In my setting, there is a ban on nuclear detonations greater than one kiloton in magnitude. This ban is enforced with a supranationally controlled strategic nuclear arsenal capable of essentially annihilating a country; one North Korea-esque nation violated it while being invaded by their equivalent of South Korea (read: nuked their capital city with a 635-kiloton boosted fission weapon) and was promptly wiped off the map.

As one might imagine, the folks in this setting are less concerned with the radiological effects, death toll, and proliferation risks of nuclear weapons then we are in real life. This means that sub-kiloton weapons are often used in warfare, primarily against tanks (two hundred+ ton, nuke-resistant abominations), against hardened positions (usually undergrounded or armored with a half-meter of lead), against aircraft (lots of anti-flash white), and during naval warfare (countermeasures up the wazoo, from nuclear depth charges to fission reactor-powered laser point defense systems).

The problem with Davy Crockett-sized nuclear (fission, of course; these are much too small to be fusion bombs) munitions flying everywhere is that you have to build a lot of them. So, this begs the questions:

How quickly and in what volumes can sub-kiloton nuclear bombs be manufactured, assuming a 21st-century industrial base and adequate knowledge of how to construct them?

Are there any production bottlenecks besides the supply of fissile elements?

  • $\begingroup$ Historically (see en.wikipedia.org/wiki/…) the US was able to increase the size of their stockpile by 16,000 weapons in the 5 years from 1955 to 1960, so 3000 or so a year net. $\endgroup$
    – Jon Custer
    Commented Nov 3, 2021 at 20:00

2 Answers 2


Supplies of fissile materials are going to be a bigger bottleneck than any other. The guidance system, unless you intend to use a direct fire rocket propelled grenade style set up is probably next on your list, followed by the precision machining capacity for making the implosion device, the fuselage, rocket motor(s), and explosive lenses in the core are all relatively easy to manufacture.

As to your first question the rest of the round can probably be assembled in at most a couple of days. Depending on the exact delivery package, an artillery shell is going to be quicker than a self-guided missile, you could knock out huge numbers of them in a medium sized factory every day. The core can probably be built just as quickly if the materials are on hand.

As a note if you build these weapons on the assumption they will be used quite quickly after the leave the final assembly line (and given that people don't seem to care too much about the background count) then you may have a wider pool of fissile elements to choose from than in current arsenals. Current nuclear weapons use long-lived isotopes partly for handling safety but also to guarantee weapon shelf-life. For example a nuke using Californium-251 would need only slightly more than half as much fuel as a Plutonium-239 core to achieve criticality but will have a very short shelf life as the Californium has a half-life thousands of times shorter.

  • $\begingroup$ 898 years is still pretty good, IMO. Most of these things are going to be fired within a year of assembly. $\endgroup$
    Commented Oct 30, 2021 at 5:26
  • $\begingroup$ @KEY_ABRADE Shelf life depends on the effect of decay on the continued viability of the weapon, not just loss of fuel but the activity of it's daughter isotopes etc... $\endgroup$
    – Ash
    Commented Oct 30, 2021 at 5:29
  • $\begingroup$ Production of that Cf-251 is going to be an issue. You have 13 neutron captures with a yield well under 1%. And what's the temperature of the core??? $\endgroup$ Commented Oct 31, 2021 at 3:43
  • $\begingroup$ @LorenPechtel I wasn't thinking of Californium as a sole fuel rather thinking of it as an example of a material you wouldn't generally use for weapons that may be worth purifying if you're not planning to stockpile your nukes for any great length of time. $\endgroup$
    – Ash
    Commented Oct 31, 2021 at 6:27
  • $\begingroup$ You don't purify Californium, you produce it by neutron bombardment starting with U-238. Note that in the process you go through Pu-239. The steps past Pu-239 waste far more than the higher critical mass of Pu-239. $\endgroup$ Commented Oct 31, 2021 at 22:36

Aside from the fissile material, there is nothing "special" or extraordnarily difficult about bulding a nuclear bomb.

The hard parts are:

  1. Sourcing the fissile material itself, in sufficient quantity and purity.
  2. Designing the bomb in the first place. One needs some rather fancy theories and calculations to get it just right and predictable.
  3. A robust, reliable, utterly dependable delivery system. Which includes inventory tracking to make sure you don't misplace the eggs!

If you can provide me with an unlimited source of the fissile material, cast and polished to the exact shape and purity it needs, I can build you a factory that will churn out 1000 of them per day. With enough budget, I could possibly have your factory ready within 2 years. It's a very simple manufacturing process, for example much less mechanically and engineering challenge than building a turbofan engine for an airliner. Frankly the only real challenge is the security aspect, with a throughput of 1000/day, an inventory loss of 0.001% would mean 3 missing devices per year.

Your bottleneck is purely and only the sourcing, purification and shaping of the fissile material. And of these, the purification is the stickler.


You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .