Remind me - why 10T field?

[TallDave]

Anyways... on the density question...

I started writing something a couple posts ago about not being sure I agreed with Rick that electron density increased toward the center, but then I realized that must just be a result of their time-averaged positions.

I'm a little staggered by the implication there. Not only do we have best confinement at beta = 1, we actually operate at something like beta * 100 (i.e. what would be beta*100 conditions if it were at the edge) in the overall volume? I had been assuming 2.5e22 was the average density, not the minimum. I can see why people are skeptical.

Is electron behavior really stochastic? There must be a study on this somewhere...

Hey, I just had an idea -- given that we know ITER's output, we can work backwards from Rick's 62,500 and then solve the radius for 100MW net. I'm very interested to see how closely that matches our WB-6 extrapolations. (Argh, torus versus sphere. OK, there's a lot fudge in here anyway, I'm just going to ignore that. And ion convergence.)

[KitemanSA]

This is something I'm not 100% clear on. Earlier on I'd assumed what you seem to -- exterior density is the same over the whole volume outside of the WB. But the more I think about it, the more I think the electrons stay by the Magrid and any ions follow the electrons, so pumping is just for fusion products and stray neutrals.

Due to the charge on the MaGrid, the ions should shoot to the chamber wall as soon as they exit a cusp and become "stray neutrals".

[TallDave]

This is something I'm not 100% clear on. Earlier on I'd assumed what you seem to -- exterior density is the same over the whole volume outside of the WB. But the more I think about it, the more I think the electrons stay by the Magrid and any ions follow the electrons, so pumping is just for fusion products and stray neutrals.

Due to the charge on the MaGrid, the ions should shoot to the chamber wall as soon as they exit a cusp and become "stray neutrals".

But presumably that would depend on how many electrons are out there. If the machine is e-driven, that implies there may be more electrons than the Magrid would offset. Art's calculation seemed to imply that was the case. Ions might make up some of the difference, with the excess of upscattered electrons making it to the wall and freedom.

OTOH, Joel's simulation said there would be few if any ions out there. So maybe not. Oh, for an experiment...

[KitemanSA]

IIRC, Dr. N. said that there was a large electron current into the well and out thru the cusps back to chamber wall. This is one part of the process to keep the system mono-energetic. (Large is a relative term, and since there is little energy spent, the power is quite low).

In order for there to be a flow into the well with enough energy to keep the well at a specified depth, the MaGrid must retain sufficient charge GREATER than the excess negative charge within it to provide the electrons their energy. Thus, assuming a 10keV well, the MaGrid would be charged to about +12kV relative to the wall/electron guns. This charge will certainly drive the ions to the wall, no?

[D Tibbets]

I think (in my humble openion) that ChrisMB is confused about the numbers.

I didn't think I was, but I am after your post! I didn't understand any of that! So, what are you saying is the density at the edge in WB?? An order of mag lower than tokamak?

The density on the edge of the Wiffleball- the area inside the magrids where the magnetic field from the magrids is excluded, is the density in question. And, it is my understanding that this can be a maximum of ~ 10^22 particles /M^3. This number is limited not by the magnetic field strength (provided you can continue to push the field strength, at least theoretically to levels well above 10 Tesla), but by the wiffleball traping factor, as this sets the ratio of the internal Wiffleball contained plasma density and the exterior plasma/gas density that cannot exceed ~ 10^19 particles/ M^3 due to arcing concerns. When I say increasing EFFECTIVE density through convergence, I am referring to the effect of having the ions on orbits that carry all of them through a smaller core area on each pass. This increases the chances of fusion on each orbit of the ion. I am not sure this represents an actual density increase. That is why I include the qualifier 'effective'. Because of the dynamic nature of the ion speeds and dwell time in various shells within the Wiffleball, I'm not sure (read that as clueless) how instantaneous measurements of density would play out. For that reason I think mostly in terms of the average ion (and electron) density within the Wiffleball inducing the pressure against the Magnetic field , I think this would be legitimate in the case where the ion density within the Wiffleball is evenly distributed and there is no confluence, like in the model by Joel presented last fall at the IEC confrence. More detailed descriptions may exist, but I'm not sure they would change anything, at least in this example.

