We need a new film on YouTube explaining the Polywell.

[mattman]

Ok,

I fixed the electron orbit problem.

Adding the cage problem is not an issue.

There is so much debate over this energy plot:




I think this is critical. So much of Rider argument centered on Radiation losses. Alex Klien, Tom Ligon, Bussard and many other people have discussed this two temperature scenario.

It is simple - if the electrons and ions can be two temperatures, you can optimize to lower radiation losses and raise fusion rates.

Rider argued that the ions cannot be two temperatures. But he never said anything about ions and electrons. There is such a thing as non-thermal plasma where ions and electrons have two temps. But much is still unknown.

Also I find it hard to believe the temp distribution would not be some kind of bell curve. Also, we know that we need LOTS of electrons for every ion to make the (-) voltage. So the ratios should be skewed.

[hanelyp]

What Rider wrote sorely neglected the potential well resulting from the charged plasma.

Taking the plasma as a whole and assuming a hydrogen plasma I expect ions and electrons to have similar energy distributions. But taking the core or edge in isolation you get different energy distributions. Also very important, while charged particles are in a low energy region their energy distribution is compressed. The compressed energy spread suppresses the high energy tail responsible for much of the radiation loss, and practically the entirety of ion losses in a polywell.

[happyjack27]

there's the issue of kinetic energy transfer from the ions to the electrons in the core, through static electric force. since the ions are traveling fast, to neutralize the spatial voltage changes, electrons would be compelled to travel with then in order to compensate. due to the great difference in mass, the ions are going to speed up the electrons much more than the electrons slow down the ions.

however, there are many counter-points to this, such as:
a) since the ions are travelling so fast, they spend very little time near these cold electrons, so the time window for this to happen is small.
b) the electrons outnumber the ions in the machine, and much more so in the core, so their energies would be more controlled by each other than the ions
c) so what happens if an ELECTRON gains kinetic energy - well like a satellite orbiting a planet, it gets "upscattered" before it can even complete a single orbit. and once it's upscattered it's:
c.1) no longer in the fusion region, so it's about as pertinent to bremthr-whatever losses as my aunt sue. (who is also not in the fusion region.)
c.2) now having been up-scattered, where magnetic forces dominant, it will either find it's way back into the core through magnetic / lorentz forces, and the only trajectories that lead to and stay in the core are ipso-facto "cold" ones, or it will leave the system as an "electron loss". and in either case it will not return to the fusion region (except in passing very briefly) without, by the same logic that got it there, being cold again.

so it seems to me that this essentially takes care of ion-to-electron kinetic energy transfer in the fusion region, as far as bremsth-whatever losses are concerned.

EDIT: "c) so what happens if an ion gains... " mean to say "so what happens if an ELECTRON gains"

[KitemanSA]

Hmm, I always pictured the ions as going much more slowly at max KE than the electron, maybe almost as slowly as the electrons at min KE. Does anyone know the velocities off hand?

[mattman]

Yes:




For WB6

http://thepolywellblog.blogspot.com/201 ... g-wb6.html


I have made the case for this distribution before. Remember Ions are 35,461 times larger and 3,626 more massive than electrons.

It's like marbles flying through atmospheric dust.

[D Tibbets]

The transfer of KE between particles of significantly different mass is not very efficient. This has many ramifications. For instance a 1 AU neutron hitting a ~218 AU lead nucleus only transferrs ~ 1/218 of it's KE to the heavire nucleus. This is why neutron moderators need to be light- something with a lot of hydrogen atoms, etc. The same applies between electrons and hydrogen isotopes. The deuterium nucleus (positive ion) has ~ 8,000 times the mass of an electron. Therefor only about 1/8,000 of the KE of the electron is transferred to the ion. The inverse is also true, an ion will knock an electron away without much loss of it;s KE. Talking of velocity you have to consider momentum which is 1/2 MV^2. Thus at the same KE an electron is traveling about 60 or even 90 times faster than the hydrogen ion.

