thread for segments files and parameters for simulation runs

Discuss how polywell fusion works; share theoretical questions and answers.

Moderators: tonybarry, MSimon

rjaypeters
Posts: 869
Joined: Fri Aug 20, 2010 2:04 pm
Location: Summerville SC, USA

Post by rjaypeters »

KitmanSA, Like these?

No gap:
Image

Gapped:
Image

Notes: Casing thickness 0.15m, midpoint of straight segments circumscribed on a 1m radius sphere. Zero radius turns.

If you can convince someone to run the simulation, I'll complete one or both and provide the segment list(s).

P.S. I'll even fix the gaps so they are even.
"Aqaba! By Land!" T. E. Lawrence

R. Peters

KitemanSA
Posts: 6179
Joined: Sun Sep 28, 2008 3:05 pm
Location: OlyPen WA

Post by KitemanSA »

Gapped, thank you!.

rjaypeters
Posts: 869
Joined: Fri Aug 20, 2010 2:04 pm
Location: Summerville SC, USA

Post by rjaypeters »

rjaypeters wrote:If you can convince someone to run the simulation, I'll complete one or both and provide the segment list(s).
Aw, who am I kidding? I'll complete the gapped and probably no-gapped just for something to do.
"Aqaba! By Land!" T. E. Lawrence

R. Peters

D Tibbets
Posts: 2775
Joined: Thu Jun 26, 2008 6:52 am

Post by D Tibbets »

happyjack27 wrote:tibbets 4-coil position space:

http://www.youtube.com/watch?v=JY9Lb6Bx6T0

i ran it at 1E5.292 amp turns, 1E-9.035 net charge. this is the standard amp turns i use for 15cm radius, but lower plamsa density. it corresponds to about 7 tesla on a standard wb-6 config. later in the sim i try turning the mag field strength up a bit and the plasma density down about an order of magnitude to get better confinement. doesn't make a whole lot of difference.

EDIT: eee.. i forgot that compression really lowers the resolution so i have to up the point size. uploading a new one now, which should be clearer. also shows it from more different angles.

The simulations are difficult to interpret as the views and parameters may not match. But still the views are perhaps interesting. The 4 grid bowed design seems to show more losses through the face centered point cusps. I am surprised that the two end 'corner cusps do not seem to leak as much. The 'funny cusps between the coils do not show much leakage, though this may be a matter of perspective. In any one area there is not much visible leakage, but it may be spread out over the entire length of the 'funny' cusp.

From a simple extrapolation I would expect the four center point cusps to each leak ~ 2X the leakage if the coils were circular like in WB6. How this compares to this model visualization is uncertain. One consideration is that there are 4 of these cusps, verses the 6 in WB6. So while the per cusp leakage is greater, the total leakage through these cusps is closer. Also, there are only two corner cusps, instead of the 8 in WB6, so the total leakage needs to be added from all of the cusps, and how that would compare is not obvious from these visualizations.

Another point is the quasi-sphericity of the Wiffleball border. Bussard claimed that increasing this might improve performance by a factor of ~3-5X. The visualizations seem to show some spikes towards the face centered point cusps, but little deformation otherwise. Compare this to a visualization of WB6.

There are three considerations:

1) primary magnetic confinement (electrons) that determines the Wiffleball Trapping Factor, and thus the obtainable density.

2) Recirculation- which determines the dominating electron input power needed.

3) Sphericity- which effects fusion efficiency (convergence)

The bowed design seems to improve #3. Uncertain how it effects #1 and 2.



A picture to illustrate my shaky speculations. Derived from U-tube videos from happyjack27:

Image

Dan Tibbets
To error is human... and I'm very human.

D Tibbets
Posts: 2775
Joined: Thu Jun 26, 2008 6:52 am

Post by D Tibbets »

One modification to the 4 bowed grid design would be to shorten and open the curvature of the grids until they were ~ 1.5 times as long (along the curve, where they are currently twice as long) as they are wide. This may result in end 'corner' cusps having ~ the same minor diameter of the face point cusps. This would exchange leakage sites somewhat, but an optimum balance may be achieved while maintaining or improving the overall quasi- sphericity of the Wiffleball border.

Or not :?

Dan Tibbets
To error is human... and I'm very human.

