A few questions on Polywell facts and figures.

[TallDave]

Well, you said before you envisioned Polywell as having a ball of electrons confined in the core, presumably with ions falling into and out of it like in a fusor.

The link shows the virtual wells that can form in a fusor. Polywell uses a virtual well (formed by the electrons).

[chrismb]

Yep. That's how I saw it, but my misunderstanding has been corrected now. It does not appear that there is a ball of electrons, in excess of the ions around it, at the centre.

[bcglorf]

Yep. That's how I saw it, but my misunderstanding has been corrected now. It does not appear that there is a ball of electrons, in excess of the ions around it, at the centre.

I'd curious why you don't believe there will be an excess of electrons in the centre. From everything Bussard ever said the entire concept revolved around magnetic confinement of electrons so that ions could be contained electrostatically. Even if the core region is quasi neutral, the ions are still being confined by the electron ball in the centre, just near the capacity of the electron ball's ability to do so. I don't see any reason one can expect any success from the polywell design if the magnets are expected to do anything to confine the ions beyond the 'proxy' of keeping the electrons confined.<again by confined I mean loosely and possibly requiring more electron injection to replace losses than you'll ever get back from fusion>

[Betruger]

I don't understand what you mean by this:

And, oh yeah, it [the tokamak] is a globally unstable flow topology.

Or this:

If the wiffleball effect is a true spherical inversion of the magnetic field carried by the plasma, as posited here,
viewtopic.php?t=650&postdays=0&postorder=asc&start=60
then the cusps effectively close, and it is game on with a near perfect field topology for containment being satisfied.

Art, yes, the tokomak is so brilliant it has taken already 50 years to demonstrate that it might be possible to build one that might one day work, given many more billions. ... For now, it has proven to be a blind alley of holy grail proportions, how many more resources does one throw at it to prove such a brilliant idea can actually work?

And that despite the fact that - given only the concepts and a good knowledge of theoretical plasma physics - the tokamak beats the polywell hands down. What does that say about the chances of successfully scaling a polywell up by a factor of a million?

I don't want to get out of line here (I'm nowhere near qualified in the proper physics), but I don't think the Tokamak and Polywell are comparable that way. The Tok has much more time and money into study and development than the Polywell. One of the reasons the Tok seems to continue being funded is that it always seems within reach, and yet it's been so many years and so much funding without actualy reaching it.
Given only the concepts and knowledge of theoretical physics, the Tok ought to be a done deal already, and the Polywell ought to never have gotten so far.

Now, if only we could see the experimental data.. This could be settled properly.

[chrismb]

I'd curious why you don't believe there will be an excess of electrons in the centre.

Because of the first page of this thread, and particularly Art Carlson's post of Mon Dec 15, 2008 4:28 pm, which no-one seemed to correct otherwise.

I have been arguing in this thread that if there were a bunch of electrons at the middle, they would be pulled out towards the ions equally and oppositely to the ions they are supposedly pulling in. This sets a limit to the number of ions that can be pulled in.

That bit of the argument is fine as it does not necessarily conflict with anything at low ion counts. I don't have a big problem with that, per se, but if you do the calculation at fusion power outputs then there will just be too many ions that need to be pulled in for the electrons to cope.

To be frank with you, I have to say, and I am sorry to be less that scientific and less than friendly sounding (I really am sorry, faith in Polywell is high and I would never want to discourage), but if you work through my argument you'll see it is all gets a bit silly. The electrons will have such a high mobility in the electron/ion mix (it's got to be a real, genuine plasma, nothing special, but I will offer up the possibility that it is some non-maxwellian mix) that as soon as a charge discontinuity exists, the electrons will swarm around it like flies around hot jam until they neutralise it. This will happen at close on to light speed - we are talking about an electric current passing through a substance of almost zero resistance here! The electrons will have surrounded and neutralised the field of an ion before the ion has moved a micron. I see absolutely no reason why a charge would be held up at the centre and I see no reason why there would be a field gradient. This will be a thermalised plasma - essentially instantly.

