I had the idea of a nuke that is scaled down for roughly 1 inch tall cockroaches. Why would they make one at all? Because it sounds cool. Would it effect them since they are roaches? Probably not.

enter image description here

But I am not here to ask about the whys of a mininuke, but the effects of detonating one.

Would it destroy the mansion/house? How far would the effect of the nuke reach? How deadly is it to a human nearby or nextdoor?

Here is some context:

  1. I assume there are a wide range of nukes, so to make the question less vague let us say the mininuke is pretty much a 1/72 scale of either nukes used on Hiroshima and Nagasaki
  2. They work despite whatever laws saying they should not.
  3. The nuked area is a mansion. I have no specific specs of it, but you could just use spec of an average mansion or even a house if need be.

Again, this is not meant to ask IF this nuke would actually work, but if it did, how bad would it wreck things.

  • 16
    $\begingroup$ If they work despite whatever laws saying they should not then, quite obviously, they will have whatever effect you want them to have. There is no way to actually compute the effect of something which cannot happen. Otherwise, I sincerely hope that you are able to compute 15,000 tons of TNT Hiroshima bomb equivalent divided by 72 cubed, to obtain a naive, incorrect but, given the requirements, believable approximation of the explosive power of the impossible device. $\endgroup$
    – AlexP
    Commented Apr 19, 2023 at 18:30
  • 4
    $\begingroup$ So you think this should be the answer given to all who ask about implausible worldbuilding question? Alright boys, pack it up, no need for input, just do what you want. @AlexP $\endgroup$ Commented Apr 19, 2023 at 18:55
  • 9
    $\begingroup$ No, this isn't possible. The laws of physics say that plutonium has a critical mass, below which it doesn't go boom. You can't scale down a nuclear reactor or nuclear weapon below that point. Read up on "suitcase nukes" for details. en.wikipedia.org/wiki/Suitcase_nuclear_device $\endgroup$ Commented Apr 19, 2023 at 19:08
  • 5
    $\begingroup$ As I read it, the question is something like "according to physics, how does this work?" and "disregard the fact physics does not allow it". If it doesn't work, you can't explain or simulate it, only option left is imagination. $\endgroup$
    – Matthieu
    Commented Apr 20, 2023 at 7:15
  • 5
    $\begingroup$ Ignore everyone who says this cannot happen. You did not tag your question science-based or hard-science, so there is no obligation for the solution to work "in reality." What I don't understand is why dividing the consequences of known detonations by 72 isn't enough to solve your problem? If I recall, the Tsar Bomba had a blast radius of about 20 miles. Reduce the Tsar Bomba by a factor of 72 and the blast radius becomes 0.278 miles or about 1500 feet. Doesn't that solve your problem? $\endgroup$
    – JBH
    Commented Apr 20, 2023 at 14:17

4 Answers 4


Blow up Tom Cruise House.

enter image description here

First we will see how big the Hiroshima explosion was.

From Wikipedia:

In Hiroshima, almost everything within 1.6 kilometres (1.0 mi) of the point directly under the explosion was completely destroyed, except for about 50 heavily reinforced, earthquake-resistant concrete buildings, only the shells of which remained standing.

Of course the destruction was much larger than this radius. But we are only interested in flattening buildings. So the above will be enough.

Your cockroach bomb is 72 times smaller. So scale down the blast radius by a factor of 72. You have a destruction radius of 1600/72 = 22.2 metres. The circle has an area of $\pi r^2 = \pi * 22^2 = 1520$ square metres.

Tom Cruise is small but his house is big.

enter image description here

A cursory search claims his house in Colorado is 930 square metres big. The same order of magnitude as the above number. So the bomb will flatten the Tom Cruise house.

  • 2
    $\begingroup$ I wonder though.... if a nuclear weapon could still explode at 72 times smaller than normal, would it still have the same punch, but in a smaller area? Or would the punch itself also be 72 times smaller? e.g. it goes off in the living room and maybe knocks over a lamp? Or perhaps the energy has a certain minimum punch anyway, because the energy of fusion/fission is always going to have a certain initial velocity... $\endgroup$
    – JamieB
    Commented Apr 19, 2023 at 20:28
  • 2
    $\begingroup$ Looking at it a different way, if we assume that (despite the laws of physics) the bomb functions identically to the Little Boy bomb, it would have a yield of 30000000lb/373248 == 80lb of TNT. That's a big enough bomb for home demolition. $\endgroup$
    – jdunlop
    Commented Apr 19, 2023 at 20:41
  • $\begingroup$ A mininuke of 1/72 scale? Mind you, this will lower the fissile material's content by 72^3 = 373248 times, making the explosion smaller by at least this factor. And having a nuke with power of less than 1 kT involves risks of a "nuclear fizzle" or failure to actually provide a supercritical chain reaction for long enough. BTW @jdunlop a nuke with nuclear yield of 80lb of TNT would require more than that to start chain reaction, so that would no longer be a nuke. $\endgroup$
    – Vesper
    Commented Apr 20, 2023 at 6:54
  • 2
    $\begingroup$ @Vesper - explicitly, from the OP: "They work despite whatever laws saying they should not." I am aware that it absolutely would not work. So is the OP. That is not a consideration. $\endgroup$
    – jdunlop
    Commented Apr 20, 2023 at 7:09
  • 1
    $\begingroup$ One thing to point out is that the destructive power of a nuke is not easy to predict. The bomb dropped on Nagasaki was on paper more powerful but in practice was far less damaging. The difference was that Little Boy detonated over flat open terrain, whereas Fat Man was dropped over a valley that contained the blast wave to the industrial part of the city. $\endgroup$
    – hszmv
    Commented Apr 21, 2023 at 17:06