D Tibbets
Posts: 2775
Joined: Thu Jun 26, 2008 6:52 am

Post by D Tibbets »

Another concideration for these sims. I assume that the grids are lines in these sims rather than real electromagnets with finite crossections. WB6 electromagnet cans minor diameter (thickness) were ~ 17% of the overall diameter. This would effectively place the center/ strongest portion of the field in the center of the can crossection. The distance to the edge/ corner/ virtual point cusps would therefor be effectively greater while the distance to the face centered point cusps would be less. Because of the inverse square law the effect may be more than supposed. Another way to look at it is to use the distance from the surface of the can to the cusp to determine the B field strength in the cusp, rather than a line representing the magnet. An example: 1 meter magrid radius, 20 cm minor radius. 1 Tesla B-field strength at the can surface. Assuming the magnets are separated by 10 mm. The distance to the center of edge cusps would be 5 mm. The distance to the center face cusps would be 50 cm minus the can thickness (20cm) or 30 cm. 1 (B field at can surface)/ 30cm^2 is 1/900 while if the 50 cm distance is used, the number would be 1/2500. If my reasoning is intact, this correction would imply that the face center point cusp B fields would actually be ~ 3 times as strong in proportion to the edge cusp B field strengths. And as the cusp leakage is, I think, logrithmically related to the B field strength, the relative leakages of the cusps would be even more offset.

This is one reason why I have wondered if making the cans more oval in crossection would be more beneficial. The inner edge of the wire windings loaded can edge would be closer to the face center cusp, allowing the relative strength compared to the edge cusps to be closer. Two possible problems with this is that there is more magrid surface area exposed to x-rays or neutrons, and there is more surface area exposed to ExB drift/ transport of electrons to the magrid surfaces. The first problem may be mitigated by the proportionate increase in internal can area available for cooling plumbing. In the second case, so long as recirculation was not so effective that it approached ExB confinement limits, you would still be ahead. Bussard, etel* gave loss ratios for cross field transport electron losses to cusp losses (with recirculation) of ~ 1/100 to 1/10. That leaves a lot of wiggle room unless recirculation can be significantly improved.

Also, if the thicker magrid can reduces primary face point cusp losses, without penalizing edge cusp losses, that should translate into a higher Wiffleball Trapping Factor (up to the ExB drift limit) - which increase the density inside the machine, and thus the fusion rate squared.

* Or perhaps I should refer to EMC^2, etel as I don't know if Bussard was involved in the 2008 patent application writeup where this is stated.

Dan Tibbets
To error is human... and I'm very human.

rjaypeters
Posts: 869
Joined: Fri Aug 20, 2010 2:04 pm
Location: Summerville SC, USA

Post by rjaypeters »

Truncated icosahedron with straight segments:

Image Partial image shown because the almost PERFECTion of this concept is too brilliant for our eyes. :)

Note: 0.15m casing diameter (0.2m is too big IMO for so many coils) with 6mm gap.