An externally charged set of electrodes will do nothing to influence this. I have asked directly, first hand, the question of whether holding the surrounding walls of a fusion plasma chamber has any effect to scientists at Culham (JET) and I have been told that this *has been tried out*. Holding up the walls of a plasma chamber has no beneficial effect on confinement, it does not create some magic electric gradient in a plasma and the electrons do not migrate to the centre, or anywhere. It's just a plasma in a vessel with walls at a differential potential to the plasma. This is what I have been told - and I accept it as true.

So I don't know how to say it in any less direct terms, but I do not accept that you will end up with anything but a thermalised plasma in a Polywell, one way or the other. The electrons MUST be pulled out to the ions. The electrons and ions are pulled TOWARDS each other. There is nothing to hold the electrons at the centre. This has to happen.

It MUST be as easy for ions to go in the opposite direction to the paths of the electrons without a magnetic field. There is no mechanism that can stop free ions and free electrons from pulling each other towards each other except a magnetic field. hence I thought there was a 1T magnetic field penetrating to these central electrons. But I am told not. there is nothing there. So.. no field, then the electrons and ions must pull each other along the same path, and the electrons will meet the ion before you even notice the ion has moved.

EVEN IF there was a 1T field holding the electrons at the centre, then that field STILL wouldn't hold the electrons in place, so there's no point even trying that either.

Well, there it is.

best regards,

Chris MB.

[bcglorf]

I have been arguing in this thread that if there were a bunch of electrons at the middle, they would be pulled out towards the ions equally and oppositely to the ions they are supposedly pulling in. This sets a limit to the number of ions that can be pulled in.

This is something Art stated more eloquently than I probably will. The ions are a shell around the electrons and so the electrons don't see any charge from them. Even if there where a single electron surrounded by a dense shell of ions, it would be strongly pulled in EVERY direction, and would have 0 net force on it from the shell of ions.

The ions on the other hand are sitting on the outside of a sphere/shell/wiffleball of electrons, and will be strongly attracted to the center of that negative charge. The ions will orbit through this core until they fuse or are lost for any myriad of other reasons, primarily upscatter/thermalizing through ion-ion collisions. (setting aside other major things like brems for the moment)

If I'm not completely ignorant I believe Art's objections come in regarding effectiveness of electron confinement and rate of ion-ion thermalization, but not the above basic principals?

[Art Carlson]

There's a big difference. The polywell plasma stays put but loses particles through the cusps. The toroidal plasma moves as a whole across the magnetic field. It's like the difference between putting a leaky balloon in a cage and putting a balloon in a bucket of water. If you estimate the time scales, the plasma is lost from a torus (unless there is a rotational transform) a factor of R/delta_sh ~ 1000 times faster than from a cusp of comparable size.

The difference is therefore the toroidal plasma will migrate across magnetic surfaces, whereas the Polywell plasma would migrate across the magnetic surfaces AND down the cusps.

I get the impression that so much attention is paid to the cusp losses that it is forgotten that there is also migration straight across the magnetic surfaces, no more nor less than with any other magnetic surface of the same strength.

To say there isn't and to dive into a discussion on cusp losses instead is to confuscate the fact that this is still a magnetic 'confinement' which is a cross-product force and cannot 'confine' with a normal reaction force. This was Bussard's principle objection, but perhaps he should have highlighted the fact that it also applies equally to every steradian of the wiffleball, not just the cusps.

Let me say it again. There is a qualitative difference between a toroidal plasma and a polywell, related to the fact that the confining fields of a polywell are all convex to the plasma, while for a toroidal plasma some field lines are concave.
Is the polywell in equilibrium? Do the jXB forces balance the pressure forces everywhere on the surface? Yes!
Is a toroidal plasma in equilibrium? No! The jXB force on the inner midplane (at r=R-a, z=0) is larger than the jXB force on the outer midplane (at r=R+a, z=0).
This qualitative difference results in a quantitative difference in the configuration lifetime of a few orders of magnitude.

Of course, once you add a rotational transform to your toroidal plasma and make a tokamak, a stellarator, or a reversed-field pinch, you have a new ball game, with the toroidal machine a clear winner. Still, the problems that remain with toroidal confinement are largely the residual effects of the bit of nasty curvature.

You say "magnetic 'confinement' ... is a cross-product force and cannot 'confine' with a normal reaction force". I can't make any sense out of that statement.

