Symmetry Solutions

[Rezwan]

Your thoughts on the asymmetry of tungsten pins, the knife edge solution, and other factors of symmetry.

[Brian H]

Seems like any asymmetries would/will grow exponentially, from minute beginnings. Is it possible that some kind of dynamic compensation will ultimately be necessary for reliable operation?

[Breakable]

It either has to be dynamic, healing or non-solid-state. Liquid state on the other hand would need a separate project to implement, actually any fix would.

[Breakable]

Another option is to raise damage resistance:
making liquefaction/vaporization temperature much higher,
reduce the material resistance,
distributing the effect over higher area,
making temperature distribution more instant.
So the technical approacher I see are:
1)changing the material,
2)making it less thin,
3)super-cool the blade-edge down to superconducting temperature to distribute temperature instantly make vaporization harder due to low temperature and reduce the electrical resistance,
4)spin the blade edge at a high speed to make distribution more even.

I would go with
1)Material to nickel-iron, chromium, yttrium, iridium, platinum, tungsten, or palladium 2) see if you can reduce the thinning of the edge
leaving 3,4 to better funded stages

[Breakable]

It seems raising damage resistance is more affordable than any kind of/healing/adaptation/liquid state electrode

[mchargue]

Wouldn’t the ‘high points’ on a circular, knife-edged conductor wear faster than surrounding low points? That is to say, if the material wears, I would think that the areas that experience the most wear would be those that conduct first, and the most: the high points. Wouldn’t that be a self-correcting situation?

If it is, to some extent, maybe the effect could be amplified so that the knife-edged conductive ring would self-adjust.

Pat

[Breakable]

It depends what kind of formations the discharge would form. It can form “U” type shapes, or it can form “^” type shapes.

[mchargue]

Breakable wrote: It depends what kind of formations the discharge would form. It can form “U” type shapes, or it can form “^” type shapes.

Well, two ‘U’ shapes, side by side, ma e a ‘^’.

The gist is, either the erosion can be eliminated, or it can be managed. Elimination is, I think, unlikely, and I think the question is how to best manage the erosion. Take the conductor out, and turn it every two weeks?

I was thinking that there might be some way of taking advantage of the erosion to focus its effects it in such a way so as to remove the ‘high points’ from the conductor.

Maybe there are material properties of the conductor that can assist this, or maybe some other means can be found to focus current flow through the high points so as to evaporate them. Thus, by using either the passive means, or active means, you’d preferentially wear down the high points until the entire conductor was back in true.

I can think of a few experiments I’d try alredy.

[msmith]

Some of the posted photographs of the inducer coil show it’s windings to be unevenly spaced.
Might the resulting magnetic field contribute to sheath asymmetry?

[jamesr]

How about having a laser inside the chamber that can scan the edge for imperfections, then give high power pulses to blast away any peaks. So you’ll end up eroding it all faster, but the laser system could fire & repolish the surface in-situ, then all you need to do is evacuate the chamber to get rid of the waste.

[Brian H]

jamesr wrote: How about having a laser inside the chamber that can scan the edge for imperfections, then give high power pulses to blast away any peaks. So you’ll end up eroding it all faster, but the laser system could fire & repolish the surface in-situ, then all you need to do is evacuate the chamber to get rid of the waste.

After each imperfection (I assume a dip in the knife-edge) developed, would it be likely that the next discharge would preferentially come from the dip or the neighbouring peak? I.e., would the flaw grow or shrink?

[markus7]

Rezwan wrote: Your thoughts on the asymmetry of tungsten pins, the knife edge solution, and other factors of symmetry.

As I understand it, the source of the plasma sheath asymmetry problem is in the asymmetry of the initial current sheath on the surface of the insulator before it gets “kicked off” and driven down the annulus.

I am wondering if such asymmetries are inevitable with all designs that initiate the current sheath across an insulator as the present test model does. Even with a ‘perfect’ knife edge, I expect current filaments (even full up vacuum arcs?) will progressively pit the cathode, creating cathode ‘hot spots’, and deposit the vaporized metal on the insulator surface adjacent to the pits. The longer the device runs, the more asymmetric the initial current sheath is likely to be.

Have other methods of initiating the current sheet, perhaps “spark plug plasma jets” of some kind ever been tried? I am imagining one “spark plug plasma jet” for each cathode outer rod (16 total) aimed to fire across the annulus to the anode to initiate sheath current flow. Thus, the initial current sheet is created across the annulus, not across the surface of the insulator.

My preference, of course, would be for the simple knife edge cathode base to work.

EDIT: I wonder if such an arrangement of 16 “spark plug plasma jets”, in addition to improving current sheath symmetry, might make it possible to eliminate using the main capacitor switches to trigger pulses. That is, the 16 outer cathode rods would, while running, be connected to the capacitors, and the “spark plug plasma jets” would take over the function of the vacuum switches, making them redundant. The function of the vacuum switches would be part of the fusor unit.

