The status of our nuclear weapons -- an interesting rant
I subscribe to a number of email lists and on one of them, a participant uses the following signature line:
Any day you get to play with nuclear warheads, is a good day!
OK - so they probably know a thing or two about these puppies.
Someone asked them about the status of our weapons today and they replied with the following excellent rant:
Easy enough. The second you put a nuclear device together, it starts to deteriorate. Chemical changes in explosives and electronics, metal gets brittle from neutron bombardment, etc. As time goes on, reliability declines.
In the old days, the AEC would do constant monitoring of the stockpile and they'd do what were called "Stockpile validation shots", a test firing of a nuke to see if it would work or not and what changes in reliability and yield would occur over the period of storage. They also came up with repair and upgrade kits and these would be tested by live fire too.
Then Bush I, decided to supplant testing by substituting use of computer models. This is nice if you know what is going on inside a bomb during storage. They did test shots of bombs of various ages and used those to calibrate the models. But there's this catch. When the bomb becomes older than the oldest bomb used to calibrate the model, the model doesn't work anymore. And you can't learn anything new from a model. It just extrapolates what you already know.
We were going to get around that, by replacing bombs that reached a certain age. The Chinese, for example, do this every seven years. Build a bomb, store it, and then decommission and scrap it and build another bomb.
That would have worked for us too, but Clinton closed the production lines at PANTEX. Now we build no new bombs and the newest bombs we've got, are older than the oldest ones that we used to calibrate the models.
And a lot of what you see in the deterioration of a nuclear warhead isn't exactly what you'd call predictable either. Look at a list of the bombs that we've built and the service dates and the dates that they were declared obsolete and withdrawn, and you find that some bombs didn't last longer than two years in our inventory. And many of the problems that resulted in their being withdrawn, were never solved.
And the punchline here is that bombs are not as simple as that little diagram you see in an encyclopedia someplace. Your typical modern bomb, say the W-80, which can be considered to be a modernised B-61, has over three thousand parts in it. That's a lot of complexity there and a lot to go wrong. And losing any one of em turns our nukes into an expensive piece of nuclear junk. And that's everything from the inflight pit insertion mechanism, to the Tritium Booster to the Permissive Action Links and the Strike Enable Plug. More stuff to go wrong than you can name, especially when most of those parts have numbers but no other nomenclature. (Remember the story of the Mk14 Torpedo? That was a whole lot less complicated than a nuke and you didn't have to deal with changes due to neutron bombardment, but it was still remarkably difficult to debug.)
And reliability in general is a contra-intuitive deal. For example, suppose you have a device consisting of five parts that are each ninety percent reliable. Most people would assume that the gadget itself is ninety percent reliable as a result, right? Nope-- dead wrong. Fifty Nine Percent. (.5 X .5 X .5. X .5= 0.59 or 59% where 1.0=100%.) So you can imagine how complex that calculation would be for something like a B-61 or a W-80.
The net result is that our deterrence doesn't rest on assured destruction if we use the bombs, but uncertainty as to what will happen. You can deal with a certainty. Uncertainties on the other hand, generally make politicians risk averse, and often the technicians that advise them, even more so.
In short, Clinton's decision amounted to a slow process of unilateral disarmament. And he did this at the same time that he let the Chinese take whatever they wanted of what should have remained, highly classified nuclear secrets. A paper on nuclear metallurgy might be boring to you, but it made for some rather exciting reading for the guys at Lop Nor and the various institutes that support it.
That's why Bush wanted to manufacture the B-61-11 as new production and why they were pushing for the Reliable Replacement Warhead. Not only do our enemies not know what our bombs will do, but we don't know ourselves! The Chinese do, however, know what theirs will do because we let them steal the designs and the statistical performance models and they replace every seven years. That should scare somebody. It does me.
Awhile back, in another discussion elsewhere, my opposite number was demanding that Washington secure Israel's future by nuking Iran. I pointed out that we couldn't do this. I also, just to give him an idea of how iffy the whole thing was, came up with a thumbnail sketch of a plan that would have to be followed if we wanted to use our bombs while still managing the risks involved. (Besides, they've got their own copy of our B-61 which Kissenger gifted them with in order to bribe them into not testing their own. And that didn't work completely, given the South African shot over the South Atlantic.)
Here's the catch. There would have to be a stockpile validation shot for the bombs, probably B-61s since they're our most numerous and reliable type, and we would have to do that during the mission in order to avoid telegraphing to the world that we intended to use them.
If you do an atmospheric detonation, VELA type sensors in orbit will detect the EMP from the detonation. If you do an underground shot, a nuclear test has a seismic signiature that is unique. Nobody's going to mistake it for anything else.
So what would have to happen, is that after we've launched the planes, we'd have to fire a stockpile validation shot with multiple bombs for a statistically valid sample and if enough of the bombs worked, we'd have to give the final release to the bombers at the failsafe point. So you're talking about an incredibly complex operation several hours in length where we bomb up the planes, send them with tankers to the failsafe point, detonate a bunch of bombs, analyze some complex data from sensors attached to light pipes and the like, decide whether the bombs will work, and then give permission for the crews to actually fly their penetration routes, deliver their bombs and in some cases fly to a recovery area inside enemy lines to bail out and wait for the Air Force to send modified C-130s in to get them. (During the Cold War, the Brits had a different idea. They were going to send in the RAF Regiment to the recovery areas and the downed aircrew would augment the RAF Regiment and fight on the ground. The assumption was that there wasn't going to be enough of Jolly old Englande to come back to! A close friend of mine was the Bombadier on a Handley Page Victor after being posted from Canberras, and that would have been his fate if the ball had dropped.)
Anyway, that's the thumbnail view. Hope it helps.
Shame on Clinton (as well as Carter who shut down the nuclear fuels reprocessing plants)
The Mark XIV had
stunning reliability problems at the outset of WWII although these were eventually ironed out and it served our Navy for 40 years.
Posted by DaveH at March 12, 2008 10:48 PM
| TrackBack