Electron current in WB6 is confusing

[D Tibbets]

Electron current in WB6 is confusing for me.
Since there is nothing else to distract me, I decided I would share my confusion concerning the electron current needed in WB6. There are several mentions (in the final results for WB6, and a graph) that show electron gun currents of 10 - 40 Amps. This current was drawn from Marine batteries at ~ 12 volts (or more if several batteries were linked serially, instead of only in parallel). The injected electrons gained their ~ 12,000 eV energy by the high current (presumably again at ~ 10-40 amps- at least before arc breakdown shorted the grid casings to ground), and ~12,000 positive volts biased magrid (provided by high voltage capacitor banks).

If the current was ~ 40 amps, this would be ~ 10^20 electrons per second (rounded down).
I'll assume the neutral fill gas was at ~ 1 millitorr (~10^-6 atm) as that was nearing the point where arcing started. That would represent ~ 10^19 atoms/ M^3.
This gas was introduced and mostly ionized in less than 1 millisecond. The injected electron current during this time would be ~ 10^17 electrons/ millisecond.

The initial atoms were ionized by these injected electrons, but most of the ionization came from secondary cascading ionizations by these cool secondary electrons (I've heard mention that these new secondary electrons are initially at ~ 100 eV).
In this context the total electron current during this ionization time would be ~ 4,000 amps. at an equivalent energy of ~ 101eV (except see below) There are ~ 100 secondary electrons for each concurrently electron gun injected electron. The numbers of these secondary electrons and their relative energy matches the number*energy of the injected electrons. This makes sense (I think). It also implies that the resultant ions have not picked up much energy before they begin their fall into the potential well..

The questions begin:

How many injected electrons are actually present in the machine at the start of this process. Space charge limits would not allow for much charge buildup. If the 1 part per million limit on charge difference applies with the ions present, would the pre ion density of electrons be limited to this ratio, irregardless of the leakage/ feed rates? IE: How many pure electrons could a Wiffleball contain? The magnetic shielding presumably helps some but my feeble understanding is that this charge buildup very quickly reaches intolorable levels.

The secondary ionization cascade reportedly takes less than 0.1 millisecond, while non ionized gas leaking out of the machine can lead to arcing within ~ 1 millisecond time frames. So the window for operating the machine is less than 1 millisecond (apparently ~ 0.2 milliseconds). The timing issue that I cannot figure out is how the injected electrons heat up the cold ionization derived secondary electrons within this time frame. If the injected electron current is ~ 10^17 electrons per millisecond, and there are ~ 10^19 cold secondary electrons, then how do you heat these cold electrons to near the temperature of the injected electrons (10-12,000 eV) Mixing 1 hot electron with ~ 100 cold electrons should result in only a modest increase in the average energy (~ 101 eV.
What am I missing? There must be a lot of hot electrons present at the start of the gas puffing, and this starting excess must provide the numbers needed, but the question again, is how many of these startup injected electrons can be retained against the coulomb pressure?

If the Beta=1 Wiffleball is formed by the electron gun current before the gas puff, then the secondary electrons and ions must very quickly push the system out of the Beta= 1 condition. I don't see how intricate and quick control of the various currents could maintain the appropriate charged particle density while at the same time having the necessary hot electron population excess that drives the heating of the cold electrons. I suppose that if the cusps quickly open up as Beta=1 is passed, the system may be self regulating. IE: once you gain sufficient charged partical flows into the machine (electrons or electrons+ions) to reach Beta= 1 (at a corresponding B field and energy level), any excess is is quickly bled off. This would imply that the Wiffleball would be stable so long as this minimal flow is maintained. The excess is automatically shed till the optimal Wiffleball condition is again approached. But, it is very important when you consider the density buildup outside the machine that leads to arcing. Fine tuning of these knobs are most important for managing vacuum pumping requirements and of course net efficiency.
I'm not sure, but I think that the Beta =1 tests results where the current and voltage were maintained while the B field strength was varied reflects this. As Beta =1 was reached the PMT measured light intensity maxed out and then fell again, demonstrating rapid falloff in containment on either side of Bets=1 conditions.

Alternately, if you start with a Beta=1 generating plasma, and you suddenly dump in 100 times as much charged particles, but the resultant plasma is initially 100 times cooler, then the density * particle energy / B field ratio would not change. Mmmm... but still, that would leave the problem of heating the cool secondary electrons, and the change in Beta as they are heated...

Dan Tibbets

[chrismb]

I didn't think injected electrons gained such energies. I though they magik'd their way into the centre by forces as yet not fully contemplated by mortal soul.

