But if we're going to talk about the 3 accidents, we should note that they are quite different.
Chernobyl was a positive pressure reactor. Only the Russians were ~~dumb~~interested in building these. Everyone else was afraid because their ability to explode and instead chose reactors that couldn't explode. The accident happened not just because of this, but because the xenon spike and not a great understanding of the science.
TMI had a bunch of things go on at once. Relief valve stuck open but indicating close. Not enough water during SCRAM. Turning off ECCS due to bad indications. Then the hydrogen explosion. Resulting in about a dental x-ray's worth of radiation for a limited number of people. We now have hydrogen absorbers and have upgraded reactors to ensure these events don't happen again. [0] (this channel also talks about all the events).
Fukushima was caused by the largest earthquake and tsunami ever recorded in the region (so lower bounding to 1k year event) and one of the largest ever recorded anywhere. The reactors were designed to withstand quakes and tsunamis. The tsunami and earthquake were much larger than anyone expected to ever happen in the region. Reactor got destroyed and it is hard to track where all the material went, which is why they have the region closed down (background radiation levels aren't so bad, but you don't want to risk consumption of reactor material. Similar to why the surrounding area of Chernobyl is closed).
The reason I bring all these up is because we should note that even Gen 2 reactors have been retrofitted with improvements from these events, and have been this way for a long time.
Indeed? I was of the understanding that the shutdown mechanisms were specifically for halting the chain reaction, and largely a solved problem with mature tech, and it was containing the decay heat of Gen 2 designs that was the trickier problem.
I am happy to be wrong about the latter, though perhaps not as happy to be wrong about the former.
I'm not particularly worried about shutdown mechanisms, but more so about containment of decay heat after an emergency shutdown event. By my understanding, our current reactor designs still produce decay heat for a time after shut down, and that heat still needs to be dealt with. And it was this decay heat and the failure to control it that lead to the Three Mile Island and Fukushima incidents.
Decay heat isn't a concern for naval reactors because 1) they're much smaller than a commercial power reactor, with order(s) of magnitude less output, and 2) they have the ocean to use as a heat sink to prevent overheating. Micro-reactors probably also have the first advantage, with outputs ranging from 1-20 MW. However, 60-1200 kW is still not as easily shrugged off when you don't have an ocean to dump heat into. IME, 60-1200kW is enough to heat steel, if not to melting, to certainly a loss of strength, and that does give me some worry about damage to containment.
So I guess my main concern is thus: Are these things being designed to be able to safely shed their worst-case heat completely passively in an emergency?
If so, awesome, and I look forward to the possibilities these designs open up.