Study of Incident Involving Fuel Leak in North Carolina State College Reactor



Curtis A Nelson, Director, Division of
Inspection

Marvin M. Mann, Assistant Director for compliance

STUDY OF INCIDENT INVOLVING FUEL LEAK IN NORTH CAROLINA
STATE COLLEGE
REACTOR

SYMBOL:INS:MMM

I spent the day with Dr. C. K. Beck in Raleigh on June 16, 1955.
As you know, Herman Roth of Oak Ridge had reported to us that the
core of the Raleigh reactor had developed a moderate leak; and that
it was estimated at the time that about 100 cc's of fuel solution
had made its way out of the core. After I had a short talk with
Dr. Beck, he took me over the reactor installation. We then spent
the rest of the day (with the staff) constructing as best we could the
chronological history of the incident, which follows:

May 6, 1955: On this date, Dr. Beck himself operated the reactor
for several hours at a power of approximately 6
kilowatts. This is the last date on which the
reactor was operated at power.

May 9, 1955: It was noted that the pressure in the core had in-
creased over the weekend. Normal pressure was said
to be between 5 and 15 inches of water below atmos-
pheric pressure. In this occasion there was no
apparent reason for the pressure increase, so investi-
gation was begun. The core pressure was returned to
normal, that is, approximately 15 inches below atmos-
pheric pressure and on subsequent readings during the
day it was observed to rise slowly.

May 10, 1955: Pressure had returned to atmospheric by morning
Samples were taken from the gas in the aluminum
envelope that surrounds the stainless steel core and
evidence of fission products was found. Subsequently,
it was determined that the primary activity was
Xe-133 and its magnitude was estimated at one microcurie
per cc. It was presumed then that fission product gases
were leaking from the gas recombination system, there
being no evidence that liquid had leaked out of the
core vessel. Note: There is a small wire suspended
between the core and the aluminum envelope very near
the bottom of the two containers. When liquid enters
this space then the resistance to ground from this wire
drops off sharply.


[page 2]

During the next few days confirmatory tests were
made of the activity and its magnitude. A detailed
procedure was formulated for disposal of the radio-
active gas. During all this time the core was at
atmospheric pressure.

May 20, 1955: Execution of the procedure just referred to was begun
and consisted of the following: --- Gas from both the
envelope and the core was bled slowly to the main
exhaust stack for the reactor building. Normal flow
to the stack is 20,000 cfm and the magnitude of the
active gas flow was controlled in such a way as to
maintain activity of the stack effluent at less than
10-9 microcuries per cc. This rate was based on the
conservative assumption that all the activity was from
iodine 131, although it had been shown earlier that the
ratio of Xe-133 to iodine 131 was greater than 1000.

May 28, 1955: A slight increase in the gamma activity of the cooling
water was noted. This is water that flows through coils
of 1/4" stainless steel tubing in the core. This is a
"once through" cooling system but there had been no
actual flow for some days. The activity noted was that
from a sample taken from a hold up tank, which is about
60 feet from the reactor. The next step was to insti-
tute flow for a sufficient time to obtain a sample which
had actually come from the reactor core. It was found
that activity of this sample was about 25 times geater
than that of the first. The cooling coils were then
put under pressure of a approximately 55 pounds and the
system was closed off. However, no appreciablc drop in
pressure occurred over a period of several hours and it
was concluded that although a leak apparently existed in
the cooling coils, it must be rather small.

May 31, 1955: The presence of uranium in the cooling water samples
taken late in the evening of May 28 was established by
alpha counts.

June 2, 1955: It appeared on this date that the purge was essentially
complete, that is, the activity of the stack effluent
was now approximately that of normal background. However,
purging continued and is still going on.


[page 3]

June 3, 1955: On this day, Dr. Beck decided to execute a criticality
test as a check, via critical rod position, on both
earlier and current measurements which indicated that
the liquid level in the core had not changed. The
critical rod position was found to be normal within
limits of reasonable error.

June 6, 1955: It was found that the "wet wire monitor" mentioned
in the May 10th note, indicated the presence of liquid
outisde the core, that is, between the core and the
envelope. The liquid level was checked and found to be
low by about 4 millimeters, indicating leakage of
approximately 250 cc's of fuel solution into the envelope.
Samples of cooling water showed somewhat greater activity
than on May 28th. It was then decided that the situation
was serious enough to warrant transfer of the fuel solu-
tion to storage containers which had been placed in a
room below the reactor for this purpose. This was done
on June 7.

Between June 7 and June 10, that portion of the concrete
shield immediately above the core and plenum was removed
and the shim rods and control rod drives and supports
were taken out so as to yield access to the reactor core.
Incidentally, the tips of the shim rods, which in shut down
condition reside in the narrow space between core and
envelope are covered with a thin deposit which is quite
active.

On June 10, the cooling coils were drained and analysis
of samples resulted in Wentification of a number of
fission products. It would appear that, in truth, at
least two leaks developed in this reactor system; one
in the cooling system, and the other in the stainless
steel core. On the basis of experience, one might feel
inclined to blame the leaks on faulty welds, although it
has been shown that a water solution of uranium sulphate
can actually corrode through stainless steel plate at
points of local surface imperfection.

My own view of this situation is that the safest, and indeed very probably
the cheapest remedy is that of complete replacement of the core assembly.
At the same time, I think that a unique opportunity exists to learn from
this incident and I would therefore recommend that careful and thorough
disassembly of this reactor core be made in some location, probably
0ak Ridge where adequate facilities exist, for handling and study of
highly radioactive equipment.


[page 4]

The other problem of the moment is that of formulating a safe and
adequate procedure for removal of the reactor core, recovery of what
liquid has leaked into the envelope, and shipment to the point of
disassembly and study, I suggested to Dr. Beck, that if he were
interested in calling on the experience of Chalk River personnel in
this regard that I will be glad to arrange contact for him.

Dr. Beck has already been in contact with Babcock and Wilcox and with
North American Aviation in regard to construction of a new core
assembly and I believe he has also requested Oak Ridge to consider
building a replacement. I think he is wise to look into the possibility
of a commercial supplier for this assembly, and at first glance it would
appear that both Babcock and Wilcox and North American Aviation are
excellent choices. Personally, I think that North American is parti-
cularly apt in view of their past and current experience with water
boiler reactors.

Dr. Beck is writing detailed report of this incident to be sent
to Dr. Roth, and his agreed to send us a copy

cc: Dr. C. K. Beck
N. C. State College
Raleigh, N. C.