Dan Tibbets

[D Tibbets]

This is something I'm not 100% clear on. Earlier on I'd assumed what you seem to -- exterior density is the same over the whole volume outside of the WB. But the more I think about it, the more I think the electrons stay by the Magrid and any ions follow the electrons, so pumping is just for fusion products and stray neutrals. I'm not sure this makes sense though. Bussard doesn't seem to really suggest one way or the other (but maybe he was thinking electrons only on the outside) -- though now that I think about it some more, it does sound a bit more like an ETW machine looked at that way, and Bussard did once say PW was most like an ETW...

I think I need to study ETWs some more.

What is an ETW machine?

Electrons, once they pass through a cusp into the volume outside the magrid, have to go somewhere. After all, in WB6 at Beta=1 there was ~ 40 amps of electron current. If the electrons are not recirculated (already accounted for in the 40 amps of current) they have to go somewhere or very quickly accumulate to ridiculous numbers. They can reach the positive charged magrid casings through cross field transport, they can hit the nubs (hinted to be a significant sink based on what Nebel said about hot spots), or they can hit the walls or other structures outside the magrid. The distance from the magrid to the vacuum vessel walls (Faraday cage in WB6) may contribute significantly to the recirculation numbers if the electrons can travel from one cusp to another. I suspect though, that as A. Carlson insisted, the electrons leaving cusps are on such long field lines that they hit the walls before much curvature. Then add all the guns and other objects in the area and there are numerous sinks for the electrons.

Dan Tibbets

[KitemanSA]

What is an ETW machine?

Elmore-Tuck-Watson. The Polywell is a magnetically protected ETW machine with WB enhancement. Hirsh-Farnsworth machines accelerate ions with the grid. ETWs accelerate electrons which develops a well which accelerates ions. That is what the Polywell does. It is a modified ETW machine.

[TallDave]

Kite,

Yes, the current has to go one of two places: wall or Magrid. I think the Magrid dominates, via cross-field transport. I'll have to check over Rick's comments again, I know he said the hotspots were on the Magrid but I'm not sure what he specifically said about the relative electron currents.

I don't think the electrons get their energy from the Magrid so much as from the guns. The Magrid is just the bottom of their well. I think the charge on the Magrid pulls the electrons in: it's the anode in the e-current. There will always be a pull because we're always losing electrons to the Magrid.

It's true that, absent any electrons, any ions on the outside would go flying off. I'm not sure that's true if there is a counterbalancing number of electrons... and then I get lost in chicken-and-egg reasoning. If there are enough electrons to offset the Magrid then it can't confine them... ah, wait, maybe that's it. Two things could restore the balance, a flow into the wall or some ions showing up. The e-guns for the outside are the cusps, so an excess might stop them up -- hey, that sounds like electron cusp-plugging.

I assume electron behavior on the outside is not stochastic. I wonder what the electrostatic potential shape looks like outside? You have the Magrid at +, then outside there a swirl of electrons trying to get to it... do you get a small virtual cathode there as well? I suppose it doesn't have to be very deep, just enough of a valley to trap some slow ions between Magrid and wall...

But I'm thinking Joel and Bussard are probably right. Time to look at ETWs again.

[chrismb]

Yes... but why in God's name would you think we plan to have 10 MeV electron guns?

You don't! That's my point! I have been trying, oh so trying, to point out that this is an internal inconsistency with the notion that beta=1 can be achieved. If it can be achieved, then you'd not want 10T. sigh!.... I will condede defeat on me trying to get the point across!

[chrismb]

I think I need to study ETWs some more.

Yes, you do. Please note that they are the same configuration as Polywell and should therefore, in theory, work just as well excepting for greater losses.

...Guess what! THEY DON'T WORK!!!! NOT A SINGLE NEUTRON EVER PRODUCED!

[TallDave]

You don't! That's my point! I have been trying, oh so trying, to point out that this is an internal inconsistency with the notion that beta=1 can be achieved. If it can be achieved, then you'd not want 10T.

Again, this point doesn't make much sense. You held density constant for no apparent reason, in contradiction to all PW theory I've seen, and got a nonsensical well depth at 10T. Well, sure, GIGO. We don't want beta=1 so we can have 10MeV wells, we want it for the density.

...Guess what! THEY DON'T WORK!!!! NOT A SINGLE NEUTRON EVER PRODUCED!

ETWs have certainly produced deep wells. I don't know for sure that anyone measured fusion from them (maybe Tom can say), but we've measured lots of fusion from PWs, a modified ETW.

