[raydiatian]

One thing I don’t understand about fusion is the mechanics of gain factors. If we can achieve a fusion reactor with a very small gain factor of say 1.01, is that sufficient to kick off an energy revolution, or do we need something more extreme like 10x or 100x?

I suppose it boils down to what the “saturation threshold” of nuclear reactors is, where you can’t pump more energy in without breaking the thing.

In any case, what are the benchmarks that engineers are shooting for?

[XorNot]

Q=10 is believed to be necessary to make it economically viable.

[raydiatian]

Is there any insight into how long a road from 1 to 10 it is?

[XorNot]

So the assumption is that once you exceed 1 with some reactor architecture, you've more or less got it provided you can build a bigger containment vessel.

ITER is as large as it is because within reasonable magnetic field strengths, you still need a certain amount of distance to curve the charged particles back towards the centre - and probability means you have a distribution, some but not all will make it. Bigger vacuum vessel, the more particles you can loop back into your reactor.

So, if we can provably get over 1, and validate our plasma behaviour models, we can then derive the correct engineering equations to target a specific Q factor when building the production type reactors.

[xupybd]

So we might have to go much bigger than ITER?

[XorNot]

Yes. The ITER successor is DEMO which is larger and meant to be the production power plant design. The idea is ITER proves out the plasma control physics and scaling laws for fusion, so you can then build DEMO, get the predicted results and say "right, this is how a fusion plant looks". ITER will do Q=10 if we're right anyway, but the scaling laws are all volume based: the real trouble with fusion is that building gigantic vacuum vessels is very hard to do, but we're at the limits of magnets (unless MIT can get HTSCs working, but even then, you still benefit from volume).

[1] https://physics.stackexchange.com/questions/175830/nuclear-f...

[DennisP]

Tokamaks scale with the square of plasma volume but the fourth power of magnetic field strength. The REBCO used by CFS should let them get ITER-level output from a reactor a tenth as large.

[willis936]

I like to show this graphic. Q>1 is in the "we don't know but we have a good idea" territory. Self-heating (burning) plasma is an untested regime. That will change soon.

https://twitter.com/jb_fusion/status/1506964692627034118

[DennisP]

Not too long, given scaling laws. In tokamaks for example, fusion output is proportional to the square of plasma volume, and the fourth power of magnetic field strength.

[nukemandan]

The lifetime cost being on par will. No power factor in itself will do that.