Stellarator class fusion reactor

[PB666]

 

The design is totally something.

Edited December 19, 2015 by PB666

[Scotius]

Yes. But that "something" can not be named without breaking savoir-vivre and forum rules

[musicpenguin]

By something, I think he means Extremely Radical.

[PB666]

By something, I think he means Extremely Radical.

Maybe? I like the interior volume, the exterior looks like something from one of those Sci-films where people have rebuilt their spaceship from space junk.

[magnemoe]

Maybe? I like the interior volume, the exterior looks like something from one of those Sci-films where people have rebuilt their spaceship from space junk.

So does lots of vacuum chambers equipment. Lots of heavy metal, loads of bolts and parts everywhere in all directions.

Worked with one in the university for vacuum deposit.

[Elthy]

I think the Stellarator is agood alternative to the tokamak, since it doesnt need to be pulsed which means it can run without an interruption every 30min.

[Finox]

It's an interesting idea, I can see how it would have been shoved aside for the simpler tokamak designs, hard to imagine anyone making a really effective stellarator using slide-rules and drafting tables. Interesting timing, The Economist had a recent article on stellarators: Warning! pay wall ahead.

http://www.economist.com/news/science-and-technology/21676752-research-fusion-has-gone-down-blind-alley-means-escape-may-now-be

[Nuke]

reactor designs like this make me like the polywell more.

[PB666]

This question is a bit bugging me. Why if the french are about to bring a working Fusion reactor on line, is everybody and their brother now suddenly interested in bringing much lower tech reactors, particularly reactors based on 30 or 40 year old designs.......could not have this been done a couple of decades ago before the big body politics sank billioons and billioins into fusion reactors that barely work?

Edit, btw mag , i have a lyophilizer, you can make them look pretty, but they loose their function. If you want to freeze dry many samples at once the do sort of look rigged, particularly when loaded...........but on the other hand when you have a flask hand spun in dry-ice plasticized ethanol (very very cold glass) and you need a port and they are all occupied, those ports become awfully attractive.

Edited October 27, 2015 by PB666

[sgt_flyer]

Well, ITER would most likely allow them to test how to mitigate high energy neutron damage within a simple shape. (And maybe how to harness those neutrons to make more tritium to fuel the reactors)

If you can find the best solution / materials for that within a simple shape, you can then think to adapt it for much more complicated devices. So while the french would make those researchs, other people can try to create those more complicated experimental reactors. In the end, if we want to get working commercial fusion reactors as soon as possible, we can't stay focused on only 1 single technology, when the final design might benefit from advances discovered between the various designs.

Afterwards,ITER will remain an experimental test bed for high energy plasmas and containment of those (at one point, you need experimental data anyway to infirm or confirm various hypothesis)

Edited October 27, 2015 by sgt_flyer

[Elthy]

The Stellarator design only got possible with strong supercomputers, it simpoly want possible to build such complicated coils before. Its not "lower tech" than Tokamak, its simply different, it has advantages and disadvantages.

[PB666]

That then begs the next question. Given that these atoms are in plasma, and given that all elements in plasma are charged and that neutrons have no charge. Why cant they use HV to roughly polarize the atoms so that when the decay they have at least some prevectoring over the neutron motion. It seems this neutron erosion argument has been up there since the mid 1980s. Hasn't any of these wonderful particle physicist come up with a method of making neutrons behave more efficiently in a reactor?

[SargeRho]

Aneutronic fusion.

[PB666]

All you need is a cheep source of Tritium. There is a considerable amount in the waste water of nuclear power plants, but there is no drive to isolate the tritium below $30,000,00/kg. The capture efficiency can be improved by substituting D20 for H20 or compressed D2 for H2 in the neutron capture zone of a reactor. The problem is that waste water handling gets even more expensive. Some of the reactors have had Lithium 6 and 7 rods inserted but the problem is that 3H escaped forcing the shut down of the neutron capture operations.

[SargeRho]

There's a better way to produce tritium: Lithium. But D-T fusion isn't aneutronic. D-He3, p-Li7, D-Li6, p-B11 fusion, etc, those are Aneutronic.

