Wendelstein 7-X

For the discussion of the sciences. Physics problems, chemistry equations, biology weirdness, it all goes here.

Moderators: gmalivuk, Moderators General, Prelates

User avatar
Izawwlgood
WINNING
Posts: 18686
Joined: Mon Nov 19, 2007 3:55 pm UTC
Location: There may be lovelier lovelies...

Wendelstein 7-X

Postby Izawwlgood » Tue Dec 15, 2015 8:07 pm UTC

Some info here.

I was wondering if you physics/engineering folk could give some thoughts on this neat sounding plasma containment system. I know it's not meant to be a sustainable fusion reaction, but I'm wondering how promising you all think it is for a notch in the 'things we know about plasma containment' side of things.
... with gigantic melancholies and gigantic mirth, to tread the jeweled thrones of the Earth under his sandalled feet.

User avatar
Neil_Boekend
Posts: 3220
Joined: Fri Mar 01, 2013 6:35 am UTC
Location: Yes.

Re: Wendelstein 7-X

Postby Neil_Boekend » Wed Dec 16, 2015 11:33 am UTC

Basically that is a Tokamak with a different solution for the drift problem (the plasma tends to drift outward from the center of the torus to the outer wall).

If memory serves, and it rarely does, the solution in the Tokamak design has the distinct advantage that the current in the coils designed to create the plasma current that solves the drift problem must keep increasing during a shot. Since there is always a limit to the current the system can provide this means that the Tokamak design can only work in a pulsed mode. One of the proposed solutions is to build multiple reactors besides each other and use them as cylinders in a car engine in order to provide more stable power, probably combined with some kind of storage. This means that the reactors will be smaller and thus less efficient.

Ergo, from my limited point of view the Steralator design would seem more practical. However, many of the plasma stability problems encountered in Tokamak designs like the ITER would likely also occur in a Steralator design. Much research and testing in the ITER would also enlighten things for a large Steralator design.
And maybe the golden solution is a combination of both, as is usually the case. A twisted magnetic field as in the Steralator, but with the plasma currents from the Tokamak design that would seem to excel in correcting small problems quickly, before they escalate into big problems.

Anyway, I am happy that the 7-X is built and is being tested on, even if it is an expensive project. We need fusion, or we'll drown.

As a side note, the first source for the wikipedia article claims that the plasma mixture is "Hydrogen, Deuterium". If I assume that that hydrogen is protium (cause that list is actually rather silly) and this is a list of reaction fuels we get a seemingly nice looking reaction of 1H+2H=>3He (stable), where the 3He carries the energy as kinetic energy, ergo it speeds away. Fast. This is a problem, because of the law of conservation of momentum. The helium nucleus cannot gain momentum out of nowhere, so something must be speeding the other way, carrying with it half of the energy. Is it a neutrino? Then half of the precious energy is going into space at near light speed, and there is not much you can feasibly do about it.
Mikeski wrote:A "What If" update is never late. Nor is it early. It is posted precisely when it should be.

patzer's signature wrote:
flicky1991 wrote:I'm being quoted too much!

he/him/his

User avatar
sevenperforce
Posts: 658
Joined: Wed Feb 04, 2015 8:01 am UTC

Re: Wendelstein 7-X

Postby sevenperforce » Wed Dec 16, 2015 1:34 pm UTC

It should be noted that this design requires the use of an ITER-style tokamak to produce the plasma which it then confines, increasing complexity somewhat.

I'm not sure if that "Hydrogen, Deuterium" is intended to mean Protium/Deuterium or is merely listing Deuterium as the isotope of hydrogen being used.

What would be some of the advantages/disadvantages of using Lithium Deuteride as the fuel for a combined clean-fission/fusion reactor?

User avatar
Neil_Boekend
Posts: 3220
Joined: Fri Mar 01, 2013 6:35 am UTC
Location: Yes.

Re: Wendelstein 7-X

Postby Neil_Boekend » Wed Dec 16, 2015 1:48 pm UTC

sevenperforce wrote:I'm not sure if that "Hydrogen, Deuterium" is intended to mean Protium/Deuterium or is merely listing Deuterium as the isotope of hydrogen being used.
Of course! Then it could be just a good old 2H+2H=>3He+fast neutron, which makes more sense, cause then the He and the neutron can each carry half the kinetic energy, for a neutral momentum.
Mikeski wrote:A "What If" update is never late. Nor is it early. It is posted precisely when it should be.

patzer's signature wrote:
flicky1991 wrote:I'm being quoted too much!

he/him/his

User avatar
LucasBrown
Posts: 299
Joined: Thu Apr 15, 2010 2:57 am UTC
Location: Poway, CA

Re: Wendelstein 7-X

Postby LucasBrown » Thu Dec 17, 2015 3:36 am UTC

If the helium and the neutron carry equal kinetic energies, then their momenta will have different magnitudes.