The smallest nuke is .01-1 kilotons, compared to 15-20 for Hiroshima and Nagasaki

enter image description here

This should be enough for your needs. A nuke with a power like that is gonna wreck almost any house.

Even just 10 tons of TNT is enough to wreck a house. A typical missile is just something like 40 kilograms of explosive. 10 tons of explosive? You'll easily destroy any small target.

Using this calculator, we see it would wreck a lot.

I put in 0.3 kilotons.

Peak overpressure:20 psi Distance from the explosion site: 0.1 Kilometers Damage and injuries:Heavily built concrete buildings are severely damaged or demolished

Peak overpressure:10 psi Distance from the explosion site: 0.2 Kilometers Damage and injuries:Reinforced concrete buildings are severely damaged or demolished. Most people are killed.

Peak overpressure:5 psi Distance from the explosion site: 0.3 Kilometers Damage and injuries:Most buildings collapse. Injuries are universal, fatalities are widespread.

Peak overpressure:3 psi Distance from the explosion site: 0.4 Kilometers Damage and injuries: Residential structures collapse. Serious injuries are common, fatalities may occur.

Peak overpressure:1 psi Distance from the explosion site: 1.0 Kilometers Damage and injuries: Window glass shatters Light injuries from fragments occur.

If you're within 100 meters you'll likely be killed and everything will be wrecked, if you're within 200 meters most will be killed, if you're within 300 meters you'll be injured and maybe killed.

  • $\begingroup$ For very small nukes, the primary effects change. At 300 tons, the dominant cause of death is radiation, not overpressure: the 500 REM radius for death by radiation sickness is 0.7 kilometers. (The radiation radius grows much slower than either the thermal radius or the blast radius; above 300 kilotons, it's inside the fireball.) $\endgroup$
    – Mark
    Commented Apr 20, 2023 at 3:27
  • $\begingroup$ This is not "the smallest human nuke", this is "a magical cockroach-sized nuke". $\endgroup$
    – jdunlop
    Commented Apr 20, 2023 at 7:09
  • $\begingroup$ They're just concerned about killing a mansion, so radiation at greater distances isn't important. It's a cockroach sized nuke that's comparable in power to modern tactical nukes, hence the comparison. $\endgroup$
    – Nepene Nep
    Commented Apr 20, 2023 at 13:14
  • $\begingroup$ And the calculation you're doing is irrespective of the scale involved in the OP's request. 0.1-03 kT is not a 1:72 scale replica of Little Boy. It's not a valid comparison. $\endgroup$
    – jdunlop
    Commented Apr 20, 2023 at 17:26
  • $\begingroup$ OP didn't say which criteria we would use to scale it down, so I scaled down explosiveness. $\endgroup$
    – Nepene Nep
    Commented Apr 20, 2023 at 17:34

The simple answer, if you managed to detonate a 1/72 scale copy of a nuke (say, the Fat Man) with similar burnup fractions to the original, is that you'll get approximately 1/733 of the yield. In the case of Fat Man, with an original yield of ~21kT, you have a scaled yield of ~56kg of TNT. That's not a whole lot, but the physics package on has about fiftieth of a gram of Pu-239 and a third of a gram of U-238 so you can't expect anything earth shattering. You probably won't see a mushroom cloud developing.

The actual effects of the device going off are likely to be a little different than a bomb of either a conventional or nuclear flavor. In a nuclear explosion, the fission reactions yield a lot of hard x-rays which are strongly absorbed by the surrounding matter (usually air), producing a fireball which then expands rapidly producing a blast. The initial size of the "fireball" is likely to be larger than 1/72 of the diameter of the real nuclear fireball. The mini fireball will have a lower energy density than the big one due to the x-ray flux being reduced in the smaller explosion, but x-ray attenuation lengths remaining the same. Quite what this will look like I'm unsure, but probably you'll get a bright flash and sufficient material nearby will absorb enough x-rays to be vaporized to produce an explosion of expanding gas, so it will probably be close enough to a regular explosion for comparison purposes. A micro nuke probably won't be very effective in an air burst, but inside a building or underground it should work well enough.

