thread for segments files and parameters for simulation runs

[happyjack27]

How do you model the Faraday cage that is encompassing the whole region?Is it grounded or held at set voltage?

Do the ions that make it tho the cage get removed from the sim.? Do the electrons that make it to the MaGrid get removed?

after a certain radius (currently 3x the magrid radius) they're considered "lost" and recycled through the egun(s). so i imagine that would count as a grounded (spherical) faraday cage.

Can the rate at which those particles are removed be measured? (This would be a measure of loss rate)

not yet. but i know that's an important thing to measure. i want to do graph their kinetic energy, for magrid losses and chamber losses separately.

[happyjack27]

here's the iterated octahedron magrid again, with all the new stuff i've added to the sim since i first ran an electron-only sim on it.

parameters are:

3m radius
-10E-7.736 coloumbs net plasma space charge
10E-4.783 coloumbs total plasma space charge
10E7.236 amp turns

http://www.youtube.com/watch?v=JmBWuGnQnmQ

[happyjack27]

important thing to note in these more recent sims:

the electron timescales have been slowed substantially compared to the ion time scales, in order to get a good resolution on their trajectory. that means, however, that they will not react nearly as quickly to electrostatic divergences introduced by the changing ion positions. e.g. they will not "screen" charges nearly as well in the short-term. all-in-all, you'll probably see a little more ion thermalization in the sim than if the electron timescale wasn't slowed down relative to the ion timescale. however, without modifying the timescale ratio, i'd have to slow down the entire simulation substantially. so this is sort of a compromise.

[rjaypeters]

Deformed coils:

