reddit: We are nuclear fusion researchers, ask us anything

[D Tibbets]

Concerning direct conversion, it is an extremely widely used and successful process. I am referring of course to ion rocket engines, Cathode ray tubes, etc, etc.
An ion gains kinetic energy from a set of electrodes or space charge. The reverse, the gaining of potential energy by decelerating ions (or electrons) is the exact same process, only in reverse You cannot have one without the other. I don't think anyone will claim that classic TV didn't work. Also, I suspect similar processes are used in many vacuum tubes. This was one of Bussard's complaints. Many modern physicists suffer from a training blind spot. They have little understanding of vacuum tube practical technology. Much of the Polywell principles are founded on established vacuum tube technology.

Changing direction of a charged particles does not generate (or consume) energy (other than some inefficiencies/ processes like cyclotron radiation or bremsstruhlung radiation). Certain manipulations may be desirable or necessary in certain systems like a tokamak, but again a reminder that the Polywell is a much different machine. The magrid is an already proven direct energy converter, at least if you accept that electron circulation is real. I think the real question with direct conversion is not whether it is possible, but whether it can be done efficiently and practically. In the Tokamak, the diverter may fall under this consideration. Just diverting edge plasma flow in a tokamak is going to be very difficult as the heating of the diverter structure will be severe. I have seen one paper that proposed applying direct conversion aspects , not just as a power harvesting effort, but to cool the plasma before it contacts the diverter, thus sparing it from much of the heating problems.

Dan Tibbets

[hanelyp]

For those actually open to learning:

Two simple questions:
1. where is the uniform current distribution in TOKAMAKs?

The whole of the plasma is a conductor. Unless the impressed electric field is high frequency, the induced electric current will be through the bulk of this conductor.

2. Where you saw the talk on inside mag distribution in plasma devices? If to recall that plasma is strongly diamagnetic and as a rule pushes out power lines outside. Have you ever read about frozen-in power lines?

Beta < 40% in a tokomak. Plasma pressure < 40% of magnetic pressure. The magnetic field penetrates the volume of the plasma. Diamagnetic means the magnetic field is reduced, not eliminated.

One problem described in pre-tokomak toroidal magnetic field devices is an uneven spiraling in a non-uniform magnetic field, sending electrons one way and ions the other.

If the toroidal field were completely excluded from the plasma, or didn't vary with radius, there would not be the charge separation described for a proto-stellerator/tokomak minus poloidal field.

If the plasma current were confined to the outer skin, the net magnetic field would not be twisted as needed to prevent charge separation.

As for where I'm getting my picture, I've read a lot, and fill in gaps working from first principle electromagnetic theory.

[Joseph Chikva]

For those actually open to learning:

Two simple questions:
1. where is the uniform current distribution in TOKAMAKs?

The whole of the plasma is a conductor. Unless the impressed electric field is high frequency, the induced electric current will be through the bulk of this conductor.

2. Where you saw the talk on inside mag distribution in plasma devices? If to recall that plasma is strongly diamagnetic and as a rule pushes out power lines outside. Have you ever read about frozen-in power lines?

Beta < 40% in a tokomak. Plasma pressure < 40% of magnetic pressure. The magnetic field penetrates the volume of the plasma. Diamagnetic means the magnetic field is reduced, not eliminated.

One problem described in pre-tokomak toroidal magnetic field devices is an uneven spiraling in a non-uniform magnetic field, sending electrons one way and ions the other.

If the toroidal field were completely excluded from the plasma, or didn't vary with radius, there would not be the charge separation described for a proto-stellerator/tokomak minus poloidal field.

If the plasma current were confined to the outer skin, the net magnetic field would not be twisted as needed to prevent charge separation.

As for where I'm getting my picture, I've read a lot, and fill in gaps working from first principle electromagnetic theory.

Wrong statement that time dependent currents have uniform distribution in good conductors. Even in superconductors skin-effect is observed.
This is the first what Google gave me on key words: "skin-effect superconductors"

The Mattis–Bardeen theory was derived to explain the anomalous skin effect of superconductors.

And nobody considers mag field distribution inside plasma. As if we talk about beta, so we talk about gas pressure from inside and magnetic pressure from outside. Regardless to value of beta: 0.00000000000001, 0.4 or 1
But for your reference, plasma is a diamagnetic.

Diamagnetism is the property of an object or material which causes it to create a magnetic field in opposition to an externally applied magnetic field.

and you cannot say does this reduces or eliminates the field.
If you want to go in depth into TOKAMAK theory, there are many more basic things you should study before.
Fill the gaps and good luck.

