Serious Scientific Answers to Absurd Hypothetical questions

[TheSaint]

2 hours ago, ARS said:

Would it be possible for gunpowder-era civilization to make a working nuclear bomb by using a cannonball-sized ball of uranium encased by layer of explosives (To force fission) connected to a fuse? How powerful it would be around it's size?

Like this one?

(For added fun, throw it from trebuchet as a makeshift tactical nuke)

It would be much more plausible for them to use the uranium and gunpowder to make a gun-type warhead, which is far simpler to engineer than an implosion-type warhead.

But, yeah, enriching uranium with medieval technology would be the very definition of impossible.

[DerekL1963]

4 minutes ago, TheSaint said:

It would be much more plausible for them to use the uranium and gunpowder to make a gun-type warhead, which is far simpler to engineer than an implosion-type warhead.

If by "much more plausible" you mean "still laughably implausible", sure.   Gunpower of that era, essentially black powder, isn't going to generate enough force to propel the masses together fast enough...  It'll fizzle.  But the real problem, as @DAL59 pointed out, is enrichment.  That's all but impossible in the gunpowder era.  They might be able to get a thermal diffusion plant running - if they can figure out how to handle fluorine.  But it's still going to require enormous inputs of conventional fuel.

[ARS]

Might be a silly question, but I wonder what's the feasibility of using conventional heavy aircraft design as an SSTO (say a B52 or An-225's fuselage, tail, wings and engine layout)? How effective it would be in aerodynamic standpoint during ascending, orbiting and reentry? (Disregard engine power, assume the engine is able to carry that plane to orbit)

[Gaarst]

24 minutes ago, ARS said:

Might be a silly question, but I wonder what's the feasibility of using conventional heavy aircraft design as an SSTO (say a B52 or An-225's fuselage, tail, wings and engine layout)? How effective it would be in aerodynamic standpoint during ascending, orbiting and reentry? (Disregard engine power, assume the engine is able to carry that plane to orbit)

Absolutely not effective at all.

In KSP you don't care about the wings shape or the sturdiness of a plane, it pretty much works whatever you do, looks usually don't impact performance that much. In real life it's pretty much the complete opposite: the requirements for subsonic and supersonic flight are very different (and hypersonic is something else altogether) and a cargo plane optimised for subsonic cruise would not be able to fly at supersonic speeds (mostly because it would fall apart before crossing Mach 1). Supersonic flight requires specific designs and materials to fly efficiently (see how different the Concorde is from any other airliner), and the faster you go, the more extreme these requirements get.

Another problem is fuel efficiency: LEO requires over 9 km/s of dV. For an SSTO you want to optimise every possible aspect of your craft to squeeze the necessary dV, altogether an SSTO capable craft is built very differently from any other aircraft.

[DDE]

@ARS @Gaarst There’s this curious case of Faget’s completely straight-winged DC-3... so curious I can’t really comment on it.

[p1t1o]

14 hours ago, ARS said:

Would it be possible for gunpowder-era civilization to make a working nuclear bomb by using a cannonball-sized ball of uranium encased by layer of explosives (To force fission) connected to a fuse? How powerful it would be around it's size?

<snip>

Like this one?

(For added fun, throw it from trebuchet as a makeshift tactical nuke)

The range of assumptions and conditions needed to bring this about without exporting a fully-realised 20th century knowledge base and the required materials is....a stretch, but I'll go with it

 

Whether or not you could get enough energy from a gunpowder explosion is one for the mathematicians - but it does seem doubtful, the speed of detonation is very much slower than a true high explosive and the energy density much lower. It is very possible that it just takes too long for gunpowder to explode for it to even "fizzle" (a nuclear, but incomplete, detonation) and you'll probably just get some shrapnel and a hot ball of plutonium.

The gross design concepts of a implosion fission device are not that esoteric, but there are difficulties that are not at first obvious:

The timing of the detonations around the sphere must be very exact to the microsecond, its very unlikely that this could be achieved with gunpowder fuses.