Code: Select all

//Deformed coils, 15 deg. central angle
//-y coil deformed begins
32
0.0, -1.351, 1.032, 0.0, -0.203, -1.343, 1.022
-0.203, -1.343, 1.022, -0.409, -1.322, 0.988
-0.409, -1.322, 0.988, -0.614, -1.292, 0.919
-0.614, -1.292, 0.919, -0.797, -1.273, 0.797
-0.797, -1.273, 0.797, -0.937, -1.273, 0.626
-0.937, -1.273, 0.626, -1.041, -1.273, 0.431
-1.041, -1.273, 0.431, -1.105, -1.273, 0.220
-1.105, -1.273, 0.220, -1.127, -1.273, 0.0
-1.127, -1.273, 0.0, -1.105, -1.273, -0.220
-1.105, -1.273, -0.220, -1.041, -1.273, -0.431
-1.041, -1.273, -0.431, -0.937, -1.273, -0.626
-0.937, -1.273, -0.626, -0.797, -1.273, -0.797
-0.797, -1.273, -0.797, -0.626, -1.273, -0.937
-0.626, -1.273, -0.937, -0.431, -1.273, -1.041
-0.431, -1.273, -1.041, -0.220, -1.273, -1.105
-0.220, -1.273, -1.105, 0.0, -1.273, -1.127
0.0, -1.273, -1.127, 0.220, -1.273, -1.105
0.220, -1.273, -1.105, 0.431, -1.273, -1.041
0.431, -1.273, -1.041, 0.626, -1.273, -0.937
0.626, -1.273, -0.937, 0.797, -1.273, -0.797
0.797, -1.273, -0.797, 0.937, -1.273, -0.626
0.937, -1.273, -0.626, 1.041, -1.273, -0.431
1.041, -1.273, -0.431, 1.105, -1.273, -0.220
1.105, -1.273, -0.220, 1.127, -1.273, 0.0
1.127, -1.273, 0.0, 1.105, -1.273, 0.220
1.105, -1.273, 0.220, 1.041, -1.273, 0.431
1.041, -1.273, 0.431, 0.937, -1.273, 0.626
0.937, -1.273, 0.626, 0.797, -1.273, 0.797
0.797, -1.273, 0.797, 0.614, -1.292, 0.919
0.614, -1.292, 0.919, 0.409, -1.322, 0.988
0.409, -1.322, 0.988, 0.208, -1.343, 1.022
0.208, -1.343, 1.022, 0.0, -1.351, 1.032
//-y coil deformed ends
//+z coil deformed begins
32
0.0, -1.032. 1.351, 0.203, -1.022, 1.343
0.203, -1.022, 1.343, 0.409, -0.988, 1.322
0.409, -0.988, 1.322, 0.614, -0.919, 1.292
0.614, -0.919, 1.292, 0.797, -0.797, 1.273
0.797, -0.797, 1.273, 0.937, -0.626, 1.273
0.937, -0.626, 1.273, 1.041, -0.431, 1.273
1.041, -0.431, 1.273, 1.105, -0.220, 1.273
1.105, -0.220, 1.273, 1.127, 0.0, 1.273
1.127, 0.0, 1.273, 1.105, 0.220, 1.273
1.105, 0.220, 1.273, 1.041, 0.431, 1.273
1.041, 0.431, 1.273, 0.937, 0.626, 1.273
0.937, 0.626, 1.273, 0.797, 0.797, 1.273
0.797, 0.797, 1.273, 0.626, 0.937, 1.273
0.626, 0.937, 1.273, 0.431, 1.041, 1.273
0.431, 1.041, 1.273, 0.220, 1.105, 1.273
0.220, 1.105, 1.273, 0.0, 1.127, 1.273
0.0, 1.127, 1.273, -0.220, 1.105, 1.273
-0.220, 1.105, 1.273, -0.431, 1.041, 1.273
-0.431, 1.041, 1.273, -0.626, 0.937, 1.273
-0.626, 0.937, 1.273, -0.797, 0.797, 1.273
-0.797, 0.797, 1.273, -0.937, 0.626, 1.273
-0.937, 0.626, 1.273, -1.041, 0.431, 1.273
-1.041, 0.431, 1.273, -1.105, 0.220, 1.273
-1.105, 0.220, 1.273, -1.127, 0.0, 1.273
-1.127, 0.0, 1.273, -1.105, -0.220, 1.273
-1.105, -0.220, 1.273, -1.041, -0.431, 1.273
-1.041, -0.431, 1.273, -0.937, -0.626, 1.273
-0.937, -0.626, 1.273, -0.797, -0.797, 1.273
-0.797, -0.797, 1.273, -0.614, -0.919, 1.292
-0.614, -0.919, 1.292, -0.409, -0.988, 1.322
-0.409, -0.988, 1.322, -0.203, -1.022, 1.343
-0.203, -1.022, 1.343, 0.0, -1.032, 1.351
//+z coil deformed ends

happyjack 27,
Can you see the coils in your simulation before things go to far? I. e. can you see errors in the coils before you start the run?

Everyone,
Open the pool for whether the above deformations make a difference.

[happyjack27]

happyjack 27,
Can you see the coils in your simulation before things go to far? I. e. can you see errors in the coils before you start the run?

Everyone,
Open the pool for whether the above deformations make a difference.

yea, the coils show up right away. i'll let you know if anything's off.

i'm betting it won't make a difference, as far as the eye can see, at least. but to really get an accurate in depth measure i need to add "diagnostics" to the sim, which i'm working on. then we can see exact measures of e.g. particle losses and well depth. but i'd wager that we wouldn't see a substantial difference there, either.

i think i got all the code for phase space visualization up. i suppose i shall see tonight. either way i'll check how the coils look and run it if they're right, and tell you if they're not.

[rjaypeters]

Excellent.

[rjaypeters]

Inverse WB-6, although it is probably too literal and LOSSY:

I expect we really want this octahedron:


While creating the octahedron, I noticed the radius for all of my six-coil designs was 0.1m too big (i.e. 3.2m "empty" volume instead of 3.0m).

Do I need to fix the deformed coil code in my previous post above?

I have modeled the octahedron at 3.0m empty volume (but not pointilized), do you want 3.0m empty volume or 3.2 or does it matter?

[happyjack27]

I expect we really want this octahedron:

yep.

it has shorter line cusps and more regular faces. so i suspect it will confine better.