[tomclarke]

For those actually open to learning:

Two simple questions:
1. where is the uniform current distribution in TOKAMAKs?

The whole of the plasma is a conductor. Unless the impressed electric field is high frequency, the induced electric current will be through the bulk of this conductor.

2. Where you saw the talk on inside mag distribution in plasma devices? If to recall that plasma is strongly diamagnetic and as a rule pushes out power lines outside. Have you ever read about frozen-in power lines?

Beta < 40% in a tokomak. Plasma pressure < 40% of magnetic pressure. The magnetic field penetrates the volume of the plasma. Diamagnetic means the magnetic field is reduced, not eliminated.

One problem described in pre-tokomak toroidal magnetic field devices is an uneven spiraling in a non-uniform magnetic field, sending electrons one way and ions the other.

If the toroidal field were completely excluded from the plasma, or didn't vary with radius, there would not be the charge separation described for a proto-stellerator/tokomak minus poloidal field.

If the plasma current were confined to the outer skin, the net magnetic field would not be twisted as needed to prevent charge separation.

As for where I'm getting my picture, I've read a lot, and fill in gaps working from first principle electromagnetic theory.

Wrong statement that time dependent currents have uniform distribution in good conductors. Even in superconductors skin-effect is observed.
This is the first what Google gave me on key words: "skin-effect superconductors"

The Mattis–Bardeen theory was derived to explain the anomalous skin effect of superconductors.

And nobody considers mag field distribution inside plasma. As if we talk about beta, so we talk about gas pressure from inside and magnetic pressure from outside. Regardless to value of beta: 0.00000000000001, 0.4 or 1
But for your reference, plasma is a diamagnetic.

Diamagnetism is the property of an object or material which causes it to create a magnetic field in opposition to an externally applied magnetic field.

and you cannot say does this reduces or eliminates the field.
If you want to go in depth into TOKAMAK theory, there are many more basic things you should study before.
Fill the gaps and good luck.

I'm following this argument and cannot say that either side convinces.

But skin effect is larger in good conductors. And hanleyp was saying specifically that in slowly time-varying fields skin effect would not be significant, so your arguments here are not correct.

Of course, I believe that plasma studies are incredibly complex, and that there is not good evidence that polywell will be immune from plasma instability issues at high beta. Perhaps there will be clear experimental data from EMC2.

[Joseph Chikva]

But skin effect is larger in good conductors. And hanleyp was saying specifically that in slowly time-varying fields skin effect would not be significant, so your arguments here are not correct.

hanleyp is saying about uniform current distribution that not corresponds to true. As initially there is the skin effect in TOKAMAKs, then bootstrap current driven by gradient of density.

[Joseph Chikva]

But skin effect is larger in good conductors. And hanleyp was saying specifically that in slowly time-varying fields skin effect would not be significant, so your arguments here are not correct.

hanleyp is saying about uniform current distribution that not corresponds to true. As initially there is the skin effect in TOKAMAKs, then bootstrap current driven by gradient of density.

By the way, plasma in TOKAMAK is a so good conductor that 0.5-2V loop voltage drives megaamperes order current.
But nevertheless:

Measurements of the time evolution of the current-density distribution in ASDEX show that lower-hybrid current drive leads to broader profiles

Where is uniformity?

[KitemanSA]

Where is uniformity?

Then again, where is the thread that gives 1/r(minor)?

[D Tibbets]

Direct conversion is an extremely widely used and successful process. I am referring of course to ion rocket engines, Cathode ray tubes, etc, etc.

I agree. Ion rocket motors are extremely widly used.

I am referring .....................Cathode ray tubes, etc, etc.

So. you think that stream of alphas or particles or plasma should bombard some plate without deceleration?

Concerning the second point, I do not think alphas should bombard some plate before deceleration. They should fly by some negatively charged plate and thus give up a portion of their energy. This could be a ~ 2 million volt plate and thus absorb ~ 2 million eV of KE from the alpha. This may represent ~ 1/4th of the KE of the mixture of alphas. By adding more plates additional energy could be harvested. It would be next to impossible to to collect all of the KE but ~ 80-90% may be reasonable. The trick is to ground the decelerated alphas, before the next plate becomes dominate and actually accelerates the alpha. It involves a complex mixture of geometry, spacing "diverter" magnetic fields, shielding and Gauss Law effects may be used just as with the Magrid and electrons. Note that the pos.magrid actually accelerates the alphas once they are outside the magrid. But outlying venetian blind type electrodes that are negatively charged (combined with Guass law considerations and / or clever use of grounding plates and magnetic fields, etc).