The explosives are not just "in layers", in an implosion device there are two types of explosive, with tow detonation speeds, and blocks of the two types are machined into precise shapes (often termed "lenses" as the geometry focuses a wave) so that the outwardly expanding spherical shock fronts from all of the seperate detonation points smoothly merge into a single spherical, inwardly converging shock front. The precision required to manufacture this (and the mathematics required to figure it all out) are very probably beyond gunpowder era technologies.

 

An implosion device is a precision made machine, with very many parts and many esoteric operating principles. The general diagram - fissile core, layers of explosive, casing and fuses, is elegant for sure, but deceptively simple.

Even the core, in a mature device, has many seperate components, to maximise efficiency/yield and minimise chance of failure (for example, your design omits a tamper and an initiator).

A modern, fully-featured weapon is among the most highly advanced pieces of machinery on the planet.

 

A poorly built device with insufficient precision could have any effect from a totally non-nuclear detonation of just the explosives, to a "dirty" explosion which spreads fissile material, to a "fizzle", a partial nuclear detonation which could have a yield almost anywhere in the range from 0% to 100% intended. With gunpowder-era precision (and, well, gunpowder) I would wager it is strongly skewed towards the first example (non-nuclear).

This is why there is little danger in declassifying the gross details of a fission devices design - its not the physics or the design which is prohibitively difficult, its the manufacture and acquisition of the materials.

 

A gun-type device is much easier to build, and seeing as little-boy was fired with cordite (essentially modern gunpowder) I dont see why you couldnt build one to use gunpowder if a similar barrel velocity can be obtained. Precision requirements are far less. There are some details about the design of the core they'd need to know, but the main design parameter is barrel velocity, there is a minimum that needs to be met to avoid a "fizzle".

Not sure about that though - anyone know the differences in performance between cordite and gunpowder?

 

Edited November 6, 2017 by p1t1o

[James Kerman]

39 minutes ago, p1t1o said:

Not sure about that though - anyone know the differences in performance between cordite and gunpowder?

You can pack a musket barrel full of black powder and fire it with no ill effects besides a sore shoulder and sore ears (if you didn't use protection).
Try the same thing with modern smokeless powder and you wont have hands or a face.

Black powder cant burn through all the grains in the barrel fast enough and a large amount of it gets ejected unburnt.

Edited November 6, 2017 by James Kerman
Edited for more information.

[YNM]

16 hours ago, ARS said:

Would it be possible for gunpowder-era civilization to make a working nuclear bomb by using a cannonball-sized ball of uranium encased by layer of explosives (To force fission) connected to a fuse? How powerful it would be around it's size?

Like this one?

(For added fun, throw it from trebuchet as a makeshift tactical nuke)

If it has critical mass, the encapsulation itself is more than enough.

The big problem with nuclear devices are enriching. I don't think they could make enough gas centrifuge and powering all of it up.

The only use of common explosives etc. in nuclear weapons is only to induce criticality, by closing a previously open gap by either compressed gas or sometimes a solid.

Edited November 6, 2017 by YNM

[p1t1o]

1 hour ago, James Kerman said:

You can pack a musket barrel full of black powder and fire it with no ill effects besides a sore shoulder and sore ears (if you didn't use protection).
Try the same thing with modern smokeless powder and you wont have hands or a face.

Black powder cant burn through all the grains in the barrel fast enough and a large amount of it gets ejected unburnt.

What about muzzle velocity? Did a bit of searching and the velocity was only about 300m/s, that's achievable with gunpowder right?

 

2 minutes ago, YNM said:

If it has critical mass, the encapsulation itself is more than enough.

The big problem with nuclear devices are enriching. I don't think they could make enough gas centrifuge and powering all of it up.

If you have a solid sphere of a critical mass, it has already detonated.

The challenge is creating a weapon that is sub-critical when not in use, and super-critical on command.