[sgt_flyer]

Well, lithium protection screens could help for both managing those pesky high energy neutrons and breed new tritium in D-T fusion reactions i think it's even among the research they'll do with ITER - the problem is getting a good design (and keeping lithium solid while submitted to these neutron bombardements & near a superhot plasma)

[PB666]

Well, lithium protection screens could help for both managing those pesky high energy neutrons and breed new tritium in D-T fusion reactions i think it's even among the research they'll do with ITER - the problem is getting a good design (and keeping lithium solid while submitted to these neutron bombardements & near a superhot plasma)

You have to place the lithium in a low neutron absorbing metal casing, by its very nature neutron absorption is going to cause it to decay to something else, else being helium and tritium. Something like an 56Fe-62Ni alloy might be suitable for the casing. Also since helium and tritium would both evolve as a gas there would need to be some way of separting and releasing. The lithium could be in a liquid state it melts at 180'C so a liquid gas phase separter would suffice to release both.

- - - Updated - - -

That is correct. But Tritium is much more easily reacted. 3He availability is quite poor so i think you prolly want to avoid that on, lithium-7 requires a very high energy proton.

[Rune]

There's a better way to produce tritium: Lithium. But D-T fusion isn't aneutronic. D-He3, p-Li7, D-Li6, p-B11 fusion, etc, those are Aneutronic.

And you can't have those reactions happening without other neutron-generating fusion reactions taking place right beside them. Even in p-B11 fusion, if I understands the stuff correctly, you get significant neutrons produced by Bremsstrahlung. Neutron radiation handling will be a thing in any nuclear reactor, fusion or fission, albeit the magnitude varies wildly.

Rune. Nitpick of the day!

[fredinno]

You have to place the lithium in a low neutron absorbing metal casing, by its very nature neutron absorption is going to cause it to decay to something else, else being helium and tritium. Something like an 56Fe-62Ni alloy might be suitable for the casing. Also since helium and tritium would both evolve as a gas there would need to be some way of separting and releasing. The lithium could be in a liquid state it melts at 180'C so a liquid gas phase separter would suffice to release both.

- - - Updated - - -

That is correct. But Tritium is much more easily reacted. 3He availability is quite poor so i think you prolly want to avoid that on, lithium-7 requires a very high energy proton.

You can make He3 by making tritium, reacting it with O2 to form H2O, then waiting a decade or so for the stuff to decay to He3, then funnel the formed He and O2 gas (which bubbles off slowly). Then, react the O2 with more Tritrium, or regular hydrogen, which is either looped back into the system, or vented.

Since He is unreactive, you can then store the gas in a specially-designed container, waiting for its time to be shipped.

You know, kind of like Alcohol.

[PB666]

You can make He3 by making tritium, reacting it with O2 to form H2O, then waiting a decade or so for the stuff to decay to He3, then funnel the formed He and O2 gas (which bubbles off slowly). Then, react the O2 with more Tritrium, or regular hydrogen, which is either looped back into the system, or vented.

Since He is unreactive, you can then store the gas in a specially-designed container, waiting for its time to be shipped.

You know, kind of like Alcohol.

only half converts in 12 years, 3/4ths in 24 years.

[ZetaX]

Aneutronic is only a solution if you somehow get that He-3 (etc.). For example, He-3 is incredible rare; you can breed it (e.g. from Tritium as mentioned before), but you definitely do not want to depend on other reactors for building sustainable fusion. Some mention the moon as a possible source, but this is still pretty SciFi.

On the other end: it is actually a yet not completely solved problem how to actually get enough neutrons in D-T fusion. Every fusion process emits at most one neutron, and you need one neutron to create new Tritium; not yet accounting for losses due to decay, hitting something else or a few other reasons. I think the ballpark figures varied from 1.3 to 2 neutrons you need to create per fusion process, depending on many things, including the reactor type.

Some elements can act as neutron multiplicators, though, and that's what saves the day. But this was, as far as i know, only tested in laboratories; a working design for an entire reactor does not yet exist.