User avatar
Zamfir
I built a novelty castle, the irony was lost on some.
Posts: 7459
Joined: Wed Aug 27, 2008 2:43 pm UTC
Location: Nederland

Re: Wendelstein 7-X

Postby Zamfir » Thu Dec 17, 2015 1:24 pm UTC

As Neil says, this design avoids the pulse problem of tokamaks. The required twist is built into the magnets, not generated by a current.

Tradtionally, such a twisted magnetic field was deeply imperfect compared to to the nice field of a tokamak. The last generations of stellarators have computationally optimized magnets. This closes much of that gap.

Trouble is, those optimized fields require strangely shaped magnets, built to very high precision. That turned out to be crazy hard to build (and expensive), at industrial instead of laboratory scale. The Wendelstein 7x is a tad smaller than 1980s tokamaks like JET or JT60, but at double the cost. This despite decades of enormous advances in magnet technology. A similar project in the US got canned because of the horrible cost overruns.

So the big, open question is: what if you scale up yet another step, to ITER-size? By extrapolation, you get mind-bogglingly expensive magnets. But if they have solved the crucial production issues also for larger sizes, then future designs will likely incorporate this method. There is some argument that DEMO should be based on the Wendelstein concept.
Last edited by Zamfir on Thu Dec 17, 2015 3:02 pm UTC, edited 1 time in total.

User avatar
Izawwlgood
WINNING
Posts: 18686
Joined: Mon Nov 19, 2007 3:55 pm UTC
Location: There may be lovelier lovelies...

Re: Wendelstein 7-X

Postby Izawwlgood » Thu Dec 17, 2015 3:00 pm UTC

The biggest cost was in custom machining the magnets? That's somewhat surprising - I wouldn't have thought producing the raw components was the pricy part, but rather the design and assembly times (precision required, etc)

Are there any other applications for magnets created? I know they're specifically designed for generating a ribboned torus, but do they have applications in, say MRIs?

Is this an example of the prototype being stupid expensive, while subsequent models significantly less so, or are these magnets always going to break the bank? And what makes them so expensive - I thought they were precise electromagnets?
... with gigantic melancholies and gigantic mirth, to tread the jeweled thrones of the Earth under his sandalled feet.

User avatar
Zamfir
I built a novelty castle, the irony was lost on some.
Posts: 7459
Joined: Wed Aug 27, 2008 2:43 pm UTC
Location: Nederland

Re: Wendelstein 7-X

Postby Zamfir » Thu Dec 17, 2015 3:55 pm UTC

Is this an example of the prototype being stupid expensive, while subsequent models significantly less so, or are these magnets always going to break the bank?

Some of both, and it's hard to tell in advance how much cheaper even these magnets could become over time. Let alone for the not-even-designed magnets of a DEMO-sized stellarator.

And what makes them so expensive - I thought they were precise electromagnets?

Yeah, sure, and those are expensive if you want them big and precise enough. Remember how the LHC crashed? Or consider how rockets are still expensive, even though they are series-produced and little more than a metal tube with a turbo-pump. Scale+precision=$$$

This is a good presentation, with lots of pictures as well:
http://www.iter.org/doc/www/content/com/Lists/Stories/Attachments/680/ITER_W7X.pdf

There's a list of issues with the magnets, which turned up when the magnets were finished. Keep in mind that these were issues that turned up after they had spent a decade on R&D and production, and thought they were ready. They are symptoms of a design that is so demanding, that every little flaw is a critical showstopper instead of minor fix.
1. deviations and damages of SC strands
2. voids in cast steel coil casings
3. geometrical deviations in coil casings
4. residuals of Cu-SS soldering flux
5. Al and SS welds to be requalified
6. quench detection cable damage
7. insulation faults in the coil header
8. danger of shorts in the coil header


The project eventually got delayed by 10 years in total. Some directly due to the magnets, some indirectly. They basically discovered that the whole project was not up to the challenge, and had to be reorganized all the way through.

Its American counterpart, the NCSX, ran into very similar issues, which suggests that the problems were not due to the Wendelstein team. Similar technical difficulties, but also the program management that outgrew the capabilities of a research lab. The NCSX got scrapped instead of reorganized at a larger scale.

EDIT: don;t get me wrong, it's a very promising machine. There is just a good reason no one built one before on this scale, and there is still an uncertain road to useful scale.

User avatar
Izawwlgood
WINNING
Posts: 18686
Joined: Mon Nov 19, 2007 3:55 pm UTC
Location: There may be lovelier lovelies...

Re: Wendelstein 7-X

Postby Izawwlgood » Thu Dec 17, 2015 5:43 pm UTC

I guess this is my utter lack of familiarity with the field showing - I didn't realize the magnets had to be that precise, given how strong I thought they were. I thought it was... I dunno, if I'm working with a liter of buffer, I only need to be accurate +/- 1ml for most of what I'm adding to things.

That was an interesting read - I'm sure that most prototype construction projects have this sort of nightmare, and can only hope that over time we get better at this sort of thing and improve upon our ability to do so.

