Magnets for fusion energy: A revolutionary manufacturing met

Point out news stories, on the net or in mainstream media, related to polywell fusion.

Moderators: tonybarry, MSimon

Post Reply
Grumalg
Posts: 24
Joined: Fri Feb 27, 2009 5:11 pm

Magnets for fusion energy: A revolutionary manufacturing met

Post by Grumalg »

http://phys.org/news/2014-07-magnets-fu ... ethod.html

The National Institute for Fusion Science (NIFS), of the National Institutes of Natural Sciences (NINS) in Japan, has achieved an electrical current of 100,000 amperes, which is by far the highest in the world, by using the new idea of assembling the state-of-the-art yttrium-based high-temperature superconducting tapes to fabricate a large-scale magnet conductor.

NIFS is undertaking the development of a high-temperature superconducting coil that is appropriate for the fusion reactor magnet. Using the state-of-the-art yttrium-based high-temperature superconducting tapes which have been developed and produced in Japan through the new thinking that simply stacks the tapes, NIFS manufactured a conductor of exceptional mechanical strength. For the conductor joints, which are important for the production of the large-scale coils, NIFS developed low-resistance joint technology through collaborative research with Tohoku University. As a result of the prototype conductor test, at the absolute temperature of 20 degrees Kelvin (minus 253 degrees Celsius) the electrical current exceeds 100,000 amperes. The overall current density exceeds 40 A/mm2 including the jackets, and this value is of practical use for manufacturing large-scale fusion reactor magnets. This result is of global importance. We use 54 yttrium-based high-temperature superconducting tapes. Each tape is 10 mm in width and 0.2 mm in thickness.
.....

GIThruster
Posts: 4686
Joined: Tue May 25, 2010 8:17 pm

Re: Magnets for fusion energy: A revolutionary manufacturing

Post by GIThruster »

Pretty sure this is just the latest from Superpower. Were they bought out by the Japanese Furukawa co.? I thought they started here in Michigan or some such.
"Courage is not just a virtue, but the form of every virtue at the testing point." C. S. Lewis

D Tibbets
Posts: 2775
Joined: Thu Jun 26, 2008 6:52 am

Re: Magnets for fusion energy: A revolutionary manufacturing

Post by D Tibbets »

I may easily be missing something, but this seems underwhelming. Reported 40 A/ mm squared is 4000 A/ cm squared. This is a number that has been reported for years for high temperature superconductors.Cold super conductors (~4 degrees K/ liquid helium boiling point) may have reached 8,000 amps per cm squared. Cooled copper can easily exceed this capacity, though obvously the current has to be continually replaced and the cooling needs to be robust..

I suspect at least an order of magnitude higher amp capacity is needed, or alternately, small wires / ribbons that can be wraped into many windings to generate the necessary amp turns, while tolerating the high surface B fields on the superconductor surfacess. Generating 10 Tesla in the huge magnets of a Tokamak is challenging. Smaller magnets such as in a Polywell or FRC (?) may actually be easier as surface B fields are proportionatly greater in smaller diameter magnets. I think there is a strength dependance that is the inverse square of the diameter (or radius). ie: 100,000 amp turns might generate can surface B fields of 0.1 T in a 1 meter magnet, but only 0.001 T in a 10 meter magnet. Obvously you have more room for windings and coolent in a larger can, but only id the conductor/ super conductor lends itself to high current densities, or many windings.

In WB6 there were 200 turns and ~ 1,000 amps for B fields of ~ 0.1 to 0.3 Tesla in 30 cm diameter magnets. That means the small diameter wires (guess 0.2 mm) was carrying 1000 A or ~25,000 A per cm squared. This is a current density ~ 5-10 X that of available superconductors. How much cooling is nessisary to maintain this current steady state is unknown by me, but the formulas are straight forward. Also, keep in mind that WB6 magnets were uncooled. WB4 was water cooled and I don'tknow how long they might have been run at several thousand Gauss strengths. If cooled to ~liquid nitrogen temperatures copper conducts up to ~ 8 times better, and nearly 20 times better at liquid helium temperatures. There is considerable growth potential for copper wires. I don't know if there is much growth potential for the superconductors.

An example (perhaps reasonable) JET copper magnets consumed upwards of a GW of power. I think they were water cooled (?). If liquid nitrogen cooled, they would have only have needed ~ 130 MW of power (increased turns allowing for constant amp turn numbers). Superconductors would have required zero current once powered up and possibly significantly less active cooling, but only if the necessary wires fit in the can. If you cannot get the necessary amp turns, the magnet power input question and the cooling issues are meaningless, as the machine will not work.