***

Critical mass is dependant on geometry, a critical mass when a solid sphere can be made sub-critical by forming it into a hollow sphere - this is a dangerous design (but WAS used in several service weapons) however, as accident could easily make it critical.

Modern weapons use a solid sphere of sub-critical mass, which is compressed violently enough to crush the sphere to a denser crystalline phase, thus pushing it into a smaller, super-critical sphere.

Some compact designs are rumoured (thats all we've got) to use a "prolate" or "lozenge-shaped" core using a more advanced implosion system.

[wumpus]

30 minutes ago, YNM said:

_{The big problem with nuclear devices are enriching. I don't think they could make enough gas centrifuge and powering all of it up.}

For values of "gunpowder" that are extremely close to the development of nitroglycerin, extremely large centrifuges aren't a problem, but you are likely stuck with liquids (and probably not flourine).  If there's one thing steam did right, it was building things *big*.  Of course, since the end result was a few kg of U₂₃₅ I'm not sure that you could get it done (especially not without multiple criticality accidents).

I'm also not convinced that you can't use multiple gunpowder explosions to get maximum speed of you uranium.  You will still be limited to subsonic speeds, but it should be going relatively fast.  You might also have to use gunpowder to propel both sides.

Like all steam-era tech, this is going to be *big*.  I'm guessing you build the thing into the keel of a ship, then send it like an old-school "fireship" to detonate in a hostile port.

I think the OP needs to read/watch more James Burke's Connections and other works.  Even if someone in 1850 suddenly received a "complete guide to building a nuclear weapon", they would continually be running into problems of "insert unobtanium here".  Attempting to enrich uranium only knowing the mass difference between U₂₃₈ and U₂₃₅ just isn't going to let a victorian engineer build a reactor.

[DerekL1963]

27 minutes ago, p1t1o said:

What about muzzle velocity? Did a bit of searching and the velocity was only about 300m/s, that's achievable with gunpowder right?

Retracting what I said above, it does look like 300m/s is achievable.  However, due to gunpowder's slow burn rate I suspect the weapon will be much larger than Little Boy as the projectile has to reach full insertion speed before it reaches the criticality zone.   Because both portions of the core have to be near criticality for any reasonable efficiency, that zone extends some distance up the barrel. In the case of Little Boy, the core went critical while the projectile was still 25cm from the target.  (The reference is unclear as to whether that measurement is from face-to-face or from face-to-fully seated.)

[p1t1o]

Oh I have a question - is there a formal definition of "gunpowder era"?

When I think "gunpowder era" my mind goes to flintlocks and muzzle loading cannon, so like around 1600ish?

But cordite was not invented until 1889 and some sources have mentions of gunpowder-like things dating back to like 500AD.

Thats a wide period. It is going to be easier to build a nuke in 1888 than in 1088 after all. 1888 only predates nukes by a handful of decades, easily within a single lifespan.

In fact, the "gunpowder era" ends after the invention of high explosives, so a legit nuke could (if some key research was done just a single generation earlier, much of the groundwork was already present by this time) have been built technically within the gunpowder era. Its not that much of a stretch to imagine a Manhattan-style crash project could have been pulled off just a few decades early.

Its not quite as romantic as lobbing spiked, cast-iron tac-nukes with a trebuchet (gonna need to up-rate those...) but still.....one could have had a cowboy with a nuke on his back, how's that? Or Sherlock Holmes (first appearance 1887) could have had one in his basement.

 

Edited November 6, 2017 by p1t1o

[Green Baron]

16 minutes ago, p1t1o said:

Oh I have a question - is there a formal definition of "gunpowder era"?

 

I think it was invented in the 13th century in Europe accidentally when searching for a way to turn charcoal into gold (those alchemists :-)). It appears earlier in China, but afaik not BC.

It is still used today in fireworks, together with salts for colouration, and in traditional musket shooting, where people cast their own bullets and fabricate their own charges for the front-loading guns and do things like shoot at a lake surface to hit a target on the side (ricochet, but not accidental).