So, questions -

My limited understanding of the design is that the ribbon, the twisting of the magnetic field, is to eliminate deviations in the magnetic field caused by a standard, untwisted, torus, producing a magnetic field that is more homogenous. Is that correct? And this will allow greater control of the plasma, as each magnet is as evenly stressed, and the field itself will be more consistent?

You guys are talking about velocities? Can you elaborate on that? I didn't realize that speed of the plasma had much to do with the project other than 'hot things move quickly'?

Would there be any benefits to modularly linking the whole design? I.e., a figure-8 consisting of two of these?
... with gigantic melancholies and gigantic mirth, to tread the jeweled thrones of the Earth under his sandalled feet.

User avatar
Zamfir
I built a novelty castle, the irony was lost on some.
Posts: 7459
Joined: Wed Aug 27, 2008 2:43 pm UTC
Location: Nederland

Re: Wendelstein 7-X

Postby Zamfir » Thu Dec 17, 2015 7:29 pm UTC


My limited understanding of the design is that the ribbon, the twisting of the magnetic field, is to eliminate deviations in the magnetic field caused by a standard, untwisted, torus, producing a magnetic field that is more homogenous. Is that correct? And this will allow greater control of the plasma, as each magnet is as evenly stressed, and the field itself will be more consistent?

Not quite. The issue is 'drift', the tendency of the particles to deflect away from the field lines. Eventually they escape from the torus. There are multiple reasons for that, I think they all count. One is that particles follow tiny helical paths around the field lines, and the toroidal field decreases in strength outwards. So the force on the 'upwards' part of the tiny helix is slightly different than in the 'down' side, since those are at different radii of the torus. This ends up as net force parallel to the axis of the torus. There's also a related effect due to the centripetal force. And as the positive and negative particles separate to the top and bottom of the torus, they start expelling themselves outwards. The net result is that the plasma cloud 'swells'.

If the field lines are helical on the large scale, the particles still follow their tiny helices around the lines. The drift effects from above still apply, but they get 'mixed', if the particle make multiple cycles around within their typical drift time. Half of the time they drift to the inside of the torus tube, half of the time they drift to the outside. Averaged over 1 cycle, they stay close to the same distance from the tube center.

This is far from perfect. There are higher order effects that still cause drift, and local instabilities that blow up to losses. But it cancels out the big first order loss effect. Without it, there's hardly any containment at all.

As you can imagine, it cancels out more if the large helical orbits are smooth helical orbits, full of symmetry. The original stellarators were only rough resemblances of that, while tokamaks do that naturally. The new stellarators are designed, through difficult simulations, to have a reasonably good twisted field. Specifically, the helical field lines should form a series of shell surfaces nested within each other, with particles staying on the same surface cycle after cycle. Every imperfection in design or construction moves particles from one shell to another, and to another in the next cycle, until they drift out of the field.

The problem of a tokamak is that it relies on a constant current through the plasma in circumferential direction, which is generated by a increasing magnetic field, which is generated by an increasing current in a central solenoid... Until that current is maxed out.

Warning: I am not a plasma expert and plasma are ridiculously complex, so there might be errors in the above.

That was an interesting read - I'm sure that most prototype construction projects have this sort of nightmare, and can only hope that over time we get better at this sort of thing and improve upon our ability to do so.

Definitely, it's not like every iteration would cost a billion. At the same time, keep in mind the size. If this machine could be turned into a power plant (it couldn't), it would perhaps deliver 5MW net electric power. A predictable powerplant without fuel needs, no emissions and normal maintenance requirements is worth, say, about 6 Million$/MW. You'll need a quarter of that just for the "ordinary" powerplant parts.

User avatar
Izawwlgood
WINNING
Posts: 18686
Joined: Mon Nov 19, 2007 3:55 pm UTC
Location: There may be lovelier lovelies...

Re: Wendelstein 7-X

Postby Izawwlgood » Sun Dec 20, 2015 6:29 pm UTC

Zamfir wrote:If this machine could be turned into a power plant (it couldn't), it would perhaps deliver 5MW net electric power.
What is the reason that it can't be turned into a power plant? Assuming the design allows plasma containment (big assumption? little assumption?) what would be required to get power from this machine? Making it bigger?
... with gigantic melancholies and gigantic mirth, to tread the jeweled thrones of the Earth under his sandalled feet.

User avatar
Zamfir
I built a novelty castle, the irony was lost on some.
Posts: 7459
Joined: Wed Aug 27, 2008 2:43 pm UTC
Location: Nederland

Re: Wendelstein 7-X

Postby Zamfir » Sun Dec 20, 2015 7:07 pm UTC

Well, it's not intended for any fusion at all. The PR department is remarkably good at hiding that ;-) It is only an experiment in plasma control. AFAIK, it cannot handle the radioactivity from fusion reactions, as a cost-saving choice.

The deeper reason is that at this size, the heat losses would always be greater than the potential heat creation from fusion. Too much escape area compared to the volume. I am not sure exactly how fundamental this is, but it's the reason why people are building the zillion euro ITER, instead of an improved JET.

