Spheroidal Foci and POPS?

[MSimon]

That is an interesting thought. Can the ions recirculate outside the grid with sufficient electron shielding? How much is sufficient?

[KitemanSA]

The wiffleball has holes in it. If we have plasma soup then the plasma just falls out the holes in a horribly lossy way as Art keeps trying to make everyone understand.

Some people seem to think that the holes magically seal up to give us a nice pretty quasi sphere. That just flat contradicts what Dr Bussard had to say about electron recirculation, but if you can handwave that away, Art has his cannons charged and ready.

So if the device works we don't have plasma soup. We have structure.

Ions are heavier than electrons and take longer to stop, so the first piece of structure is that the wiffleball is covered by a layer of ions. That should make for some interesting calculations.

This does not necessarily follow. The wiffleball keeps the electrons in (mainly) and the electrons form a virtual negative electrode in the middle, and the VE keeps the positive ions in, stopping them before(?) they reach the wiffleball.

The second piece of structure is that the plasma falls out of the holes in jets, actually tubular double sheets, focussed by the magrid.

Past statements by Dr. N have suggested to me that the ions do NOT leave the MaGrid in any appreciable numbers. The electrons do, but are returned rapidly by the MaGrid charge.

Proving that those jets turn back on themselves and return the plasma to the wiffleball, rather than spraying it all over the wall, is this week's exercise...

Mayhaps an exercise without purpose?

The clues are Rick's insistence on collective mechanisms and Art's insistence that where the electrons go, the ions follow and vice versa.

But at the end of one of those long discertations by Art, Dr. N seemed to disagree with Art's "wither they goest" hypothesis. His statement (IIRC) was that the Polywell is quasi-neutral, NOT ambipolar.

Maybe we should build one and find out. I think an improved MPG would work very nicely and be fairly easy to build.

[TallDave]

Kiteman addressed this all pretty well, imho, but let me add...

Then it works by magic, right? Cool...

Not sure what you're talking about. Wells aren't magic.

The wiffleball has holes in it. If we have plasma soup then the plasma just falls out the holes in a horribly lossy way as Art keeps trying to make everyone understand.

If by "plasma soup" you mean a plasma that is locally quasineutral everywhere, yes, but why would you expect that? IEC devices form wells.

Some people seem to think that the holes magically seal up to give us a nice pretty quasi sphere.

It isn't magic, it's a property resulting from the electron pressure.

alexjrgreen wrote:
The second piece of structure is that the plasma falls out of the holes in jets, actually tubular double sheets, focussed by the magrid.

Past statements by Dr. N have suggested to me that the ions do NOT leave the MaGrid in any appreciable numbers. The electrons do, but are returned rapidly by the MaGrid charge.

I'll just add "I second that."

Ions are heavier than electrons and take longer to stop, so the first piece of structure is that the wiffleball is covered by a layer of ions. That should make for some interesting calculations.

It's not just the mass. For the electrons, the Magrid is the bottom of the well. They spend all their time bouncing around trying to get to it, some making it to the cusps where they oscillate. The ions want to get to the center, and spend their time oscillating around it. I think this is where your picture of things differs from ours. Bussard says the ions are injected at the edge of the well, and Chacon argues they don't upscatter out.

Art's picture of "ions follow electrons" ignores the well and the fact the plasma wants to spit out electrons. For our picture, it's like putting a one-inch trough at the top of a 30 foot deep well and then wondering why all the water at the bottom doesn't come rushing out.

It probably depends on what those cusps really look like. If they're small and spiky I don't think they can pull ions out. If they're big and cone-y maybe they can. (I'm guessing, based on the WB-8 contract, that the WB-7 results rule out ions streaming out the cusps, but maybe Rick's just a smooth talker.)

[D Tibbets]

It's not just the mass. For the electrons, the Magrid is the bottom of the well. They spend all their time bouncing around trying to get to it, some making it to the cusps where they oscillate. The ions want to get to the center, and spend their time oscillating around it. I think this is where your picture of things differs from ours. Bussard says the ions are injected at the edge of the well, and Chacon argues they don't upscatter out.
.......

