[cyberax]

> ITER doesn't use high temperature superconductors.

It does, for high-current buses that interface with regular resistive power distribution. They are also planned for some auxiliary components (like the neutral beam injectors).

> ITER has been criticized since early days as a dead end, for example because of its enormous size relative to the power produced.

ITER is NOT designed for power generation. It's essentially a lab experiment to see how plasma behaves in magnetic confinement and test various technologies.

That's why ITER was designed with a very conservative approach to reduce the technical risk. We don't need it to be compact, this can come later. We just need it to work.

And yes, it is necessary. Plasma behavior can't be simulated numerically or analytically. It always provides surprises, sometimes even good ones: https://en.wikipedia.org/wiki/High-confinement_mode

[pfdietz]

> ITER is NOT designed for power generation. It's essentially a lab experiment to see how plasma behaves in magnetic confinement and test various technologies.

That's the go-to excuse. But if you look at DEMO, it's power density is not enormously greater. ITER is so far out of the running that DEMO (or PROTO, etc.) will be too.

We're learning a great deal about something that's largely irrelevant.

[cyberax]

DEMO concept sketches are completely obsolete at this point. It's not going to look anything like this.

They're based on the state-of-the art from about 2005. Since then, a lot of improvements happened. A more realistic power plant design is going to use a thinner center column (because of better superconducting magnets), resulting in a smaller cryostat volume. Possibly high-TC magnets.

It can also be made more compact, if neutral beams can be used to suppress some plasma instabilities.

[Calwestjobs]

misunderstanding about ITER what you some people doing is that it is just one thing, it is not.

ITER is not only facility in france it is multitude of manufacturing capabilities all over the globe which build parts for ITER and all future power plants.

[pfdietz]

It's my understanding that neutron wall loading of DEMO concepts had been trending downward (due to materials limits), the opposite of the trend you're trying to portray there. And in no future world is the power density of DEMO going to be anywhere close to that of a fission reactor.

[cyberax]

Neutron loading is not the limiting factor (for now), it's the magnetic field pressure and its homogeneity. That's actually what is driving the humongous size of the ITER.

To solve the homogeneity problem, you need the central column to be as thin as possible. That's where a lot of the recent advancements can help.

Also, fission research and fusion are actually aligned in designing materials that can tolerate more displacements per atom.

> And in no future world is the power density of DEMO going to be anywhere close to that of a fission reactor.

That's for sure. Modern fission reactors are close to magic, with the amount of heat they produce for a given volume.

[pfdietz]

However, if neutron loading has gone down, and the reactor is the same size, the volumetric power density must also have gone down.

The comment about fission and neutron dpa is misleading. The neutron damage issue is much less bothersome in fission reactors.

Fission produces about 3% of its energy in neutrons, vs. 80% in the DT fusion reaction. The spectrum of fission neutrons is much softer, with a peak around 1 Mev, vs. 14 MeV for DT neutrons. The DT neutrons are above threshold for (n,2n) reactions in most materials, and have much higher cross section for (n,p) and (n,alpha) reactions. The latter is particularly troublesome, as helium accumulates inside materials, forming microscopic very high pressure bubbles that rip the materials apart.

But it's even comparatively worse for fusion than that. In a PWR (for example), the core is carefully designed so that the only parts exposed to unmoderated neutrons are the fuel rods and the replaceable parts of the fuel rod bundles. The latter provide structural support for the fuel rods and are removed along with the fuel rods when the fuel is spent. The actual core supports for the fuel bundles are well away from where the chain reaction is occurring, shielded by water. The mean free path of a fission neutron in water is just a few centimeters, so their energy is quickly dissipated before reaching these components.

So, exposure of permanent reactor components to fast neutrons is essentially a non-issue in PWRs. Even control rods are not exposed much; reactivity is controlled by boric acid dissolved in the water (BWRs do it somewhat differently.)

This same strategy cannot be used in a fusion reactor; the plasma facing surfaces are exposed to the full, unshielded brunt of the DT neutron flux. Maybe a few cm of liquid lithium could be flowed along some surfaces? This is a stretch, particularly in a toroidal reactor.

[adgjlsfhk1]

is a conservative approach useful if it is taking 30 years to build?