Why people are so optimistical to Polywell?

[D Tibbets]

Joseph Chikva, I'm not sure of the level of understanding you are seeking, but the details of the Polywell have been studied in depth by Bussard and presumably others. My understanding is superficial at best. But, I have not yet seen any issue that has not been addressed. There may be arguments about the validity of certain claims, but there are no glaring holes in the understanding of the various physics that has not been addressed (successfully, of not, based on contested assumptions).

No particles disappear unaccountably from the system. They are upscattered, downscattered, side scattered or restored by simple particle collision processes, or more complex group interactions, waves, etc. Magnetic , space charge and local collisions all play a role in a dynamic fashion. Static interpretations can be misleading.
If you are thinking that upscattered electrons disappear, they don't. They leave through the cusps after a few thousand passes inside the magrid, just like all the other electrons (except those that manage to diffuse through the magnetic fields to the magrid coil metal casings). The difference is that the upscattered electrons that are not accelerated back into the magrid at the KE determined by the positive charge on the magrid, instead go to a grounded surface where they are either absorbed and leave as a current in the grounded casing (or by depleting somewhat the charge on the magrid if the power supply isn't keeping up). If the upscattered electron has too much residual KE it might cause some sputtering of wall materials. This sputtered gas would need to be removed quickly by the vacuum pumping system. The same applies to escaped fuel ions, and in a similar manner, the fusion ions. This is another possible advantage of direct conversion. Most of these fusion ions would be decelerated to much lower energies before they hit something. Sputtering, etc. will be much more manageable.

As for upscattered ions, you need to appreciate that at the densities and sizes, and KE of the ions during most of their transit have a MFP greater than the diameter of the machine and thus local conditions near the Wiffleball edge where the ions slow to very slow speeds (KE), allows for the ions to thermalize very quickly in this region. The important point is that this thermalization is around a very low KE, so compared to the maximum energy as they fall back into the potential well, this thermal spread is minimal, and resets on each pass of the ion.

As for lateral scattering - increased angular momentum, it is anticipated that with some degree of convergence, most Coulomb collisions will occur in this denser area (except for the edge annealing region where the local colisionality of the cold ions would dominate- but at very low scattering speeds), and collisions near the center of a sphere cannot impart much angular momentum.

Of course not all ions are confined and conditioned in this manner. Some just have too much energy, and may escape if they hit a cusp and fly to the vessel wall, stealing energy from the magrid in doing so. A mitigating factor is that even if an upscattered ion escapes the potential well, it will be turned by the magnetic containment (unless it hits a cusp- ~ one chance in a thousand) and reenter the annealing region- essentially the annealing forces would have two chances to slow the upscattered ion. But, these losses are apparently much less than the electron losses. Based on my appreciation of confinement times, electrons are lost (even with recirculation) at a rate ~ 10 to 20 times faster. And my understanding is that most ions will fuse before they escape in this fashion. According to Bussard, this resultant fusion ion escape may actually increase the negative potential well as their electrons are left behind (in a gas puffed system, or where multiple Z fuel like boron is injected only partially ionized). This may require careful management of the electron gun current at lower levels to compensate.


The Polywell seems to borrow the advantages of magnetic confinement (of electrons), and electrostatic confinement of ions with the potential well formation, while at the same time seemingly avoiding most of the disadvantages of each. This is remarkable, but based on claims and admittedly sparse data, it apparently can work.

Dan Tibbets

[D Tibbets]

The electrons are neither absent nor thermal in the center. Ideally, the electrons have a KE of ~ 10,000 eV (in this example) at the outer border of the Wiffleball, but at the center their KE would mostly be much closer to 0 eV, while the ion KE distribution would be the reverse of this.

I'm thinking that electrons would have a somewhat broader than maxwelian distribution in the center. The absolute width of the energy distribution wouldn't vary much between high energy edge and low energy center, while the relative width would vary from near mono-energetic to broader than thermal. Ions would do the same with positions reversed.