User avatar
Link
Posts: 1356
Joined: Sat Mar 07, 2009 11:33 am UTC
Location: ᘝᓄᘈᖉᐣ
Contact:

Re: Wendelstein 7-X

Postby Link » Sun Dec 20, 2015 8:12 pm UTC

Zamfir wrote:Well, it's not intended for any fusion at all.
Wait, what? Then why are they calling it a fusion device?

The site mentions they don't need to produce an energy-yielding plasma, but I can't find anything that says they can't do fusion at all. 2 minutes of internet browsing suggests that p-p or p-D fusion should be easily doable (i.e. will happen unless you take measures to avoid it) at 60-130 MK. Not at break-even, perhaps, but still.

User avatar
Izawwlgood
WINNING
Posts: 18686
Joined: Mon Nov 19, 2007 3:55 pm UTC
Location: There may be lovelier lovelies...

Re: Wendelstein 7-X

Postby Izawwlgood » Sun Dec 20, 2015 10:54 pm UTC

In terms of scale of difficulty, where does 'fusing plasma' fall relative to 'containing plasma'? Where does 'extracting energy from fused plasma' fall?
... with gigantic melancholies and gigantic mirth, to tread the jeweled thrones of the Earth under his sandalled feet.

Frenetic Pony
Posts: 177
Joined: Fri Jan 27, 2012 6:31 am UTC

Re: Wendelstein 7-X

Postby Frenetic Pony » Mon Dec 21, 2015 4:10 am UTC

Izawwlgood wrote:In terms of scale of difficulty, where does 'fusing plasma' fall relative to 'containing plasma'? Where does 'extracting energy from fused plasma' fall?


Containing plasma for a long time = very hard. Getting energy from superhot plasma, well that's a bit easier.

User avatar
Zamfir
I built a novelty castle, the irony was lost on some.
Posts: 7459
Joined: Wed Aug 27, 2008 2:43 pm UTC
Location: Nederland

Re: Wendelstein 7-X

Postby Zamfir » Mon Dec 21, 2015 7:43 am UTC

The site mentions they don't need to produce an energy-yielding plasma, but I can't find anything that says they can't do fusion at all. 2 minutes of internet browsing suggests that p-p or p-D fusion should be easily doable (i.e. will happen unless you take measures to avoid it) at 60-130 MK. Not at break-even, perhaps, but still.

Yeah, they try to avoid it as much as possible. The problem is not to achieve fusion, but that the machine is not designed to withstand much in the way of neutron fluxes. And they need constant access to the machine, it can't get activated too much.

From Wiki:
Für den Normalbetrieb ist Wasserstoff als Arbeitsgas vorgesehen. Darüber hinaus sollen Experimente mit Deuterium durchgeführt werden, um auf die Eigenschaften eines Plasmagemisches aus Deuterium und Tritium zu extrapolieren. Dabei können in geringem Maße Fusionsreaktionen zwischen Deuterium-Kernen auftreten, bei denen Neutronen freigesetzt werden.

Rough translation:
Hydrogen is planned as working gas for normal operations. On top of that, there will be experiments with deuterium, in order to extrapolate to the properties of a deuterium-tritium plasma mixture [i.e., the mixture for reactors, Zamfir]. A small amount of fusion reactions betwen deuterium nuclei can occur, which will liberate neutrons.

So technically it will do a tiny bit of fusion, but only as an undesired side effect.

A lot of the work for ITER is in materials research, finding materials that don't detoriate much under the neutron bombardment. The Germans decided (wisely, I presume) to avoid that headache for this machine, and let ITER work on the actual fusion part of the equation. The two approaches might then be merged at a later date in DEMO.

User avatar
Izawwlgood
WINNING
Posts: 18686
Joined: Mon Nov 19, 2007 3:55 pm UTC
Location: There may be lovelier lovelies...

Re: Wendelstein 7-X

Postby Izawwlgood » Mon Dec 21, 2015 1:44 pm UTC

Zamfir wrote:Yeah, they try to avoid it as much as possible. The problem is not to achieve fusion, but that the machine is not designed to withstand much in the way of neutron fluxes. And they need constant access to the machine, it can't get activated too much.
I'm confused - I was under the impression that you could generate a plasma that was still VERY far from reaching fusion temperatures? Surely 'avoiding fusion' is an easy task?
... with gigantic melancholies and gigantic mirth, to tread the jeweled thrones of the Earth under his sandalled feet.

User avatar
sevenperforce
Posts: 658
Joined: Wed Feb 04, 2015 8:01 am UTC

Re: Wendelstein 7-X

Postby sevenperforce » Mon Dec 21, 2015 2:14 pm UTC

Izawwlgood wrote:
Zamfir wrote:Yeah, they try to avoid it as much as possible. The problem is not to achieve fusion, but that the machine is not designed to withstand much in the way of neutron fluxes. And they need constant access to the machine, it can't get activated too much.
I'm confused - I was under the impression that you could generate a plasma that was still VERY far from reaching fusion temperatures? Surely 'avoiding fusion' is an easy task?

