Fold-up Polywell

[DeltaV]

When you need to travel with a fusion reactor in your carry-on bag...





(From http://www.polyhedra.jp/polyhedron/js/zindex.html )

The coils for these arrangements interfere if you try to fold them...

[hanelyp]

It's harder to see in the first 2, but I think those also have interference when folding.

[DeltaV]

You're right. I should have viewed it with IrfanView (hit 'G' to stop animation).

The solution seems obvious...
Conform the coils to lie in nested spherical shells, with small differences in diameters.
But that may not work either since the triangles seem to hit near the corners, where the pivots are located.

Maybe raypeters or tombo can do a CAD kinematics study?
Juno may already have solved it -- hard to tell with the transparent linkages:
http://www.polyhedra.jp/polyhedron/js/z12photo.html
Looks like it could work if you allowed the coils to be non-rigid.
Would like to see a mag field sim during expansion, if I only had the time...

[DeltaV]

This one is really interesting:
http://www.polyhedra.jp/polyhedron/js/5656/zindex.html

[D Tibbets]

As an exercise in geometry, this may be interesting. But for a real reactor it is ,um, meaningless. Keep in mind that each magnet face is a separate assembly that has to have cooling plumbing, power leads, structural supports (against the powerful repulsion between all magnets, a robust and cooled vacuum vessel, etc.

Dan Tibbets

[DeltaV]

As an exercise in geometry, this may be interesting. But for a real reactor it is ,um, meaningless. Keep in mind that each magnet face is a separate assembly that has to have cooling plumbing, power leads, structural supports (against the powerful repulsion between all magnets, a robust and cooled vacuum vessel, etc.

Dan Tibbets

Dan...
Cooling is not a problem for pulsed machines.
Power can be transferred through slip rings or rotary-inductor/-capacitor couplers. [EDIT: capacitors... better for high-power transients]
Could also use permanent magnets.
Magnet repulsion acting on the sphere means that the forces are primarily tension. Best case. Thank God it's not bending moment.
But the real secret is a good acronym.

[KitemanSA]

Keep in mind that each magnet face is a separate assembly that has to have cooling plumbing, power leads, structural supports (against the powerful repulsion between all magnets, a robust and cooled vacuum vessel, etc.

Ummm, no it doesn't need to be. Remember the MPG.

[DeltaV]

For those really long trips you'll need to maximize your packing efficiency...
https://www.youtube.com/watch?v=KKs9vkYxAps
http://www.mathematicaguidebooks.org/so ... ding_t.mov

[D Tibbets]

Tension can be managed with robust connections between magnets, but I think interconnects between magnets is recognized as undesirable. Wall standoffs with associated power feeds, and cooling is most likely the best solution by far. Couplings inside the vacuum chamber are fine but each introduces another site for leaking. Leaks, even tiny leaks in high vacuum systems is always a major consideration.


Permanent magnets have been, and may remain useful, mostly for demonstrating certain magnetic geometries and interactions. But for any hope of approaching breakeven, they are anathema. There are always cusps leading directly into magnetic surfaces.

For research purposes in pulsed or short duration machines, active and vigorous cooling may not be needed. But for any power producing pulsed or steady state machine, coolent flow will be need to be significant, even if superconductors are not used- the Ohmic heating by the resistance in copper wires generate a huge amount of heat. Then there are the X-rays and neutron bombardment, and the charged particle impacts due to ExB transport, etc.

Even ignoring any considerations of the structure of the Polywell, the vacuum vessel needs to be robust and up to several times the dimensions of the magrid. If you have a rigid vacuum vessel that limits the size of the structure, the innards are immaterial. The magrid may be broken down for ease of assembly and transport- ie each magnet with associated connections and supports may be bolted onto the vacuum vessel much as in the current WB8 design. But a folded up assembly of multiple magnets adds complexity, if it is doable at all.

Dan Tibbets

[KitemanSA]

Well, the first one won't work.

[DeltaV]

Well, the first one won't work.

I assume that you are referring to N vs S directions. Yes, the facet arrangement will need to allow all inward (or outward) poles (even if with the expense of intercoil gaps and "funny" cusps). But the general idea of rapid rotational compression has a certain steam-punk appeal for pulsed operation.

Consider a Project Orion inspired arrangement using an even number of rapidly counter-rotating, truncated icosahedra. Space... already in vacuum. Inject fuel puffs. Spinning polyhedra with nozzles on one end (symmetry axes aligned with spin axes) are simultaneously, inertially actuated into compression using disc brakes. Torques cancel. Ka-boom-boom.

Ooops. Not synchronized...

Ka-b-boom.

Almost got it...

Ka-boom.

Perfect!