Yes, though I'm uncertain if you are interpreting this as an advantage or disadvantage. The ion edge annealing, or core electron annealing if it occurs definitely would thermalize to a large degree in these locals. But, just to reinforce the point, the thermalized spread in these regions is about a very low number. The spread may be +/- 100 times the average, but if the average KE is only 1 eV, the thermalized spread would be ~ 0 to 100 eV. This would be for all piratical purposes be a monoenergetic starting point for the next pass.
I have not seen this electron annealing mentioned in the Polywell literature so if it occurs I suspect it is much less significant than the ion annealing. Either that or it is irrelevant because the electron lifetime/ thermalization time ratio is already adequate.

Dan Tibbets

[TallDave]

The electrons are neither absent nor thermal in the center. Ideally, the electrons have a KE of ~ 10,000 eV (in this example) at the outer border of the Wiffleball, but at the center their KE would mostly be much closer to 0 eV, while the ion KE distribution would be the reverse of this.

I've been thinking about this and re-reading some of Rick's comments and I think I probably had a wrong conception of this -- electron behavior is probably stochastic across the interior rather than fast at the edges and slow in the middle. I don't think the electrons can both create the well and see the well in a way that creates the well -- it would be like climbing by putting your feet on your hands. In other words, their speed is roughly uniform, and the center has a well (for the ions) simply because the center is the electrons' average position.

There may seem to be a discrepency in the above numbers, but I believe Nebel was comparing a D-T Tokamak to a D-D burning Polywell at optimal/ achievable KE crossection levels.

It was an "all else being equal" kind of comparison. Roughly speaking, 250 times greater density means 62,500 times greater power, whatever fuel/temp you're using. It's basically a result of being a higher-beta device -- with similar magnet strength, PWs should have much greater density. (Whether their transport would be acceptable is another question...)

[93143]

I've been thinking about this and re-reading some of Rick's comments and I think I probably had a wrong conception of this -- electron behavior is probably stochastic across the interior rather than fast at the edges and slow in the middle. I don't think the electrons can both create the well and see the well in a way that creates the well -- it would be like climbing by putting your feet on your hands. In other words, their speed is roughly uniform, and the center has a well (for the ions) simply because the center is the electrons' average position.

No, a potential well is a potential well. If it exists, the electrons will see it.

An individual electron doesn't react to its own charge. But it does react to the charges of other electrons.

[D Tibbets]

The electrons are neither absent nor thermal in the center. Ideally, the electrons have a KE of ~ 10,000 eV (in this example) at the outer border of the Wiffleball, but at the center their KE would mostly be much closer to 0 eV, while the ion KE distribution would be the reverse of this.

I've been thinking about this and re-reading some of Rick's comments and I think I probably had a wrong conception of this -- electron behavior is probably stochastic across the interior rather than fast at the edges and slow in the middle. I don't think the electrons can both create the well and see the well in a way that creates the well -- it would be like climbing by putting your feet on your hands. In other words, their speed is roughly uniform, and the center has a well (for the ions) simply because the center is the electrons' average position.

Also, I don't know how this would fit within the arguments about the Bremsstrulung being less due to slower electrons in the center. Perhaps the electron dragging above plays a role. Perhaps , even though there is not a large radial nature to the electron movements, there is a differential in the electron velocity.
Perhaps ....

...

