posted on Jan, 27 2007 @ 04:19 PM
Ok, here's my relatively uninformed two cents worth.
1) POGO finds some interesting things at times. But they are no more likely to present such events in a neutral light than PETA is to present a study
that eating meat might be good for you. You have to understand that carping about anything the gubmint does is what POGO is. And if they find
something they can build into a dramatic moment, maybe they'll get a bit more funding.
POGO is to government as CSPI is to food, you can always bet that their "studies" will show that meat is pure evil, even if their study was
scientifically invalid and they can't get it published in a peer-reviewed journal. PETA will blare it out anyway, because they ARE CSPI. In a similar
way, POGO can always be counted on to yap about something if the government's involved, because that's what pays for their power bills at the end of
the month. That doesn't mean that they didn't find something interesting, but you have to understand that trying to get you all freaked out about it
is SOP.
2) Pantex is an interesting place. They do have fairly long work hours, though I'm not sure you can accurately ascribe what happened to the long work
hours. It's a tense place due to the work involved, everyone lives on Ibuprofen and Pepcid. There's nothing like having to be around these things
while they're being tinkered with to get the hair up on your back. I never liked pulling special WSA duty during maintenance. I can't imagine
working on one. The bomb techs have really creepy senses of humor, and I'm sure it's related to the stress, although most of them won't admit to
it.
The DOE once described the atmosphere at Pantex as being "perennially truculent", which I'd tend to agree with having been in and out of it a few
times visiting. They get a steady bait of citations for fistfights in the plant. So, it doesn't surprise me a lot to discover that they take any
opportunity to gripe about the hours, the management etc. I notice that POGO didn't describe who the "knowledgeable experts" were.
My guess, too, is that they're making a creative use of the term "explode" not in the sense of the bomb successfully detonating in a nuclear
fashion but in the sense of the high explosive charges going off. I imagine someone at POGO knows the difference but the UPI guys might not, a lot of
journalists don't really understand much about it.
3) "What really happened"
Ok. When you're disassembling a W56 there's this bible of procedures that you're supposed to follow. You're not supposed to deviate at the tech
level. If something unexpected happens, you hit the kill switch and get the nukies down there.
Anyways, on March 30, 2005, they were disassembling this W56. Now, W56's have a freaky two-component midbody rad shield that is threaded together.
The bimetal joint tends to not ever unscrew again after a while. So, how you get the physics package out of the midbody is, you use a hydraulic ram to
shove it out the ass end of the warhead.
Now, you've already removed the secondary (no thermonuclear explosion) and the boost system, initiator, and trigger package so there's not a chance
of a traditional detonation. Also, the pit vents are open where the boost injection used to be, so getting a clean compression on what's left of it
is nearly impossible if you DID still have the thing in one piece.
But there's no way to get the pit out unless you either unscrew the shield halves, which ain't happenin', or shove the guts out the end with a ram
into the open.
They put the W56 on the ram and hit the "go" button. You're supposed to manually turn this jackscrew in that sets the pressure, and you're
supposed to creep up on it so that it breaks loose without any extra pressure. It went up to about 1500 pounds and popped, so they thought it had come
loose. That's about where it normally happens. At that point, you raise the ram on the run-up gear (which uses a motor to spin the jackscrew) and the
guts slide out. They raised the ram and the case broke, then the ram safety limit clutch released the ram. Whoops! So they stopped and called a
conference. The HE had fractured, not come loose like they'd thought, and it was still in there. On top of which, the case had cracked.
Now if you thought you could get a yield of any sort before, it was down the drain now, because the HE was no longer intact and the shield was split.
But they put the obligatory chain in the pit vent in order to "safe" what remained of the physics package. At that point, even IF the HE went off
the obstruction in the pit center would prevent a dense enough assembly for criticality (it guarantees a Munroe jet of enormous magnitude, like
spurting water out of your hands in the tub), but the neutron absorber in the chain would prevent achieving criticality anyway. It wasn't going to do
anything at that point but splat around the lab and ruin a piece of the facility.
They called in some WER guys and designers from LANL to look at the mess. A new procedure was written, which basically said, resecure it, and put the
ram to it again, it's not going to detonate. Now, the usual point at which the explosive charges pop out the rear is in the neighborhood of 1500-1600
pounds. WER set a recommended limit of 3700 lbf for the second run. Pantex's limit is 3600 lbf.