Edited November 6, 2017 by Green Baron

[p1t1o]

On 11/4/2017 at 1:07 PM, Earthlinger said:

If you could magnetize the hull of a tank so that it had an extremely negative charge, and somehow magnetized incoming bullets so that they acquired a similar negative charge, would this essentially create a sort of repelling-thingy force shield?

The like charges would repel, and the bullets would slow down the closer they got to the hull. With a strong enough charge, they might even be stopped completely and flung in the other direction.

Well first off, you kinda answered your own question, of course it would work. Its just a question of magnitude and whether it could be made to be sufficiently practical for use.

You've mixed up magnetic and electric charges there I think, and magnets wont work as well because all (non-exotic) magnets have to have both poles, but a negatively charged bullet would be repelled by a negatively charged tank, and there is a magnitude which would produce enough force to stop a bullet for sure. But it might also have many other, less desirable effects.

Your version probably isnt practical, but its not a ludicrous idea at all:

https://en.wikipedia.org/wiki/Reactive_armour#Electric_reactive_armour

 

**edit**

Now that I think about it, there IS a magnitude of charge - theoretically - that could stop a like-charged bullet, but Im not sure if it is possible to GIVE an object that much charge. The only way I know to charge a physical object is to manipulate the population of electrons present in some manner or other. Strip some off to give it positive charge, squeeze some more in to give it a negative charge. But electric repulsion or attraction is very powerful. At some point, the negative charge is enough that electrons themselves are repelled and they just...ping off. Similarly, strip enough away, give something enough of a positive charge, and it will simply rip electrons off surrounding materials.

 

Edited November 6, 2017 by p1t1o

[ARS]

The maker of that picture claims that this is a medieval-era primitive tac nukes. Initially I'm just thinking "whoa really? Medieval? That's too primitive for something like this (relatively) simple", so I uplifted the question into gunpowder era. But then after reading the answers then I understand it would most likely goes fizzle and not gonna have a full-blown nuclear explosion due to the primitive and imprecise nature of earlier gunpowder, not to mention the difficulty in enrichment of uranium

But then, back in the design. If we could built that nuke design with a technology several step forward (say WW2 weapon tech, where the first nuclear bomb were used), could it explode into a proper nuclear explosion? At that size (2,5 m ball), how powerful an approximate power of it's explosion in kiloton? (Not sure if it reaches megaton, I doubt)

[p1t1o]

5 minutes ago, ARS said:

The maker of that picture claims that this is a medieval-era primitive tac nukes. Initially I'm just thinking "whoa really? Medieval? That's too primitive for something like this (relatively) simple", so I uplifted the question into gunpowder era. But then after reading the answers then I understand it would most likely goes fizzle and not gonna have a full-blown nuclear explosion due to the primitive and imprecise nature of earlier gunpowder, not to mention the difficulty in enrichment of uranium

But then, back in the design. If we could built that nuke design with a technology several step forward (say WW2 weapon tech, where the first nuclear bomb were used), could it explode into a proper nuclear explosion? At that size (2,5 m ball), how powerful an approximate power of it's explosion in kiloton? (Not sure if it reaches megaton, I doubt)

Well the "Fat Man" device was 1.5m in diameter, almost all of it weapon, and its yield was roughly 20kt.

Its not a simple question at all to try and extrapolate that up to 2.5m, but you get an idea of the ball park - its not going to be a Megaton, we can say that much. Reasonable guess would be anywhere within a range from 30-100kt, gut feeling.

Just for reference, in 1959 the RAF deployed a very large fission-only device that was about the same size as Fat Man but with a 400kt yield, and a modern thermonuclear warhead can do 475kt in a device 21inches across (but a bit longer) - it is said that the diameter [of modern thermonuclear weapons] is largely set by the diameter of the fission primary, which will have (very roughly) 5-10% of the final yield.

