> Theory dictates that with stronger magnets, the reactor can be scaled down (with the square/cube, can't remember exactly), and thus cost and time to develop
Here's the quick summary:
B: magnetic field strength
R: length scale
Fusion rate ∝ (plasma pressure)^2 ∝ B^4
Energy gain (Q) ∝ R^1.3 B^3
Power density ∝ R B^4
Cost ∝ R^3
So, say for example you're targeting a fixed Q. Doubling the magnetic field strength results in R1 = R0 / 2^(3/1.3) = 0.2 R0. And 0.2 R0 translates to 1/(0.2)^3 = 0.008 = 0.8% the cost.
The scaling is absolutely insane, and a stronger magnetic field has other advantages (such as making plasma instability far less of a concern), though structural loads can be an issue (that, at least is a relatively straightforward engineering problem).
If you take 12T for ITER and 20T for SPARC, that's not actually 2x, it's 1.67, which translates to 30% the size and 3% the cost (and time). It should also be noted that this is just rough, order-of magnitude estimation, but it should be broadly accurate.
> It's even more subtle than that, in fact this is really one of the things we've studies at MIT, is that there's other things that come in terms of benefits, particularly when you make the magnetic field very high, it basically starts to tame, just all of the whole suite of plasma instabilities that exist.
Here's the quick summary:
B: magnetic field strength
R: length scale
Fusion rate ∝ (plasma pressure)^2 ∝ B^4
Energy gain (Q) ∝ R^1.3 B^3
Power density ∝ R B^4
Cost ∝ R^3
So, say for example you're targeting a fixed Q. Doubling the magnetic field strength results in R1 = R0 / 2^(3/1.3) = 0.2 R0. And 0.2 R0 translates to 1/(0.2)^3 = 0.008 = 0.8% the cost.
The scaling is absolutely insane, and a stronger magnetic field has other advantages (such as making plasma instability far less of a concern), though structural loads can be an issue (that, at least is a relatively straightforward engineering problem).
If you take 12T for ITER and 20T for SPARC, that's not actually 2x, it's 1.67, which translates to 30% the size and 3% the cost (and time). It should also be noted that this is just rough, order-of magnitude estimation, but it should be broadly accurate.
For a more detailed explanation: https://youtu.be/KkpqA8yG9T4