[MSimon]

Chris,

Here is a nice paper on well formation:

http://wwwsoc.nii.ac.jp/aesj/division/f ... hikawa.pdf

I had my doubts until I had read that.

[chrismb]

Let me say it again. There is a qualitative difference between a toroidal plasma and a polywell, related to the fact that the confining fields of a polywell are all convex to the plasma, while for a toroidal plasma some field lines are concave.

hmm... the posts and papers I have read suggest that a stable wiffleball would be the height of Polywell success - which is concave, is it not?? (and a 3D curvature at that rather than a toroid's 2D curvature).

So is a wiffleball a good idea, or a bad idea? Me confused!

You say "magnetic 'confinement' ... is a cross-product force and cannot 'confine' with a normal reaction force". I can't make any sense out of that statement.

This was one of Bussard's own quotes, maybe transliterated from my memory. I quite liked the expression! A normal reaction is one that works back against the motion of an object trying to get past, e.g. gas molecules bouncing off the insides of a gas cylinder. The cylinder pushes back on the gas molecule precisely and exactly so as to oppose its motion.

A magnetic field is a cross-product field and does not push back against the direction of motion but pushes sideways. Imagine that instead of having bars on a prisoner's cell, we just fit a geat big fan that pushes air at high speed across the opening of the cell door. That way the prisoner will get pushed sideways as he tries to get out. That's a 'cross-product' force. It may give the guy trying to get out a hard time, but it doesn't directly oppose him so he can get out if he runs through the generated wind quick enough.

In point of fact, I would say it is technically wrong to say a tokamak is a 'magnetically confined' plasma - it is actually confined by the divertors as this is the physical part that provides a reaction force to particles in the 'scrape-off-layer' and that stops particles from migrating any further away from the plasma.

[apouliot]

The tokamak and the BFR are 2 totally different beast.

I know most people here have an understanding of how the polywell work. But I'm trying to make an explanation back from the basic.

The Tokamak use magnetic field to direct along a toroid path and force ion together by having such an intense magnetic field. Since ion are heavy and you need huge pressure for fusion to occur the need for strong magnetic field is present.

On a standard (IEC)fusor reactor there is a grid that form the cathode. That grids was the principal source of loss for fusor because of the impact, you need a grid nearly 98% invisible if I remember right. That is impossible from using available material.

For the BFR the magnetic field is there to hold the electron in the center of the reactor. When there is enough electron in the center, the magnetic field is deformed by the electron own magnetic field. This create the wiffleBall effect. This create a virtual cathode that allow removing the grid of the fusor. The electron that get out of the wiffleball are recaptured and brought back by the magnetic field toward the center. The actual wiffleball is rather far from the magnet. the farther the better.

All those electron form a virtual cathode. The potential well is formed from a positive charge on the outer wall(anode) and the electron forming the negative potential.

When injecting the ion in the system now you have a electric field created between the outer wall and the wiffleball. Since the ion are push by the electrical field they converge toward the center. You up the potential on the outer wall to increase the ion speed at the center. The ion going toward the center now have great speed. There are 2 possibility: either impact with another ion to get fusion or not. If the ion don't hit another ion there is recirculation happening. The ion go through the center toward the wall and get push back by the outer wall toward the center again.

To keep that dynamic you need 1 condition you must keep the proportion of electron higher than the quantity of ion in the center. There is no need for a lot of difference just enough.

What the wiffleball really allow is that the "grid" it create is invisible, that should allow to get more energy from the system than injected in it, when scaled appropriately.

[TallDave]

Thanks apouliot, one of the best descriptions I've seen.

[chrismb]

Can I please clarify this description;

Do
A) ions go from the edge of the device and accelerate all the way down a potential gradient, through the wiffleball's magnetic surface and into the ball of electrons. If it does not collide, it comes back out, out of the wiffle ball and back up to the high potential on the other side of the device?

or

B) ions are all reciprocating back-and-forth across and within the wiffleball?

The problem is that if A) then the ion will never get there as it has to cross all these magnetic surfaces and its gyroradius is far too small for it to ever reach the centre, or B) the electrons and ions will be in some form of quasi neutrality and will thus settle into a 'confined' plasma state.

You can't have it both ways. You can't say that there's no mag field in the wiffleball, but then on the outside expect ions to queue up patiently in line as they funnel down into it from the outside.