I admit I don't have a clear picture of the electron distibution. The initial injection would certainly be radial towards the center- not precisely the center as the injection through the cusps results in some dispersion. As the electrons would deflect off of the magnetic surface at a small angle, and this would tend to increase their angular momentum on each pass till a 'balance' was reached- pretty much a random bouncing around within the Wiffleball. This would fit with what A. Carlson claimed, and I eventually discovered that Bussard said in his Google talk. But, then Bussard said that the electrons are dragged towards the center somewhat by the ions. I'm not sure why they would not be drawn outward at the same rate (actually I do- the difference in inertia between the ions and electrons), but Bussard said this action is what creates the parabolic well from the initial square potential well. So there are two forces that causes deviation from a pure square potential,- the residual injection/ recirculation dominate radial vectors and the weak inward coupling between the ions and electrons. I suspect the latter is by far the greater effect. The electrons would be mostly radial on their first pass, but with 1000-10000 passes before recirculation, this would not last long.
How this relates to the formation and maintainance of a deep potential well is confusing. Also, visualizing how this 'bag' of electrons confines the ions before they can reach the wiffleball border- or just beyond it. Visualizing the electron cloud as a series of shells helps. Also, if the ions are introduced a modest distance within the Wiffleball border, I can see them as remaining within this Wiffleball border (their potential well peak is inside the Wiffleball border, not at the Wiffleball border. This makes sense for a gas puffed system, but I'm not sure how it would be managed with an ion gun without sticking the ion gun nozzle within the Wiffleball and this would presumably be a huge electron loss portal.

Also, I don't know how this would fit in with arguements about two stream instabilities- unless that concerns mostly energy (upscattering and downscattering) issues. Also, I don't know how it would fit with Bremsstrulung suppression arguments.

Dan Tibbets

[Joseph Chikva]

Why nobody reflects?
Why it is so much assumptions of the people who are not in the core a plasma physicists? Why theoretical articles and results of the led experiments are a little quoted?
Unlike for example TOKAMAKs.
I see only sketchy data about loss of particles and the strong braking radiation but a lot of reasonings on MaGrid design.
But significant loss of particles and the strong braking radiation may be only as a result of instabilities and then secondary emission of heavy ions into plasma.

[Joseph Chikva]

Of course not all ions are confined and conditioned in this manner. Some just have too much energy, and may escape if they hit a cusp and fly to the vessel wall, stealing energy from the magrid in doing so.

I do not see any other possibility for ions to have too much energy besides instability.
I am repeating once again: instability's wave accelerates some ions at the expense of energies of others.

Also you see the advantage for Polywell in its spherical geometry. I know one very effective way to fight with instabilities - creation of a strong longitudinal magnetic field that dramatically expands the stability area. In spherical geometry it is impossible.

[93143]

Why it is so much assumptions of the people who are not in the core a plasma physicists? Why theoretical articles and results of the led experiments are a little quoted?
Unlike for example TOKAMAKs.

We don't have a lot of hard data. Unlike tokamaks, there's basically just the one research group that's been working on Polywell, and the U.S. Navy has been preventing them from publishing. There was a brief window in the data embargo in 2006-2007 when the Navy defunded the project, but the release was sparse, and the Navy has re-funded the project and reinstated the gag order.

I do not see any other possibility for ions to have too much energy besides instability.

Thermalization.

The ion energy varies dramatically between different regions in the plasma. Low-energy thermalization is supposed to dominate over high-energy thermalization, but it's still possible for a given ion to pick up a significant amount of energy via collisions in a high-energy region and shoot right through the outermost annealing zone without stopping.

Also, if the system uses ion guns, the ion energy at the edge of the wiffleball may be significant, because the ions have to cross the potential between the gun (presumably on the inner surface of the magrid) and the wiffleball to get in.

...

Are you a plasma physicist?

[Joseph Chikva]

Why it is so much assumptions of the people who are not in the core a plasma physicists? Why theoretical articles and results of the led experiments are a little quoted?
Unlike for example TOKAMAKs.

We don't have a lot of hard data. Unlike tokamaks, there's basically just the one research group that's been working on Polywell, and the U.S. Navy has been preventing them from publishing. There was a brief window in the data embargo in 2006-2007 when the Navy defunded the project, but the release was sparse, and the Navy has re-funded the project and reinstated the gag order.

I do not see any other possibility for ions to have too much energy besides instability.

Thermalization.

