Just a nitpick - the US had gun-type uranium weapons into the 1980’s. I did inspections of the components. Soviets probably into the 1990’s.
What made them so unsafe? What model relaced them?
If I have this wrong, someone on this board who’s actually handled nuclear weapons (is this the coolest place, or what?) will correct me, but among other things, they were not built with any of the interlocks and other safety systems that modern weapons have. Although a quick review of the Nuclear Weapons Databook shows that even into the 1960s some weapons were protected with little more than a mechanical combination lock ( :eek: ), Little Boy didn’t even have that. It certainly lacked the Permissive Action Link that all modern weapons have that permanently disables the device if incorrect attempts to detonate or tamper with it are made. Also, I suspect that LB’s explosives were not very stable.
The real problem with LB (and almost any gun design), as Hail Ants says, is that it was inefficient. It’s really hard to slam two subcritical pieces of U-235 together fast enough to get a good chain reaction going before the force of the reaction blows all the matter apart. Only a small fraction of your fuel is converted to energy before the reaction is halted. So it’s very wasteful of the frightfully expensive U-235 that is needed for the gun design. Plutonium, which is cheaper to produce, won’t work in the gun system, because it tends to pre-detonate.
That’s why we had literally dozens of different implosion designs, but only four gun-type designs.
The Mark 8 was the production version of Little Boy. It was produced in 1952 and 1953 and retired in 1957, after 40 were made.
The Mark 9 was a 280mm howitzer shell (80 produced), and the Mark 11 was a bomb for the Navy. Forty of them were made.
Those four (LB, Mark 8, 9, 11) were the only gun-type weapons the U.S. ever built, and only 165 units were produced, compared to hundreds of thousands of implosion bombs.
One of my favorite stories from the Manhattan project concerns the designing of LB. When they were trying to figure out how big it would be, at first it looked like it would be too heavy to be carried by any bomber available at the time. The initial design of the barrel of the gun had been based on previous cannons that fired a shell with comparable weight and speed as the U-235 “bullet.” But then someone realized that those guns were expected to be fired more than once.
Oops. I missed the S33 howitzer shell, made from 1957 through 1965 and retired around 1984. One thousand of them were made. (The other three post-LB gun weapons were all taken out of service by 1957.)
So 1,165 gun-type bombs of five designs were actually produced.
Thanks for the information. My own knowledge of US nuclear weapons predates the publication of that book, and was filtered through both civilian and military security thinking. I am glad I don’t have to shoot myself.
Tris
The alternative is the implosion design. To make these work requires that explosives in many locations be initiated at EXACTLY the same time. This is rather difficult to do intentionally, and darned near impossible to do accidently. So if an unfortunate lightning strike hits the weapon, for example, the conventional explosives may detonate, and scatter nasty radiactive material around, but you will NOT have nuclear explosion. In this type of device, if you keep the triggering circuits from working, the device really can’t make a big bang.
The gun type requires no such precision in it’s triggering. If ANYTHING causes the CE to go, you are going to have a nuclear event.
The first nuclear detonation (the trinity test) was an implosion device. The scientists were very sure the gun device would work, so the first “test” of that design occured over Hiroshima. The implosion device was much fussier, so warranted testing prior to deployment at Nagasaki.
Besides the danger of accidentally firing the gun portion of the device, I read that there was a real danger of the device going critical if it was immersed in water. Water is a neutron moderator.
Bomb goes in cage. Cage goes in water. Shark’s in water. Our shark.
Mmmmm… flash-fried sushi. 
What about the bombs accidentally lost by the US Military, such as the ones off the coasts of North Carolina, Georgia, and Greenland?
I had not heard mks57’s claim about immersion in water being dangerous before, but the bombs you mention were AFAIK implosion devices, not gun weapons. (Although I don’t know enough physics to know why – or if – a gun bomb would be more dangerous in water than an implosion bomb.)
In any case, didn’t they recover the ones in NC and GA?
I believe all of those weapons had implosion primaries, which are much safer than a gun-assembly device. Many things have to work very precisely for an implosion device to achieve criticality.
The gun-type 8" howitzer shell had a “projectile” and “target rings” of U-235. The shapes and mass of each were such that they were sub-critical as long as not rapidly assembled together - a propellant charge driving the projectile into the cavity inside the rings. I believe some other designs had the rings or donuts driven over the donut hole! These components along with the shell body, fuze and propellant charge (to drive the elements together) were shipped and stored separately as one important safety. They were only combined when the codes came down and items were delivered to the gun crew. The shell had a large mechanical combination lock that prevented access to the interior to install components. The scientists “knew” the design would work from some accidents and near misses during research. All the separate components were a security headache but until assembly; you had a relatively safe system. The gun-type round was replaced with an 8" projectile with an implosion device and an electric/physical permissive action link. PAL device.
Note: Kennedy misspoke the name during a national address so the acronym was changed to match the speech. (rolls eyes)
Anyway, the uranium sections were subject to problems. Uranium corrodes requiring frequent inspections and the shapes weren’t all that sub-critical. The projectiles and rings would spall off small chunks of metal during radioactive decay. You had to clean and measure spalled areas as gaps would be opening.
The two 8" systems overlapped for a short period and by 1992, the US Army was getting out of the nuclear business.