Translating FRCs

[Art Carlson]

It's called brainstorming. Weeding out the bad ideas is a subsequent step.

OK, fair play. I'll try to avoid the negative pull-down on the comment and replace 'can/could' with 'may conceivably' when I read it again.

I don't really know how elastic FRC bounces are. Actually, I think if you throw them at a metal plate or at each other, they tend to thermalize and stick. If you send them into a gently increasing magnetic field, then they should decelerate and turn around. But that would take you as much length as accelerating them in the first place, so you haven't gained anything. I got on this tack because I thought I saw a similar concept sometime in the distant past, but I may have gotten something confused.

I think I'll give up the bounce idea and try to figure out what will happen if you try to make an FRC go around a corner.

[choff]

Can a FRC be run inside a Stellarator, or would that be counterproductive to confinement?

[Art Carlson]

Can a FRC be run inside a Stellarator, or would that be counterproductive to confinement?

If you don't have a close-fitting shell, then an FRC is unstable to tilting and at best marginally stable to translations. I'm pretty sure a curved filed would spit out an FRC (to the low-field side).

There have been experiments that successfully shoot an FRC across a magnetic field, the idea being to develop a new method to refuel tokamaks,.

[chrismb]

Can a FRC be run inside a Stellarator, or would that be counterproductive to confinement?

If you don't have a close-fitting shell, then an FRC is unstable to tilting and at best marginally stable to translations. I'm pretty sure a curved filed would spit out an FRC (to the low-field side).

There have been experiments that successfully shoot an FRC across a magnetic field, the idea being to develop a new method to refuel tokamaks,.

Sounds clearer-headed to me now. I would suggest that it will merely act with gyroscopic precession - in the best case! An FRC going around the corner will do so 'non-uniformly' if it isn't rotating and would break up due to differential compression and Rayleigh instabilities across the thing. But if it is rotating, then its rotational axis will tend to twist away from the major circumference under a crossed product translation. In fact, what I'm writing sounds almost good enough to be a description of Zeta instabilites (if one considers a Zeta plasma as a series of FRC packets in a circle).

[BenTC]

although as magnetic confinement discovered 60 years ago, bending a magnetic field can have unexpected and grave consequences.

Could you expand on that - its a bit broad to search on. I'm curious and will probably learn something new.

[MSimon]

although as magnetic confinement discovered 60 years ago, bending a magnetic field can have unexpected and grave consequences.

Could you expand on that - its a bit broad to search on. I'm curious and will probably learn something new.

Start with ELM (Edge Localized Modes) and then go on to Alfven Waves.

That should lead you to other things.

[Art Carlson]

although as magnetic confinement discovered 60 years ago, bending a magnetic field can have unexpected and grave consequences.

Could you expand on that - its a bit broad to search on. I'm curious and will probably learn something new.

Start with ELM (Edge Localized Modes) and then go on to Alfven Waves.

That should lead you to other things.

I was thinking of something much more fundamental. (I wish I could just refer you to Wikipedia, but they hardly touch the issue, and get even that little bit wrong.) A straight magnetic field, as in a theta-pinch, does a good job of confining a plasma in the transverse direction, but the ends are leaky. The obvious solution is to bend the ends around to connect to each other. But charged particles in a non-uniform field (both curvature and gradients) are found to drift across the field. Since the direction of drift depends on the charge of the particles, electrons and ions move in opposite direction. You might hope that the electric field that builds up with the charge separation helps hold them together, but in fact particles react to crossed electric and magnetic field by drifting in a new direction, in this case outward, away from the axis of rotational symmetry. To save the day you have to give the field lines a twist so that each field line spends some of its time on the outside, lossing plasma to flux surfaces farther out, and some of its time on the inside, shoveling plasma back into flux surfaces closer to the minor axis. (There are other ways you can look at this problem and come to the same result.)

You have to get all this equilibrium physics straight before you can even start to think about stability.

(edit: OK. I made a stab at fixing up the Wikipedia article.)

[jmc]

The obvious advantage of translating the thing and allowing it two compress itself is that the nudges each coild gives the FRC to get it to the required energy exert less force on each coil, than a compressing linar and disipate less energy in one place allowing the accelerating coils to be used again and again.