The ion energy varies dramatically between different regions in the plasma. Low-energy thermalization is supposed to dominate over high-energy thermalization, but it's still possible for a given ion to pick up a significant amount of energy via collisions in a high-energy region and shoot right through the outermost annealing zone without stopping.

Also, if the system uses ion guns, the ion energy at the edge of the wiffleball may be significant, because the ions have to cross the potential between the gun (presumably on the inner surface of the magrid) and the wiffleball to get in.

...

Are you a plasma physicist?

Not, I am not a plasma physicist too. I am Ph.D. mechanical engineer.
So, I can't judge but can put the questions within the limits of my restricted knowledge. Also certainly to assume something.
And that limited knowledge which I have was acquired by self-education. Because I invented a new Method of nuclear fusion and interested in fusion problems.
No classic plasma physics education.

And I assume:
By my opinion thermalization can not give ions the energy high enough for leaving the potential well.
I think that together with significant thermalization and so - big enough difference between of arrange velocities of electrons, two-stream instability appears and accelerates some ions with deceleration of others.
Accelerated ions will have KE big enough for leaving potential well.

[Joseph Chikva]

We don't have a lot of hard data. Unlike tokamaks, there's basically just the one research group that's been working on Polywell, and the U.S. Navy has been preventing them from publishing. There was a brief window in the data embargo in 2006-2007 when the Navy defunded the project, but the release was sparse, and the Navy has re-funded the project and reinstated the gag order.

In addition:
It would be interesting to me to see braking radiation spectrum analysis for example.
This will confirm or refute my hypothesis.

And if my hypothesis is true, many heavy ions will be in plasma.
I do not think that the navy lab did not use spectroscope. It is easy to use it - and that is a standard procedure for over already 50 years.

[chrismb]

the Navy has re-funded the project and reinstated the gag order.

WRONG!

That perversion of the truth needs to be corrected every time it is repeated.

[93143]

Well... to be honest I forgot about that. It's hard to tell what's going on (not going to dredge up any conspiracy theories here, except to mention that I found some of them halfway plausible), but the upshot is that the only people currently researching Polywell in a reasonably well-funded, properly-staffed lab (as opposed to a one-man garage project) are not releasing any information. So it'll be a while before anything as specific as a bremsstrahlung spectrum is available...

[Joseph Chikva]

Well... to be honest I forgot about that. It's hard to tell what's going on (not going to dredge up any conspiracy theories here, except to mention that I found some of them halfway plausible), but the upshot is that the only people currently researching Polywell in a reasonably well-funded, properly-staffed lab (as opposed to a one-man garage project) are not releasing any information. So it'll be a while before anything as specific as a bremsstrahlung spectrum is available...

Yes. Too little data to be sure. But a lot of questions.

[KitemanSA]

the Navy has re-funded the project and reinstated the gag order.

WRONG!

That perversion of the truth needs to be corrected every time it is repeated.

Unless, of course, EMC2 used whatever FOIA loophole allowed them to fullfill the gag order?

This may be one of those issues that otherwise useless historians will argue about for millenia.

[TallDave]

I've been thinking about this and re-reading some of Rick's comments and I think I probably had a wrong conception of this -- electron behavior is probably stochastic across the interior rather than fast at the edges and slow in the middle. I don't think the electrons can both create the well and see the well in a way that creates the well -- it would be like climbing by putting your feet on your hands. In other words, their speed is roughly uniform, and the center has a well (for the ions) simply because the center is the electrons' average position.

No, a potential well is a potential well. If it exists, the electrons will see it.

An individual electron doesn't react to its own charge. But it does react to the charges of other electrons.

Yes... I was thinking about this after I posted it last night and I realized that was only true for the first electrons to transit the WB, which don't see a well (since it doesn't exist yet) and subsequent electrons must be affected by the existing well.

But I do think we can say that it can't be true that the electrons both slow down in the center because of the well and the well is formed because they slow down in the center, as that would still be circular logic that fails to explain where the well came from in the first place. A small point, but one that caused me some muddle in the past when thinking about the electron behavior.