Talk-Polywell.org

hi,

some questions occur to me, would someone mind trying to answer?

1) should we be cooling the target fuel ions?
2) if so, how? (laser/microwave cooling, electron cooling?)
3) should we expect WB Polywell to work in a zero gravity environment exactly as it does in a moderate or high acceleration environment? eg. space shuttle applications - considerations?
4) of what significance are fluid and gas behaviours to IEC (vs Magnetically) confined plasma stabilities (instabities) generally? what are the (significant?) couplings?
5) how do these figure, if at all, in the WB Polywell configuration/regime?
6) in IEC fusor demonstrations i have seen (eg. around fusor.net), (at least) two 'modes' of plasma are observed - 'jet' and 'star' configurations. do these figure in the WB Polywell configuration/regime also, or not?
7) if they do, which is preferable?

thanks

I'll take a stab:

1) no
2) n/a
3) yes, earth gravity is a very small force on the particles
4) too much complexity and not enough data imho; WB-8 may tell us more
5) not too badly, I hope
6) If I remember right, star mode was something that helped them avoid collisions with the physical cathode grid. Since the cathode in a Polywell is virtual, star mode per se doesn't really apply, though Polywells do focus at the center.
7) I think whether you want a jet depends on the application

TallDave wrote:I'll take a stab:

1) no
2) n/a
3) yes, earth gravity is a very small force on the particles
4) too much complexity and not enough data imho; WB-8 may tell us more
5) not too badly, I hope
6) If I remember right, star mode was something that helped them avoid collisions with the physical cathode grid. Since the cathode in a Polywell is virtual, star mode per se doesn't really apply, though Polywells do focus at the center.
7) I think whether you want a jet depends on the application

hmm, thanks for that talldave.

the only response i have some uneasyness about, is the one you seem most certain of - that is 'cooling fuel ions is a bad thing is true'.

i've chased through other threads dealing with the issue:

viewtopic.php?t=960&highlight=ion+temperature
viewtopic.php?t=1062&highlight=ion+temperature
viewtopic.php?t=833&highlight=ion+temperature
viewtopic.php?t=1115&highlight=ion+temperature
viewtopic.php?t=1223&highlight=ion+temperature
viewtopic.php?t=1228&highlight=ion+temperature
viewtopic.php?t=1234&highlight=ion+temperature
viewtopic.php?t=1255&highlight=ion+temperature

and am still unsure.

do we consider the WB as a (focused?) beam-beam device, or a marginally / quasineutral thermalisation (essentially gausian + filter) device?

how about localised or 'oriented' (edge) cooling, say at the edge of either the outer or inner well? what order of magnitude would this likely have on the scattering cross section (though the core orbits) and silmilarly how should this affect the power balance of the machine?

any couple with the (scale-relative) size of the machine?

can we introduce spin polarised ions into the machine/regime?

would your answer be the same?

Heh, sorry if I gave the impression of being certain on #1. It's my semi-educated opinion.

There may be an annealing effect at the edge for ions at low energy (which has been discussed quite a bit). I'm still not sure why you would want to cool them, though. It seems so counterintuitive.

I'm not sure what you mean by "spin polarized."

At this point, I'm waiting to see what we find from WB-8 for any scaling questions.

I'm still not sure why you would want to cool them, though. It seems so counterintuitive.

Wasn't it simply to have that local thermalization act as a way of prolonging the time before ions thermalize globally?

We know for certain that collisions in the core are rapidly thermalizing the ions globally. Collisions in the edge when the ions are at low energy though would counter act that to an extent. If it is a strong enough effect then we might reach Bussard's point where on average, ions fuse before thermalizing.

bcglorf wrote:I'm still not sure why you would want to cool them, though. It seems so counterintuitive.

Wasn't it simply to have that local thermalization act as a way of prolonging the time before ions thermalize globally?

We know for certain that collisions in the core are rapidly thermalizing the ions globally. Collisions in the edge when the ions are at low energy though would counter act that to an extent. If it is a strong enough effect then we might reach Bussard's point where on average, ions fuse before thermalizing.