Code: Select all

//Truncated Icosahedron straight hexagons and pentagons, 1m central radius
//Large - 20 coils x 6 segments, small - 12 coils x 5 segments
//Large coil circumferences, ideal - 1.891m, segmented = 1.716
//Small coil circumferences, ideal - 1.307m, segmented = 1.235m
//large +x, +z coil begins - #1
6
0.613, 0.0, 0.803, 0.479, 0.248, 0.854
0.479, 0.248, 0.854, 0.211, 0.248, 0.956
0.211, 0.248, 0.956, 0.078, 0.0, 1.007
0.078, 0.0, 1.007, 0.212, -0.248, 0.956
0.212, -0.248, 0.956, 0.479, -0.248, 0.854
0.479, -0.248, 0.854, 0.613, 0.0, 0.803
//large +x, +z coil ends - #1
//large +x, -z coil begins - #2
6
0.613, 0.0, -0.803, 0.479, -0.248, -0.854
0.479, -0.248, -0.854, 0.211, -0.248, -0.956
0.211, -0.248, -0.956, 0.078, 0.0, -1.007
0.078, 0.0, -1.007, 0.212, 0.248, -0.956
0.212, 0.248, -0.956, 0.479, 0.248, -0.854
0.479, 0.248, -0.854, 0.613, 0.0, -0.803
//large +x,-z coil ends - #2
//large -x, +z coil begins - #3
6
-0.079, 0.0, 1.007, -0.213, 0.248, 0.956
-0.213, 0.248, 0.956, -0.480, 0.248, 0.854
-0.480, 0.248, 0.854, -0.614, 0.0, 0.803
-0.614, 0.0, 0.803, -0.480, -0.248, 0.854
-0.480, -0.248, 0.854, -0.212, -0.248, 0.956
-0.212, -0.248, 0.956, -0.079, 0.0, 1.007
//large -x, +z coil ends - #3
//large -x, -z coil begins - #4
6
-0.079, 0.0, -1.007, -0.213, -0.248, -0.956
-0.213, -0.248, -0.956, -0.480, -0.248, -0.854
-0.480, -0.248, -0.854, -0.613, 0.0, -0.803
-0.613, 0.0, -0.803, -0.479, 0.248, -0.854
-0.479, 0.248, -0.854, -0.212, 0.248, -0.956
-0.212, 0.248, -0.956, -0.079, 0.0, -1.007
//large -x, -z coil ends - #4
//large +x, +y coil begins - #5
6
1.007, 0.079, 0.0, 0.956, 0.213, -0.248
0.956, 0.213, -0.248, 0.854, 0.48, -0.248
0.854, 0.48, -0.248, 0.803, 0.613, 0.0
0.803, 0.613, 0.0, 0.854, 0.48, 0.248
0.854, 0.48, 0.248, 0.956, 0.212, 0.248
0.956, 0.212, 0.248, 1.007, 0.079, 0.0
//large +x, +y coil ends - #5
//large +x, -y coil begins - #6
6
1.007, -0.078, 0.0, 0.956, -0.212, 0.248
0.956, -0.212, 0.248, 0.854, -0.479, 0.248
0.854, -0.479, 0.248, 0.803, -0.613, 0.0
0.803, -0.613, 0.0, 0.854, -0.479, -0.248
0.854, -0.479, -0.248, 0.956, -0.211, -0.248
0.956, -0.211, -0.248, 1.007, -0.078, 0.0
//large +x, -y coil ends - #6
//large -x, +y coil begins - #7
6
-1.007, 0.078, 0.0, -0.956, 0.212, 0.248
-0.956, 0.212, 0.248, -0.854, 0.479, 0.248
-0.854, 0.479, 0.248, -0.803, 0.613, 0.0
-0.803, 0.613, 0.0, -0.854, 0.479, -0.248
-0.854, 0.479, -0.248, -0.956, 0.211, -0.248
-0.956, 0.211, -0.248, -1.007, 0.078, 0.0
//large -x, +y coil ends - #7
//large -x, -y coil begins - #8
6
-1.007, -0.078, 0.0, -0.956, -0.212, -0.248
-0.956, -0.212, -0.248, -0.854, -0.48, -0.248
-0.854, -0.48, -0.248, -0.803, -0.613, 0.0
-0.803, -0.613, 0.0, -0.854, -0.479, 0.248
-0.854, -0.479, 0.248, -0.956, -0.212, 0.248
-0.956, -0.212, 0.248, -1.007, -0.078, 0.0
//large -x, -y coil ends - #8
//large +y, +z coil begins - #9
6
0.0, 0.803, 0.613, 0.248, 0.854, 0.479
0.248, 0.854, 0.479, 0.248, 0.956, 0.212
0.248, 0.956, 0.212, 0.0, 1.007, 0.078
0.0, 1.007, 0.078, -0.248, 0.956, 0.212
-0.248, 0.956, 0.212, -0.248, 0.854, 0.479
-0.248, 0.854, 0.479, 0.0, 0.803, 0.613
//large +y, +z coil ends - #9
//large +y, -z coil begins - #10
6
0.0, 1.007, -0.078, 0.248, 0.956, -0.212
0.248, 0.956, -0.212, 0.248, 0.854, -0.48
0.248, 0.854, -0.48, 0.0, 0.803, -0.613
0.0, 0.803, -0.613, -0.248, 0.854, -0.479
-0.248, 0.854, -0.479, -0.248, 0.956, -0.212
-0.248, 0.956, -0.212, 0.0, 1.007, -0.078
