An Important Question About Polywell

Discuss how polywell fusion works; share theoretical questions and answers.

Moderators: tonybarry, MSimon

Post Reply
vahid
Posts: 33
Joined: Tue Aug 17, 2010 3:33 pm

An Important Question About Polywell

Post by vahid »

When a large number of electrons orbit around Polywell Magrid they can insulate electric field of Magrid then the Potential Well Depth is reduced.

Isn't it?
Engineering Is the Art of Making What You Want from What You Can Get at a Profit. ( MSimon )

MSimon
Posts: 14334
Joined: Mon Jul 16, 2007 7:37 pm
Location: Rockford, Illinois
Contact:

Re: An Important Question About Polywell

Post by MSimon »

vahid wrote:When a large number of electrons orbit around Polywell Magrid they can insulate electric field of Magrid then the Potential Well Depth is reduced.

Isn't it?
Yes. Typically to 80% of the applied voltage.
Engineering is the art of making what you want from what you can get at a profit.

JohnFul
Posts: 84
Joined: Sat Feb 27, 2010 7:18 pm
Location: Augusta, Georgia USA

Post by JohnFul »

Good question, and good answer. I vote we move the thread to theory however.

Tom Ligon
Posts: 1871
Joined: Wed Aug 22, 2007 1:23 am
Location: Northern Virginia
Contact:

Post by Tom Ligon »

I would not say "insulate" is the right word. A Polywell runs on Poisson's Equation, or its specific vacuum tube form the Child-Langmuir Law, particularly the spherical form of Langmuir and Blodgett, 1924. Child-Langmuir Law results in a condition called space-charge limitation, a condition in which electron flow masks out the potential gradient produced by the electrodes of a vacuum tube such as a reflex diode. That gives the maximum current the tube can carry.

A Polywell will behave much the same, complicated by the fact that ions carrying positive charge are also available.

Electrons acting alone will reach a density limit in the center of the machine when their kinetic energy gained by acceleration by the magrid is countered by Coulomb repulsion. At that point their kinetic energy is spent and the potential well is at its maximum depth. From the electron's perspective, this is really a "potential hill", and the hill can be thought of almost as a pile of electrons. The maximum height is set by the energy the electrons have available to spend climbing it. The potential hill limits the electron population ... when the pile is as big as the potential allows, electrons can go in only as fast as others are lost.

Throw ions into the mix and things get more complicated. The ions see the electron "potential hill" the other way around, as the "potential well" we are trying to achieve. They counter the coulomb repulsion of the electrons and allow more to reach the center. One of the keys to a Polywell is limiting the ion population so that this effect is minimized, in order to be sure the electrons in the center of the machine are really at low kinetic energy.

MSimon
Posts: 14334
Joined: Mon Jul 16, 2007 7:37 pm
Location: Rockford, Illinois
Contact:

Post by MSimon »

I'm moving this to theory. In case anyone asks.
Engineering is the art of making what you want from what you can get at a profit.

D Tibbets
Posts: 2775
Joined: Thu Jun 26, 2008 6:52 am

Post by D Tibbets »

As Tom Ligon pointed out the potential well is limited below thre drive potential assuming at least some electron confinement (more than one pass) is achieved. The mention of the ion interaction manifests as the vertual central anode. I think Bussard placed a limit on the ion population so that this virtual anode does not exceed ~ 20% of the negative potential well and is dependent on both the ion electron ratio overall and the local conditions. If there is ion confluence a central vertual anode will form. This is a relative anode, not an anode in the sense that it has a positive potential relative to ground, it is just that the negative potential becomes less near the center.

The interactions are complex and confusing (at least for me) and Tom Ligon's description is one of the best I have seen.

Dan Tibbets
To error is human... and I'm very human.

vahid
Posts: 33
Joined: Tue Aug 17, 2010 3:33 pm

Post by vahid »

MSimon wrote:I'm moving this to theory. In case anyone asks.
Thank you.
Engineering Is the Art of Making What You Want from What You Can Get at a Profit. ( MSimon )

vahid
Posts: 33
Joined: Tue Aug 17, 2010 3:33 pm

Post by vahid »

Tom Ligon, MSimon, D Tibbets wrote: .................................
Thank you so much for your complete and useful answers. You know that electrical current is the best insulation for electric and magnetic fields. Also it is considered an adequate protection against the emission of electromagnetic fields. Dr Bussard predicted that in a 100 MW Polywell reactor, more than 10 ^ 14 A of electrons rotate around the magnetic coils of Magrid. And I thought that this ultra high current of electrons, can insulate the electric field of Magrid then potential well depth will be lower than 80%.
Engineering Is the Art of Making What You Want from What You Can Get at a Profit. ( MSimon )

D Tibbets
Posts: 2775
Joined: Thu Jun 26, 2008 6:52 am

Post by D Tibbets »

Lets see... If WB 6 had an electron current input of ~ 40 amps at Beta=1, and the electrons circulated (and recirculated) for ~ 100,000 passes (minimum) then the circulating amps would be ~ 10^7 amps. This would scale up to ~ 10^9 circulating amps in a 3 meter machine (using the r^2 input scaling with corresponding B field scaling- 3 meter diameter with 10 Tesla fields). Where does the extra 10^5 electron current come from?

Mentioning that electron current is an excellent insulator against magnetic fields would seem to be another enlightening way of describing the exclusion of an external magnetic field from the confined electron space, and thus be a reasoned argument for the Wiffleball formation. The expansion of the Wiffleball would seem to be a natural consecuence of excess electron current added to the system, supported by the claim that as the electron cloud volume expands it drives up the Beta in a feedback mechanism until the Beta=1 limit is reached. So long as you can provide an input electron current greater than the leakage rate at the initial condition (very low beta) electron accumulation and resultant Wiffleball formation is a natural consequence.

