magrid configuration brainstorming

[drmike]

DrMike, when you're done with welding you can use the welder to power the coils. High current, low voltage, robust, short tolerant. Not too bad if you can do what you want with a few hundred amps. It would take a huge parking lot full of them to jump start a pw100. You probably have though of this too, but it might make a nice dual use.

Exactly what I was thinking! If the coils short, the power supply will have no problem vaporizing the short

I'm scavenging parts and reading up on a processor I haven't played with before. I got the processor because the eval board was the cheapest I could find. If I screw things up and blow up the board I want to minimize my losses....

It's interesting though - to get good control and power efficiency you need to go to high voltage low current and then transform back to low voltage high current. The first stage does power factor correction efficiently, the second stage does power delivery efficiently. So not only will I have a magnet supply, I'll have the basics for a high voltage supply too - I just need to change the second stage to be a bigger voltage boost instead of a current control boost.

I can tell my kids like it too. This weekend I tried to suggest all sorts of things to do. They came back with no, that's too boring. Then I suggested we build something that can blow things up, and we had an hour of fun looking at how to build tesla coils and high power co2 lasers.
It all comes back to power eventually, but I think controlled power is more useful.

[tombo]

OK Well, this seems to be where the wires go according to the pattern of B curling around straight segments.
http://i299.photobucket.com/albums/mm31 ... _wires.jpg
But it does not make sense and I could not get any colors to show even after hours of messing with light sources.
Is this anything like what you intended?

Yes, pyromania has certainly evolved into our gene pool.
Cave-kids who were not fascinated by fire did not make it. (pun not intended but I'll leave it anyway.)
Also, show one even as young as 1 year a real hammer (or any other tool) and a toy hammer and they can tell the difference.
They will take the real one EVERY time. (boy or girl)
I spent a lot of time carefully watching them swing the real ones.

[drmike]

NICE! I really don't know why the colors didn't come out - oh wait a minute!

There's a file that holds the colors: arrow_color_map and forgot to upload it before. Curses!!

Crap, this stuff is too complicated. I was so happy I got anything to work I completely forgot that the color data for surfaces is defined in different places. And it gets more complicated if you want to add surface texture - that goes in yet another set of files.

I'm surprised the program didn't complain - but I do have the color file in the same directory as the data file. I think there is a command in the 0bj file that tells it to go look for the color file. Crap! See if copying that over gets the colors.

[tombo]

OK Now I'm seeing color.
http://i299.photobucket.com/albums/mm31 ... 2blend.jpg
But the wires aren't located where I expected them to be.
I tried to center them by hand in the centers of the straight sets of closed circular field arrows.
I see 2 bent wires one in the plane and one 90 deg out of the plane.
Is that where you expected the wires to be?
It is not where I expected to see them.
How can I post a copy of the file so you can play with it?

[drmike]

I think I did a corner, but with 2 bends next to each other. I'd have to look at the code more carefully. Best thing to do is go offline and communicate by e-mail so we can nail it. I can't draw in the pipes, but I can tell you where they are (it's pretty amazing to me you've figured out blender that well!) I'll send a pm with my email address and we can go from there.

[Solo]

I've got an idea, after reading Dolan's review paper. In most other devices, rather than charging the actual magnetic coils, positively charged rings are placed around the cusp throats and negatively charged repeller plates are placed outside the cusps at a small distance, wide enough to cover the flux lines coming through the rings at the cusp throats.

If you apply this to the polywell, it bounds the electron sheath, and concerns about the surface conformity of the magnet coils are removed. This would allow the formation of a magnetic grid that was not conformal at all points to the magnetic field. In particular, take Tombo's octahedron coil (the one made out of a single tube), and merge the corners (insulating the wires as necessary to preserve the electric circuit). The result will be a magnetic field with 8 'triangle' cusps, which ought to be more like point cusps than otherwise. That is, on the surface of a sphere around the machine, the centers of the sides will be local field minimums in two dimensions.

The triangularity will probably make the corners more leaky than the point cusps in a spindle cusp machine, due to the lack of symmetry. But the cusps will definitely not be line cusps, and all the cusps will be similar (whereas the current truncube design has corner cusps and face cusps that are definitely different animals). Using only 'triangle' cusps should mean that repeller plates + rings would be smaller than ones meant for a line cusp, as well.