//large +y, -z coil ends - #10
//large -y, +z coil begins - #11
6
0.0, -1.007, 0.078, 0.248, -0.956, 0.212
0.248, -0.956, 0.212, 0.247, -0.854, 0.48
0.247, -0.854, 0.48, 0.0, -0.803, 0.613
0.0, -0.803, 0.613, -0.248, -0.854, 0.479
-0.248, -0.854, 0.479, -0.248, -0.956, 0.212
-0.248, -0.956, 0.212, 0.0, -1.007, 0.078
//large -y, +z coil ends - #11
//large -y, -z coil begins - #12
6
0.0, -1.007, -0.078, -0.248, -0.956, -0.212
-0.248, -0.956, -0.212, -0.248, -0.854, -0.480
-0.248, -0.854, -0.480, 0.0, -0.803, -0.613
0.0, -0.803, -0.613, 0.248, -0.854, -0.479 
0.248, -0.854, -0.479, 0.248, -0.956, -0.212
0.248, -0.956, -0.212, 0.0, -1.007, -0.078
//large -y, -z coil ends - #12
//large +x, +y, +z coil begins - #13
6
0.775, 0.528, 0.374, 0.59, 0.745, 0.343
0.59, 0.745, 0.343, 0.374, 0.776, 0.528
0.374, 0.776, 0.528, 0.343, 0.591, 0.744
0.343, 0.591, 0.744, 0.528, 0.375, 0.776
0.528, 0.375, 0.776, 0.744, 0.343, 0.591
0.744, 0.343, 0.591, 0.775, 0.528, 0.374
//large +x, +y, +z coil ends - #13
//large +x, -y, +z coil begins - #14
6
0.591, -0.744, 0.343, 0.776, -0.528, 0.375
0.776, -0.528, 0.375, 0.744, -0.343, 0.591
0.744, -0.343, 0.591, 0.527, -0.375, 0.776
0.527, -0.375, 0.776, 0.342, -0.591, 0.744
0.342, -0.591, 0.744, 0.374, -0.776, 0.528
0.374, -0.776, 0.528, 0.591, -0.744, 0.343
//large +x, -y, +z coil ends - #14
//large -x, +y, +z coil begins - #15
6
-0.744, 0.343, 0.591, -0.528, 0.375, 0.776
-0.528, 0.375, 0.776, -0.343, 0.591, 0.744
-0.343, 0.591, 0.744, -0.375, 0.776, 0.527
-0.375, 0.776, 0.527, -0.591, 0.744, 0.343
-0.591, 0.744, 0.343, -0.776, 0.527, 0.374
-0.776, 0.527, 0.374, -0.744, 0.343, 0.591
//large -x, +y, +z coil ends - #15
//large -x, -y, +z coil begins - #16
6
-0.744, -0.343, 0.591, -0.776, -0.528, 0.374
-0.776, -0.528, 0.374, -0.591, -0.744, 0.343
-0.591, -0.744, 0.343, -0.375, -0.775, 0.528
-0.375, -0.775, 0.528, -0.343, -0.59, 0.744
-0.343, -0.59, 0.744, -0.528, -0.374, 0.776
-0.528, -0.374, 0.776, -0.744, -0.343, 0.591
//large -x, -y, +z coil ends - #16
//large +x, +y, -z coil begins - #17
6
0.591, 0.744, -0.343, 0.776, 0.528, -0.375
0.776, 0.528, -0.375, 0.744, 0.343, -0.591
0.744, 0.343, -0.591, 0.527, 0.375, -0.776
0.527, 0.375, -0.776, 0.343, 0.591, -0.744
0.343, 0.591, -0.744, 0.374, 0.776, -0.528
0.374, 0.776, -0.528, 0.591, 0.744, -0.343
//large +x, +y, -z coil ends - #17
//large +x, -y, -z coil begins - #18
6
0.591, -0.744, -0.343, 0.375, -0.775, -0.528
0.375, -0.775, -0.528, 0.343, -0.59, -0.744 
0.343, -0.59, -0.744, 0.528, -0.374, -0.776
0.528, -0.374, -0.776, 0.744, -0.343, -0.591
0.744, -0.343, -0.591, 0.776, -0.528, -0.374
0.776, -0.528, -0.374, 0.591, -0.744, -0.343
//large +x, -y, -z coil ends - #18
//large -x, +y, -z coil begins - #19
6
-0.591, 0.744, -0.343, -0.375, 0.775, -0.528
-0.375, 0.775, -0.528, -0.343, 0.59, -0.744
-0.343, 0.59, -0.744, -0.528, 0.374, -0.776
-0.528, 0.374, -0.776, -0.744, 0.343, -0.591
-0.744, 0.343, -0.591, -0.776, 0.528, -0.374
-0.776, 0.528, -0.374, -0.591, 0.744, -0.343
//large -x, +y, -z coil ends - #19
//large -x, -y, -z coil begins - #20
6
-0.591, -0.744, -0.343, -0.776, -0.528, -0.375
-0.776, -0.528, -0.375, -0.744, -0.343, -0.591
-0.744, -0.343, -0.591, -0.527, -0.375, -0.776
-0.527, -0.375, -0.776, -0.342, -0.591, -0.744
-0.342, -0.591, -0.744, -0.374, -0.776, -0.528
-0.374, -0.776, -0.528, -0.591, -0.744, -0.343
//large -x, -y, -z coil ends - #20
//small +x, +y coil begins -#21
5
