Lockheed's compact fusion reactor question

[mfb]

Higher magnetic fields are challenging. ITER has up to 13.5 T at its coils, which is the maximum Freidberg considers in his paper for current coil materials. YBCO can handle 20 T, fine - but that is just a 50% increase, and making coils out of high-temperature superconductors is a huge mess. It was considered for an LHC upgrade, but they chose to keep using Nb₃Sn. It is still planned to go to 20 T in the dipole magnets if a larger ring will ever be built, but that is a field over a few centimeters, not a few meters. Cable links (but not coils) use HTS in some places (article).
Freidberg's paper predicts costs that increase substantially with increasing magnetic field strength (figure 7b). It also predicts lower costs/MW for larger power plants (8b).

Stellarators could be an answer. We'll see what Wendelstein 7-X does.

 

[the_wolfman]

One of my few criticism of Freidberg's papers is the cost function that he uses. I think it's a very good paper, and I know that he chose to use simple models clarity. But there are costs and economic risks associated with large scale projects, like ITER, that his cost function does not account for.

For example 4 of the poloidal field coils at ITER are so big that the have to be manufactured on site. You simply cannot transport 500 ton magnets across the world. So not only do they have to build the magnets on site, but they have to build the magnet winding facility on site. There is a significant savings to be had if you can shrink the size of the reactor, such that the magnets can be made at a preexisting factory and then shipped to the construction site. Freidberg's cost function does not account for this.

Another simple cost is interest on a loan. If a company has to take out a loan to build the power plant, then the amount of interest they accrue depends, in part, on how long it takes to build the power plant. Large power plants simply take longer to build, and thus they'll proportionally accrue more interest. Additionally there is a large economic risk associated with taking out a large loan that takes a long time to pay off. Freidberg's cost function does not account for these costs or risks.

In terms of a research reactor there are costs incurred when you do build an international experiment. The would be a significant savings if we could shrink the size of an experimental rector to one where the USA would be completely willing to fund it on it's own. Freidberg's cost function does not account for this either.

You're correct that going to high fields means that we need more structural materials which have an additional cost. So going to high field is not a magic bullet.
Superconducting magnets are not my field of expertise, but the engineers at MIT are very excited about REBCO superconducting tapes. The way they talk about these tapes its sounds like they are very easy to work with. Maybe you've had other experiences?

 

[mfb]

Well, I only see it for the LHC magnets. The general feeling is "we can probably do it, but it will be very complicated and not cheap". The current design uses conventional superconductors outside and HTS only in the core for the highest field strengths.
The tapes don't seem to scale well to larger coils and have problems with quenches. Here is a presentation discussing this, slide 25 has the planned design for 20 Tesla dipole magnets with an aperture of 2 centimeters.

 

[rootone]

If it was possible to produce enough ultracooled material in the Bose-Einstein condensate state,
that might help to reduce the size of some reactor parameters.
The trouble with that is you probably need a massive and dangerous industrial plant to make enough of it.

 

[etudiant]

Well, I only see it for the LHC magnets. The general feeling is "we can probably do it, but it will be very complicated and not cheap". The current design uses conventional superconductors outside and HTS only in the core for the highest field strengths.
The tapes don't seem to scale well to larger coils and have problems with quenches. Here is a presentation discussing this, slide 25 has the planned design for 20 Tesla dipole magnets with an aperture of 2 centimeters.

Thank you, mfb, for this excellent link.
It is the kind of summary that reflects real work done by serious and dedicated researchers, stuff that is rarely seen any more.
I'm glad that there is an active focus on managing failures (quenches) gracefully, but given CERN's past experiences that welcome focus is probably natural.
In this context, I've heard nothing at all about the magnets proposed for the Lockheed design. Do they not see it as an issue?

 

[mfb]

If it was possible to produce enough ultracooled material in the Bose-Einstein condensate state,
that might help to reduce the size of some reactor parameters.
The trouble with that is you probably need a massive and dangerous industrial plant to make enough of it.

Where do you expect Bose-Einstein condensates to be useful, and where do you see a danger from them?

It is the kind of summary that reflects real work done by serious and dedicated researchers, stuff that is rarely seen any more.

Every publication, every conference, every other meeting has that.

Quenches are a necessary part of the commissioning - you need them to improve the maximal field strength of coils.

I don't the field strength Lockheed wants to use.

 

[etudiant]

Where do you expect Bose-Einstein condensates to be useful, and where do you see a danger from them?Every publication, every conference, every other meeting has that.

Quenches are a necessary part of the commissioning - you need them to improve the maximal field strength of coils.

I don't the field strength Lockheed wants to use.

It is good that these summaries are produced. What is missing is wider dissemination of the results.
I do routinely skim the various journals such as Science News or Technology Review, there has not been such a document referenced, much less actually linked.
It is as if the field were operating under a security blanket or the leaders were ashamed of what they were doing. There is no effort to celebrate gains or to create some sense of the potential. Given that fusion seems a lot greener than covering the earth with windmills and solar panels, that reticence is incomprehensible to me.

 

[rootone]

Where do you expect Bose-Einstein condensates to be useful, and where do you see a danger from them?

I'll try dig out a reference, but I do know that I read somewhere that Bose-Einstein condensates are a form of matter that can only exist near to absolute zero.
Matter in that state so I gathered can undergo fusion without the need of extreme pressure, (containment).
If that is actually true then you don't need the megawatts of power needed just to get the reactor started up, everything can be downsized.
However the industrial scale production of such material is definitely not feasible with present technology.

I think if that was feasible, the careful handling and application of the resulting material would be risky.
If for some reason there was a problem in transporting it very quickly, could it explosively undergo a phase transition to 'normal' helium or whatever.

 

[mfb]

I'll try dig out a reference, but I do know that I read somewhere that Bose-Einstein condensates are a form of matter that can only exist near to absolute zero.

That is clear and doesn't need a reference.

Matter in that state so I gathered can undergo fusion without the need of extreme pressure, (containment).

Who claims that?