Command and Control [video]

I am in complete agreement, with the exception of this statement:

> there is virtually no way for the nuke to "fail" in a way that would be even remotely close to a detonation sequence

That is not true, at least not universally. Nuclear bombs are, by definition, designed to go off, and as a result there is a category of failure modes that could lead to inadvertent detonation. Basically, a failure that produces the same command that would cause the normal detonation sequence to begin, could cause an unintended detonation. The Mk. 39 bomb involved in the 1961 North Carolina B-52 crash [0] reportedly came close to detonating in this manner.

Additionally, there are weapons designs that are not "one point safe", such that a failure which causes ignition of the high explosive could lead to a nuclear detonation. Weapons considered "one point safe" require multiple, carefully timed or simultaneous ignitions of the high explosive. The last non-one-point-safe weapons in the US arsenal were built in the 50s but not disassembled until recently; the W56 was reportedly not one point safe [1], or at least not provably so to the safety margins now recognized as reasonable. I don't know of any information in the unclassified literature about whether Soviet weapons are universally one-point-safe or not.

It would not surprise me if there are a variety of weapon designs still floating around in the world that aren't one-point-safe. Single compression "gun type" bombs, like the Fat Man design, would require additional engineering to make one-point-safe beyond what is minimally required to get them to detonate, which makes me suspect that some marginal nuclear powers probably don't bother.

[0]: https://en.wikipedia.org/wiki/1961_Goldsboro_B-52_crash

[1]: https://web.archive.org/web/20070110205223/http://www.pogo.o...

>> an accidental detonation especially through brute force that would physically damage the warhead is about as likely as a tornado ...

Even if you do damage the warhead, even if you get the high explosives of the warhead to detonate, that detonation will not be symmetrical. If it isn't perfect, the shell doesn't collapse evenly --> no critical mass --> no nuclear fission --> no nuclear fusion == a week dirty bomb. No mushroom cloud. The bomb's brain has to be involved. It has to want to go off.

You refer to the spreading of fissile material.

The original post refers to an "explosion 600 times more powerful than Hiroshima" -- which would obviously refer to a Nuclear detonation.

So yeah, I have a problem with that based on the literature available to the public regarding the safety of these weapons.

Note that your two examples didn't involve any nuclear material, as the fissile cores were kept separately and weren't installed in the bombs at the time.

I don't disagree that they are dangerous objects. They are radioactive material surrounded by explosives atop a rocket full of other explosives. But the 600x description means a full and proper thermonuclear detonation. That is a very different beast than an on-pad/in-silo explosion possibly releasing a small amount of radioactive material in an isolated location, a 'dirty bomb' at best. The reference to a large nuclear mushroom cloud is improper. It was never a possibility short of someone actually wanting it[1].

[1] There is much evidence that these things could have been set off deliberately (see "111111") but that isn't what this film is about.

I did, and found no particular cause for concern with regard to safety. Past practice lacked in some regards, I agree. But, as I recall the book, even Schlosser conceded that those issues have been effectively addressed, and that the balance between agility of response and effective C&C had been swung pretty firmly in the latter direction even by the time of the 1980 Titan accident, to say nothing of today.

I've read the book. The issue is not the trigger mechanisms/lack of security - we are talking about the durability of the warheads and accidental detonation.