Yeah, I imagine you can generate plasma that's far from fusion temperature easily enough, but the farther you get from actual fusion temperatures, the less useful it is. As far as research is concerned, anyway. Their goal seems to be getting more familiar with manipulating and containing plasma that is near fusion temperature.

User avatar
Izawwlgood
WINNING
Posts: 18686
Joined: Mon Nov 19, 2007 3:55 pm UTC
Location: There may be lovelier lovelies...

Re: Wendelstein 7-X

Postby Izawwlgood » Mon Dec 21, 2015 2:56 pm UTC

Right - I was under the impression that the entire machine was a test bed for better plasma containment techniques/technologies... as Zamfir said :)
... with gigantic melancholies and gigantic mirth, to tread the jeweled thrones of the Earth under his sandalled feet.

User avatar
Neil_Boekend
Posts: 3220
Joined: Fri Mar 01, 2013 6:35 am UTC
Location: Yes.

Re: Wendelstein 7-X

Postby Neil_Boekend » Mon Dec 21, 2015 3:33 pm UTC

That impression is right. It's just that testing with cold plasma is not very useful because it does not behave like really really hot plasma. They edge towards fusion to get the temperature right for the right plasma behavior. Some fusion will occur at these temperatures but not enough to damage the vessel. They probably monitor the neutron flux and stop heating the plasma when it rises too much. The size of the vessel indicates that there will probably not be sustained fusion without external heating (I do hope they did more extensive calculations) because the volume (=heat generation through fusion) is too little for the surface area (=heat loss through the wall).

I have the equipment at home to make cold plasma. It won't get too hot to touch the glass envelope with my hand, although the glass envelope does get slightly warm. They are not overly expensive.
Mikeski wrote:A "What If" update is never late. Nor is it early. It is posted precisely when it should be.

patzer's signature wrote:
flicky1991 wrote:I'm being quoted too much!

he/him/his

User avatar
Izawwlgood
WINNING
Posts: 18686
Joined: Mon Nov 19, 2007 3:55 pm UTC
Location: There may be lovelier lovelies...

Re: Wendelstein 7-X

Postby Izawwlgood » Mon Dec 21, 2015 8:33 pm UTC

Does the difficulty in containing plasma increase as a function of the plasma's temperature? Are temperatures in excess of fusion particularly difficult due to... emissions? Increased released energy?

Actually - how does one extract energy from a torus of plasma that is undergoing some rate of fusion? Once you go past the critical temperature for fusion, does the energy released reduce the requirements for additional heating, since the reaction is now heating itself?
... with gigantic melancholies and gigantic mirth, to tread the jeweled thrones of the Earth under his sandalled feet.

User avatar
PM 2Ring
Posts: 3646
Joined: Mon Jan 26, 2009 3:19 pm UTC
Location: Mid north coast, NSW, Australia

Re: Wendelstein 7-X

Postby PM 2Ring » Tue Dec 22, 2015 6:13 am UTC

There's a lot more to plasma control than mere containment. Plasmas are tricky things.

A sample of gas can (generally) be adequately modeled by the ideal gas equation, PV = nRT, where P is pressure, V is volume, T is temperature, n is the number of moles of gas, and R is the ideal gas constant. This model treats the gas molecules as perfectly elastic point-like particles that don't interact with each other apart from exchanging momentum & KE via collision. Note that P, V & T are simple scalar quantities. An ideal gas "stores" energy in its pressure (which is proportional to density for a homogenous gas) and temperature.

However, a plasma is much more complicated because it's electromagnetically active, so energy in a plasma can be shuttled between pressure, temperature, the electric field, and the magnetic field. Note that both electric and magnetic field strength are vectors, and the presence of electromagnetic field lines in a plasma means that you can't treat it as a homogenous blob, you have to deal with the dynamic geometry & topology of those field lines. The study of electromagnetically active fluids is called magnetohydrodynamics; as you can see from that Wikipedia article it's not a simple topic. :) And when the plasma is so hot that you have to take nuclear forces into account then it gets even more complicated.

User avatar
Zamfir
I built a novelty castle, the irony was lost on some.
Posts: 7459
Joined: Wed Aug 27, 2008 2:43 pm UTC
Location: Nederland

Re: Wendelstein 7-X

Postby Zamfir » Tue Dec 22, 2015 11:12 am UTC

Does the difficulty in containing plasma increase as a function of the plasma's temperature?

The required magnetic field increases with temperature (as a square root), so it's a cost driver. It's similar (but different in the details) to the behaviour of a regular gas: higher temperature in a fixed volume leads to higher pressure, higher pressure requires stronger forces to contain. And apart from the magnets, higher temperatures imply higher heat fluxes in every heated part of the installation, which is also a major headache.

Are temperatures in excess of fusion particularly difficult due to... emissions? Increased released energy?