Things can become confusing, but one of the basic tenets of plasma physics is that electrons and ions do not exchange energy well when considering a one on one collision basis. Thermalization processes are greatly dominated by collisions amone the same species,.ie: electrons and electrons or ions and ions, but not electrons and ions. Space charge is a different matter. The extremely small excess of electrons in the system can establish a space charge/ potential well that can effect the ions greatly.

Ions are accelerated by either the potential well or by inter ion close encounters (Coulomb collisions) Such colse encounters with electrons have much less effect on the ions. Ions that are upscattered do travel to greater radii from the center and enter the emagnetic domain where they exit a cusp or turn due to the Lorentz force of the B field. It is reconized that this turning is not directly towards the center if the ion came from that direction. There is a half spiral completed which defocuses the ion central focus, and this needs to be accounted for. In WB6 conditions the ion- ion MFP collision distance is great enough that in most of the machine (mantle) the ions are not colliding. Most of the collisions occur in the core where up scattering and down scattering will ocur but not not much transverse scattering. At the edge the collisions increase greatly because Coulomb collision cross section scales as 1/ 1.75 power of the temperature). Thus ion- ion collisions are greatly increased and Maxwell- Boltzman thermalized distributions are very quickly established. But it is important to remember that this is a localized process and the thermalized distribution about the average temperature here (near 0 eV) is tiny in relation to the energy of the ions as they fall down the potential well to the core. Thus the term Annealing on each pass.

The electrons follow a similar trend except the geometry is reversed, the slow, quickly thermalizing region is in the core while the fast hot low collisionality occurs more on the edge. The time slices and resident densities in the regions along with the relavence to total volume has not been described for the electrons as it has for the ions.

I mentioned that electrons and ions do not trade energy well, but it is not totally insignificant. In Bussard's GOOGLE talk he pointed this out. The potential well is initially square, but once ions are introduced, as they travel inward they tug (gently) on the electrons and this leads to a parabolic potential well. The ions can tug on the electrons gently because during the brief times they are close together (very brief at fusion temperatures) the electron is very light and carries a -1 charge. The ion is relatively very heavy with a +1 charge. The electrons motion will be effected much more due to this inertia difference. Electron tugging on the ions is not absent but it is trivial compared to the space charge effects.

Bremsstruhlung is somewhat different process. I do not fully understand it, but this breaking radiation seems to come from the light electron whipping around a heavy ion at high speed. It may be due to both the speed of the electron and the tightness of this semi orbit. It is similar to cyclotron radiation except in the Polywell the gyroradii of intrest is much larger than the electrostatic ion electron Bremmstruhlung collision and thus the radiation is proportionatly less. I have seen where the cyclotron radiation losses in the Polywell are only ~ 1 % of the Bremmstruhlung (or total?). I have the impression that cyclotron losses in a Tokamak is a more significant problem.

I suspect Bremmstruhlung radiation is similar to gravitational wave emmision in an orbital system. But as the electromagnetic forces are much stronger than the gravitational forces the magnitude of the effects are much different.

If the distance of the electron orbit about an ion/ nucleus results in greater radiation emmision as the radius decreases, what prevents the electron from losing it's KE at a ever increasing rate till it merges with the nucleus? Well, I suppose quantum mechanics comes into play, and things like the Pauli exclusion principle comes into play.

Why does fast electrons seem to be the culprit? It would seem that having an electron and ion approaching each other towards a common intercept point wouldn't matter which was traveling and which was not. Here momentum and the speed difference comes into play. A ion with 10 KeV of energy may be traveling at ~ 100 KM/s, and electron would be traveling at ~ 10,000 KM/s. The electrons are zipping around while the ions are almost stationary relatively speaking. If the ions are at the top or bottom of their potential well, the number does not change much relative to the electrons speed. But the electrons speed changes much more significantly as it's KE changes. Thus in high ion density regions (ideally the core) if the electrons speed can be minimized the density * speed product of the Bremmstruhlung can be minimized compared to a population that had the same average speed throughout the volume of the machine. At least that is my reasoning.
I have seen a formula for Bremmstruhlung losses scaling as the temperature of the electrons (proportional to speed) to the 1.75 power. I'm assuming it also scales with the density of charged particles to the second power (or is it the 4th power?). Thus interpretations (like Rider's) where average temperatures distributed homogeneously throughout the volume gives considerable differences in results from an inhomogeneous distribution of energies. There are several ways the problem could be set up and I believe this has been done in some of the EMC2 papers. The differences are apparently game changers, though not profound. Otherwise playing with diluted Boron mixtures would not be needed. Confluence of ions to a point (very small core area) in the center has never been a possibility or desired. Disruptive central virtual anode formation limits the best reasonable confluence (or central focus) possible.