Would it destroy the mansion/house? How far would the effect of the nuke reach? How deadly is it to a human nearby or nextdoor?

The amount of energy yielded by the reaction is something like a Mark 81 general purpose bomb, a "250lb" device with 44kg of explosives in its warhead. People have been using explosive weapons on this scale for quite some time, with plenty of bombs of that size being air-dropped over the last 100 years or so. Here's an image of the old Surrey County Hall, which was struck by a single 250lb bomb in 1939:

An example of the damage done by a 250lb bomb to an unreinforced civilian two storye building. The side of the building had been destroyed, and the roof has collapsed, and all the visible internal rooms are filled with rubble

Assuming similar blast effects, the amount of destruction is probably similar. It won't necessarily level an entire mansion, but it will probably cause serious damage and people inside the building at the time of the blast are obviously in grave danger, mostly because the building will likely become structurally unsound and bits of it are going to collapse. Building styles which use lighter materials and more wood will suffer worse, especially because the initial flash of radiation will cause combustible materials in close to the blast and in line of sight to ignite in a way that regular explosions would not.

The prompt radiation yield and fallout will be low, but I can't tell how low and working it out seems too hard. The total amount of nuclear material is low, so fallout should be readily containable and probably won't be a significant health threat, though breathing the dust after the building collapses is probably bad news.

They work despite whatever laws saying they should not.

As luck would have it, some work has been done on subcritical nuclear reactions. Remember that a critical mass of fissile material is considered critical at regular pressures... nukes used various means to compress a subcritical assembly to make it go foom. If you can compress a small fissile assembly enough it can achieve criticality, but the devices you need for this (massive electrical coils or lasers) are probably impractical to fit in a pocketable bomb.

Here's a thing though: antimatter catalyzed microfission. Originally thought up for pulsed spacecraft propulsion as a way to scale down the old Orion concept. According to Project Rho, the 3 gram thermonuclear fuel pellets imagined in the ICAN-II spacecraft design would be detonated by injecting them with 160 femtograms of antiprotons via a small particle accelerator. These would annihilate with some of the uranium nucleons, but the yield of the annihilation would be insignificant. Those uranium nuclei, robbed of some of their nucleons and suddenly having an excess of energy would then fission and generate a lot of neutrons, enough to trigger a reaction which could then ignite the fusion part of the pellet to deliver ~71 tonnes of TNT equivalent yield... a substantially more powerful blast. This would have been similar to the real-world W48 warhead, and there are probably some images of these being tested but I haven't managed to find out. The well-known Davy Crockett device had a mere 20 tonne yield, but there's plenty of information about that. A 71 tonne blast would reduce even quite a large mansion to rubble and leave quite a crater behind. The (non-nuclear) Operation Blowdown blast (45 tonnes TNT) might be a reasonable basis for comparison. You'd almost certainly get a mashroom cloud from this.

So anyway... if you can make a pocket-sized antimatter trap and fill it (and keep it filled long enough), you can have a micronuke work just fine. The antimatter is left as an exercise for the reader. I don't think CERN will scale down so well.


Short answer: you might be able to make a small nuke, but the technical difficulties are huge.

In an article entitled "Facts and Fallacies of World War III" in the July, 1961 edition of Popular Science magazine, the claim was made that "A californium atomic bomb need be no bigger than a pistol bullet. You could build a hand-held six-shooter to fire bullets that would explode on contact with the force of 10 tons of TNT."

This story surfaces from time to time. There are claims that the USSR made it too. Weirdly enough, most accounts refer to Californium 251, which is fairly stable. The one to go for would be Californium 253 with a half-life of about 18 days. If you had an atomic warship, it might be possible to continuously make and purify Ca253, and continuously make fresh bullets and recycle the older ones. This was believed to have a critical mass a bit over 20 grams, though no-one had tested it. A nuclear rifle bullet would replace much heavier naval cannons.

The scientists in the sixties did many mad things, but even they could not get this to work. If your roaches want to attack a house, they could use termites or wood rot, or poison the inhabitants...

  • $\begingroup$ My question was not about how realistic it was as a concept, more so how much damage it would do assuming they managed to do it. I mean this is a verse with upright talking cockroaches that may or may not be in a simulation. $\endgroup$ Commented Apr 21, 2023 at 12:29
  • $\begingroup$ Californium-251 and 252 have critical masses of a few kilos. No-one has any useful information on the 253 isotope, because it is pretty unstable, so making a critical assembly is probably impractical. Californium is extremely expensive due to the difficult of synthesizing the stuff, and the tiny quantities that can be made. $\endgroup$ Commented May 2, 2023 at 18:04
  • $\begingroup$ @StarfishPrime I used to work with nuclear stuff. The gadget I described was seriously considered in the sixties, even if it never came to anything. The only press articles I can find are clearly heavily scrambled, and the Ca253 reconstruction is my own. Love the name BTW. $\endgroup$ Commented May 3, 2023 at 11:10

You must log in to answer this question.

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