Actually I think he said fuse or get lost thru upscattering before thermalizing.

bcglof,

Yes, I understand and agree with the annealing to keep the distirbution closer to monoenergetic, I 'm just not sure we need cooling on top of that. Sorry, I realized later how I wrote it was unclear.

re: relative reduction in overall thermalisation cf 'localised' velocity accros well regions - agreed

sustaining mono-energetic distributions - agreed.

annealing - i have no idea - in fact, use of the word, unneccessarily complicates / obfuscates matters to my mind. i have an idea of annealing processes in eg. metals and other crystaline materials, i also have a good idea of the use of simulated annealing in mathematics.

i can see how either interpretation (where they differ), might help us, but i would rather reduce the phenomenon back into a simpler analytical frame. (eg. a double-well harmonic oscilator).

re. cooling - though i obviously dont know the answer myself, my logic for considering it was as follows:

in the Pollywell we essentially have two concentric ion-electron shells spaced around a central core (which oscilates +/-, but is net -)

(see. eg. the Dolan paper 1994 - http://mr-fusion.hellblazer.com/pdfs/ma ... nement.pdf or
Matsuuma - http://mr-fusion.hellblazer.com/pdfs/io ... lasmas.pdf or Yoshikawa - http://mr-fusion.hellblazer.com/pdfs/me ... ofiles.pdf)

(various) collisions happen throughout the device, but are differentiated (skewed at least) about these shells.

we prefer only collisions that form part of neutronic fusion processes - high velocity (inellastic) ion collisions, that are (ideally) head-on.

(note: profile of neutron production from double-well, is greatest from within the inner shell (or core), then outwards, reduces, then increases again within the outer shell (though not as strongly as the core, it covers a greater area), then reduces again exponentially outwards to the wall).

seems to me that there are (at least) two approaches or modes we can consider: a) take a bottle of ions and pump energy in until, statistically, they start colliding in all directions with appropriate force, or b) hold a small dense population of ions relatively still, and hit it with (charge) directed (focused) energy (ions).

it seems to me, sort of both views/phenomena can obtain in Polywell, although it is not predominently (supposed to be) a thermal / Tokamak type device.

by changing thermal gradient at the shell (well) edge), I suggest we would be favouring one 'mode' of collisionality, over all others.

i would therefore expect, net output down, but efficiency, up.

this is not to mention any thermo-acoustic effects we might like to consider, nor POPS, nor even more exotic Pauli / BEC type mechanisms - though these are all most interesting.

(my personal hunch, is that Polywell may perform best (highest net output) in pulse-mode, rather than steady-state, where it sort of fizzles along).

as to spin polarisation - bit of a wild-card, but interesting to consider.

here's a nifty proposal from brave Japanese proponents: http://www.quark.kj.yamagata-u.ac.jp/is ... Fusion.pdf

a more detailed paper exists - which i've mislaid at present. anyway - factor of 2.5 to 3 improvements in cross-sections observed)

(see also, eg:
http://accelconf.web.cern.ch/accelconf/ ... /MPE01.PDF , http://www.riken.go.jp/lab-www/library/ ... /4_067.pdf, http://www.springerlink.com/content/thq3150280981387/, http://www.sparc.lkb.ens.fr/IMG/pdf/Surzhykov_A-4.pdf,
http://www.am.qub.ac.uk/users/g.gribaki ... 223201.pdf
http://plasma-gate.weizmann.ac.il/Pubs/PRA52726.pdf - interest in the the concept grows, most recently you might recall, nano-fusion in palladium/platinum, was it? - another thread anyway)

rcain,

I think Rick Nebel has a more eloquent description of the "edge annealing" process here somewhere. Basically the ions outside the core are slow and have a lot more collisions so they thermalize amongst themselves.

Thanks for the link, spin polarization is interesting. They seem to be saying about $5M USD for the whole shebang, which is reasonable for a WB-100 add-on and might be justified for a 3x increase in fusions. I wonder if Rick is aware of this.

The physics are over my head and I don't have time to do more than skim them, but Dr Nebel has only 4 posts on the forum where the word "thermalize" appears.

hi talldave

re. slow ions thermalising at edge (/outwards) - that was my understanding also - undesirable for our purposes.

would be very interested indeed to see what rick nebel thinks is going on here, and what to do about it / how best to use it.

re. spin - i feel sure he is aware (vis. POPS), just more concerned at present with getting a handle on effects of greater dominance to overall WB power balance, scaling, etc.

do you know if theres anything (a model) on pulse-mode operation?

Betruger wrote:The physics are over my head and I don't have time to do more than skim them, but Dr Nebel has only 4 posts on the forum where the word "thermalize" appears.

Actually, I don't know if he used that word in his description. That was just my description.