This goes back to around-about my first question here (ne'er answered); how does the Brillouin limit square up with the envisaged operation of this thing?

[D Tibbets]

A few quick links concerning Brillouin limit:
http://www.ncbi.nlm.nih.gov/pubmed/10054206

Title only



http://www.osti.gov/bridge/purl.cover.j ... bviewable/

They report exceeding the Brillouin limit by 35 X in a Penning trap type of machine.



http://www.apam.columbia.edu/CNT/public ... ylimit.pdf
Discussion of the Brillouin limit in a magnetic Torus. I am presuming that quasi-spherical magnetic, cyclindrical, etc. containment is significantly different as suggested by the second link.

Dan Tibbets

[chrismb]

I'm well aware of this work, thanks, and that the budget was cut for it. By the logic of this board, that means it didn't work.

Anyhows, whether the Brillouin limit may or may not be exceed-able for another experiment is neither here nor there. There are physical reasons to 'observe it' in one's calculations. A diamagnetic wiffleball *means* that it is at the Brillouin limit.

This is drifting from my point; why do you think the electrons gain energy? It is the ions accelerated by the fields, the electrons are being repelled. If anything, they will have to loose energy.

[93143]

I may be misunderstanding your confusion, but...

The electrons are dropped in from outside the magrid, which is positively charged (ie: maintained at a positive potential with respect to the electron sources) and thus accelerates them considerably. Once they get inside, then they start to decelerate.

If we're dealing with a gas puffer system in its startup transient, the hot electrons will start an ionization cascade, and transfer some of their energy to the cold electrons that are mostly busy knocking each other off fuel atoms. Apparently you can at least form a potential well this way, and probably make some fusion, which implies that enough hot electrons stay hot enough to penetrate some distance into the wiffleball... it also suggests to me somewhat of a rationale for the use of ion guns... anyway, the fusion transients in WB-6 were a big mess in terms of physics; I don't have nearly the understanding necessary to clear up the issue.

[chrismb]

So how do new electrons get in, once the -ve electrostatic well is formed?

[D Tibbets]

"How do new electrons get in?"--
In the gass puffer system the ionized gas provides one positive ion for each negative electron so the contribution to excess electron is a wash, or at least that is my understanding. If you can stuff in x amounts of free electrons into the machine, you can stuff the same amount in if there are a gazillion particles of a neutral plasma already present. The same charge conditions would result. What I am unsure of is what is the density of pure electrons can be contained. Using the 1 part per million of the target non neutral plasma that the Polywell is supposed to operate at. As the secondary electrons are claimed to be quickly heated by the hot injected electrons to near the injection energy, I'm assuming the density of the injected electrons must exceed the secondary electrons (and corresponding ion density) ~ 100 to 1. Alternately, given enough time a trickle of hot electrons could do the job, but not within the time frame that WB6 operated at.

It occurs to me that I may be confused (again ). about the cold hot ratio, so a quick comparison:
1 C (100 eV cool electron) + 1 H (10,000 eV hot electron) yields avg temp of 5,050 eV
1 C +2+H yields avg temp of 6,700 eV
1 C +3 H yields avg temp of 7,600 eV
1 C + 10 H yields avg temp of ~ 9,000 eV

So, my assumption of 100:1 ratios needed was off by about an order of magnitude, perhaps ~ 10 to 1 ratios are enough. This eases the limits, but doesn't eliminate them. . Again the question seems to be how many pure electrons are stored in the machine at the time of the gas puff. lf the injected electron density is at some tolerable density limit within the quasispherical magnetic field, the gas puff molecules must be limited to ~ 1/10th of this in order for the ionization and secondary electron heating to occur within the WB6 time limits. At least that is my understanding.

If ~ 10^19 ion- cool electron pairs are introduced/ created, then ~ 10^20 hot injected electrons must be in residence. This implies that the magnetic field was able to store ~ 1 second of the electron gun current of ~ 40 amps. or ~ 10^20 hot electrons. My question is if this is possible, and/ or is my reasoning unreasonable.
If the non neutral plasma has an excess of one electron per million ion/ electron pairs, how do you account for the much greater electron dominance before and at the the time of the creation of the fusion plasma?
And, as the confinement time (even with recirculation) was only a few milliseconds, how could the above concentration have been obtained without much higher input electron currents?
I must be missing something. This dynamic situation is confusing. As mentioned, ion guns would make things much more 'reasonable'.

Dan Tibbets

[chrismb]

Interesting as your conversation with yourself is, I still do not understand why an electron in an electron gun and an ion formed on the edge of the wiffleball both experience forces that pull them in the same direction, into the wiffleball. The electron gun must fire electrons of the same energy as the well potential, i.e. they have the energy to get to the middle.