Now, they were supposed to go to the second ram unit and recalibrate it from stem to stern, test it with dummy loads and make sure it was in perfect
shape, then move the busted warhead over to that press. After all, the WER guys just stroked you a SIER for a one time 3700 pound limit. You should at
least make sure, since you're going outside normal procedures, that the press is actually DOING that. You'd think that, wouldn't you? Guess
what...they didn't do it. They used the same press it was already stuck in.
So, on Monday, April 25, they went back to the same press. Now, WER had just issued a SIER for 3700 pounds. But Pantex issued an internal HAR that
limited the attempt to 3600 pounds. Keep that in mind, the LANL guys had just said that 3700 pounds was still in the safety limits and a good number
to use for this try as a maximum ram setting.
They hit the button. They ran it up to about 3300 pounds force...and the safety clutch let go. Now, it's supposed to be good for the max you have
set. But it let go 250 pounds early. No dice, the weapon's still in one piece.
They replace part of the clutch assembly and try to sort of half-ass recalibrate it. Tuesday dawns.
Hit it...3350. Hit it...3350. Hit it...3350.
Now, they're not supposed to be doing this at all. They should have been on a known good ram that's been calibrated and tested for the maximum ram
pressure. And if it doesn't work the very first time, they should freaking stop and rethink everything. But now they're p--sed off at it. It's been
stuck in there for a month.
The supervisor says, use the run up gear. Now, you have to understand sort of how it works, there's this run up gear that you use to advance the
jackscrew on the ram really fast in order to chuck the weapon and to raise the package out of the midcase once it separates. Normally, you run up to
near where you want to be and hand crank it the rest of the way, so you have some control. There's a safety clutch that's supposed to release the
whole thing if you exceed the 3600 pound Pantex limit. But the clutch takes time to release. If you use the run-up gear, it will apply more force than
normal before it gets a chance to release. The clutch is really intended to be an oversight on the hand crank.
So they cut it off and let it cool down, another trick, the clutch will release faster than normal if it's hot. Then after it got cool, they spun it
up and hit the run up gear and kept it engaged the whole way. Bam! Crunch! and out came the package from the midcase. Problem solved.
Now, the maximum allowed pressure was exceeded. The ram gauges say that the peak was 3750 lbf. That's 50 big whopper pounds force over the LANL/WER
recommendation. So, yes, technically it's over the limit, by about 1.3%. LANL had a hefty safety margin in their number, so 3750 is not over the
danger line. But it is over both the Pantex procedure limit and the SIER limit.
The fecal matter hits the fan at Pantex on Wednesday afternoon after the reports start to be written. Pantex management begins to deliver live kittens
from every orifice. Everyone involved gets their anuses removed. Pantex does a safety shutdown. Fingers point, curses fly, truculence reigns. The
press operators blame the supervisor, the supervisor blames management, management blames the operators. The press operators dash off a
"whistleblower" letter to DOE trying to dodge the responsibility by using the whistleblower protection act. That, incidentally, is one reason why
"A spokesperson for the Energy Department declined to respond to safety complaints in the letter."...it's really not that much protection to you if
you byotch about the safety of the plant after you just used the run-up gear to smack a bomb apart. In the end, there was plenty of blame to
share.
1) they should have specially set up, tested and calibrated the second ram for this run instead of continuing production on it
2) the supervisor should have never tried more than once before moving to the other press
3) at any rate, he should have stopped and called another meeting when the Monday afternoon attempt failed
4) and he should never ever have taken repeated tries Tuesday the 26th
5) the thing with the runup gear was just a travesty, there's no real control in that mode. He wouldn't have had to, either, if he'd used a press
that was working.
6) the operators should have refused to do it
7) there was some auxiliary finger pointing that the supervisor and the press operators developed their little run-up gear trick on the side and
waited for the process and tooling engineers to leave the area before trying it. However, the same engineers sure weren't stopping them from repeated
retries or using the same press.
In the end, the $110,000 fine was:
$55,000 for not switching to a different, tested and calibrated press when they were told to
$55,000 for the run-up gear trick and the multiple retries, which were both a violation of procedures
A suspended fine for not having a better way of determining when the explosives actually separated than listening for it
A suspended fine for not having the fixture calibrated and working correctly anyway
A $110K fine is peanuts anyway, Pantex gets something like $25-40 million a year in performance bonuses. It's cheaper to take the fine and a
violation cite than to wait and lose the bonus. Pantex gets 100K level fines every year. This was not an unusual fine amount for simple procedural
violations. W56's are a pain in the butt anyway, I think they get fines every year for the HE charges fracturing, the Pantex guys are going to burn
them anyway so they generally tape the assembly together with duct tape and keep going. If they get caught, it's a fine in the 100K range.