Indrek wrote:Recirculation is essential. But I don't think it says anywhere recirculation means "making circles around coils". I think there's some confusion here.
Electrons do move through between the coils, so open system is essential (with high potential coils), that's true. The simulators show that as well. But they move back-and-forth, rather than round and round.
- Indrek
That was my understanding too.
OK, I think I see where the confusion arises: the electrons follow the field lines, but not forever. They can jump across and recirculate back in along a field line they didn't start on. The fields only push them at 90 degree angles to the grid, so depending on pitch, velocity, etc, they may jump off of one.
Indrek's double coil simulation makes me wonder what it would look like if the second coil were smaller, say half the diameter, with the same current but with the current in the same direction as the main coil.
The smaller diameter coil would make the field intensity on the centerline higher, so the mirror ratio would be better.
This should reduce losses, particularly of the problematic electrons traveling near & parallel to the centerline
The help could be countered by what the field lines look like between the 2 coil sets.
If it works we would only have to pay for the higher fields in strategic locations.
-Tom Boydston-
"If we knew what we were doing, it wouldn’t be called research, would it?" ~Albert Einstein
tombo wrote:Indrek's double coil simulation makes me wonder what it would look like if the second coil were smaller, say half the diameter, with the same current but with the current in the same direction as the main coil.
The smaller diameter coil would make the field intensity on the centerline higher, so the mirror ratio would be better.
This should reduce losses, particularly of the problematic electrons traveling near & parallel to the centerline
The help could be countered by what the field lines look like between the 2 coil sets.
If it works we would only have to pay for the higher fields in strategic locations.
The problem with smaller coils is ion impingement. The coils should probably in fact be larger to be in the shadow of the reactor coils. I think such additions can wait, however. As Dr. Mike says - we should start out as simple as possible.
Engineering is the art of making what you want from what you can get at a profit.
Going to larger coils defeats the purpose of increasing the B field with a limited current.
If larger diameter coils could carry enough current to create the larger B field it would be better to simply apply that to the main coils.
Of course complications are to avoided whenever possible.
I thought this line of inquiry was to squeeze a little more confinement into the system.
Yes, the ions are troublesome.
-Tom Boydston-
"If we knew what we were doing, it wouldn’t be called research, would it?" ~Albert Einstein
Looking at the picture of the operating WB7 at http://www.emc2fusion.org/
It looks like the coils are extremely close to the chamber wall.
I can't see much room for electrons to recirculate outside the WB.
How far outside should we expect them to travel?
Some threads say meters. Some say the recirculation is all inside.
Dr B. wrote that the recirculation was needed to return the electrons that escaped back to the inside of the WB, so they must be going out and then back in.
Circulation within the Magrid or around the cusps on the inside (important as it may be) is not RE-circulation to my mind.
-Tom Boydston-
"If we knew what we were doing, it wouldn’t be called research, would it?" ~Albert Einstein
In theory (no upscattering) electrons should travel one reaction space radius outside the reaction space minus the voltage lost in well creation. So they should never hit a wall that is at at ground potential.
Ions of course are trapped in the reaction space by electrostatic charge.
Engineering is the art of making what you want from what you can get at a profit.
So starting at the inside, they fall down the potential well toward the positive magrid, then continue beyond it under their momentum until they reach the starting E field potential outside the magrid where they stop and fall back toward the positive magrid and past it back into the device. (where hopefully they will collide with something and stay)
That picture sure looks like the wall is much closer than that to the coils.
If the coils are 10" dia it looks like they are maybe 2" from the inside rim of the port.
I suppose it could be an optical illusion.
-Tom Boydston-
"If we knew what we were doing, it wouldn’t be called research, would it?" ~Albert Einstein
I would think it wouldn't need to be far. Once they get outside they can't see the electron well anymore, and the field just pushes them at 90 degrees to the coil. Their main influence at that point is the positive magrid.
Let's see if I understand this. Starting from a charged-up well in steady state:
1) Electrons get injected through a cusp. They have to have enough energy to overcome the electrostatic repulsion of the E-field between the magrid and the virtual cathode.
2) Once through the cusp, they rapidly thermalize, which is actually good, because they get knocked off of the B-field lines emanating from the cusp. Otherwise, they'd just go rocketing out another cusp and we'd have lousy containment.
3) So now we've got pretty hot, electrostatically uncomfortable electrons bouncing off of the wiffle-field and trying to avoid their friends. However, every so often one of them hits a cusp and gets out of the field.
