut-down.
(7) Cooling water activity is very low, only slightly above background after
reaching counting room.
(8) Use iron-
used as cement
(9) Removal of 4 1/4 x 4 1/4 graphite plug next to reactor is worth less than
(10) Three inch fission chamber has a "sensitivity" of 3 microamps/KW in
(1) Portable Beam Traps
Several beam traps were constructed of discarded oil drums. These were
mounted horizontally on casters and were filled with B4C and paraffin in
granular form with 4" of lead in the rear. The size of the drums used was
about 2 ft long x 18" in diameter. These have proved to be very useful.
Larger sizes were also noted.
(2) Portable Neutron Shields
Borax and paraffin make very good portable shields. This should probably
be placed in metal cans to reduce fire hazard of too much exposed paraffin.
(3) Specimen Holders
The specimen holders should be of either Be, special magnesium free iron
or Al. Plastic holders are not too good. The special magnesium free iron is
obtained from iron carbonyl and is very good. Al is fair but builds up quite
a bit of induced activity.
The exposure holes are 4" from reactor and are placed in a 1 1/2" x 4"
opening in the graphite as indicated in sketch below. These holes are worth
less than 0.1 gram of U235.