I'm not sure if that discriptin is complete and I' not sure if my modification makes any difference, but-
The charged magrid is the bottom of the potential well if you ignore the blocking effect of the magnetic fields (except at the cusps). A better visualization is that the Wiffleball and the magrid forms a hill with a wall around it before you reach the base. This wall has some holes that the balls (electrons) can escape through, and through which the cannons at the hill base fire balls up the hill (electron injection, using energy (gunpowder)). Recirculation is represented by springs at the base of the wall that efficiently bounces the balls hitting it back with no additional energy needed. Elsewhere the balls acumulate along the walls till they can drain through the holes. This is what would happen if there was no elastic bouncing off the Wiffleball surface. Add that bouncing effect and the balls will not accumulate at the wall, but continue bouncing untill they find a hole.

It just occured to me that a pinball machine might be a good analogy. The tilt of the board represents the potential well. The ball injector represents the 'magrid' ball accelerater. The bumpers represent the Wiffleball border. The posts are removed (no unshielded surfaces) The flippers represent a combination of the cusps and recirculation. Then add more balls to make it more fun and confusing. Make the bumper layout more circular, with small open ports between some of them, which almost always drains down to the flippers. Make the center of your board the highest point, add additional complexity , restore the posts (with bumpers) to represent ions clustered near the center or dispersed radially or randomely, continue adding complexity to your hearts content....

Dan Tibbets

[MSimon]

I remember playing on machines that would release one or two extra balls if you hit the right combinations. Exciting. Probably a Williams machine.

The Bally's were for gambling and tended to go through a ball pretty fast.

[alexjrgreen]

Some people seem to think that the holes magically seal up to give us a nice pretty quasi sphere.

It isn't magic, it's a property resulting from the electron pressure.

Go back and review Indrek's work. You can see how big the holes are.

alexjrgreen wrote:
The second piece of structure is that the plasma falls out of the holes in jets, actually tubular double sheets, focussed by the magrid.

Past statements by Dr. N have suggested to me that the ions do NOT leave the MaGrid in any appreciable numbers. The electrons do, but are returned rapidly by the MaGrid charge.

I'll just add "I second that."

So ions are easier to contain than electrons? Dr Bussard had something to say about that...

[Art Carlson]

The clues are Rick's insistence on collective mechanisms and Art's insistence that where the electrons go, the ions follow and vice versa.

But at the end of one of those long discertations by Art, Dr. N seemed to disagree with Art's "wither they goest" hypothesis. His statement (IIRC) was that the Polywell is quasi-neutral, NOT ambipolar.

I have never insisted on ambipolarity. My technical specialty is Langmuir probes, which live from the fact that the flux of electrons depends on the electric potentials, whereas the ion flux usually doesn't. The importance of quasi-neutrality is that you don't have the freedom to stop the ions at one position with an inwardly directed electric field and then stop the electrons farther out with an outwardly directed electric field.

[KitemanSA]

...

The second piece of structure is that the plasma falls out of the holes in jets, actually tubular double sheets, focussed by the magrid.

Past statements by Dr. N have suggested to me that the ions do NOT leave the MaGrid in any appreciable numbers. The electrons do, but are returned rapidly by the MaGrid charge.
....

So ions are easier to contain than electrons? Dr Bussard had something to say about that...

Given that the ions are (or have been) introduced inside the well and only gain enough energy to leave it via upscattering, and given that the ions "anneal" thru maxwellization at the low energy edge of the well, then yes, they are esier to contain. That is the WHOLE PURPOSE of the Polywell, to contain fusile ions easily.

[alexjrgreen]

the flux of electrons depends on the electric potentials

So the electrons fall out of the hole in the wiffleball and head for the magrid. Being forced to miss it by the magnetic field, they pass it, stop, and oscillate back to the wiffleball. Indrek and MSimon think they oscillate out and back several times.

whereas the ion flux usually doesn't. The importance of quasi-neutrality is that you don't have the freedom to stop the ions at one position with an inwardly directed electric field and then stop the electrons farther out with an outwardly directed electric field.

So the ions leave the wiffleball with the electrons, and return to it with them as well.

