I am curious: will an electric field from a capacitor repel electrons which are "flying" towards the field?
My thought process is: what if the electron rich plate of a capacitor was used as part of the polywell device itself? There would be the electron rich plate separated by a dielectric, then an electron deficient plate on the outside of the polywell.
The goal of the thought process is: use the inside structural material as one plate in a capacitor in order to repel electrons swirling around the magrid. This is to be used where the magrid is not: structural connections and enclosure which allows for vacuum.
I was thinking of ways to divert the stray electron cloud without having to use electromagnets everywhere.
All thoughts are very much welcome, even the ones that point out how much of an idiot I am. :-)
e
Electric field VS electrons in motion = who wins?
Hmmm. MrE, you may wish to view Joe Strout's resources page. There, you will find many answers to your questions. However, in brief, the electron cloud could be diverted by either electrostatic fields or by magnetic fields.
As it turns out, magnetic fields have an advantage over electric fields - they "turn" the electrons by 90 degrees from incident rather than repelling them. This allows the magnetic field to circulate electrons without having them be lost forever. The electrons move around the magnets in the polywell in an endless loop. If electrostatics were used - and they are, in traditional IEC reactors - see the University of Wisconsin's pages on IEC - then we would either confine the electrons forever or lose them forever, rather than keeping them in the loop. The efficiency of such a reactor cannot be much more than 1e-5 (and we would ideally like 1e0). Whereas polywell is much more likely to exceed breakeven.
Regards,
Tony Barry
PS I would not claim to understand a great deal of the physics, but I follow what I can with interest.
As it turns out, magnetic fields have an advantage over electric fields - they "turn" the electrons by 90 degrees from incident rather than repelling them. This allows the magnetic field to circulate electrons without having them be lost forever. The electrons move around the magnets in the polywell in an endless loop. If electrostatics were used - and they are, in traditional IEC reactors - see the University of Wisconsin's pages on IEC - then we would either confine the electrons forever or lose them forever, rather than keeping them in the loop. The efficiency of such a reactor cannot be much more than 1e-5 (and we would ideally like 1e0). Whereas polywell is much more likely to exceed breakeven.
Regards,
Tony Barry
PS I would not claim to understand a great deal of the physics, but I follow what I can with interest.
Well MrE, join the club of idiots 
I've spent the last 24 hours beating my head on a "simple" problem of getting electrons from the source into the center of the polywell. With no plasma. This is a basic "electron gun" or "magnetron" problem - it is just a different geometry. My fundamental problem is that I am used to thinking about particles not full beams. I keep getting zero for my answer, and I know I'm doing something really stupid!!
I think it is worth while to first understand a 1D problem. Even a 1D single particle problem.
Take an electron in a fixed, uniform magnetic field and work out the equations of motion. Then add an electric field. What happens when E is aligned with B and what happens when E is perpendicular to B?
The critical thing to do is to check to see if your answer makes sense. At some point you have to do an experiment and measure things - some times it does, some times you are back in the club of idiots! But that's the best way to learn.
I've spent the last 24 hours beating my head on a "simple" problem of getting electrons from the source into the center of the polywell. With no plasma. This is a basic "electron gun" or "magnetron" problem - it is just a different geometry. My fundamental problem is that I am used to thinking about particles not full beams. I keep getting zero for my answer, and I know I'm doing something really stupid!!
I think it is worth while to first understand a 1D problem. Even a 1D single particle problem.
Take an electron in a fixed, uniform magnetic field and work out the equations of motion. Then add an electric field. What happens when E is aligned with B and what happens when E is perpendicular to B?
The critical thing to do is to check to see if your answer makes sense. At some point you have to do an experiment and measure things - some times it does, some times you are back in the club of idiots! But that's the best way to learn.
You know drmike, I was thinking of a way to inject electrons into the core. Assuming an octagon shaped grid with 6 sides(even the truncated dodecahedron but I am playing with a simpler shape).
You are thinking of a coil gun type injection system, why don't you use a slightly cup shaped portion (that can move the outer ridge to focus the e-beam like a directional antenna) as the coil?