This question is a bit bugging me. Why if the french are about to bring a working Fusion reactor on line, is everybody and their brother now suddenly interested in bringing much lower tech reactors, particularly reactors based on 30 or 40 year old designs.......could not have this been done a couple of decades ago before the big body politics sank billioons and billioins into fusion reactors that barely work?

Peope are not "suddenly" interested in bringing those around. Politics is slow. Wendelstein 7-X was essentially conceived more than 20 years ago. If you build something this complex and huge, you have to:

- convince politics that it is a good idea via several channels

- convince the public that it is not a too bad idea and safe

- get enough money from politics, universities, research grants and industry

- find a place to build it (more politics and bureaucracy)

- do tons of simulations to get ready for the next point (possibly requires months or years on supercomputers and also involve )

- plan the thing itself with sufficient detail

- get the permits (more of that horrible bureaucracy)

- plan how to actually get the parts; also note that this requires tons of politics again, as everyone giving you money or else will have another say (e.g. Wendelstein 7-X's magnets were built in France, not because they are great at that, but mostly because where the money comes from and flows to)

- get offers from companies on who might build what cheapest/best

- actually order the stuff (tons of really lengthy contracts)

- actually assemble it (ever tried to build a sub-millimeter-perfect deformed ring-like shape with 50 very weird-looking magnets around it? now add that is supposed to have hundreds of other intruments)

- turn it on (cooling down, like for the LHC, takes quite long, for example)

(and finally: - do the experiments (some of which cannot happen at the beginning, e.g. those that might irradiate the reactor's casing when it is mainly about plasma physics))

- don't forget the repeatedly tell the scientifically illiterate public and protesters that you are still not building a doomsday device or worse

Most of these steps cannot start before the ones before are (almost) finished. I actually find it miraculous that it only took 20 years or so in this case...

Last note on politics as an example what kind of things are relevant here: Wendelstein 7-X is built where it is (Greifswald) due to some programs intended to economically help the former DDR.

only half converts in 12 years, 3/4ths in 24 years.

I am pretty sure he knows that. Seperating Tritium from normal hydrogen is almost trivial. Apart from a huge factor (3) between the mass of their nuclei, they even have significantly distinct chemical properties (most prominently the melting point of "water").

Edited October 28, 2015 by ZetaX

[PB666]

The issue is not what ITER is doing or how fast it is doing it, thats a whole different set of questions. In the past few months we have seen no less than three sort of "we can do better, smaller, cheaper" reactor concept, which is all fine and dandy BUT given the fact its been 60 years and nobody yet has a working reactor.

You cannot convince that this is not doable because they lack computer modeling capability. Computers have not suddenly got mega more powerful.

The critical issue with Tritium is that trutium has in the US a reged dereg status. Its been largely dereged relative to other radionucleotides BUT the limits have been set low, so effectively its still regulated as a chemical env. contaminant. That is the problem is you are doing steam distillation, because steam is highly corrosive and difficult to contain.

[Rune]

Aneutronic is only a solution if you somehow get that He-3 (etc.).

And you have the reactor capable of fusing it in a sustained fashion, which we currently don't. And if we did, said reactor would also fuse D-D at the same time, producing plenty of neutrons.

Rune. Anaeutronic fusion is a myth if you take it literally. It's less-neutronic fusion.

[Elthy]

Computers have not suddenly got mega more powerful.

They did! According to wikipedia (i think ion this case a valid source) the best supercomputer of 1996 had 368 Gigaflops, the fastest at the moment (build in 2013) has 34 Petaflops, a factor of 100000 in 17 years.

[PB666]

17 years is not sudden, and I had a 3.6 ghz processor back in 2003 and now 12 years later my brand new machine maxes out at 3.7 , no change of speed there. The only real change is in parallel processing and 64 bit IS and OS, and GPUs.

I was suprised to learn the other day that my L1 cache is still a measley 256k, that is only twice the memory of my apple IIe c.1986.

As per the other thread, i have projects sitting around for the last decade or so waiting for a processor to come forth affordably so that i can do then. I wouldn't be going through the effort to learn C and 64 bit intel assembly if that processor was out there.

BTW, the in-silico analysis, if its not chained to actuals, you can keep it.

Edited October 28, 2015 by PB666