S.T.A.R. - Evolution in progress

For posts specifically relating to fusor design, construction, and operation.
User avatar
Steven Sesselmann
Posts: 2111
Joined: Thu Aug 11, 2005 1:50 am
Real name: Steven Sesselmann
Location: Sydney - Australia
Contact:

S.T.A.R. - Evolution in progress

Post by Steven Sesselmann » Sat Oct 04, 2008 12:39 am

As Frank S. said, the route to fusion is through evolution and not by revolution.

After taking a few weeks off from my STAR construction and focusing on my business for a change, which was badly needed, I am now back at it again.

For those of you who are new to the forum, S.T.A.R. is a fusor with a twist, it is a hollow cathode collider where ions are trapped, moving up and down through an electrostatic potential energy well. See my old posts for more information.

So far the S.T.A.R. reactor has proven that it can make neutrons, but both the numbers and efficiency have been lower than a standard Farnsworth Fusor. My prototypes have suffered from a number of problems that owe their existence to my inexperience with vacuum systems, voltage systems and neutron detection systems.

Fortunately I am learning by trial and error, and things are going a bit smoother now.

The problems have been narrowed down, and there are now two main problems to overcome, one should be easy, the other might be a little bit harder.

PROBLEM 1
The vacuum seal between the dielectric tubes and the cathode continue to be problematic. In air these junctions would be adequate, but because the STAR cathode is immersed in transformer oil, it is a bigger problem. Even the tiniest amount of oil entering the vacuum system results in a carbonized conducting deposit inside the glass tubes. As soon as the build up forms, electrons or ions are attracted to the tube walls, causing x rays and erosion of the glass. More importantly neutron production stops completely when this happens. I believe that this problem can be solved by using ceramic tubes and brazing/welding them directly to the cathode. This can be done professionally by Kyocera in Singapore, they have a division who do this kind of thing. I had a quote from them indicating it would cost around $2000.00

PROBLEM 2
Poor beam focus, ions enter one side of the cathode, and spread out, resulting in ion loss to the edges of the apperture on the opposite side. STAR depends on good recirculation, thereby allowing the beam to build up to fusion capable densities.

I am working on a small modification now, which I hope will improve performance. The metal centering ring in the KF10 seal has been replaced with a machined Macor ceramic sleeve. The ceramic sleeve now extends right into the cathode, thereby changing the shape of the electrostatic field lines.

With the ceramic sleeve inserted, the field lines will extend from the inside surface of the cathode, sharply turning into the tube and terminating in the anode reflector, which is an inverted cone at the thick end of the glass tube.

I am hoping that the ions will accellerate along the field lines down the tube, but where the field lines do a 90˚ bend at the cathode, the ions will have enough momentum to continue in a straight line.

Worth a try, and if it works, I might go for a pair of professionally made tubes.

A few pics to clarify my ramblings..

Steven

http://www.beeresearch.com.au
Attachments
DSCN2536.jpg
DSCN2536.jpg (35.61 KiB) Viewed 1704 times
DSCN2535.jpg
DSCN2535.jpg (41.07 KiB) Viewed 1704 times
DSCN2534.jpg
DSCN2534.jpg (43.58 KiB) Viewed 1704 times
http://www.gammaspectacular.com - Gamma Spectrometry Systems
https://www.researchgate.net/profile/Steven_Sesselmann - Various papers and patents on RG

DaveC
Posts: 2346
Joined: Sat Jun 30, 2001 5:13 am
Real name:

Re: S.T.A.R. - Evolution in progress

Post by DaveC » Sat Oct 04, 2008 7:05 am

Steven -

Thanks as always, for your stimulating and thought provoking work, as well as great pictures.

Regarding improving the ion focus: The two holes in your central sphere, make two aperture lenses positioned at the ends of the insulating ion tubes.

See if you can lay hold on the book, "Building Scientific Apparatus" by John H. Moore, Chistopher C Davis and Michael A Coplan, Published by Addison Wesley 1989.

Chapter 5 (2nd Ed) on Charged Particle Optics, has a lot of useful information on electrostatic lensing. They quote a basic source by Harting and Read... for some of their formulae.