[Art Carlson]

Solo, I think you're forgetting that the 6 lines from the center to the vertices of the octohedron will have zero field along their entire length. They will confine the plasma much worse than any point cusp, if at all. That is why Bussard and Nebel insist that the circular coils in their cubic machines must not touch each other.

[drmike]

Solo - I'm not done reading the whole thing, but I think we can generalize your statement. Create independent magnetic and electric field shapes and hunt for some optimum.

But now, let's generalize further. Suppose that instead of static fields we use dynamic ones, like a motor has a dynamic magnetic field. So every where the plasma tries to leak out the field moves just enough to reduce the bulk of it.

Pulsed, static fields are all that have ever been done. Maybe we need to get way outside the box.

[Art Carlson]

Pulsed, static fields are all that have ever been done. Maybe we need to get way outside the box.

• There was some work on FRCs in Australia to drive current by rotating a magnetic field at a frequency that allowed it to drag the electrons better than the ions.
• Tokamaks often use rf waves to drive current (as well as for heating).
• I believe schemes have been tried to plug the ends of mirror machines using rf power.

rf fields are not usually a very efficient way to transfer momentum (Those photons just aren't very heavy!), and that is what you need in order to plug a hole.

[hanelyp]

Regarding magrid coils conformal to the magnetic field: The magrid doesn't have to be conformal, but an electron reaching any field line intercepting the coils is quickly lost to the coil. Thus you want to intercept the minimum of field lines given your coil cross section, which ends up a conformal shape.

Also, coils conformal to the field don't have corners which concentrate electric fields and invite discharge.

Shifting topic, for a thermal (D-D) reactor you might consider coils that cover most of the sphere, with lots of cross section to produce a large magnetic field, and just enough of a hole in the center and gap to support recirculation.

[Solo]

the 6 lines from the center to the vertices of the octohedron will have zero field along their entire length

Rats. So much for that idea. If you look at those lines, you have two coil segments coming in with currents toward the corner, two away, so by symmetry the field on the axis cancels. I'm annoyed that I can't visualize it, though: it seems like the fields reinforce as you bring the coil corners closer, so why should the field go to zero suddenly when the coils merge? Oh well.

Hanyley: Hmm. That's Bussard's line, and it made sense. I've just been wondering about it lately, though, since electrostatically plugged cusp machines intentionally intercept field lines. The rationale is that it keeps the width of the electron flow in the cusp down, and hence the space charge (no ions in cusp, basically). The space charge would reduce the potential in the cusp, so that the ions would not have to climb all the way to coil potential to escape the well, but only to some lower value.

The plugged cusp machines have cusp electron flows limited to 1cm or so wide, yet they expect 100kV or more difference between the limiters and the center of the cusp! Think about a magrid coil 30 cm around. Something must be limiting the density of un-neutralized electrons, or else there's almost no potential well! It just makes me wonder if Bussard's claim of 0.8-0.9 for well depth/drive voltage is misleading. He might not have measured the potential in the cusp throat, just assuming it was about as high as the coils.

Or, perhaps the high-beta/whiffleball effect changes the field lines so that the plasma sheath naturally comes to a small diameter in the cusp throats. Beats me.

[rcain]

what happens if yu make the whole magrid structure oblate - like a like a rubgy(-foot) ball?

ragarding it simply as a cavity surface - then use MHD flow? use the cusps to extend the well?

[icarus]

A rugby foot ball is prolate (spheroid).

A discus is oblate (spheroid).

http://en.wikipedia.org/wiki/Spheroid

An interesting idea though ... and then rotate the fields at some enormous rate (about the geometric axis) to "eliminate" four of the point cusps and four of the line/point cusps?

[tonybarry]

I thought of the spinning idea, but the problem is that mass inertia in the plasma wiffleball interior is a very small force when compared with the electrostatic forces. Consequently the rotational velocity would need to be huge to give the smoothing effect, and any irregularity in the magnetic field would induce enormous currents in the plasma.

Regards,
Tony Barry

[tombo]

Intriguing.

But, hoo-boy if rnebel considers plasma sheaths to be "complicated" ...

Those currents of course are how you get a handle on the rotation rate.

Lets see, it is spinning like a dynamo and pulsing with the POPS glowing with brehmstralung (sp?) and heat and has the plasma cutting through the field lines.
It looks like a movie special effect.

Although, since the cusps are small the rotation rate might not need to be all that high to disrupt flow out of them.