Has this been discussed?

[TDPerk]

So I was just wondering if the following had been considered as shapes for grids (cross sections attached as a file).

The black arrows represent magnetizing current flow through a charged column, the column having the cross section shown by the heavy green lines.

The field lines would have the wrong shape, not convex towards the focus in the main, but the lack of any but two cusps might make up for it?

I'm looking now for the shareware software thread to model it...

Or a thread where the shape has been discussed?

[DeltaV]

Interesting idea.

I Googled "magnetic field inside a hollow conductor" and got this for a non-magnetic conductor --

http://www.ndt-ed.org/EducationResource ... Fields.htm



Apparently, an additional centerline conductor is needed to get any magnetic field at all in the hollow region --



In this picture, the central conductor is non-magnetic and the outer conductor is magnetic.

I think the same will hold as you bulge out the cylinder at arbitrary cross-sections.

[TDPerk]

I'll have to look very carefully at that. It seems to be at odds with the known construction of several EMP generators, which depend on the explosive axial forced collapse of hollow conductors to emit stupendous field strengths from the closing far end--away from the explosive initiation--albeit briefly.

[DeltaV]

I was assuming DC. For varying current, things get much more complicated.

The current in an EMP device will, most assuredly, be varying.

[D Tibbets]

I see two possible problems. First a two dimensional drawing doesn't reveal directly the current flow. At the throats the current density is greatest, as the radius of the sphere increases so does the area that the current is flowing through, which implies less current flow per unit of surface area and thus lower magnetic fields, unless you are correcting with increasing wires as the radius expands so that the surface area to amp turns remains constant or within some desired gradient. It might be an engineering nightmare.

Secondly, all magnetic fields must be convex towards the plasma, but at the throats the fields would be concave towards the central axis. You not only need to consider orientation towards the center of the vessel but also the orientation towards the plasma locally (in the throat or regions where the magnetic field changes direction , such as in the second drawing. Macro instabilities may be a problem.

Dan Tibbets.

[mattman]

Hello,

So the current and B field looks like this?

[hanelyp]

I don't think so, not if that's a cross section of a rotationally symmetric system. The magnetic field inside the structure would be zero.

[TDPerk]

Since the field in a hollow cylinder is not zero:

http://physics.bu.edu/~duffy/semester2/ ... inder.html

I am wondering the following:

I've seen several years ago links to freeware which would permit the simulation of field lines, electron trapping density/distribution, and the presence of ions held in well.

It was many years ago and I cannot find the links.

Could someone who knows them repost them, please?

[DeltaV]

Since the field in a hollow cylinder is not zero:

http://physics.bu.edu/~duffy/semester2/ ... inder.html

Inside the hollow part of the cylinder the magnetic field is zero (an amperian loop encloses no current)

[TDPerk]

I think the java may be hinky. I've seen the slope representing the field strength starting a r=0 be non zero, and I've seen it snap tot he inner radii of the conductor on being refreshed.

Darn it.

[DeltaV]

I permanently uninstalled Java after the recent security scandal.