EDIT: Ah, here we go:

2. The general rule of thumb on ion collisions is that ion collisions in the core add angular momentum to the ions (thermalization) while collisions in the edge remove angular momentum. The reasons edge collisions remove angular momentum is that as the ions reach their radial turning point, their angular velocity exceeds the radial velocity. Consequently, thermalization takes energy from the angular direction and puts it in the radial direction. The collision rate gets big because the velocities are small. The upshot of this is that if you want to look at the effect of collisions on ions, you have to do something like bounce-averaged Fokker-Planck where you take into account the collisions at all points in the ion orbit. There are two places this is discussed in the literature. The original work is:

M. ROSENBERG and N. A. KRALL, Phys. Fluids B, 4,1788 (1992)

There is also a 1993 paper by these authors (I don’t have good access to Journals here) but I think the correct one is the one above. Of course, the Chacon paper also looks at this.

viewtopic.php?p=7798&highlight=slow#7798

TallDave wrote:

2. The general rule of thumb on ion collisions is that ion collisions in the core add angular momentum to the ions (thermalization) while collisions in the edge remove angular momentum. The reasons edge collisions remove angular momentum is that as the ions reach their radial turning point, their angular velocity exceeds the radial velocity. Consequently, thermalization takes energy from the angular direction and puts it in the radial direction. The collision rate gets big because the velocities are small. The upshot of this is that if you want to look at the effect of collisions on ions, you have to do something like bounce-averaged Fokker-Planck where you take into account the collisions at all points in the ion orbit. There are two places this is discussed in the literature. The original work is:

M. ROSENBERG and N. A. KRALL, Phys. Fluids B, 4,1788 (1992)

There is also a 1993 paper by these authors (I don’t have good access to Journals here) but I think the correct one is the one above. Of course, the Chacon paper also looks at this.

Let me try to play with this (without pretending to do a publishable analysis).

Imagine a polywell with two ion species and purely radial velocities. All collisions will be in the center of the machine, so they won't create any angular momentum. They will, however, change the radial velocities. For example, if the mass difference is very large, the heavy particle will be slowed down a bit and the light particle will bounce back with a little extra energy. When we follow these particles back out to the edge, we will find that they have a spread of radial velocities but still no azimuthal velocity. Collisions at the edge will tend to isotropize the distribution. Viola, reverse annealing!

hi Art

that doesnt sound good.

so you think it should just heat up and fall over shortly afterwards?

if i'm understanding correctly?

Art Carlson wrote:

TallDave wrote:

2. The general rule of thumb on ion collisions is that ion collisions in the core add angular momentum to the ions (thermalization) while collisions in the edge remove angular momentum. The reasons edge collisions remove angular momentum is that as the ions reach their radial turning point, their angular velocity exceeds the radial velocity. Consequently, thermalization takes energy from the angular direction and puts it in the radial direction. The collision rate gets big because the velocities are small. The upshot of this is that if you want to look at the effect of collisions on ions, you have to do something like bounce-averaged Fokker-Planck where you take into account the collisions at all points in the ion orbit. There are two places this is discussed in the literature. The original work is:

M. ROSENBERG and N. A. KRALL, Phys. Fluids B, 4,1788 (1992)

There is also a 1993 paper by these authors (I don’t have good access to Journals here) but I think the correct one is the one above. Of course, the Chacon paper also looks at this.

Let me try to play with this (without pretending to do a publishable analysis).

Imagine a polywell with two ion species and purely radial velocities. All collisions will be in the center of the machine, so they won't create any angular momentum. They will, however, change the radial velocities. For example, if the mass difference is very large, the heavy particle will be slowed down a bit and the light particle will bounce back with a little extra energy. When we follow these particles back out to the edge, we will find that they have a spread of radial velocities but still no azimuthal velocity. Collisions at the edge will tend to isotropize the distribution. Viola, reverse annealing!

Mm... Are you saying the up and down scattering of radial ions vectors will increase transverse (non radial) scattering in the periphery?. Except for tremendously up scattered ions the radial speeds will still be much less than transverse speeds at the turn around points of most of the ions. How much likely radial scattering would be needed to significantly change things. I suppose it would depend on the thickness of the contributing 'layers' approaching the Wiffleball border and their relative positive and negative contributions. Also, keep in mind that the up scattered ions wouldn't go much further outward due to the magnetic field unless they hit a cusp, then they might be lost to the system (hopefully with minimal loss of the input energy), or to put it another way - the upscattered ions would be preferentially extracted from the system. Down scattered ions wouldn't reach the periphery so should not effect things much (if most of the proposed anneling process occurs close to the Wiffleball border). Of course the downscatered ions themselfs would not benifit from the edge anneling, so there would be a compromise. But, if the downscattered ions are considered in isolation - any ions reaching a given height would annel with like ions. Is that why some tool like bounce-averaged Fokker-Planck analysis (what ever that is) is needed due to the extreamly complex interactions of ions with each other? I'm guessing that things would get more interesting when several different ions (like P-B11) are involved . Since they are closer in mass to each other the interactions may be more complex than scattering reactions with electrons ( which I understand are much less significant to the ions per collision- they can be ignored). Extreamly upscattered ions (ie- charged fusion products) would complicate things more, but R. Nebrl claimed that they left the system through a cusp after ~ 1000 passes with close to thier original energy, so I'm assumeing they have not had much oppertunity to scatter with other fuel ions.


Dan Tibbets