No single nuclear weapon detonation ever hit 100,000,000 tons of TNT. There are designs for weapons in that range, and single example at 1/2 that power. However, such designs are impractical as using the same material to carpet bomb an area is more effective.

PS: Before the H-Bomb nuclear weapons where not actually cost effective per ton. Their 'advantage' over high explosives was portability and fear.

The Tsar Bomba was supposed to be in this range, but the allegedly the designer of the weapon started to fear his own creation and decided to tamper with it and tone down the yield.

https://en.wikipedia.org/wiki/Tsar_Bomba

The Tsar Bomba is a 100MT design, there's zero doubt it could achieve that yield. However, like most thermonuclear weapons roughly half of that yield would have come from fission in the natural uranium casing. Essentially every neutron from a fusion reaction has the ability to cause fissioning of U-238, and when you do all the math it works out to release as much energy as the fusion reactions.

But that would have meant that the test would have released 50MT worth of fission byproduct fallout, about an order of magnitude more than the roughly 7-8MT worth of fallout accidentally retired released in the castle bravo test. Since the USSR's test sites were on land, such a release was daunting, so instead they changed the device for the test and used an inert material for the casing. The end result was that the 50MT test was comparatively very "clean" with most of the yield coming from fusion.

I kind of remember reading that somebody involved in the creation / testing of that bomb made a rough estimate of the area covered by the fallout and decided to ditch the U-238 damper on the second (third?) stage.

Also, getting the plane dropping the bomb to a safe distance before the explosion was kind of an act as it was, if the bomb had gone off at twice the yield, it probably would have killed the crew.

(All this is from rather fuzzy memory, though! If somebody has less fuzzy information, I'd be happy to stand corrected!)

From my own fuzzy memories that's correct - a full 100MT device would have been a pretty awful weapon, I've seen it described as something that could have utterly destroyed Belgium but that would have killed people in Russia with fallout.

I seem to recall that the Tsar Bomb had multiple tertiary stages?

Nitpick: it's tamper not damper.

Also without a fissioning tamper round it's secondary and tertiary stages the Tsar Bomb actually was a "fusion bomb" in that a very high proportion of its yield came from fusion. In most H-bomb designs most of the yield comes from the fissioning of the tamper round the secondary.

With bombs of this type you can chain things on several times and really amp up the yield. The theoretical limit, if there is one, is absolutely nuts, like 500MT or more.

The problem is that beyond a certain yield most of the explosive force punches up through the atmosphere and is wasted. Apart from shock and awe, not sure what the utility of something that big would ever be.

Additionally since a spherical explosion of double the diameter takes eight times as much energy to create, you quickly run into diminishing returns. Lots of smaller bombs is more effective, hence MIRVs: https://en.wikipedia.org/wiki/Multiple_independently_targeta...

Nuclear weapons are so awesome and so utterly insane at the same time.

One pedantic correction is that the term is 'tamper' not 'damper'.

No idea why you have been downvoted

Surely the pre-thermonuclear bombs also had a big advantage in ease of delivery? One bomber could do what previously took a thousand, which means that a bomber fleet could destroy a bunch of cities simultaneously, instead of hitting a couple of cities per week as in the Second World War.

The super-large bombs weren't entirely impractical. Hitting an area with multiple smaller bombs is indeed far more effective, if you can accurately deliver a bunch of smaller bombs at once. There was a period where that wasn't a given, and they tried to make up for inaccuracy with explosive power. The 100Mt bomb was never deployed (or, indeed, built other than the one de-rated test article) but both the US and USSR deployed 25Mt bombs.

Carpet bombing a larger area means you need even less precision. Picture throwing a 10 darts at a dart board with a 2 foot radius, vs 1 dart with a 4 foot radius.

Still, even at half design yield the Tsar Bomba had an instantaneous power output equal to 1.4% of the sun, albeit for a few nanoseconds at most.