0.573, 0.804, 0.2, 0.697, 0.728, 0.0
0.697, 0.728, 0.0, 0.574, 0.804, -0.2
0.574, 0.804, -0.2, 0.374, 0.928, -0.124
0.374, 0.928, -0.124, 0.373, 0.928, 0.123
0.373, 0.928, 0.123, 0.573, 0.804, 0.2
//small +x, +y coil ends -#21
//small +x, -y coil begins -#22
5
0.697, -0.728, 0.0, 0.574, -0.804, 0.2
0.574, -0.804, 0.2, 0.374, -0.928, 0.124
0.374, -0.928, 0.124, 0.373, -0.928, -0.123
0.373, -0.928, -0.123, 0.573, -0.804, -0.2
0.573, -0.804, -0.2, 0.697, -0.728, 0.0
//small +x, -y coil ends -#22
//small -x, +y coil begins -#23
5
-0.697, 0.728, 0.0, -0.574, 0.804, 0.2
-0.574, 0.804, 0.2, -0.374, 0.928, 0.124
-0.374, 0.928, 0.124, -0.373, 0.928, -0.123
-0.373, 0.928, -0.123, -0.573, 0.804, -0.2
-0.573, 0.804, -0.2, -0.697, 0.728, 0.0
//small -x, +y coil ends -#23
//small -x, -y coil begins -#24
5
-0.697, -0.728, 0.0, -0.574, -0.804, -0.2
-0.574, -0.804, -0.2, -0.374, -0.928, -0.124
-0.374, -0.928, -0.124, -0.373, -0.928, 0.123
-0.373, -0.928, 0.123, -0.573, -0.804, 0.2
-0.573, -0.804, 0.2, -0.697, -0.728, 0.0
//small -x, -y coil ends -#24
//small +y, +z coil begins -#25
5
0.0, 0.697, 0.728, -0.2, 0.574, 0.804
-0.2, 0.574, 0.804, -0.124, 0.374, 0.928
-0.124, 0.374, 0.928, 0.123, 0.373, 0.928
0.123, 0.373, 0.928, 0.2, 0.573, 0.804
0.2, 0.573, 0.804, 0.0, 0.697, 0.728
//small +y, +z coil ends -#25
//small +y, -z coil begins -#26
5
0.0, 0.697, -0.728, 0.2, 0.574, -0.804
0.2, 0.574, -0.804, 0.124, 0.374, -0.928
0.124, 0.374, -0.928, -0.123, 0.373, -0.928
-0.123, 0.373, -0.928, -0.2, 0.573, -0.804
-0.2, 0.573, -0.804, 0.0, 0.697, -0.728
//small +y, -z coil ends -#26
//small -y, +z coil begins -#27
5
0.0, -0.697, 0.728, 0.2, -0.574, 0.804
0.2, -0.574, 0.804, 0.124, -0.374, 0.928
0.124, -0.374, 0.928, -0.123, -0.373, 0.928
-0.123, -0.373, 0.928, -0.2, -0.573, 0.804
-0.2, -0.573, 0.804, 0.0, -0.697, 0.728
//small -y, +z coil ends -#27
//small -y, -z coil begins -#28
5
0.0, -0.697, -0.728, -0.2, -0.574, -0.804
-0.2, -0.574, -0.804, -0.124, -0.374, -0.928
-0.124, -0.374, -0.928, 0.123, -0.373, -0.928
0.123, -0.373, -0.928, 0.2, -0.573, -0.804
0.2, -0.573, -0.804, 0.0, -0.697, -0.728
//small -y, -z coil ends -#28
//small +x, +z coil begins -#29
5
0.728, 0.0, 0.697, 0.804, -0.2, 0.574
0.804, -0.2, 0.574, 0.928, -0.124, 0.374
0.928, -0.124, 0.374, 0.928, 0.123, 0.373
0.928, 0.123, 0.373, 0.804, 0.2, 0.573
0.804, 0.2, 0.573, 0.728, 0.0, 0.697
//small +x, +z coil ends -#29
//small +x, -z coil begins -#30
5
0.728, 0.0, -0.697, 0.804, 0.2, -0.574
0.804, 0.2, -0.574, 0.928, 0.124, -0.374
0.928, 0.124, -0.374, 0.928, -0.123, -0.373
0.928, -0.123, -0.373, 0.804, -0.2, -0.573
0.804, -0.2, -0.573, 0.728, 0.0, -0.697
//small +x, -z coil ends -#30
//small -x, +z coil begins -#31
5
-0.728, 0.0, 0.697, -0.804, 0.2, 0.574
-0.804, 0.2, 0.574, -0.928, 0.124, 0.374
-0.928, 0.124, 0.374, -0.928, -0.123, 0.373
-0.928, -0.123, 0.373, -0.804, -0.2, 0.573
-0.804, -0.2, 0.573, -0.728, 0.0, 0.697
//small -x, +z coil ends -#31
//small -x, -z coil begins -#32
5
-0.728, 0.0, -0.697, -0.804, -0.2, -0.574
-0.804, -0.2, -0.574, -0.928, -0.124, -0.374
-0.928, -0.124, -0.374, -0.928, 0.123, -0.373
-0.928, 0.123, -0.373, -0.804, 0.2, -0.573
-0.804, 0.2, -0.573, -0.728, 0.0, -0.697
//small -x, -z coil ends -#32
To my simple mind, this configuration should be pretty close to the two long posts (return to Coruscant),take up a lot less memory and probably run faster, too!
"Aqaba! By Land!" T. E. Lawrence