Also, it's not quite true that Polywells are just ETWs with better shielding. The other main problem with ETWs aside from grid collisons is that you need a large excess of electrons, which limits the fusion you can produce. Polywells only need a 1:1+1E-6 ratio.

[chrismb]

Again, this point doesn't make much sense. You held density constant for no apparent reason, in contradiction to all PW theory I've seen, and got a nonsensical well depth at 10T. Well, sure, GIGO. We don't want beta=1 so we can have 10MeV wells, we want it for the density.

'Fraid I have given up trying to explain this. It would not be a 10MeV well depth, it would be a 10MeV temp electron[-rich?] plasma - at whatever well depth you put it. And you have no need for 10T if you are only going to run at beta=0.01 when you could achieve beta=1.

Why run 10T and beta=0.01 if you can get the same density by running 0.1T and beta=1? Maybe I should've tried to push that as the core of my question rather than refer to plasma temp, which seems to have confused you. Plasma temp and well depth are two different things.

we've measured lots of fusion from PWs, a modified ETW.

Again, it is a pure speculation that it has come from actual ion-ion fusions. This is a dangerous presumption, just as when Zeta and the other early experiments jumped to the conclusion they'd done thermonuclear fusion when they'd just done a bit of pinching instability. The rate of fusion indicated is in keeping with a presumption of fast neutrals hitting the chamber wall. There is, as yet, everything to prove and nothing can be stated as known from polywell yet.

Once statistically significant neutron counts are obtained and those counts are shown to be fully isotropic, then there may be an assumption of WB-produced neutrons.

Also, it's not quite true that Polywells are just ETWs with better shielding. The other main problem with ETWs aside from grid collisons is that you need a large excess of electrons, which limits the fusion you can produce. Polywells only need a 1:1+1E-6 ratio.

I don't understand why this is so. By what means do you come to these conclusions?

** Why do electrons limit the rate of fusion?
** Why are the accelerating/confining E-fields in an ETW worse than in a Polywell, if the electron ratios are the same?

[TallDave]

chrismb,

I don't think you understand the basics of IEC fusion. How do you imagine ions get accelerated to MeV temps without an MeV well? You're making less and less sense here. But that's beside the main point anyway: why are you holding density constant???

Why run 10T and beta=0.01 if you can get the same density by running 0.1T and beta=1?

Why would you ever run beta = .01? You want to maximize the density because of the power gains, and you don't get a wiffleball at .01 anyway. And no, beta = 1 at 10T doesn't create a 10MeV plasma, it creates a denser plasma at 10T, still at temp of 10KeV, than you had at .1T. I don't know why you have so much trouble comprehending these relatively basic concepts. Did you read Valencia?

It's very unlikely the neutrons measured are from fast neutrals, unless you think that's where we got all IEC neutrons ever created. I'm not sure where you think they're coming from, either; there are no background neutrals within the WB because the ionization temps are very low compared to any well. You keep confusing this with a fusor.

The ETW electron problem is that you need a large excess to accelerate the ions, because it isn't very efficient. I think you run into space charge limits. Tom wrote a piece about this a while back.

[TallDave]

I think chris does raise one question we might try to answer: at what point (of density) does ion pressure overwhelm electrostatic confinement?


I'm not entirely sure how to get from KeV to pressure/force, but the question does seem physically sensible.

So Mr. Deuterium decides to take a dive into the ol' Polywell swimming hole. He strips off his electron and jumps over the edge of the electrostatic cliff. Wheee! What is his acceleration, assuming a 10KeV well? And how does that compare to the opposing force from pressure, at, say, 1E+25 density?

I'm guessing the counterforce from pressure has to be very small by comparison, or the fusion rates would increase more slowly with density at some point and I've never seen a graph of that. But maybe they do slow down at some point, I'm not 100% sure.

[TallDave]

I am still not sure what the issue is. From what I see, at β=1, you can have either a little VERY hot plasma, a LOT of hot plasma, or somewhere in the middle. Also, it seems folks think that the most beneficial condition for Polywell is a LOT of hot plasma.

So why does chrismb keep harping on the "little VERY hot plasma" condition as proof that Polywell won't work?

Yep, that's the mystery.

I hope we can get some FAQ answers out of this at least. I look forward to someday answering 99% of questions from noobies by telling them to go FAQ themselves.