While creating the octahedron, I noticed the radius for all of my six-coil designs was 0.1m too big (i.e. 3.2m "empty" volume instead of 3.0m).

Do I need to fix the deformed coil code in my previous post above?

I have modeled the octahedron at 3.0m empty volume (but not pointilized), do you want 3.0m empty volume or 3.2 or does it matter?

i can scale the whole magrid arbitrarily with a single number. so having things in terms of "1" probably makes scaling the easiest. i've been doing coil radius = 1, but i imagine the radius of a sphere that touches all the coils = 1 would probably be better and more portable between geometries. that would simply be sqrt(coil radius^2 + distance from center of coil midplane to absolute center^2) = 1. that would probably be ideal. as that would makes scaling simplest. but i can scale anyways. so just tell me what scaling factor you want. i've been running 3m "circumscribed" sphere radius as i understand that's theoretically just past break even for pb11. so as is should be just rigaht with a scaling factor of 1. but if it's easy for you to scale everything might as well make it 1m empty volume. it's up to you, really.

[happyjack27]

eh, looks like my code changes didn't work - rendering's broken. i just love opengl errors: "GLError: invalid operation". thanks, you think you can give me any less information? dumb it down for me maybe? it's kind of hard to sort out what's relevant and what's not in all that. anyways, though i have no idea why that didn't work, i have a different idea which should work. but in any case i can't sim until then, which will be tommorow. maybe longer. if it's anything like i envision it'll be worth the wait.

[93143]

i just love opengl errors: "GLError: invalid operation". thanks, you think you can give me any less information? dumb it down for me maybe?

Heh... I tried an all-up test of my multiphase code today, and it instantly gave me one line of output and quit:

Segmentation fault

It's going to be a fun evening...

[rjaypeters]

...i've been running 3m "circumscribed" sphere radius as i understand that's theoretically just past break even for pb11. so as is should be just rigaht with a scaling factor of 1. but if it's easy for you to scale everything might as well make it 1m empty volume. it's up to you, really.

Because it's an integer, I'd rather circumscribe on a 3m diameter sphere (leaving an empty volume of 2.8m). Further, I think a 1m empty volume won't serve our purposes as well (see your comment on break-even for pB11).

I'll get to work. Thank you for your understanding. I'll fix the bent rings first.

[happyjack27]

you mean radius, i presume. my point is i have a place in the config file where i just enter a number, say "3", and a 1m polywell becomes a 3m polywell. i have that number set to 3 and it would be easier if i never had to change it when switching between configurations, so it would be easier if all configurations where unit polywells.

[rjaypeters]

happyjack27,

For analysis purposes, I will happily use a 3m radius.

In my previous work I wrote and meant 3m diameter. From Magrid configuration brainstorming:

The coil thickness information is useful. Not to put too fine a point on it, I asked two other questions:
Should I been using a 1 meter diameter, 1 meter radius or something else for the coil centers?
What spacing should the coils have?...

...MSimon did a pB&J design with ~1.5m major radius and the SC core was 29cm square, IIRC...

I have been trying for realism in my work. That attempt included a 20 cm coil thickness and, because I misread a D Tibbets post, 5 - 10mm coil spacing. I'm going to keep the coil thickness and 5-10mm spacing, mostly because it looks better and you are of the opinion the closer spacing is not important.

We may learn coil spacing is important from simulation and reality.

[happyjack27]

k. i just want to be able to easily compare everything when i have the diagnostics up and running. you know, apples to apples.

i won't have the crucial power loss and power gain statistics yet. (i've got to get density before that, for the power gain formula.) those would be the penultimate tests. still a ways off.

what's this about pb&j? i prefer ham, myself.

[rjaypeters]

happyjack27,

I'm all for the easy comparison of designs, it will be critical once statistics become part of the output.

Surely* you know the derivation of the pb&j shorthand?

Best,

Robert

*In honor of Leslie Nielson, I won't call you Shirley again.