[magnemoe]

20 hours ago, DerekL1963 said:

If by "much more plausible" you mean "still laughably implausible", sure.   Gunpower of that era, essentially black powder, isn't going to generate enough force to propel the masses together fast enough...  It'll fizzle.  But the real problem, as @DAL59 pointed out, is enrichment.  That's all but impossible in the gunpowder era.  They might be able to get a thermal diffusion plant running - if they can figure out how to handle fluorine.  But it's still going to require enormous inputs of conventional fuel.

You have to go for an little boy setup obviously, you might be able to use it with blackpowder its not so much weaker than cordite, various tricks you could use like an larger bore and longer barrel. But yes you would need chemistry and metallurgy of another level to separate u235, also lots of electricity. 
Now could you do it for WW1, assuming nuclear theory was more advanced but nothing else. Say you knew about chain reactions in 1905 and build an test reactor in 1910. 
Various greater powers saw the potential for an nuclear bomb to destroy fortifications. 

[DerekL1963]

5 minutes ago, magnemoe said:

Now could you do it for WW1, assuming nuclear theory was more advanced but nothing else. Say you knew about chain reactions in 1905 and build an test reactor in 1910. 

Maybe, maybe not.   It's not just nuclear theory you need, but chemistry and metallurgy.  (And a way to deliver the weapon - something were just barely capable of doing in WWII.)

[Zeiss Ikon]

10 hours ago, Gaarst said:

In real life it's pretty much the complete opposite: the requirements for subsonic and supersonic flight are very different (and hypersonic is something else altogether) and a cargo plane optimised for subsonic cruise would not be able to fly at supersonic speeds (mostly because it would fall apart before crossing Mach 1). 

Well, this is just plain wrong, at least at a basic level.  One of the early jet airliners (DC-8, as I recall, basically the MacDac answer to the Boeing 707) was routinely flight tested (as in, every single airframe) in a shallow dive to (IIRC) Mach 1.05, and virtually every other airliner built by Boeing, MacDac, Airbus, and whoever else since, say, 1960 had airframes that were not Mach limited -- that is, they have no grossly undesirable flight characteristics as shock waves build up over flying surfaces through the transonic regime.  None of them (until Concorde and the TU-144 -- the Boeing version never flew) could exceed Mach 1 in level flight, but any airliner that routinely operates above Mach 0.85 (which is virtually all the big players since the early 1960s -- let's say any large transport with jet engines) can exceed Mach 1 without harm.

As a counter-example, the Me-262 and Me-163 were Mach-limited airframes (despite both being extremely fast -- faster than any propeller fighter, ever).  Due to lack of knowledge about transsonic flight, both had designs that promoted compressibility problems like control reversal, stabilizer blanketing, and so forth.  The Me-163 was accidentally flown into this regime once, after igniting its rocket engine at altitude (towed up like a glider) instead of using almost all the rocket power to climb to bomber interception altitude -- the pilot lived through the incident because the tuck-under that occurred due to control reversal killed the rocket engine and the aircraft slowed to subsonic due to drag.  Both operated under flight rules prohibiting power dives.

Quote

Supersonic flight requires specific designs and materials to fly efficiently (see how different the Concorde is from any other airliner), and the faster you go, the more extreme these requirements get.

Well, yes, efficient is very much different from possible.  Even though a B-52 can exceed Mach 1 in a shallow dive at full throttle, and remain controllable, it would surely come apart at higher Mach numbers -- those long, thin wings are very flexible (ground and air photos will demonstrate how much so), and the controls aren't designed to compensate for that (modern fly-by-wire could do it, but would require additional surfaces, and B-52 doesn't have this technology).  The result would be surface flutter, which defines V_ne for most airframes built for subsonic flight.