The advantage over the general fusion pistons technique is that the coils can "feel" the FRC coming, in that the approaching FRC induces a current in them as it passes with which you could create a control system to induce a current in each coil to give it an extra kick. The pistons are seperated from the plasma target by an impenetrable wall of liquid lithium so it is hard to envisage how they could feel it coming, this would make any feedback system where sensing of the position of the FRC afects the pistons.

You wouldn't be able to make the curves too narrow otherwise the centrifugal force would squash and distort the FRC. There are ways of decreasing the linear acceleration distance:

Fz = mu dB/dz

i.e. if push up the magnetic moment of the plasmoid or increase the gradient of the field you can achieve a stronger force.

[jmc]

How cool is this??!!

http://www.niac.usra.edu/files/library/ ... Slough.pdf

[Solo]

jmc: ok, here's my take on the tapered flux conserver: It won't pull any new magnetic field lines into itself, so the field/flux already needs to be there, so it doesn't do you any good unless the conserver contracts as in MTF. Well, it may help somewhat in that it keeps the displaced flux of the FRC from being spread over the entire annular area between the seperatrix and the coils, and just confine it between the conserver and the FRC. But if the conserver decreases in diameter along the taper, and the strength of the pre-existing field doesn't increase, then the total flux inside the conserver decreases along the length. That is, field lines penetrate the flux conserver, and so the FRC will come in contact with the walls as the necessary excluded flux isn't there.

The plasmoid will have to compress itself against a large static field. But we can still get all the advantages of translation-acceleration and self-compression.

re sensing the plasma: actually, Slough's idea is that the coils don't have to sense the plasma. Since it's diamagnetic, just send your wave along and the FRC will stay in front of it.

[Axil]

In the first post on this thread, Jmc said as follows:

Why no focus on translating the plasmoid to ever higher kinetic energies through repeated magnetic squeezing at the rear and then let the FRCs own kinetic energy accomplish the compression stage.

This is a very simple idea, so it most probably won’t work. The membership on this site will be best suited to tell me why. Please do.

A packet of deuterium plasma ions with a positive charge can be magnetically accelerated in a vacuum down a long round tapering tube that is electrostatically charged to a high positive voltage. The forward motion of the plasma packet will compress and slow the packet as the diameter of the tube progressively decreases.

The positive charge on the tube walls will repel and center the particles in the plasma packet equipotentially in the radial direction and also tend to compress the packet in the axial direction with a electrostatic compressive force that goes as the square of the tubes decreasing radius.

Will the forward motion of the packet force it into a high density state as its kinetic energy propels its travel down a tube with an ever decreasing diameter that eventually decreases into a very small highly compressive diameter? Can such a high density plasma be formed to cause fusion?

If this is not a valid scenario, what process would stop this compressive process from progressing to fusion?

[D Tibbets]

In the first post on this thread, Jmc said as follows:

Why no focus on translating the plasmoid to ever higher kinetic energies through repeated magnetic squeezing at the rear and then let the FRCs own kinetic energy accomplish the compression stage.

This is a very simple idea, so it most probably won’t work. The membership on this site will be best suited to tell me why. Please do.

A packet of deuterium plasma ions with a positive charge can be magnetically accelerated in a vacuum down a long round tapering tube that is electrostatically charged to a high positive voltage. The forward motion of the plasma packet will compress and slow the packet as the diameter of the tube progressively decreases.

The positive charge on the tube walls will repel and center the particles in the plasma packet equipotentially in the radial direction and also tend to compress the packet in the axial direction with a electrostatic compressive force that goes as the square of the tubes decreasing radius.

Will the forward motion of the packet force it into a high density state as its kinetic energy propels its travel down a tube with an ever decreasing diameter that eventually decreases into a very small highly compressive diameter? Can such a high density plasma be formed to cause fusion?

If this is not a valid scenario, what process would stop this compressive process from progressing to fusion?

I'm not sure an electrostatic charge on the tube at the same position on the tube would compress the plasma- think of Gauss's law. I'm guessing an electrostatic charge applied sequentiallly along stages of the tube might accelerate the plasma down the tube as the charge would be leading the plasma ball, not surrounding it. Compression would presumably be done by the plasma ball being pulled/ pushed into stronger magnetic fields and/or being squeezed and possibly pushed by variable magnetic fields. Or not...

Dan Tibbets