[Art Carlson]

the flux of electrons depends on the electric potentials

So the electrons fall out of the hole in the wiffleball and head for the magrid. Being forced to miss it by the magnetic field, they pass it, stop, and oscillate back to the wiffleball. Indrek and MSimon think they oscillate out and back several times.

whereas the ion flux usually doesn't. The importance of quasi-neutrality is that you don't have the freedom to stop the ions at one position with an inwardly directed electric field and then stop the electrons farther out with an outwardly directed electric field.

So the ions leave the wiffleball with the electrons, and return to it with them as well.

What causes the electrons, after they pass the magrid, to stop and turn around?
What causes the ions, after they pass the magrid, to stop and turn around?
(Hint: "the electric field" cannot be the answer to both questions.)

(My apologies if I posted this twice. I am in panic mode getting ready for a big conference.)

[TallDave]

Go back and review Indrek's work. You can see how big the holes are.

With all due respect to Indrek, those are rough simulations. They're very interesting but I wouldn't depend on them for detailed or definitive answers.

So ions are easier to contain than electrons? Dr Bussard had something to say about that...

Given the same forces, no. But they're experiencing different forces. The ions are contained electrostatically by a force pulling them in, the electrons are confined magnetically against a force pulling them out.

[KitemanSA]

What causes the electrons, after they pass the magrid, to stop and turn around?
What causes the ions, after they pass the magrid, to stop and turn around?
(Hint: "the electric field" cannot be the answer to both questions.)

(My apologies if I posted this twice. I am in panic mode getting ready for a big conference.)

As far as I am aware, the ions, once they pass the MaGrid, DON'T stop and turn around. They get accelerated to the chamber wall as a loss to the system, and a potential source of neutrals that need removal to prevent arching. The point is, they don't leave the MaGrid except in minor amounts due to unannealed upscattering.

• The virtual negative electrode at the center of the well keeps the ions in.
  To a large extent, the wiffleball keeps the electrons in.
  To a minor extent, where electrons have made it out the cusps, the positive electrode of the MaGrid keeps the electrons in.

Simple. No?

[93143]

I just thought of something.

If the initial pickup of the direct conversion system is at a sufficient distance (potential-wise) from the trap grid where the field reverses, the ions that escape will come out the cusps, accelerate to the trap grid, pass it, slow down and reverse, pass the trap grid going the other way with the same energy, and head back down into the wiffleball, decelerating as they go.

Barring collisions, of course.

On the other hand, it might be difficult to actually get back in, because the ions are travelling uphill for most of the return trip, and will tend to veer off sideways...


I still think the ion population will have much lower leak rates than the electron population, and that multiple well formation will resolve the quasi-neutrality paradox without resorting to a neutral Maxwellian plasma ball with a sheath - but I don't have a simulation yet and I'm not likely to any time soon (dissipation from velocity remapping alone could swamp collisional dissipation if I'm careless/unlucky, and I have other things to tackle first before I try to make the simulation work, not to mention that I'll probably have to re-code it in C++ with MPI and run it on the school's high-performance cluster if I want to actually get anywhere. This is quasi-1D Boltzmann, by the way, and yes, it assumes a spherical wiffleball, with the cusps represented by a finite porosity. It may or may not be ready by the time we get the results from WB-8, or ever).

[D Tibbets]

The clues are Rick's insistence on collective mechanisms and Art's insistence that where the electrons go, the ions follow and vice versa.

But at the end of one of those long discertations by Art, Dr. N seemed to disagree with Art's "wither they goest" hypothesis. His statement (IIRC) was that the Polywell is quasi-neutral, NOT ambipolar.

I have never insisted on ambipolarity. My technical specialty is Langmuir probes, which live from the fact that the flux of electrons depends on the electric potentials, whereas the ion flux usually doesn't. The importance of quasi-neutrality is that you don't have the freedom to stop the ions at one position with an inwardly directed electric field and then stop the electrons farther out with an outwardly directed electric field.