Both D-D and D-T fusion produce fast neutrons that just fly out of the plasma. 'Fast' means that they move much faster(have higher energy) than the typical particles in the plasma. For a working reactor, those neutron fluxes will be very strong, without much precedent really. The neutrons end up in the surrounding masses. The capture of a high energy particle tends to cause local defects in materials, causing a gradual detoriation of most materials. Also, certain elements become radioactive in a neutron flux. Some of the nuclei become unstable isotopes that will decay again, over a longer time period. Even a trace amount of the wrong element in the surroundings can render a space uninhabitable for hours to years.

Actually - how does one extract energy from a torus of plasma that is undergoing some rate of fusion?

The plasma spews energy like a madman... The challenge is more to keep the energy in for long enough. Even an unreacting plasma will emit a lot of photonic radiation, and its relatively colder outer edges hit the walls. For a reactive plasma, you get those neutrons on top of that, that's about 80% of the total. The walls of the plasma chamber need to have many cooling channels through them just to keep them from melting. In principle, that cooling fluid can be used to generate power. You do need hot cooling fluid for that, experimental setups tend to keep the cooling fluid too cold for efficient power generation.

An extra complication is that the reactor burns expensive tritium, and ideally you would use the high neutron flux to create more tritium. There are ideas for that, using lithium as a cooling fluid that also reacts with the neutrons to form tritium. But that's rather experimental still, to my best knowledge.

Once you go past the critical temperature for fusion, does the energy released reduce the requirements for additional heating, since the reaction is now heating itself?


Yes, if the reactor is big enough to make its surface area small compared to its volume. Otherwise too much heat escapes. That's the point of the huge ITER. Keep in mind that most of the energy escapes with the neutrons, the other (helium) fusion products have to mix with the plasma to transfer their energy to it. Then you have to get them out of the mix, that's a challenge in it self. The Wendelstein has some experiments to separate helium out of a hydrogen plasma.

User avatar
Neil_Boekend
Posts: 3220
Joined: Fri Mar 01, 2013 6:35 am UTC
Location: Yes.

Re: Wendelstein 7-X

Postby Neil_Boekend » Tue Dec 22, 2015 12:32 pm UTC

One of the other advantages of the pulsed function of the Tokamak solution is that it would allow for relatively easy removal of the fusion plasma.
Mikeski wrote:A "What If" update is never late. Nor is it early. It is posted precisely when it should be.

patzer's signature wrote:
flicky1991 wrote:I'm being quoted too much!

he/him/his

User avatar
Izawwlgood
WINNING
Posts: 18686
Joined: Mon Nov 19, 2007 3:55 pm UTC
Location: There may be lovelier lovelies...

Re: Wendelstein 7-X

Postby Izawwlgood » Tue Dec 22, 2015 1:32 pm UTC

Wow. That is all quite awesome.

The wiki entry on Tokamaks talks about how the heating is done via four methods, current, injection, magnetic compression, and radiofrequency. How does one run a current through a plasma? I presume there isn't like... a cathode and an anode dipping into the plasma? I feel like I'm asking a super basic question...

As for magnetic compression - I thought the magnets used were fixed? Is there an auxilliary set of magnets that is used to make the field suddenly smaller? Like, set of magnets A for initial containment, then you switch to set of magnets B, which generate a smaller field?
... with gigantic melancholies and gigantic mirth, to tread the jeweled thrones of the Earth under his sandalled feet.

User avatar
Zamfir
I built a novelty castle, the irony was lost on some.
Posts: 7459
Joined: Wed Aug 27, 2008 2:43 pm UTC
Location: Nederland

Re: Wendelstein 7-X

Postby Zamfir » Tue Dec 22, 2015 7:51 pm UTC

How does one run a current through a plasma?

Induction, like in a transformer. The plasma ring is like a short-circuited secondary winding of a transformer, and there's an electromagnet on the central axis of the tokamak that act as the primary windings. A time-varying current through the primary windings induces an electromotive force in the plasma ring, in the direction around the ring. This results in the current that the tokamak needs for its helical field lines, and that current also helps as heater. But that heating is mostly a side effect of a current that was needed anyway. The main heaters are the particles injectors and the microwaves.

In the picture of JET, you can see the big transformer iron core with its squarish loops around the plasma. In later tokamaks, the magnetic return path goes through the air, making the transformer less visible.
Image

Image

ijuin
Posts: 817
Joined: Fri Jan 09, 2009 6:02 pm UTC

Re: Wendelstein 7-X

Postby ijuin » Wed Dec 23, 2015 8:37 am UTC

Izawwlgood wrote:As for magnetic compression - I thought the magnets used were fixed? Is there an auxilliary set of magnets that is used to make the field suddenly smaller? Like, set of magnets A for initial containment, then you switch to set of magnets B, which generate a smaller field?


The magnets repel the plasma away from the vessel walls. Increasing the strength of the field increases the force of this repulsion, thus squeezing the plasma into a smaller space.

Toffo
Posts: 163
Joined: Tue Jun 03, 2014 3:17 am UTC

Re: Wendelstein 7-X

Postby Toffo » Wed Dec 23, 2015 10:52 am UTC

Zamfir wrote:Induction, like in a transformer. The plasma ring is like a short-circuited secondary winding of a transformer, and there's an electromagnet on the central axis of the tokamak that act as the primary windings.