R. Peters

KitemanSA
Posts: 6179
Joined: Sun Sep 28, 2008 3:05 pm
Location: OlyPen WA

Post by KitemanSA »

rjaypeters wrote:Truncated icosahedron with straight segments:
....BEAUTIFUL!!!
Now THAT is "less quasi- more spherical"!

And just to be sure, since I haven't learned to read the code, ALL the coils are "North In" (or out), right?

KitemanSA
Posts: 6179
Joined: Sun Sep 28, 2008 3:05 pm
Location: OlyPen WA

Post by KitemanSA »

D Tibbets wrote: ... This would effectively place the center/ strongest portion of the field in the center of the can crossection.
Shouldn't, should it? Doesn't Guass' law apply here too?

happyjack27
Posts: 1439
Joined: Wed Jul 14, 2010 5:27 pm

Post by happyjack27 »

rjaypeters wrote: To my simple mind, this configuration should be pretty close to the two long posts (return to Coruscant),take up a lot less memory and probably run faster, too!
the time of this part is pretty much negligble compared to the 14336 particles it has to compare to. w/143 segments it takes about 1% the time. so e.g. you get 14.85 frames per second instead of 15.

anyways i'll definitely sim this one next. at 15cm, finite coil thickness. what should the coil thickness be?
Last edited by happyjack27 on Wed Dec 29, 2010 2:30 pm, edited 1 time in total.

happyjack27
Posts: 1439
Joined: Wed Jul 14, 2010 5:27 pm

Post by happyjack27 »

KitemanSA wrote:
D Tibbets wrote: ... This would effectively place the center/ strongest portion of the field in the center of the can crossection.
Shouldn't, should it? Doesn't Guass' law apply here too?
the mag fields should cancel between a pair of parallel currents, as they both produce (counter?-)clockwise fields. (likewise there's no net mag field on the inside of a wire, only along the edge.) so the strongest part of the field would be, as one would expect, just outside the wire bundle.