The only conceivable scenario is as per 'normal' plasma physics - becase the electrons are contiguous through the cusps and back around, so the well potential is present at all places in the cusps and the returning loops. So what we then have is a scenario that the electron guns can fire into any position, cusp or recirculation stream, as they are all at the same potential and magnetically confined. But we also have the situation that the magrid potential points in all directions, including outwards as well as inwards. If this were so, the ions may then head off in any-old direction.

[93143]

...no, it looks like you really did fundamentally misunderstand this device.

The magrid is held at a positive potential with respect to the electron emitters by the power supply. That means that as the negative charge inside grows, the positive charge on the magrid grows to always exceed it by enough to produce the imposed potential difference.

Once the electrons start to build up inside, the gradient of the potential reverses at the magrid, resulting in a trough or "well" inside - but even in steady-state this "well" isn't quite deep enough to get all the way back down to the electron emitter potential, so electrons from the emitters can still get anywhere in the core (barring magnetic shielding effects).

I still do not understand why an electron in an electron gun and an ion formed on the edge of the wiffleball both experience forces that pull them in the same direction, into the wiffleball.

They aren't in the same physical position. The electron emitters are at some distance outside the magrid. The ions are formed inside the magrid, at the edge of the wiffleball, at which point the electrons have already accelerated towards the magrid, missed it, entered the wiffleball, and started to decelerate towards the centre.

At least, that's how it works in a gas-puff scenario. I suspect that with ion guns (outside the magrid, like the electron emitters...?) they'd have to fire them at significant energies to get them past the magrid, since they're firing uphill... unless I've misunderstood the usage of "ion gun"...

Normal plasma physics doesn't apply here, because almost all of the particles' kinetic energy is provided by the electric potentials in the system. You can't get Debye shielding in the conventional sense under these circumstances.

[TallDave]

Good explanation, thanks for sharing. This comment from Rick has stuck in my mind and may be relevant:

My other comment was related to electrons trapped in the wiffleball. Over most of their orbit there is little or no magnetic field (i.e. Larmor radius bigger than the device size) with the electrons turning when they hit the barrier magnetic field. The electron behavior is stochastic since there are no invariants.

At least, that's how it works in a gas-puff scenario. I suspect that with ion guns (outside the magrid, like the electron emitters...?) they'd have to fire them at significant energies to get them past the magrid, since they're firing uphill... unless I've misunderstood the usage of "ion gun"...

I was really hoping we'd get a picture of WB-7.1 to help clear this up. Maybe in April.

[chrismb]

Normal plasma physics doesn't apply here

Seems that the word 'plasma' is redundant in that comment.

I still can't see it. Either ions see an attracting field while electrons see a repulsive field, or...er... ah, whatever!.. I just want to see the damned thing operating, but seems that will never happen because the more successful it is the less we see, so I gather.

[93143]

ions see an attracting field while electrons see a repulsive field

That's exactly right. Inside the magrid.

Outside the magrid there are few to no ions, and any that make it outside are repelled by the magrid and are lost. But the electron emitters are out there. They aren't guns; not in WB-6. They're headlight filaments; they emit low-energy electrons. Which then accelerate towards the positively-charged magrid.

Once those electrons pass the magrid (at which point its charge ceases to matter to them in accordance with Gauss' Law) and meet up with the neutral gas population and start ionizing it, you have the situation where the ions are attracted towards the centre of the electron-rich wiffleball and the electrons are repelled. But the electrons have already picked up considerable energy by falling towards the magrid, so they're slowing down from very high energy, which is what allows them to penetrate the wiffleball, generating the potential well that accelerates the ions from very low energy.

Better?

[chrismb]

Why don't electrons go directly to the magrid, and neutralise/hit it?

[Giorgio]

The reply to that question and many other was partially supposed to come from EMC2 secreted experiments.

I guess we will find out only when they will release some info.

[chrismb]

I guess we will find out only when they will release some info.

Indeed it appears so. And yet the prevailing logic here is the more successful they are, the less likely we are to hear.

So, essentially, that means we are never going to hear of Polywell's success.

I'm just more and more worried that they have confused beam-target fusions in the wall/magrid with volumetric fusions.

We need to see proof of isotropic fusions. Just read the history of fusion devices - constant confusions over other origins of fusion, in devices poorly understood and insufficiently instrumented. There is too much egg-on-face in the history of fusion to presume new fusion ideas work out. The probability of egg-on-face is higher than success by orders of magnitude, so it is an odds-on-favourite bet to expect failure.