A time-varying current through the primary windings induces an electromotive force in the plasma ring, in the direction around the ring.

Image



Is there an error in the picture? In a transformer one current induces an opposite current. I mean, why are the magnetic field lines going to the same direction in the primary and the secondary winding?

Toffo
Posts: 163
Joined: Tue Jun 03, 2014 3:17 am UTC

Re: Wendelstein 7-X

Postby Toffo » Wed Dec 23, 2015 11:37 am UTC

ijuin wrote:The magnets repel the plasma away from the vessel walls. Increasing the strength of the field increases the force of this repulsion, thus squeezing the plasma into a smaller space.



I'm a little bit skeptical. What magnets repel the plasma away from the vessel wall? I'm not seeing such magnets. I know that anything with a high current going through tends to implode, tokamak plasma is one such thing.

So I conclude that in Wendelstein 7-X there no compression happening at all.

User avatar
Neil_Boekend
Posts: 3220
Joined: Fri Mar 01, 2013 6:35 am UTC
Location: Yes.

Re: Wendelstein 7-X

Postby Neil_Boekend » Wed Dec 23, 2015 12:39 pm UTC

Toffo wrote:
ijuin wrote:The magnets repel the plasma away from the vessel walls. Increasing the strength of the field increases the force of this repulsion, thus squeezing the plasma into a smaller space.



I'm a little bit skeptical. What magnets repel the plasma away from the vessel wall? I'm not seeing such magnets. I know that anything with a high current going through tends to implode, tokamak plasma is one such thing.

So I conclude that in Wendelstein 7-X there no compression happening at all.

Why would anything with a high current through it implode? Current creates ohmic heating so things with high current [/i]expand[i]. The green coils cause magnetic compression and confinement, with DC current.
Image
The Wendelstein has similar coils, but with the additional feature that they provide twist in the magnetic field, solving the drift problem.
Mikeski wrote:A "What If" update is never late. Nor is it early. It is posted precisely when it should be.

patzer's signature wrote:
flicky1991 wrote:I'm being quoted too much!

he/him/his

Toffo
Posts: 163
Joined: Tue Jun 03, 2014 3:17 am UTC

Re: Wendelstein 7-X

Postby Toffo » Wed Dec 23, 2015 1:32 pm UTC

Good reading for everyone and particularly the previous poster :

https://en.wikipedia.org/wiki/Z-pinch

https://en.wikipedia.org/wiki/Pinch_%28 ... physics%29

User avatar
Neil_Boekend
Posts: 3220
Joined: Fri Mar 01, 2013 6:35 am UTC
Location: Yes.

Re: Wendelstein 7-X

Postby Neil_Boekend » Wed Dec 23, 2015 1:47 pm UTC

Indeed. That phenomenon should have occurred to me.
Mikeski wrote:A "What If" update is never late. Nor is it early. It is posted precisely when it should be.

patzer's signature wrote:
flicky1991 wrote:I'm being quoted too much!

he/him/his

User avatar
Zamfir
I built a novelty castle, the irony was lost on some.
Posts: 7459
Joined: Wed Aug 27, 2008 2:43 pm UTC
Location: Nederland

Re: Wendelstein 7-X

Postby Zamfir » Wed Dec 23, 2015 3:04 pm UTC

Is there an error in the picture?


I think the picture would be correct for the early phase of a shot, when the central solenoid's field is building down but is still in the direction of the picture. Eventually it will reverse, until it hits its other maximum.

Image
This is a picture for ITER, see how the currents in coils and plasma are in the same direction early on. They even appear to have different coils with different profiles, some more to build up the plasma current and some more to maintain it.

But I am not 100% certain of my understanding - it works just a tad different from a sinewave-fed ordinary transformer.

Hypnosifl
Posts: 257
Joined: Wed Mar 02, 2011 10:05 am UTC

Re: Wendelstein 7-X

Postby Hypnosifl » Thu Feb 04, 2016 2:09 am UTC

Looks like they had their first successful test with hydrogen (they had previously used helium in December):

http://www.nytimes.com/aponline/2016/02 ... .html?_r=0

Following nine years of construction and testing, researchers at the Max Planck Institute for Plasma Physics in Greifswald injected a tiny amount of hydrogen into a doughnut-shaped device — then zapped it with the equivalent of 6,000 microwave ovens.

The resulting super-hot gas, known as plasma, lasted just a fraction of a second before cooling down again, long enough for scientists to confidently declare the start of their experiment a success.

...

David Anderson, a professor of physics at the University of Wisconsin who isn't involved in the project, said the project in Greifswald looks promising so far.

"The impressive results obtained in the startup of the machine were remarkable," he said in an email. "This is usually a difficult and arduous process. The speed with which W7-X became operational is a testament to the care and quality of the fabrication of the device and makes a very positive statement about the stellarator concept itself. W7-X is a truly remarkable achievement and the worldwide fusion community looks forward to many exciting results."

