I get to be scary morbid!!
Back when I was a baby Nuc, the Navy slammed this one home hard, that we (the Nav) would never do something like this ourselves…
OK, first a little background:
The US Army wanted in on the nuclear game, back when no one knew just how hairy it could get, and the only realistic use they could find for nuclear power was semi-perminent power plants for remote bases (not a bad idea, conceptually…). They came up with a design that would use a pre-fab parts for the reactor and related buildings, and use local materials (dirt, gravel, etc) for shielding. OK, so far, so good.
Now having designed and set-up the prototype, the trouble starts: The Army sent their reactor technicians through a “rigorous” training program, measuring just a couple months long (rigorous? NOT! The Navy takes about two years to make a Nuc). This was in no way sufficient to properly train the techs. Next, the design had a couple of critical flaws: First , the reactor could go critical (“critical” is a self-sustaining chain reaction: The point where the reactor is actually starting to ‘work’) when only one control rod was pulled, and it would go critical before that rod was withdrawn more than about a foot. Worse, the reactor could go “prompt-critical” when only one rod was pulled to about 1 1/2 feet. Prompt-critical is a VERY BAD THING, leading to uncontrolled power excursions (think steam explosion, destroyed instrumentation, etc: BOOM). Second, the reactor need to have it’s control rods manually operated to perform certain shutdown tests, such as inserting neutron flux measuring strips. In order to do this, the Control Rod Drive Mechanisms (the machines that raise and lower the control rods in normal operation) need to be disassembled.
So:
One fine Christmas season, while LS-1 was shutdown for the holidays, and only a skeleton crew was on hand, an evolution to insert neutron flux measuring strips was conducted, by a staff of three (you don’t monkey the reactor vessel on the back shift, and certainly not with a skeleton crew!) techs with limited training. It was determined that one of the control rods was manually pulled almost completely out of the core, leading to prompt-criticality, a massive power excursion, a lethal sleet of neutrons, and one big-ass steam explosion. The reactor vessel was wrecked. Two of the technicans never made it out of the reactor building. The tech who was actually pulling the rod was found impaled to the ceiling when the control rod he on which he was working transfixed him through the pelvis. The one tech who did manage to escape was suffering life-threatening burns, life-threatening explosive trauma, and fatal radiation exposure. He lived long enough to be placed in an ambulance, but died shortly after being “rescued”. IIRC, the Army had to bury the ambulance because it was too crapped-up to be economically decontaminated.
LS-1 was demolished and buried at the National Reactor Facility near Idaho Falls. The Army got out of the reactor business.