This cup shaped portion would make the inner part of the vacuum chamber an RF emitter. The idea would be to created a cone of RF energy focused on the center of the polywell. Seeing as electrons will be sliding along the magnetic fields from the magrid an electron stream should slip in through with the flow of the existing electrons.
Imagine the primary of a tesla coil. I have seen designs that curve the outer rings of copper wire into an almost cup shape.
*scurry's to the corner with dunce cap*
You are thinking of a coil gun type injection system, why don't you use a slightly cup shaped portion (that can move the outer ridge to focus the e-beam like a directional antenna) as the coil?
This cup shaped portion would make the inner part of the vacuum chamber an RF emitter. The idea would be to created a cone of RF energy focused on the center of the polywell. Seeing as electrons will be sliding along the magnetic fields from the magrid an electron stream should slip in through with the flow of the existing electrons.
Imagine the primary of a tesla coil. I have seen designs that curve the outer rings of copper wire into an almost cup shape.
*scurry's to the corner with dunce cap*
You are on the right track. One of the things I want to do is shape the
emitter to see what difference it makes. But I also want to start with DC
solutions and work up to RF. If it isn't stable in DC, it won't work at all.
But you know, talking here really helps. I found the error in my thought
process and got the math to work. You don't need to "inject" electrons, you
just let them fall. The MaGrid is positive, so you can set the electron source
negative. The magnetic field lines help to focus the beam towards the center
and the momentum keeps them going.
The problem I'm working on is a simple (!!) approximation with only 2 coils.
I can figure out simple formulas for the magnetic and electric fields and I can
add the electron source. Then life gets messy. But it is still analytical.
Tesla and Edison fought a lot, and much of it has to do with the philosophy
on how to solve hard problems. Tesla liked theory, Edison liked experiment.
Both methods are really important, sometimes just trying things can get
you useful information. But for large systems, being able to estimate
what will actually happen is very useful.
To really appreciate your idea you need to do some back of the envelope
estimates. What are the basics - power, energy, current and voltage.
Get a feel for the concept with really simple equations. If the numbers
are obviously silly, then the idea isn't sound. From there you can try to
build an experiment and try things, or you can refine the estimate. With
today's computers, there are a lot of models you can do.
emitter to see what difference it makes. But I also want to start with DC
solutions and work up to RF. If it isn't stable in DC, it won't work at all.
But you know, talking here really helps. I found the error in my thought
process and got the math to work. You don't need to "inject" electrons, you
just let them fall. The MaGrid is positive, so you can set the electron source
negative. The magnetic field lines help to focus the beam towards the center
and the momentum keeps them going.
The problem I'm working on is a simple (!!) approximation with only 2 coils.
I can figure out simple formulas for the magnetic and electric fields and I can
add the electron source. Then life gets messy. But it is still analytical.
Tesla and Edison fought a lot, and much of it has to do with the philosophy
on how to solve hard problems. Tesla liked theory, Edison liked experiment.
Both methods are really important, sometimes just trying things can get
you useful information. But for large systems, being able to estimate
what will actually happen is very useful.
To really appreciate your idea you need to do some back of the envelope
estimates. What are the basics - power, energy, current and voltage.
Get a feel for the concept with really simple equations. If the numbers
are obviously silly, then the idea isn't sound. From there you can try to
build an experiment and try things, or you can refine the estimate. With
today's computers, there are a lot of models you can do.
The way I think about it, is if you have magnetic fieldlines with electrons on them intersecting solid surfaces, then if the potential difference of the surface and the electrons on the fieldline is a few times the electron temperature then the electrons will be repelled by the surface, if it isn't then the electrons will rapidly (almost instantaneously) collide with the surface and be lost.
When surfaces don't intersect fieldlines then you have to consider cross-field transport, which can be extremely complicated, however on the plus side, your in with a chance that the electrons will take a significant ammount of time to collide with said surface.
Just out of curiousity, how many people here are familiar with the E X B drift?
When surfaces don't intersect fieldlines then you have to consider cross-field transport, which can be extremely complicated, however on the plus side, your in with a chance that the electrons will take a significant ammount of time to collide with said surface.
Just out of curiousity, how many people here are familiar with the E X B drift?
It is a long time scale, bulk fluid phenomena. So is a pressure gradient drift. grad(p) x B sets up a similar current flow.