What's happening in your system is this. The apertures in the sphere's surface create diverging electrostatic lenses for the inbound ions as they enter the sphere. Since there is no significant field on the interior of the sphere, these trajectories remain unaltered and some (maybe a high percentage) probably hit the opposite interior surface of the sphere.

This are, no doubt, some collisions from opposing ion beams. That you are detecting some neutrons would seem to be evidence of this.

You can probably sketch in a few equipotential surfaces around the sphere and the apertures to see that they bend into the apertures creating a concave-outward potential surface. In order to focus, that surface needs to curve outward..(be convex-outward) .

The convex-outward equipotential surface will form, if a second, inner sphere at higher potential than the outer sphere were present. We're not addressing the mechanical and electrical complexities of doing this, but simply considering what would create an electrostatic lens that could focus, and preserve the essential features of your STAR fusor design.

Hope this makes some sense.

Dave Cooper


Other forms of the central electrode avoid this problem by their largely open geometry, which can allow a field to develop on the interior (albeit, not a particularly large field).
The standard spherical fusor, avoids this problem, somewhat by design, in its assumption that all ions will follow the (sort of) radially directed field lines and converge at the center. Looking at the rather broadly defined spherical convergence region, we can see that this is not really happening to any large extent.

User avatar
Steven Sesselmann
Posts: 2111
Joined: Thu Aug 11, 2005 1:50 am
Real name: Steven Sesselmann
Location: Sydney - Australia
Contact:

Re: S.T.A.R. - Evolution in progress

Post by Steven Sesselmann » Sat Oct 04, 2008 9:22 am

Dave,

Thanks for your suggestions, I have ordered that book you suggest, one of the local book shops had it on 8 day delivery. That will easily save me more than what it cost.

Your idea to have multiple concentric cathodes inside eachother has been discussed. I recall Frank suggesting this some time ago. Building this would be a bit of a challenge.

As you correctly mention, the field lines stretching down through my dielectric tubes will diverge as they enter the cathode, however in the latest config. where the dielectric tube extends into the cathode, these lines diverge at acute angles. The ion travelling at 50 kev. can't turn the corner at that speed, and should continue in a straight line through the cathode and up the other side.

Well, I hope...

Steven
http://www.gammaspectacular.com - Gamma Spectrometry Systems
https://www.researchgate.net/profile/Steven_Sesselmann - Various papers and patents on RG

User avatar
Chris Bradley
Posts: 2930
Joined: Fri May 02, 2008 11:05 am
Real name:

Re: S.T.A.R. - Evolution in progress

Post by Chris Bradley » Sat Oct 04, 2008 10:29 am

As per other threads, this is a case where the experiment will say more than the theory. In this case, I think the theory may be possibly against you, so I hope the experiment is with you! As the ions diverge some will tend to collide with the ceramic, charging it up over time and thus modifying the electric fields to the point where you don't get the acceleration down/back-up the tube as you hope for. There again, charging of those ceramic parts may prove to be in your favour if that surface charge self-stabilises to a magnitude sufficiently to centre-align ions rather than pinching them off, but not so strong that the majority of the field gradient is actually within the central electrode.

I'm not sure there will be continuous operation possible for this configuration due to this charging - but I hope you will prove me wrong!

best regards,

Chris MB.

DaveC
Posts: 2346
Joined: Sat Jun 30, 2001 5:13 am
Real name:

Re: S.T.A.R. - Evolution in progress

Post by DaveC » Sat Oct 04, 2008 6:53 pm

Steven -

The use of insulating tubes with charged particles ( which we have all discussed a lot here) is problematic, to say the least. Charge deposition and uneven leakage create random potentials and gradients which will steer the beams. Coatings and resistive materials to grade the potentials are ways that can work. In a sputtering environment such as the fusor creates, maintaining the desired down wall conductivity is dicey.

Ignoring that for the moment, I would expect, some ions in the very center of each beam to be least likely to deviate as they enter the spherical electrode. This assumes the fields around the apertures are more or less symmetric. If the holes line up on a diameter, then some portion of the beams should approach each other head-on.