4) The wayward electron follows a B-field line, pretty much radially out of the well. It's gaining energy as it heads in the positive direction in the E-field, toward the magrid.
5) Now it rockets past the (magnetically insulated) magrid, so it starts losing energy. In addition, the B-field is now bending back on itself. One of three things can happen:
a) The electron has so little energy that the E-field stops it and reverses its course, dropping it back into the well.
b) The electron has enough energy to overcome the E-field working against it but not enough energy to hop the B-field lines, so it whips around the B-field line back into the well.
c) The electron has enough energy to hop B-field lines and is lost.
6) If the electron didn't get lost through mechanism 5c), it is now falling back toward the Magrid, gaining energy. When it passes the Magrid, it starts losing energy to the E-field.
7) As we get closer and closer to the wiffle-field cusp, one of two things can happen:
a) The electron has enough energy to overcome the electrostatic potential at the wiffle-field boundary and migrates back into the herd.
b) The electron has less energy than the electrostatic potential at the boundary, and it gets pushed out of the well, to oscillate back and forth between the inside and the outside of the magrid.
8. Lather, rinse, repeat.
Some comments/questions:
A) If you've got a direct conversion anode sitting out beyond the magrid at 1.25 MV (to catch doubly-charged 2.5 MeV alphas), you are completely screwed. Very few electrons will have enough energy to overcome the electrostatic repulsion at the wiffle-boundary and will wind up doing something bad outside the magrid (neutralizing the well's E-field and/or falling into the big anode). To avoid this, you have to have a flat spot in the E-field potential curve, extending some distance from the outer edge of the magrid. The only way I can think of to do this is with a negative grid outside the magrid, but that'll send us directly to space-charge hell. The radius is gonna get bigger, too, sending us on a bonus excursion to unimaginably-big-vacuum-chamber hell.
B) You want the electrons that have just emerged from the wiffle-boundary to have a temperature profile such that:
(electrostatic potential at wiffle boundary) < (energy of electrons leaving cusps) < (energy required to escape the B-field on the outside of the magrid)
That's a fun problem for somebody who remembers what they learned in thermodynamics. However, it's instructive to consider a very cold (i.e. zero velocity) electron that's just escaped the wiffle boundary. If that electron doesn't thermalize outside the boundary and doesn't lose any energy to brem while circling the magrid, it should arrive back at the boundary with zero velocity. That's probably not quite good enough; the electron needs to penetrate the boundary far enough to thermalize, or it will just get pushed back along the cusp-line and be lost to the well, oscillating between opposite sides of the magrid.
C) There's an interesting problem at startup, which must have already been solved. Since there are no electrons in the soon-to-be wiffle-field to begin with, all electrons recirculate until there are enough of them inside the wiffle-boundary to start thermalizing. Since all injected electrons start out with very low entropy (they're all orbiting the friggin' B-field), you must have to have a lot more electrons around the magrid at startup than you do in steady-state. What kinds of transients does this impose on the machine at startup?
D) For the electrons that are energetic enough to follow the B-field line all the way around the outside of the magrid, do we have any bremmstrahlung problems? (I'd think not--they can't be very energetic out that far.)
A) If you've got a direct conversion anode sitting out beyond the magrid at 1.25 MV (to catch doubly-charged 2.5 MeV alphas), you are completely screwed.
I think you would just move it back till the inverse square law solved the problem for you. The alphas are far more energetic and will shoot right by areas where the electrons will be pulled back by the charge on the Magrid.
b) The electron has less energy than the electrostatic potential at the boundary, and it gets pushed out of the well, to oscillate back and forth between the inside and the outside of the magrid
I don't think they'll oscillate much. It looks like the reason the Wiffleball works is that it's harder to get out than to get in, due to the geometry of the field lines.
To further abuse the Wiffleball analogy, imagine each hole in the ball has a funnel with the mouth pointing out. It's easy for marbles to get back in through the wide mouth, but hard for them to get out through the narrow neck. Thus you get a high ratio of density inside versus outside.
The analogy isn't wrong - it is too simplistic. Each electron sees all the other ions and electrons in the system as well as the MaGrid and B field. As everything moves, it all affects the way everything moves. Some electrons will get smacked out cusp lines with energies high enough to hit external power grids. But the probability is really low. Just how low is not known - and that is the important factor to find out experimentally. It's too darn hard to compute correctly at high density, even with all the computing power on the planet. Today anyway. Give it a few years.
