Comment by shaqbert | Hacker News Reader

Comment by shaqbert | Hacker News Reader

Parent Full thread

Wendelstein 7-x is really not fusion research, but plasma physics research. At the energy/temperature levels and magnetic field strength this machine is operating, plasma is showing all kinds of undesirable behavior, such as turbulence, radial forces... a little bit like the storms on Jupiter. This imposes many challenges, because as soon as the plasma escapes the magnetic field and touches the vessel, it rapidly cools - ending any potential fusion.

Wendelstein's goal is to find out the viability of the stellarator concept, to see if it could be on par with the Tokamak concept, which so far have shown a better ratio of energy invested and energy won back, but come with their own bag of problems.

What the folks at Wendelstein are doing is a step by step verification of some of the hypothesis. There is this excellent 3h podcast with the scientific leader of Wendelstein [0], unfortunately it is in German. It is fascinating to hear their story on how they build this ultra complex piece of kit. The current change is the shielding of the vessel, which now permits higher energy levels and longer runs. Their long term goal is to operate at 100m K for 30 min.

Regarding the "when we should stop trying" and "it is 30 years out" adage: There has been great progress made in improving the ratio of energy invested and energy won back, the G-factor. Right now no fusion reactor is crossing the G > 1 limit. But Iter would be design to yield about G = 5. Newer designs using high temperature superconductors could even yield G > 10 with a smaller footprint. For more on the current state of fusion research, this video [1] from MIT is fantastic, albeit 1h long.

[0]: http://alternativlos.org/36/

[1]: https://youtu.be/L0KuAx1COEk?t=37m23s

almost all of fusion development is basically plasma physics research. nuclear fusion itself is known science, its the plasma containment & stabilizing that hard. so you can sustain the fusion & get positive yield.

that said I'd much prefer all the money being burnt on ITER/tokamak would be better spent by spreading the bets into exploring concepts like W7 or inertial confinement or even MIT ARC kind of efforts. Grand efforts like ITER are just ensuring fusion is always 30 years out.

My personal prediction is that we'll have fusion positive yield within a decade of when alternative energy becomes a significant portion (say 25% .. & growing) of total energy mix.

Except when stellarator starts doing longer runs and requires more hardened materials (like ITER) and when we need more instrumentation in the containment walls (like ITER) and when we need to actually plan to harvest energy from it (like ITER) and somewhere at that point someone proposes yet another concept for fusion reactor, modern and not dragged down by all that "practical" stuff. Also fusion is not 30 years away, it is likely more, after ITER we'll have DEMO, that would take many years again and only after that we may have first practical reactors running.

There really are physics that only work at very large scale, such as stars and black holes.

It may be that we can only get fusion power generation working in huge reactors like ITER. It's also a good attempt at getting many countries working together. Ideally, more big brains working together will have better ideas and results.

There's downsides to ITER too of course. But I think it's worth working on.

There are fusion reactors that are 5 cm diameter. Total size less than a microwave. They're very useful but not energy-positive, though there are reports of one that was Q=0.2, which given that that one was fridge sized was pretty good. They're the only practical way we have of producing fast neutrons.

They are critical in physics research, some medical treatments, fusion research, ... their is absolutely no even remotely practical alternative to them (they're microwave sized devices, using 4-10kw of power that replace synchotrons. Small synchotrons are basketball-field sized and need their own power station) (replace should be taken with a grain of salt since most places that have IEC fusors had no way in hell to afford a synchotron, so they are democratizing fast neutrons. Okay, that's perhaps a strong word but if you have a use for them, there's no reason why you couldn't operate one of these devices in any regular office)

(ps: given how easy, hard to detect, and deadly mistakes with fast neutrons are, please do do it in an office building at least 100m away from me. As it stands though, they're completely unregulated)

The reason we scale up is roughly:

1) Scaling laws work in favor of Q. Q should scale with something like the 3rd power of the size of the reactor. So it's much easier to build a huge Q>1 reactor.

2) Where to stick parts ? Fusion reactors require strong magnetic fields and the only real way we knew of doing that 20 years ago when these were designed was cyronically frozen. That means we need sections inside the reactor for superconductors, for crygenic cooling equipment (mostly piping).

Even disregarding that, fast neutrons will destroy any material they touch, making it brittle and crumble. Aside from bigger reactors making sure more material can get destroyed without failure, one thing that they interact with well is large volumes of water. So if at all possible, we'd like large volumes of water inside the reactor too (for other reasons too, like one strategy for extracting power). That needs space, obviously.

3) It is much easier to keep things stable if their scale is larger. There is more reaction time for the control equipment, the fields involved are larger and move slower, ...

ITER is actually a small scale model. Full scale would be DEMO, a few decades in the future :) .

Got a friend at ITAR, when I asked him how many people work there, he said about 10%

Ha, I have the vague sense that that's a joke but I'm not exactly sure what it means?

The joke is that 90% of the people aren't "working".

But how much of that G > 1 energy has ever been harvested? I get the impression that very little is being done in regards to extracting energy from a sustained fusion reaction. We can't put solar panels in there, nor can I see how one would extract heat given the cooling needs of the magnets. Is there some sort of medium that we can use to pipe heat energy out of this plasma without destroying it?

The inner surface of the Wendelstein plasma vessel is water-cooled (it has to be, or it would eventually melt), so in the future you could drive a generator from the cooling system.

There is a separate engineering problem about how to keep the plasma vessel from warming the magnets (which need be near absolute zero). In the Wendelstein this works by putting most of the device inside a vacuum, to provide better thermal isolation. So the full system is a vacuum chamber which contains the magnets which wrap around the plasma vessel which wraps around the plasma.

There is an article about the cooling system here: https://www.ipp.mpg.de/ippcms/eng/presse/pi/02_10_pi

> But how much of that G > 1 energy has ever been harvested?

None, because we haven't achieved G > 1 yet. :)

Per https://physics.stackexchange.com/questions/70209/how-is-ene..., hydrogen fusion produces helium and neutrons; neutrons are unaffected by magnetic fields, so they escape confinement and impact the reactor walls, heating them. The heat is simply vented in current test rigs, but in a working model it can be used to produce steam to power generators. That's specifically for tokamaks, but I would guess the process is similar for stellarators.

It's great how even our most advanced energy generation technologies are still just steam boilers at heart. That hasn't changed in 305 years.

I've read about some theoretical designs that would open the field to allow all negative charged particles out and use these electrons as power directly.

But first, designs would have to be able to keep the containment pressure/temperature much higher than it currently is.

https://en.wikipedia.org/wiki/Aneutronic_fusion#Energy_captu...

The Focus Fusion http://lppfusion.com/ reactor does direct conversion to electricity.

A Shame these guys are the least funded contender in this race and could do their research on a rounding error from ITAR.

Steam initially, but effect ways of converting all that heat to electricity are pretty limited at the moment.

One way to convert the heat from radioactive decay into electricity is through thermocouples (typically found in RTGs used on deep space probes and Mars exploration craft). They're incredibly inefficient though, less efficient than simply turning steam turbines.

So if the neutrons are untouched by magnetic fields, don't they also heat the magnets? I can see them passing through walls to boil water, but any heat created in the magnets would be very difficult to harvest given the temperatures involved. How much energy would it take to keep the magnets cool in the presence of a sustained (ie days) fusion reaction nearby? Generating such a flux in a bottle surrounded by things that need to be kept so cool sounds a bit of a fool's errand... but I really want practical fusion to be a thing.

Water is a good neutron absorber. Maybe the water is in front of the magnets.

Reply on news.ycombinator.com