Dan Tibbets
To error is human... and I'm very human.

prestonbarrows
Posts: 78
Joined: Sat Aug 03, 2013 4:41 pm

Re: Magnets for fusion energy: A revolutionary manufacturing

Post by prestonbarrows »

Getting 0.1T from water cooled copper on the order of 20 cm in diameter is fairly easy to do for continuous duty. This is around the 1-2kW power range. I have never seen a continuous water cooled copper coil get close to 1T, maybe someone has examples of this though. The trade off comes between the diameter of the cooling lines and the pressure drop across the coil. Getting enough water pumped through for heat removal requires fatter pipes which either reduces the number of turns or increases the magnet size, and so on. The fluid velocity also has a limit, try to push it too fast through skinny pipes and things become turbulent causing the pressure drop increase to even further.

One big practical issue with cryogenic copper in a continuous duty is the venting of the boiled cryogen. Assuming you are using nitrogen, and dumping multiple KW of power into the coils, that is going to create many many liters of gas constantly. This has to escape through your coil piping and would tend to blow out any liquid with it (think of the problems you have when water coolant boils). Again, you would tend to need fatter pipes to handle the exhaust which degrades your coil specs. Externally cooling the copper is maybe a more sensible option, though again the reservoir and exhausting system are going to kill a lot of space in the coils. Getting cryogen into the center of the coil would be an issue also. I imagine you would need some type of re-condenser system to keep up with the coolant demand.

That is one reason pulsed magnets can get away with murder. 10,000 A for only a microsecond dosen't drop that much total energy, somewhere on the order of kJ (the energy stored in your pulse capacitor bank). The total energy is then low enough to just be soaked up by a few degrees temperature rise from the copper's specific heat.

D Tibbets
Posts: 2775
Joined: Thu Jun 26, 2008 6:52 am

Re: Magnets for fusion energy: A revolutionary manufacturing

Post by D Tibbets »

Good points. Obviously a closed system would be needed, basically along the lines of a car engine with radiator. The coolant needs to be maintained in a liquid state. The only metric I know is hearsay from people like M. Simon, where estimates of ~ 40% volume of a can can be occupied by wires and the remainder is cooling plumbing. Then there are approaches like using Bitter magnet designs. Here I think cooled copper magnets have reached steady state strengths of over 20 Tesla (admittedly in a small internal region).

http://www.ru.nl/hfml/research/levitati ... -solenoid/


Then there are hybird designs

http://lss.fnal.gov/conf/C720919/p18.pdf

And, it is not like super conductors do not need aggressive cooling, at least in a nuclear fusion situation. A medical MRI machine may operate fine with a cryostat. No resistive heating, and relatively easy insulation from room temperature air on the outside. In DT fusion deeply penetrating neutrons have to be stopped before they reach the cryostat, and the resultant heat removed. This will occupy a significant volume within the can, perhaps nearly as much as would be needed for more tolerant copper windings. With P=-B 11 the external heating is more from deeply penetrating Bremsstruhlung X- rays. The point is that the internal volume available for actual windings may not be that much different between resistive copper coils and super conductors, especially when considerations of the paramount safety measures needed to prevent or handle superconductor quenching is added. The solution again ends up being the amp/ cm^2 capacity.
The game changer is either significant improvement in super conductor currant capacity and/ or super high temperature superconductors. A super conductor that works at temperatures up to several hundreds degrees C would ease cooling concerns stupendously.

PS: JET achieved magnetic fields of 3.45 T with its large copper coil magnets. This size may be comparable to the size for a breakeven or higher capacity Polywell, though with about twice the strength target.

Dan Tibbets
To error is human... and I'm very human.

prestonbarrows
Posts: 78
Joined: Sat Aug 03, 2013 4:41 pm

Re: Magnets for fusion energy: A revolutionary manufacturing

Post by prestonbarrows »

Biter magnets are pretty amazing, but can't really be used on geometries needed for tokamaks/polywell. They need to be long small bore solenoids.

I believe JET still holds the record for non-SC tokamak field strength. The design specifics of systems of that scale are out of my range of experience though.

Closed cryogen loops are a fairly common thing, especially on systems using expensive liquid helium. Typically they don't have much capacity though and just recycle the losses due to heat leaks in the cryostat on the order of watts, not active power loads of kilo- or megawatts.

The big difference between a SC tokamak in a fusion environment and a cryocooled resistive copper coil is the direction the heat load is coming from. In the former case, the heat is generated outside the coil and the majority of this is soaked up with extensive nested layers of water and nitrogen cooling in the first wall putting relatively low heat load into the helium cooled coil itself. In the latter case, there is a huge heat load directly into the cryogen leading to the boiling problems mentioned before; in addition to the external heat load if you are talking about a fusion environment.

Even with the water shield taking most of the heat load (~500MW for ITER), the cryogen still needs to be re-condensed.
ITER wrote:The ITER cryogenic system will be the largest concentrated cryogenic system in the world with an installed cooling power of 65 kW at 4.5K (helium) and 1300 kW at 80K (nitrogen).
Just to give some context to the idea of continuously cooling cryo-copper coils; even the biggest systems in the world can't handle all the much power.

Cryo-copper can work great for pulsed science experiments, but will likely never really be feasible for a CW full scale device.

Post Reply