My discussion seems to being kept juggled from one description to the other, depending on which is most suitable to deflect my particular critique of the moment, and I am now quite confused by what is intended.

[D Tibbets]

I'm confused again (or is that yet?). Your statments seem to imply that the gyroradius is the distance a particle can penitrate through a magnetic field befor it is stoped/ translated ito lateral motion. My understanding that the gyroradius is: "Radius of orbit of charged particle about a magnetic field line. Also called gyroradius, Larmor radius". ie, once traped on a magnetic field line, the gyroradius is the distance that the particle spirals around the field line, not a measure of how deep it can penitrate into a magnetic field befor it is stoped.

And, the arguments for the Polywell that I am aware of sets the magrid magnetic field to a strength nessisary to stop the electrons as they approach the positively charged magrids, and this strength is way to small to stop/ divert the much heavier ions. So, from the ions viewpoint, the magnetic field effects (weather the magrid or the opposing wiffle ball) are trivial. The ions motions are therefor almost totally dependent on the electrostic effects only.

Dan Tibbets

[Art Carlson]

You say "magnetic 'confinement' ... is a cross-product force and cannot 'confine' with a normal reaction force". I can't make any sense out of that statement.

This was one of Bussard's own quotes, maybe transliterated from my memory. I quite liked the expression! A normal reaction is one that works back against the motion of an object trying to get past, e.g. gas molecules bouncing off the insides of a gas cylinder. The cylinder pushes back on the gas molecule precisely and exactly so as to oppose its motion.

Bussard said a lot of things that don't make sense.

[A magnetic field is a cross-product field and does not push back against the direction of motion but pushes sideways. Imagine that instead of having bars on a prisoner's cell, we just fit a geat big fan that pushes air at high speed across the opening of the cell door. That way the prisoner will get pushed sideways as he tries to get out. That's a 'cross-product' force. It may give the guy trying to get out a hard time, but it doesn't directly oppose him so he can get out if he runs through the generated wind quick enough.

Your fan always works in the same direction relative to the cell. The vXB force changes its direction as v changes. It's more like giving the guy a ball to drag on his left leg. It won't stop him completely, but it'll make him walk in circles. How's he going to get out then?

In point of fact, I would say it is technically wrong to say a tokamak is a 'magnetically confined' plasma - it is actually confined by the divertors as this is the physical part that provides a reaction force to particles in the 'scrape-off-layer' and that stops particles from migrating any further away from the plasma.

This doesn't make sense. Reaction force as in Newton's third law and rockets? Maybe you should leave the technical definitions of tokamak physics to tokamak physicists.

[chrismb]

This doesn't make sense. Reaction force as in Newton's third law and rockets?

Really, and, pray tell, how much of the vacuum vessel would the plasma fill if the divertor was not there??

Your answer should then be 'up to the wall', or whatever other physical structure gets in the way of the plasma first.

If it was 'magnetically confined' then it would/could be the magnets alone that could hold in a plasma from escaping beyond a specific volume. But this doesn't happen. You could call it 'magnetic/divertor' confinement if you wish, but 'magnetic confinement' is well understood but potentially misleading as it suggests it confines like a bottle or a box does, yet - it doesn't.

'Confinement' is done by the thing which stops stuff going any further than some given point. At which point does the magnetic field in a tokamak, or in a Polywell, actually and properly stop it from migrating any further?

Maybe you should leave the technical definitions of tokamak physics to tokamak physicists.

You mean like the term 'magnetic bottle'? I feel yours is a slightly ad hominem comment and seems to demand homage to the mighty high priests of science. If I demonstrate a term can be mis-understood, it is not for anyone else to say I should have known better in the first place but for them to argue the points. You have made points, and I have responded. So the last comment of yours is redundant.

Ignorance is not knowing something, stupidity is not knowing something that should have been known. I freely admit to ignorance on a whole range of topics, even mistakes of misunderstanding. If you wish to argue that I should know what tokamak physicists mean when they say things that don't quite add up in my understanding, then feel free to argue that point.

Personally, I do not know any tokamak scientists who would refuse a discussion on a definition of a technical term. They seem quite willing to discuss. Agreement may be a different thing, of course, but that is debate for you.