A Review of Criticality Accidents A Review of Criticality Accidents

4. Oak Ridge National Laboratory, 1 February 1956 40
Uranium solution assembly; wave motion created by falling cadmium sheet; single excursion; insignificant
exposures.
In this experiment, certain reactor parameters were
being investigated by measuring stable reactor periods.
The system was a cylindrical tank (0.76 m in diameter)
filled to a depth of 130 mm with 58.9 l of UO 2 F 2
solution containing 27.7 kg of 235 U. Transfer of
solution from storage to the test cylinder was achieved
by applying air pressure to the storage vessel; flow was
controlled by a remotely operated valve in a one–half
inch diameter line. With the control switch in the
“feed” position, the valve was open and air pressure
was applied; with the switch in the “drain” position,
the valve was also open, but the air supply was turned
off and the storage vessels were vented to the atmosphere.
When the switch was in the intermediate
“neutral” position, the valve was closed and the storage
vessels were vented.
The situation was one in which the solution volume
was about 100 m l less than the critical volume. An
increment of solution was added, and the transient
period decreased rapidly to approximately 30 seconds,
where it seemed to remain constant. Shortly thereafter
the fuel control switch was placed in the drain position
and the period meter indicated a rapid decrease in
period so that the safety devices were actuated almost
simultaneously by both manual and instrument signal;
the instrument trip point had been set at a 10 second
period. Immediately thereafter the excursion occurred.
The yield was 1.6 × 10 17 fissions and, in this case, a
considerable volume of solution was forcibly ejected
from the cylinder. Post excursion tests showed that if
insufficient time were allowed for venting the operating
pressure, addition of solution to the reactor could
have continued for several seconds after the control
switch was placed in the drain position. This addition
of solution accounted for the decrease in period that
72
precipitated the scram, but the increment of solution
could not have added enough reactivity to account for
the excursion.
The reactivity of such shallow, large diameter
assemblies is very sensitive to the solution depth but
quite insensitive to changes in the diameter. For this
system, the estimated difference between delayed
criticality and prompt criticality is only 1 mm of depth.
If the effective diameter were reduced to 0.50 m, the
depth would have to be increased only 12 mm to
maintain delayed criticality. It is thought that the
falling scram, a cadmium sheet slightly deformed at
the bottom, set up a wave system that must have
converged at least once and created a superprompt
critical geometry.
In this case, the analysis was directed to finding
what reactivity insertion rate would cause a power
spike of the required yield. The analysis was then
examined to see if it contradicted any known facts. It
was found that a rate of 94 $/s was adequate to cause a
spike of 8 ms duration, which would account for the
observed yield. The maximum excess reactivity would
be about 2 $ over prompt criticality; the void volume
could be 12 times that of the Oak Ridge National
Laboratory 26 May 1954 accident (II-A. 3), thus easily
accounting for the splashing of the solution. The void
volume that results as microbubbles (caused by
disassociation of water by fission fragments) coalesce
is discussed in Power Excursions and Quenching
Mechanisms.
A laborious chemical decontamination of the
assembly room was required to clean up the ejected
solution. Slight mechanical damage was evidenced by
distortion of the bottom of the cylinder. The largest
radiation dose received was 0.6 rem.