D Tibbets
Posts: 2775
Joined: Thu Jun 26, 2008 6:52 am

Post by D Tibbets »

happyjack27 wrote:
KitemanSA wrote:
D Tibbets wrote: ... This would effectively place the center/ strongest portion of the field in the center of the can crossection.
Shouldn't, should it? Doesn't Guass' law apply here too?
the mag fields should cancel between a pair of parallel currents, as they both produce (counter?-)clockwise fields. (likewise there's no net mag field on the inside of a wire, only along the edge.) so the strongest part of the field would be, as one would expect, just outside the wire bundle.
I don't know the exact configuration of magnetic fields in a bundle of wires. It would depend in part on the separation of the wires (insulation, etc.). Consider solenoids. The wire spacing is important to how it behaves. In transformers, the number of windings is certainly important. If there was not some electromagnetic cumulative effect amoung the windings, the results would not be any different wheather you had one big winding, or 1 million small windings.

In any case my second discription of the real magrid cans still applies. The outer surfaces are seperated by 1 cm in my example and the field gradient drops accordingly. The same applies to the inner surface of the coil which is substantially closer to the center face cusp than it would be if you assume a line cross section for the magnet can, and that the can is placed the few gyroradii (1 cm at closest approach in my example) from the next can.

Dan Tibbets
To error is human... and I'm very human.

rjaypeters
Posts: 869
Joined: Fri Aug 20, 2010 2:04 pm
Location: Summerville SC, USA

Post by rjaypeters »

KitemanSA wrote:And just to be sure, since I haven't learned to read the code, ALL the coils are "North In" (or out), right?
Wouldn't have it any other way. When I enter the data for the points and then check against the CAD software, one of my steps is to ensure I am getting the points in counter-clockwise order as seen from the outside of the sphere (apparently I am not always successful).

BTW, I'd be really surprised if you are able to look at a segment list and be able to tell the orientation of the coil. That capability would just be plain...computer-like.*
happyjack27 wrote:...at 15cm, finite coil thickness. what should the coil thickness be?
Short answer, I don't know. Longer answer: I'm finally acting like alphas come out preferentially at cusps (per someone 's quote of Dr. Nebel) and we can hope for reduced heat load (also I was pretty much forced into the change because of 32 coils on a 1m sphere - otherwise it was just ugly - mostly metal around the sphere). So I don't think you should be using a different conductor diameter just because I am showing a different casing diameter.

Another way to approach the question might be: What is the required diameter of cryo-cooled copper** or superconductor necessary to achieve the fields being simulated? I hope those diameters are less than what I'm using for my models.
happyjack27 wrote:i'll definitely sim this one next
What? Don't we have a soft spot in our hearts for Coruscant's power source (even on a smaller scale)? Also, I would like to know if my semi-automatic method of coil generation works.

*In my most tired moments, I wished for a software tool that would run on my poor computer, look at a segment list and show me the 3-D representation and let me look around a little bit so I could check what I made before submitting them for simulation. It would be tiny subset of happyjack27's software, with the added capability to show current direction for each subset.

I think my desire for this visualization tool has decreased now that I have a semi-automatic method of segment generation...and I think we are almost out of the business of new concept generation, until statistics can be derived from the simulations...

** cryo-cooled copper just for comparison, if you wanted to.
"Aqaba! By Land!" T. E. Lawrence

R. Peters

rjaypeters
Posts: 869
Joined: Fri Aug 20, 2010 2:04 pm
Location: Summerville SC, USA

Post by rjaypeters »

For your desk top?Image Image

Kidding! But clearer, I think, views of what will be simulated.
"Aqaba! By Land!" T. E. Lawrence

R. Peters

happyjack27
Posts: 1439
Joined: Wed Jul 14, 2010 5:27 pm

Post by happyjack27 »

rjaypeters wrote: I think my desire for this visualization tool has decreased now that I have a semi-automatic method of segment generation...and I think we are almost out of the business of new concept generation, until statistics can be derived from the simulations...
on that note i've inadvertently forked my code into two branches that i need to merge. the one you saw in the last few sims and some untested changes towards loss visualization and the aforementioned statistics. and yeah, i'm running out of interesting things to sim, so pretty soon i'll have nothing to do _besides_ that.

anyways, i'll probably get back on that after the new years. if i can find time between watching quantum physics lectures on youtube. (ever so fascinating! the math, though... yeesh! more rules and formulas to remember. no thanks, maybe later. curious though if i could work in a cybernetic (i.e. information theoretic) interpretation... i've always wanted to do that. i can't see how to formulate it w/out the math getting unweildly, though. anyways...)

Post Reply