Does a dodecahedron really meet Bussard's requirements?

drmike · Post by **drmike** » Tue Jan 22, 2008 2:57 pm

scareduck wrote: This is what I keep coming back to: a dodecahedron has an odd number of faces off each vertex. It seems wholly inexplicable, then, that he would choose it if it didn't meet this requirement, or that he would bring this up if it were irrelevant.

So does a cube - there are 3 faces off each vertex (corner).

The 2x2-B experiment had a "baseball" coil at the end as well as a normal coil to help prevent instabilities. The field lines in the center of the mirror expand out so that was one way to help prevent problems.

The polywell is a pure mirror, but in a sphere. That is the genius of it. Looking out from the center, in every direction there is a mirror field.
The peak mirror field will be near a corner, the weakest mirror will be the center of any one coil.

It is well known that a plasma is more stable near a B field of negative radius of curvature - meaning the plasma is outside the strongest point in the field. Cusp plasmas were studied a lot in the 1940's and 50's. Again this points out the genius of polywell - the center is a 3D cusp.

It may not matter that much how many coils there are, but we know nature likes symmetry a lot. I'm betting 6 or 12 coils will be optimal for lots of reasons.

MSimon · Post by **MSimon** » Tue Jan 22, 2008 3:07 pm

Bouncing is what I would expect if some of the e-field energy is converted to gyrorotation.

Roger · Post by **Roger** » Tue Jan 22, 2008 3:49 pm

drmike wrote:
scareduck wrote: That is the genius of it. Looking out from the center, in every direction there is a mirror field.

Yes, well said, the genius, beauty, elegance, the simplicity of the Polywell. >sigh<

Here is another version. The Whiffle ball... imagine looking out from the center of the whiffle ball........

scareduck · Post by **scareduck** » Tue Jan 22, 2008 4:04 pm

Roger wrote:The Whiffle ball... imagine looking out from the center of the whiffle ball........

Preferably before the vacuum chamber has had the air evacuated.

Roger · Post by **Roger** » Tue Jan 22, 2008 4:13 pm

Silly duck, everyone knows that plastic whiffle balls dont have magnetic fields.

And no reason to put one in a vacume chamber.

HA...ha...ha..

Pol..lee..well......Pol..lee..well...... Pol..lee..well......

TallDave · Post by **TallDave** » Tue Jan 22, 2008 4:29 pm

drmike wrote:So does a cube - there are 3 faces off each vertex (corner).

Yeah, all I can think is Bussard must have been referring to the 2-dim "vertices" -- the corners of the decagons in a trunc dodec, or the octagons in the trunc cube -- rather than the 3-dim corners of the 3-dim shape (where the 2-dim decagons/octagons meet).

Looking at a lot of other Archimedean solids, most don't meet the requirement of having an even number of edges at each vertex.

TallDave · Post by **TallDave** » Tue Jan 22, 2008 4:37 pm

What you guys are saying is that generating a plasma in a polywell using a microwave so that the electron energy is about ice cold guarantees no electrons will ever escape. Should be an easy enough experiment - I bet the confinement time is a lot shorter than "forever".

I'll settle for 1E5 transits or better

MSimon · Post by **MSimon** » Tue Jan 22, 2008 4:59 pm

Dr Mike,

Didn't Dr. B say that what confounded a lot of the early simulations was zero dimensional grids and coils?

drmike · Post by **drmike** » Tue Jan 22, 2008 5:46 pm

I haven't run across that comment, but I would not be suprised. Theory is ok up to a point, but you make lots of assumptions that are just plain stupid when you compare experimental results to what you expected. That's probably why I like building experiments.

My initial thinking about "recirculation" was that electrons flow along the B lines and go around the coils and back. But looking at the paths, they don't, they mirror. That is, they get tighter and tighter circles until they bounce back from the flux compression.

So if TallDave is right and we need 1e5 bounces, then we have to keep the electron temperature cool enough so that the loss angle in the middle of each coil is smaller than it can fit through. The loss angles at the edges and corners is already smaller (i.e. it takes a lot more energy to escape there than in the center of a coil).

Cooling the electrons takes energy, but the net gain in power may be worth it.

But back to the topic - the number of coils isn't so important, but the loss angle at each point in space is. What we have to do is compare the total surface area of loss angle between the different configurations to see what will hold the electrons in better.

(the real term is "loss cone angle" if you do a web search).

More holes of smaller area may not equal fewer holes of larger area, and there may well be a net power gain. So the 12 coil configuration may win.

Roger · Post by **Roger** » Tue Jan 22, 2008 5:57 pm

drmike wrote:
More holes of smaller area may not equal fewer holes of larger area, and there may well be a net power gain. So the 12 coil configuration may win.

Back @ nasa space forum, Tom Ligon talked about this. I had asked Tom about a polywell with many coils, at least dozens to 100's. IIRC Tom was infering thats why Bussard wanted to build both, the truncated cube and truncated dodec, to see which one had fewer losses, vs better performance maybe due to less quasi and more spherical.

MSimon · Post by **MSimon** » Tue Jan 22, 2008 6:14 pm

Roger,

If I recall correctly the reduction in loss was expected to be about 3X to 5X.

In part I would expect because for a coil of a certain size the center of the coil was farther from the center of the "sphere" reducing the volume of the loss cone.

Another loss reduction mechanism was expected to be the more even magnetic field in the reaction volume.

BTW Dr. Mike,

As I said - I do have a way of asking the "right" question even if I don't rationally understand the math details.

Some times it takes two questions. ;-)

scareduck · Post by **scareduck** » Tue Jan 22, 2008 6:29 pm

drmike wrote:Cooling the electrons takes energy, but the net gain in power may be worth it.

I wonder that brehmsstrahlung from ion upscatter wouldn't do some of this naturally.

drmike · Post by **drmike** » Tue Jan 22, 2008 10:53 pm

Simon,

Sometimes I trust the math and then ask stupid questions. Eventually I figure things out too! What ever works, that's the best way.

drmike · Post by **drmike** » Tue Jan 22, 2008 10:54 pm

scareduck wrote:
drmike wrote:Cooling the electrons takes energy, but the net gain in power may be worth it.
I wonder that brehmsstrahlung from ion upscatter wouldn't do some of this naturally.

That will be fun to estimate with a model. I think it should to some extent.

scareduck · Post by **scareduck** » Tue Jan 22, 2008 11:20 pm

The big problem I see is that unless the annealing forces are real, ions will escape the core if the collision fails to result in a fusion. Then they upscatter back toward the magrid, where they will see a whole lot of electrons whirling around them. That sets up the possibility of brehmsstrahlung. If that happens enough, you're back to Q<1.