The Polywell has natural collectors, they are called the cusps. And I think Debye considerations may apply in a neutral ambipolar flow of plasma. But this does not apply to a Polywell. The electrons at thousands of eV energy are far different than alphas at millions of eV KE. They are not closely coupled with each other. They do effect each other but not in a dominate fasion.

I envision a collector grid as something like this. At increasing radii, the magrid, then a low voltage negative or grounded plate with magnetic fields that will divert the electrons to a grounded surface. They will curve in a magnetic field more do to their lower KE and lower momentum. The high energy alphas will be diverted some but to a much smaller extent. Subsequent electrodes will slow them as the alphas pass and pull away from them, until the alpha has lost enough momentum that they curve and hit a grounding plate, perhaps alternating with the charged plates. I am assuming that electron separation and collection is necessary so that they are not accelerated to millions of eV. As mentioned, this is easy because of the huge difference in momentum of the electrons compared to the alphas. Similar separation is done in ion guns, etc.

Consider the TV. The electrons do not hit the accelerating plates. That would defeat the purpose. Electrons are boiled off of an electrode, accelerated past one or more anodes, focused and guided (so that they do not hit the anodes) with magnetic fields, the electrodes then stream on to hit the phosphors on the screen. A direct conversion grid is exactly the same with some variation. Alphas escaping through a cusp are the equivalent of the charged particles near the phosphor screen. They are traveling fast due to their fusion birth KE. They are focused by the cusp into a beam with some degree of dispersion. The cathodes in this case and are decelerated, and they hit the ground with much of their kinetic energy harvested. The multiple electrodes succeed in sequentially accelerating or decelerating the charged particle due to Gauss law effects, and/or careful shielding effects.
I admit this is a complex problem, but there are many examples of practical devices. Think of linear and circular accelerators. They work with more than one electrode (from the charged particle's perspective) to sequentially impart KE to the charged particle. The reverse (deceleration) is the same.

As for what a direct conversion grid might look like consider the device produced by the MIT scientists (or was that Columbia graduates.

http://www.physics.usyd.edu.au/~khachan ... LEtalk.pdf
Look at slide (figure)15

Dan Tibbets

[Joseph Chikva]

Where is uniformity?

Then again, where is the thread that gives 1/r(minor)?

Of what?

[Joseph Chikva]

Concerning the second point, I do not think alphas should bombard some plate before deceleration. They should fly by some negatively charged plate and thus give up a portion of their energy.

Dan, decelaration of alphas possible if you have a positively charged surface and not negatively. But alphas will escape reactor together with electrons and lossing. And you should separate electrons before deceleration of alpha. And technically that is possible only after expansion after wich Debye length of stream would be big enough.

[ladajo]

Joseph, please remember that in Polywell the mechanism for "confining" fuel is different for that of electrons. But technically, the fuel is not confined, but merely "attracted". This is the main difference between Tokamak and Polywell. It is not important to hard confine the fuel. It is important to hard confine the electrons, thus sufficient fuel can be attracted towards the volume of the core energetically enough to react.

The fuel products are also "somewhat confined" by the fuel "confinement" mechanism. In the case of alphas, about 1000 passes (as estimated by Dr. Nebel).

It is irrelelvant that alphas (fuel products) and electrons will escape to differing degrees.

I would also point out that the Venetian Blind construct's greatest strenght is in that it allows capture of a range of energies. Each blind is tuned to a differnet energy band. So by the last blind, you have them all. (or as much as you want).

[Joseph Chikva]

But technically, the fuel is not confined, but merely "attracted".

Attraced for confinement like particles is confined in stars by gravitational field which is also potentional field by its nature. And what?

[ladajo]

I have to wonder if EMC2 has looked at this one.

http://www.physics.usyd.edu.au/~khachan ... LEtalk.pdf

I am assuming yes.

[Joseph Chikva]

I have to wonder if EMC2 has looked at this one.

http://www.physics.usyd.edu.au/~khachan ... LEtalk.pdf

I am assuming yes.

I saw similar somewhere. If I recall correctly, FPGeneration.

[KitemanSA]

Where is uniformity?

Then again, where is the thread that gives 1/r(minor)?

Of what?

It was your statement.