Even with machines before WB6 EMC2 claimed Wiffleball formation. The problem was that they had to poor in much higher electron currents to drive the evolution.

I'm uncertain how this relates to the potential well formation. In WB5 they tried increasing the input current to reach expected well depths, but the results were disappointing. Possibly the loss rate was increasing at nearly equal rates. The lose rate (confinement and RECIRCULATION) seems to have paramount importance in the balance. In WB5 there was not much, if any, recirulation. It was a closed box machine.

A side consideration is that with primary magnetic confinement, the electrons bounce off the magnetic fields at divergent angles from the center, especially in the near cusp regions where the field lines are approaching angles close to parallel to the center. There is no restoring force towards a radial vector, not to mention an impedance to velocity thermalization. With recirculation, I think the electrons are restored to a radial monoenergetic vector. I'm not sure if this has an effect on the potential well depth, but this resetting would change the electron thermalization time verses the input electron current to maintain it. The available thermalization time is limited by the magnetic confinement time of the electrons (assuming there are not any internal restoring forces for the electrons , unlike what is claimed for ions ('annealing').

Thus recirculation not only limits the amount of input electron current needed to maintain a desirable ion density useful for fusion, but also decreases net electron thermalization time vs electron input current levels. I think electron current necessary to overcome thermalization issues was one of Rider's major considerations of why IEC machines could not reach breakeven, especially for non D-T fuels. Recirculation may ease this concern by a factor of ~ 10-100X

Dan Tibbets
To error is human... and I'm very human.

ladajo
Posts: 6258
Joined: Thu Sep 17, 2009 11:18 pm
Location: North East Coast

Post by ladajo »

Let's not forget the idea that the machine may self generate enough electrons from Ion stripping...

vahid
Posts: 33
Joined: Tue Aug 17, 2010 3:33 pm

Post by vahid »

D Tibbets wrote:Mentioning that electron current is an excellent insulator against magnetic fields would seem to be another enlightening way of describing the exclusion of an external magnetic field from the confined electron space, and thus be a reasoned argument for the Wiffleball formation. .................................................
Thank you. It seems that I got the point.
Engineering Is the Art of Making What You Want from What You Can Get at a Profit. ( MSimon )

D Tibbets
Posts: 2775
Joined: Thu Jun 26, 2008 6:52 am

Post by D Tibbets »

ladajo wrote:Let's not forget the idea that the machine may self generate enough electrons from Ion stripping...
Indeed, in the 2008 patent application it is mentioned that with a high Z fuel (like P-B11) so many electrons may be stripped away that the input electron current may actually need to be throttled way back. This is with a neutral gas feed, not an ion gun setup (the ions are already stripped). Also, I.m not sure how the secondary electrons (born at ~ several hundred eV) get heated without a significant excess of high energy injected electrons. I suppose that if recirculation can be improved, this would be be less of a concern.

With multiple electrons released with the ionization of boron, if the fusion rate is fast enough, the high energy fusion ions may leave faster than the electron losses. This could in theory, cause too many electrrons to be left behind (the input electron current would have to be less than zero). This would obviously represent the absolute maximum (if highly improbable) performance that could be obtained, and is again mentioned in the patent application.

Dan Tibbets
To error is human... and I'm very human.

ladajo
Posts: 6258
Joined: Thu Sep 17, 2009 11:18 pm
Location: North East Coast

Post by ladajo »

Also, I.m not sure how the secondary electrons (born at ~ several hundred eV) get heated without a significant excess of high energy injected electrons.
I recall seeing something on this somewhere. Not sure. I will poke around.

D Tibbets
Posts: 2775
Joined: Thu Jun 26, 2008 6:52 am

Post by D Tibbets »

For enlightenment for those who have not endured my rambling posts or learned from other sources... There is an ionization cross section just like there are fusion or coulomb scattering cross sections dependent on the KE of the particles. Apparently the highest crossection for ionization ( at least for hydrogen) is ~ 100 eV. This means that once ~ 100 KE is reached, most of the atoms will be losing their electrons. Little will be left for higher energy ionizations. Thus the average energy of the secondary electrons (from ionization) will have an initial KE of ~ 100 eV. If one 10 KeV injected electron powers the cascading ionization of ~ 100 secondary electrons, then all of them would have an average energy of ~ 100 eV. Further injected electrons are needed to subsequently heat these 101 electrons to the potential well depth. If you do the math, you can appreciate what ion confinement time vs electron containment times are needed to allow electron heating to close to the injected electron energy (potential well voltage). This gives an idea of the confinement time ratio in WB6. If you assume 80 injected (and recirculated?) electrons are needed for each secondary electron, this would allow for the potential well to be ~ 80% of the electron injection energy (I think). This would imply that the potential well (and to a much lesser extent magnetic) ion confinement is ~ 80 times as great as the electron confinement time. This is certainly consistent with the the claims that ion losses is a miner loss mechanism compared to electron losses.
From this you can play with the densities , mean fusion collision distance/ times, cross sections etc, to calculate results. Then throw in further complications such as annealing, convergence and traveling or standing waves (POPS effects) to make things more entertaining.
High Z fuels like boron (5 released electrons if fully ionized) changes the dynamics, and would result in even more interesting results.

Dan Tibbets
To error is human... and I'm very human.

Post Reply