One of the difficult to model aspects of focused beams is the charge repulsion forces.
At some charge density, these coulomb forces will oppose any further tendency of the particles to converge. One can arm-wavingly estimate this nearest appoach distance by multiplying the beam energy by the Sine of one-half the approach angle. (If the angle is small, as it usually is, then just multiply by the angle in radians). This represents the portion of beam energy at right angles (normal) to axial energy. This is the energy that converges the beam.

At some value of "R" in the coulomb potential equation, the coulomb energy equals this converging energy and that's where things level off. With a beam of macroscopic size, it's an arm-waving calculation, but estimating on the basis of in-flight charge density is a starting point. This is more or less what Simion3D does. At the pico coulomb level, things begin to get rather interesting... scattering gets fairly strong.

Hope this is somewhat useful.

Dave Cooper

Starfire
Posts: 1482
Joined: Wed Oct 24, 2001 6:14 pm
Real name:

Re: S.T.A.R. - Evolution in progress

Post by Starfire » Sat Oct 04, 2008 11:10 pm

A bit reluctant to suggest it, but there is still the option of magnetic focus Steven and this may be an answer to the divergence of reflected Ions.

I can invisage two magnets in counter focus - one for the main Ion beam and a second for the reflected with an apex centered in the sphere - permanent MOV's may work and have the advantage of being a ready made toroid. Pulse electromagnets should also be considered - cap. dumpers would give fantastic field strength and you will learn about spark gaps and Hydrogen Thyratrons.

I watch with interest this great work - power to your elbow.

User avatar
Steven Sesselmann
Posts: 2111
Joined: Thu Aug 11, 2005 1:50 am
Real name: Steven Sesselmann
Location: Sydney - Australia
Contact:

Re: S.T.A.R. - Evolution in progress

Post by Steven Sesselmann » Sat Oct 04, 2008 11:18 pm

John, I would have tried the toroidal magnets in some form, had it not been that the hole in the center is "just" a tiny bit too small to fit over the glass tube.

After all I am knee deep in those magnets...

Next version..

In the mean time, they make great tool holders above my lathe..

Steven
http://www.gammaspectacular.com - Gamma Spectrometry Systems
https://www.researchgate.net/profile/Steven_Sesselmann - Various papers and patents on RG

User avatar
Frank Sanns
Site Admin
Posts: 1641
Joined: Fri Jun 14, 2002 6:26 pm
Real name: Frank Sanns
Location: Pittsburgh, PA USA

Re: S.T.A.R. - Evolution in progress

Post by Frank Sanns » Sun Oct 05, 2008 5:09 pm

Steven,

Magents get my vote. They work well on my gridless Pillar of Fire setup. You might be surprises how well they will work for you.

If you have a wet saw and go real slow you can half that torroid so you don't have to slip it over your tubes.

Frank Sanns

User avatar
Chris Bradley
Posts: 2930
Joined: Fri May 02, 2008 11:05 am
Real name:

Re: S.T.A.R. - Evolution in progress

Post by Chris Bradley » Sun Oct 05, 2008 5:57 pm

I'll second that. I've recommended magnets a few times for this. This STAR has a real advantage being a 1D linear approach, but the only way to help that process along is magnetic.

A simple way to explore this that may be useful is to wind some electromagnets. I would suggest two around the inlet top and bottom, but also a third around the full width of the centre of the device.

The point of such an experiment would not be to 'solve' a problem but with electromagnets you can increase the current and see if there is any effect. If there is an effect then plot the function of mag field to ..whatever effect it has. Maybe go for something stronger and possibly some custom permanent magnets if you extrapolate that out and find something beneficial could happen with stronger fields.

Just my two-pennies worth...

best regards,

Chris MB.

User avatar
Chris Bradley
Posts: 2930
Joined: Fri May 02, 2008 11:05 am
Real name:

Re: S.T.A.R. - Evolution in progress

Post by Chris Bradley » Sun Oct 05, 2008 5:59 pm

Is this to limit equatorial electron currents, Frank, or have you something more tantalising up your sleeve??

Post Reply