The Polywell is a collection of compromises with opposing effects. It is remarkable that if the claims are true, a compromise has been found that works.

Dan Tibbets

[D Tibbets]

Mattman, the hydrogen nucleus may be much larger, but I doubt it pertains as it is the electromagnetic force range that applies- thus the electron has the same range of force as the proton.

Also, a proton is ~3600 times more massive than an electron (I often round up to 4,000 just to make it easier), but reactor fuel like a deuterium atom / ion is ~ 7200 times more massive, and tritium , Helium, etc. heavier yet.

The only place a proton might be involved might be in the P - N15 reaction which is one of the steps in the CNO burning of hydrogen in stars. This reaction as championed by ChrisMB is the fastest step in this chain and at several hundred KeV it is only ~ 100 times slower than D-D fusion at similar temperatures. Admittedly a long shot, but if a Super Polywell with 100 Tesla magnetic fields comes along, who knows?

Err... I should have mentioned P-B11 or possibly something with lithium, which have better characteristics than p-N15.

Dan Tibbets

[happyjack27]

Hmm, I always pictured the ions as going much more slowly at max KE than the electron, maybe almost as slowly as the electrons at min KE. Does anyone know the velocities off hand?

yeah, well we must bear in mind that velocity and inertia are two very different things here. at the same inertia, the electron would be going over 1000 times as fast as the electron. and KE is yet another story - point being, at the same KE, the ions and electrons are travelling at very different speeds. conversely, when they quasi-match trajectories to quasi-cancel out each other's electric fields, the ion's KE is going to be way higher than the electron's. At the end of the day, from this rough picture, ions will have all the KE, and electrons will have all the speed. that still means the electrons are "cold" though. from my understanding, temperature is related to the ability to impart pressure and/or energy and/or entropy, which in all cases is related to the KE, not the velocity. and the ions are the ones with the KE.

EDIT: i just realized i flubbed that up. ke=1/2*m*v^2. so its not neccessarily the ions w/all the ke.... whatever... point is electrons are cold in the center, as far as electrons go. you put a bunch of things that repel each other in a small area, they are going. get as far away from each other as they can and stay there. like a marble at the bottom of a funnel.

[Roger]

Err... I should have mentioned P-B11 or possibly something with lithium, which have better characteristics than p-N15.

Dan Tibbets

IIRC Bussards Eagle class ship, designed to make a trip out to 550AU to place a grav lens telescope (round trip) would use L6/L6.

[hanelyp]

I mentioned that electrons and ions do not trade energy well, but it is not totally insignificant. In Bussard's GOOGLE talk he pointed this out. The potential well is initially square, but once ions are introduced, as they travel inward they tug (gently) on the electrons and this leads to a parabolic potential well. ...

A square potential well could only exist if the excess charge is confined to the edges of the plasma. With a mean free path large compared to device dimensions, I'd expect electrons scattered from the confining magnetic field to venture into the interior, limited by their energy.

Bremsstruhlung is somewhat different process. I do not fully understand it, but this breaking radiation seems to come from the light electron whipping around a heavy ion at high speed. It may be due to both the speed of the electron and the tightness of this semi orbit. It is similar to cyclotron radiation except in the Polywell the gyroradii of intrest is much larger than the electrostatic ion electron Bremmstruhlung collision and thus the radiation is proportionatly less. I have seen where the cyclotron radiation losses in the Polywell are only ~ 1 % of the Bremmstruhlung (or total?). I have the impression that cyclotron losses in a Tokamak is a more significant problem.
...
I have seen a formula for Bremmstruhlung losses scaling as the temperature of the electrons (proportional to speed) to the 1.75 power. I'm assuming it also scales with the density of charged particles to the second power (or is it the 4th power?).