Overall, though, a Concorde isn't any stronger than a 737 -- it's just better optimized for high speed, and has much more power.

if you had a power source that would let you climb at near Mach 1 in, say, a 767, you'd eventually reach an altitude where the fastest you can go is still too slow for the wings to support the weight.  With a Mach limited airframe, like a U-2, this condition was referred to as "flying in the coffin" -- you mustn't exceed Mach 0.95, but you must stay above stall speed.  Yet, at around 80,000 feet, stall speed exceeded the Mach limit of the airframe, and simply lowering the nose would cause enough speed increase to encounter undesirable/unrecoverable Mach effects.  To come down from that regime, the pilot had to very carefully reduce the throttle and manage airspeed and angle of attack to avoid stalling as he eased down from maximum altitude and Mach limit.

BTW, for those discussing the medieval nuke, it's worth mentioning that there were well known formulae for detonating substances as far back as the 15th century.  I wouldn't want to have anything to do with trying to manufacture and handle significant quantities of those (fulminates, mainly, generally sensitive to friction, shock, and heat, and what explosives handlers call "primary explosives" in that they detonate without any additional help), but they'd provide far more shock than anything else until the invention of nitroglycerin in the 19th century.  Given a high-enrichment fissionable, and the knowledge, I don't see it as impossible for a 15th century alchemist to assemble a nuke (or blow himself and half his city to hell trying).  If they have to start from natural uranium, however, they'd never get there.  As I recall, critical mass is measured in tons at natural isotope ratio.

[DDE]

11 hours ago, p1t1o said:

Your version probably isnt practical, but its not a ludicrous idea at all:

https://en.wikipedia.org/wiki/Reactive_armour#Electric_reactive_armour

Oh, hey, my most beloved canard! How much have I had to deal with energy shield proponents on the Space Engineers forums citing it.

First of all, from my best determination, it works by being a huge capacitor, huge enough to completely evaporate the kinetic penetrator or HEAT hypersonic jet that connects the inner and outer plates. Good luck with building something like that.

Second of all, it doesn’t exist outside the two Telegraph artciles a deacde apart claiming it to be in a highly experimental stage.

[mikegarrison]

9 hours ago, Zeiss Ikon said:

Well, this is just plain wrong, at least at a basic level.  One of the early jet airliners (DC-8, as I recall, basically the MacDac answer to the Boeing 707) was routinely flight tested (as in, every single airframe) in a shallow dive to (IIRC) Mach 1.05, and virtually every other airliner built by Boeing, MacDac, Airbus, and whoever else since, say, 1960 had airframes that were not Mach limited -- that is, they have no grossly undesirable flight characteristics as shock waves build up over flying surfaces through the transonic regime.  None of them (until Concorde and the TU-144 -- the Boeing version never flew) could exceed Mach 1 in level flight, but any airliner that routinely operates above Mach 0.85 (which is virtually all the big players since the early 1960s -- let's say any large transport with jet engines) can exceed Mach 1 without harm.

There is a reason why airplanes have a speed known as M_MO. That's not to say the plane will instantly fall apart if it exceeds that speed, but it it not allowed because of the Mach buffet problems. Mach buffet counts as "grossly undesirable flight characteristics".

According to this article https://www.airspacemag.com/history-of-flight/i-was-there-when-the-dc-8-went-supersonic-27846699/ the DC8 was put into a supersonic dive one time as a publicity stunt, and it was rather hairy trying to recover control enough to pull it back out of the dive. There is absolutely no reason that they would "routinely flight test" this on every airframe. What's your source for this?

Edited November 7, 2017 by mikegarrison

[p1t1o]

2 hours ago, DDE said:

Oh, hey, my most beloved canard! How much have I had to deal with energy shield proponents on the Space Engineers forums citing it.

First of all, from my best determination, it works by being a huge capacitor, huge enough to completely evaporate the kinetic penetrator or HEAT hypersonic jet that connects the inner and outer plates. Good luck with building something like that.

Second of all, it doesn’t exist outside the two Telegraph artciles a deacde apart claiming it to be in a highly experimental stage.