Press release here:

http://www.mpg.de/9926419/wendelstein7x-start

The first hydrogen plasma, which was switched on at a ceremony on 3 February 2016 attended by numerous guests from the realms of science and politics, marks the start of scientific operation of Wendelstein 7-X. At the push of a button by Federal Chancellor Angela Merkel, a 2-megawatt pulse of microwave heating transformed a tiny quantity of hydrogen gas into an extremely hot low-density hydrogen plasma. This entails separation of the electrons from the nuclei of the hydrogen atoms. Confined in the magnetic cage generated by Wendelstein 7-X, the charged particles levitate without making contact with the walls of the plasma chamber. “With a temperature of 80 million degrees and a lifetime of a quarter of a second, the device’s first hydrogen plasma has completely lived up to our expectations”, states Dr. Hans-Stephan Bosch, whose division is responsible for operation of Wendelstein 7-X.

The present initial experimentation phase will last till mid-March. The plasma vessel will then be opened in order to install carbon tiles for protecting the vessel walls and a so-called “divertor” for removing impurities. “These facilities will enable us to attain higher heating powers, higher temperatures, and longer discharges lasting up to ten seconds”, explains Professor Klinger. Successive extensions are planned until, in about four years, discharges lasting 30 minutes can be produced and it can be checked at the full heating power of 20 megawatts whether Wendelstein 7-X will achieve its optimisation targets.

ijuin
Posts: 817
Joined: Fri Jan 09, 2009 6:02 pm UTC

Re: Wendelstein 7-X

Postby ijuin » Thu Feb 04, 2016 10:40 am UTC

"A tiny amount of hydrogen"? How many milligrams are we talking about here?

KarenRei
Posts: 274
Joined: Sat Jun 16, 2012 10:48 pm UTC

Re: Wendelstein 7-X

Postby KarenRei » Thu Feb 04, 2016 2:05 pm UTC

Toffo wrote:
ijuin wrote:The magnets repel the plasma away from the vessel walls. Increasing the strength of the field increases the force of this repulsion, thus squeezing the plasma into a smaller space.


I'm a little bit skeptical. What magnets repel the plasma away from the vessel wall?


Magnets don't "repel plasma". But any charged particle moving through a magnetic field tends to move in a spiral path with a characteristic gyroradius. The reason is because charged particles moving through a magnetic field experience Lorentz force, and the direction of the force changes as the particle moves, causing it to overall average into a spiral. So they basically get pinned into spiraling around field lines. In a perfect vacuum with a single particle, it would stay confined forever. But that's obviously not the case here - whenever any two particles collide, each can get kicked into a new relative position within the field. So eventually things average drifting to the edge and colliding to the walls. The more collisions they have to take to do that, the lower the losses by this mechanism. The gyroradius is inversely proportional to the magnetic field, so the distance a particle averages moving per collision is halved for every doubling of the field, which offers a far greater than linear increase in the number of collisions required on average to migrate to the edge.

jewish_scientist
Posts: 903
Joined: Fri Feb 07, 2014 3:15 pm UTC

Re: Wendelstein 7-X

Postby jewish_scientist » Tue Feb 09, 2016 12:33 am UTC

How does one actually get the energy from a fusion reactor? I understand that the energy is stored in the atoms' mass, and that the energy is released when two atoms fuse; but what does a reactor actual do to collect and convert the released energy into electricity?
"You are not running off with Cow-Skull Man Dracula Skeletor!"
-Socrates

User avatar
Copper Bezel
Posts: 2426
Joined: Wed Oct 12, 2011 6:35 am UTC
Location: Web exclusive!

Re: Wendelstein 7-X

Postby Copper Bezel » Tue Feb 09, 2016 12:46 am UTC

Same as everything else: heat a kettle.
So much depends upon a red wheel barrow (>= XXII) but it is not going to be installed.

she / her / her

Frenetic Pony
Posts: 177
Joined: Fri Jan 27, 2012 6:31 am UTC

Re: Wendelstein 7-X

Postby Frenetic Pony » Tue Feb 09, 2016 3:19 am UTC

Copper Bezel wrote:Same as everything else: heat a kettle.


Or a longer explanation "giant ring of superhot plasma gives off a lot of heat, lets flush water around it (to cool the superconducting magnets and etc.) then dump the hot water below a turbine."

The engineering specifics are a lot more complex, but the as Copper pointed out, the idea is basically the same :D

User avatar
thoughtfully
Posts: 2253
Joined: Thu Nov 01, 2007 12:25 am UTC
Location: Minneapolis, MN
Contact:

Re: Wendelstein 7-X

Postby thoughtfully » Tue Feb 09, 2016 10:42 am UTC

Fusion reactors are a bit different because the reactant is physically separated from the reactor. There's not very much heat transfer by conduction, as I understand it. Still, there's quite a lot of neutron flux and thermal radiation getting out.
Image
Perfection is achieved, not when there is nothing more to add, but when there is nothing left to take away.
-- Antoine de Saint-Exupery


Return to “Science”

Who is online

Users browsing this forum: No registered users and 4 guests