I'm thinking electron density times ion density, or second power of plasma density assuming a fixed ion/electron ratio.

Both Bremmstruhlung and cyclotron radiation are effects of accelerating charged particles.

[KitemanSA]

Also, a proton is ~3600 times more massive than an electron (I often round up to 4,000 just to make it easier), but reactor fuel like a deuterium atom / ion is ~ 7200 times more massive, and tritium , Helium, etc. heavier yet.

Dan, a proton is more like 1800X the weight of an electron. Your "3600" is the weight comparison for a deuteron.

[D Tibbets]

Hmm, I always pictured the ions as going much more slowly at max KE than the electron, maybe almost as slowly as the electrons at min KE. Does anyone know the velocities off hand?

Taken from Bussard's comments for WB6. With a potential well of ~ 10 KeV the electrons had a speed of ~ 1 billion cm/s or ~ 10 million M/s. The ions have the same KE, but the speed is KE=1/2 MV^2, so the velocity difference is ~ the square root of the mass difference. The deuterium ion has a mass of ~ 7200 times that of the electron so the speed would be ~ 1/85th or rounded down to ~ 100,000 M/s.

These numbers could be calculated directly, but I generally use my recollection of Bussard's statement. I'm uncertain though if he was using the average velocity, the maximum velocity, etc. so I consider the numbers ball park close and of course dependent on where the charged particle is in the potential well at the time.

In WB 6 with 100,000 electron passes before loss equates into a travel distance of ~ 20-30 KM and the lifetime was a few milliseconds. That works out to ~ 10 KM/ms or 10 million M/S average velocity for the electrons.

If you take statements of ~ 20 ms lifetime for the ions (read that somewhere) and multiply that by the average (?) velocity of 100KM/s, then the ion would travel ~ 2 KM. That would be ~ 10,000 passes in WB6. This is similar to what Bussard said was required for a chance for fusion to occur (50% chance, 90% chance?).

This can be compared to ChrisMB's number of ~ 300,000 KM travel distance needed (or at least my recollection of a few hundred thousand KM) in a Tokamak to achieve good chances of fusion. Using the density difference between the two machines of 100-1000X and the fusion rate scaling as the square of the density means that in the Polywell, the confinement time in terms of ion travel distance needs to be only ~ 1/10,000th the distance or less. That would be ~ 30 KM, or as little as 0.3 KM. This shows that the ion confinement claims are consistant with sucessful fusion rates - the all important Triple Product. Confinement times per fusion event is similar, though the raw numbers are way different. This does not include thermalization distribution differences and possibly confluence, which would only help the Polywell picture, and is reflected in Nebel's estimate that the Polywell is ~ 60,000 times as energy dense as the Tokamak- at least the current low Beta Tokamak. If someone can develop a higher Beta Tokamak the numbers would change accordingly, but the Triple Product is a constant.

Dan Tibbets

[D Tibbets]

Also, a proton is ~3600 times more massive than an electron (I often round up to 4,000 just to make it easier), but reactor fuel like a deuterium atom / ion is ~ 7200 times more massive, and tritium , Helium, etc. heavier yet.

Dan, a proton is more like 1800X the weight of an electron. Your "3600" is the weight comparison for a deuteron.

Noted and checked. I originally used 60 as the inertial difference between the ion and electron. I should have stayed with it, but alas... I became fixated on the deuterium being made up of a proton and similar weight neutron, without rechecking the numbers.

The speed difference between the deuterium ion and the electron at the same KE of ~ 10 KeV would thus be closer to ~160,000 M/s for the Deuterium ion and 10,000,000 M/s for the electron. The other numbers for the ion speed and travel distance in the prevous post would need to adjusted accordingly. Instead of traveling ~ 2 KM the ion would have traveled closer to 3 KM. But, at least from the ball park perspective the comparison would not change much (it is improved in Polywell's favor).

D. Tibbets

[KitemanSA]

But the issue is, in the center where things are densest, when the ion is Vmax and the electron is Vmin, how fast are they moving?

[ladajo]

The $64 question.