Reading this several times, I realize that I don't fully realize what you are saying. I'm guessing from a strictly electrostatic viewpoint it makes sense. I think my understanding of quasineutrality is regressing again. Isn't quasineutrality a term describing the variable charge accumulations (not quite neutral) locally that is limited by Debye lengths? The plasma overall is neutral. Or, is a term describing a plasma that is not quite neutral overall? Is it interchangable? The Polywell plasma is not neutral . Due to the excess electrons, the ions experiance a increased bias towards moving towards the center; while, due to the excess of the electrons, the electrons have an increased bias towards accumulating away from the center(more so than they would based on a balanced number of positive and negative particles occillating back and forth). How this acceleration is distributed within these limits is determined by the shape of the potential well (enter here arguments about radial flows, continous high energy radial electron injection, impeaded thermalization, slow in the center - fast on the edge and visa versa- depending whether it is an electron or ion (and on the degree of central vertual anode formation)). I'm not sure whether this would keep the ion below the average Wiffleball border (assuming the ion was born there), but it should limit it's travel beyond that point. Within the Magnetic border there is effectively a stronger electric field acting to pull the ions in, and effectively stronger electric field pushing the electrons out. This unstable 'imaginary' condition is maintained by the magnetic containment of the excess number of electrons combined with their high energy bounces off of the magnetic border. The cost of doing this is minimized through the small Wiffleball holes, and recirculation of most of the electrons that do find these holes; and limited by the tolorable space charge(?), coulombs of charge that can accumulate before breakdown occurs. If this 'imaginary' distortion of the fields has any reality on the charged particals, then I assume it would be detected by a Langmuir probe( based on your comment above). I understand Tom Ligon did some work with these, and I understant that they are very tempermental beasts.

At the Wiffleball average border the electrons would act oppositaly, there would be more force pushing it outward through a cusp, till the magrid was reached. If this electron is passing an ion in this area it might tug on the ion some, but once past, it would be invisible to the ion, because it is counterbalanced by the electros traveling through the other cusps that have reached outside of that ion's radius (Gauss's Law excuse again). The complications with Debye sheeths, etc confounds me.

Hopefully my arguments are not too foolish, and my lack of the proper technical jargen does't impead my arguments too much. ie- listen to what I mean, not to what I say,

Dan Tibbets

[D Tibbets]

I just thought of something.

If the initial pickup of the direct conversion system is at a sufficient distance (potential-wise) from the trap grid where the field reverses, the ions that escape will come out the cusps, accelerate to the trap grid, pass it, slow down and reverse, pass the trap grid going the other way with the same energy, and head back down into the wiffleball, decelerating as they go.

Barring collisions, of course.

On the other hand, it might be difficult to actually get back in, because the ions are travelling uphill for most of the return trip, and will tend to veer off sideways...


I still think the ion population will have much lower leak rates than the electron population, and that multiple well formation will resolve the quasi-neutrality paradox without resorting to a neutral Maxwellian plasma ball with a sheath - but I don't have a simulation yet and I'm not likely to any time soon (dissipation from velocity remapping alone could swamp collisional dissipation if I'm careless/unlucky, and I have other things to tackle first before I try to make the simulation work, not to mention that I'll probably have to re-code it in C++ with MPI and run it on the school's high-performance cluster if I want to actually get anywhere. This is quasi-1D Boltzmann, by the way, and yes, it assumes a spherical wiffleball, with the cusps represented by a finite porosity. It may or may not be ready by the time we get the results from WB-8, or ever).

If an upscattered ion escapes confinement at (eg) 10,000 eV and is acceleratedawat by (eg) a 100,000 magrid positive charge, it would pass the first direct conversion grid/plate and be stoped/ neutralized if the conversion grid was at (eg) 111,000 negative volts. If the collection grid was at 210,001 volts and magnetically shielded, the ion would fall back and enter the cusp at an energy of 1 eV at the magrid border. But, this would have a higher internal speed (only 1 eV) compared to new ions introduced below the magrid but it's higher position at the top of the potential well might increase upscattering of other ions to a degree that it might be mildly harmfull (?). In which case, it might be better to just collect what energy you can and let it ground on structure outside the magrid. It would set the minimum needed potential on the first collection grid if you are trying to limit this energy loss. If the numbers of these escaped fuel ions is low enough compared to the fusion ions streaming outward, you might go ahead and set your first collection grid to 2 million volts, or whatever is needed to capure the energy range of the fusion ions, and accept the energy wasted as heat from the fuel ion. This might increase sputtering problems though, so you may need to place that first lower voltage collection plate back in.

Dan Tibbets