EDIT 2: Mainly for people like me who are new to this forum, relevant previous threads include Spark Plugs? https://focusfusion.org/index.php/forums/viewthread/471/ and Laser spark Plugs https://focusfusion.org/index.php/forums/viewthread/855/. Both of these threads focus on initiating the external switches. But they might also be applicable to the idea of reducing the asymmetry problem in the initial plasma sheath in the main fusor, while, as a secondary effect, eliminating the need for high tech external switches.

[TimS]

Clearly, if the knife edge on the new electrode plate works, then this problem is solved.

EDIT: I proposed bending the pins inward toward the anode to fix the asymmetry if it persists. After checking the photo in the article, I see that I misunderstood where it said that the problem pins were higher than the others. It looks like they are longer, not shorter (or more eroded). In the article, it says that they fire first because the tips are closer to the end of insulator, not the base as I had thought. In this case erosion of the pins would have gotten rid of the asymmetry even with straight pins, but that does not seem to have happened. The ends of the pins do not look eroded away to me. I wonder what feedback mechanism is making the asymmetry worse as more shots are fired? Perhaps the filament of current leading from the pin to the anode generates a force which pulls the pin further downward? Perhaps the asymmetry is caused by a tiny diffusion of some conductor from the plasma filament onto the surface of the insulator? Maybe a different composition of insulator would help.

If the problem persists with the knife edge, I wonder if there would be some way to add a slight inductance at the base of the main electrodes so that the current sheath could be created to the knife edge (or pins) all around before lifting off to the electrodes? Perhaps returning to pins and adding a slight inductance in each would slow the current rise through the first pins and allow other pins to become active? I should go back to my old basic electronics class and work out the inductance and timing of the pins as they are…

Although the previous poster’s suggestion of using ‘laser spark plugs’, completely removing the need for high speed synchronized switches and the knife edge or pins, sounds pretty cool.

I had initially thought that the negative side of the capacitors was switched into the cathode electrodes separately, instead of having the positive side of the capacitors switched and then tied together to the anode. That way the shot energy would always be divided evenly between the electrodes. I had thought the circuit worked this way when I saw all those switches, specifically in order to avoid this problem of asymmetric current through the electrodes. I spent some time yesterday confusedly looking at the focus fusion pictures on flickr (aren’t they cool!) trying to figure out why the problem occurs before I realized I had it backwards. I had been wondering how all those separate circuits were synchronized so precisely… A schematic of the focus fusion device would be great.

[markus7]

TimS wrote:

If the problem persists with the knife edge, I wonder if there would be some way to add a slight inductance at the base of the main electrodes so that the current sheath could be created to the knife edge (or pins) all around before lifting off to the electrodes? Perhaps returning to pins and adding a slight inductance in each would slow the current rise through the first pins and allow other pins to become active? I should go back to my old basic electronics class and work out the inductance and timing of the pins as they are…

Although the previous poster’s suggestion of using ‘laser spark plugs’, completely removing the need for high speed synchronized switches and the knife edge or pins, sounds pretty cool.

I had initially thought that the negative side of the capacitors was switched into the cathode electrodes separately, instead of having the positive side of the capacitors switched and then tied together to the anode. That way the shot energy would always be divided evenly between the electrodes. I had thought the circuit worked this way when I saw all those switches, specifically in order to avoid this problem of asymmetric current through the electrodes. I spent some time yesterday confusedly looking at the focus fusion pictures on flickr (aren’t they cool!) trying to figure out why the problem occurs before I realized I had it backwards. I had been wondering how all those separate circuits were synchronized so precisely… A schematic of the focus fusion device would be great.

It had also occurred to me that 16 capacitors, each firing its own cathode rod by individual, synchronized (highly synchronized!) vacuum switches would likely reduce the non-uniformity problem. There might also be an advantage in terms of reduced inductance in the separate circuits and higher peak currents, but I don’t know if that is possible. But that would be a totally new fusor design.

Could you explain what you mean by “add a slight inductance at the base of the main electrodes”?

“Inductance is the property of an electrical circuit causing voltage to be generated proportional to the rate of change in current in a circuit.”

[Henning]

markus7 wrote:
EDIT: I wonder if such an arrangement of 16 “spark plug plasma jets”, in addition to improving current sheath symmetry, might make it possible to eliminate using the main capacitor switches to trigger pulses. That is, the 16 outer cathode rods would, while running, be connected to the capacitors, and the “spark plug plasma jets” would take over the function of the vacuum switches, making them redundant. The function of the vacuum switches would be part of the fusor unit.

Generally a good idea, but I think the electrodes will shorten before the target voltage is reached, i.e. distance between cathode and anode are too short to hold back the sparks, as we try to reduce the diameter of the DPF and increase the amperage of the pulse (which goes along with the voltage). Anyone an idea how to reduce voltage and increase amperage? Well, by reducing inductance, which is somewhat one of the main goals…

[Author]

Posts