Oh it hypothetical for sure, what I meant was that "electric armour" has been seriously considered as a possibility, as far as made-up supertechnologies go, its more plausible than some. It doesnt require free neutronium for example

 

[Green Baron]

Airliners are not "routinely" tested at supersonic flight. If you browse the web you'll find out that trying so would most probably have fatal consequences. Engines choke out, profiles stall, control surfaces loose flow and fly off, frames get overloaded. Pick one (if you had enough power to cross the sound barrier, which an airliner does not have). Vne/Mmo is usually around Mach 0.9.

Edit: Vne = never-exceed-speed ... or exceed-once-speed :-)

Edited November 7, 2017 by Green Baron

[YNM]

21 hours ago, p1t1o said:

If you have a solid sphere of a critical mass, it has already detonated.

The challenge is creating a weapon that is sub-critical when not in use, and super-critical on command.

***

Critical mass is dependant on geometry, a critical mass when a solid sphere can be made sub-critical by forming it into a hollow sphere - this is a dangerous design (but WAS used in several service weapons) however, as accident could easily make it critical.

Modern weapons use a solid sphere of sub-critical mass, which is compressed violently enough to crush the sphere to a denser crystalline phase, thus pushing it into a smaller, super-critical sphere.

Well, the "demon core" criticality is enough just by closing it completely in neutron reflector. So it really depends on what they're making.

For shock compression/implosion, I think it really depends on the explosives used. But again, I don't think that's the main problem back then...

21 hours ago, wumpus said:

For values of "gunpowder" that are extremely close to the development of nitroglycerin, extremely large centrifuges aren't a problem, but you are likely stuck with liquids (and probably not flourine).  If there's one thing steam did right, it was building things *big*.  Of course, since the end result was a few kg of U₂₃₅ I'm not sure that you could get it done (especially not without multiple criticality accidents).

I'm also not convinced that you can't use multiple gunpowder explosions to get maximum speed of you uranium.  You will still be limited to subsonic speeds, but it should be going relatively fast.  You might also have to use gunpowder to propel both sides.

Like all steam-era tech, this is going to be *big*.  I'm guessing you build the thing into the keel of a ship, then send it like an old-school "fireship" to detonate in a hostile port.

I was thinking the device would be more like sneak-bomb (so put somewhere then move then detonate). The timing probably depends on a mechanical device, probably some acid-on-wire thing. I was thinking radio but I wonder how strong the radio would have been.

[p1t1o]

13 minutes ago, YNM said:

Well, the "demon core" criticality is enough just by closing it completely in neutron reflector. So it really depends on what they're making.

There are some complexities to the terms here, I think I made the same mistake, a critical mass is not necessarily a nuclear detonation. The demon core incidents were "criticality" incidents in that they were pushed to a "critical" state where the power output (and neutron production) was rapidly increase dot a high level.

A nuclear detonation is a "supercriticality" which is what we recognise as a runaway chain reaction, which is only halted by the destruction of the device. The exponentially rising "supercritical" state requires somewhat more severe conditions than those created in a laboratory experiment in order to release the sort of energies involved in a detonation, but a steady criticality can be obtained and brief supercriticality will result in the apparatus dismantling itself.

Enclosing a sub-critical mass in a neutron reflector changes the conditions present and makes sub-critical mass into a critical one, this will reach a steady, critical, state of high power output, and if it is rapid enough it will explosively deconstruct and halt the reaction.

Almost the entire structure of a fission device is designed to contain the reaction for long enough, and to make the reaction proceed fast enough, at the exact right time, that a significant amount of energy is released before the destruction halts the reaction. That is why you could never get a true nuclear detonation merely from creating a critical mass by enclosing a subcritical one [with a neutron reflector].

The "Lady Godiva device" (essentially used for the same experiments as those performed with the demon core) -  The 3 parts form a critical mass, note: no neutron reflector required. Before and after a criticality incident, the right picture shows the damage caused: