I was lucky to see the stellator construction in progress in 2011. It is easily the most amazing piece of tech I've witnessed with my own eyes. Not only it's huge, but it is also built with a stunning precision (to shape the magnetic field exactly like designed).
I took couple of more raw pictures, feel free to take a look: https://drive.google.com/folderview?id=0B41yKtv8jj2RdzdKbmxH...
If I remember my final year Gymnasium paper correctly, we've come full circle again.
The concepts for stellarators and tokamaks were invented at about the same time. Although one in the west and the other in the east.
At first, stellarators were all the rage in the west but then were replaced mostly by tokamaks. Historically, tokamaks proved to be more popular because they had simpler geometry and produced hotter plasma because of providing a better confinement. Now, with the advantages of computer simulation, designing a stellarator isn't as hard as it used to be. That's really exciting.
But that's how science works. Even if tokamaks might end up being a dead end after all, there was scientific knowledge and practical experience being gained by studying and running them which also is useful for stellarator researchers.
OTOH ITER might be an expensive dead end, but ITER is also a political project and as we all know, those often go awry.
A challenge for stellarator researchers is the complexity of the geometry, which makes theoretical models really tricky to build (there's already a lot of complexity in the 'simple' tokamak ones)
Howover, a significant pro for stellarators: magnetic geometry is already twisted (necessary condition for confinement), so it is more prone to continuous operation than tokamaks, where you have to twist the field lines (still difficult to do continuously)
These big engineering projects always leave me with feelings of awe.
It so interesting to look at the shape of the coil, nothing regular. Shapes optimized by a super computer, who knows what algorithms they've used there, what kind of search was it, how many parameters, how long did the simulation last.
I'm always amazed and impressed by the people who work on these giant projects. To do all that engineering without really knowing if it'll even work in the end must require a huge amount of confidence in both yourself and the rest of the team.
Furthermore, if the technology works eventually, you may already be retired (if not passed away!)
In my opinion, the motivation rather comes from the fascinating scientific questions behind those projects. Whether it works as an industry may be "just a plus".
Disclaimer: did a phd in tokamaks / experimental data crunching
Maybe this is pedantic but it does look like it has fivefold rotational symmetry, which makes it a lot more regular than a typical car body, for example. Still, complex geometry.
Yeah, it was probably something symmetrical in the laws that allowed them to simulate only single part of the coil - saving the computing time and lowering the search space, stacking up resulted in full coil.
I'm not sure about the laws you refer to. It seems to me that the constraint is that for a twisting ribbon-like plasma they had to have an integral number of twists or half twists in a complete circuit, not twisting too sharply, not too gradually.
> took 1.1 million working hours to assemble ... Wendelstein 7-X’s bizarrely shaped components must be put together with millimeter precision. All welding was computer controlled and monitored with laser scanners.
Sounds like a large 3d printer could really help make this more economical, assuming that all of the required materials could be printed.
The end of that first sentence you quote, "must withstand huge temperature ranges and enormous forces", kind of gives away the answer -- this isn't made from "standard" materials. 3D printing is an amazing technology that's rapidly developing, but it's not quite that far yet.
Yeah, I was thinking more in the "fusion timescale", perhaps in 15 years you would be able to print those kind of materials in molecular precision on a large scale. But let's hope it continues to advance a lot faster than fusion, so that it could help make fusion a reality.
>Approval to go ahead is expected from Germany’s nuclear regulators by the end of this month.
How could the approval of a previously unknown reactor type actually mean anything? Is it going to be something pointless like "Reactor contains no fissionable material. So OK!"?
Stellarators are hardly a "previously unknown reactor type". They were just infeasible for a long time because the computational power to compute the shapes they need to be in wasn't there yet. The concept is many decades old.
