Grid stalk (insulator) losses

[John Taylor]

There has been extensive research and discussion about grid transparency and its relation to collisional losses, but does the insulating stalk contribute to losses when ions/electrons collide with it? I know I can visibly see the results of heating on the small tungsten wires of my grid, but the insulator's ceramic has much more material to distribute the heat away without becoming incandescent.

If so, would reducing the cross-section of this component reduce some of the losses? Or are there some insulators that are less "lossy" than others?

I know this would be a very, very small contributor compared to actual grid element losses, but with enough recirculation, the insulator is bound to be in the path.

I hope to soon use a thermal imaging camera to determine different hot spots.

John

[Chris Bradley]

In principle, and one may hope, that ions colliding with an insulating stalk would charge it up and it would no longer become 'attractive' to the ions, unlike the grid whose polarity makes it the preference for the ions to head to.

Unfortunately, many many things may happen, not least of which is that the substance of the insulator parts undergoes physical changes on account of a long list of possibilities (sputtering and sputter coating, reduction in an H atmosphere, thermal degradations, &c.). These are also time dependent, so something working one day may be a problem the next.

Experimentation is the key - your set-up is your own, known only unto yourself. Try what you can and see what works best. I think the reports here are useful to build up a picture, but I suspect it is unlikely that you will find anyone here performing a series of controlled experiments on several stalk materials/sizes/configurations in several fusor set-ups of different chamber sizes and shapes.. and even if they did would it have the same bearing on your own experiment? You are working here only with anecdotes. I look forward to seeing some of your thermal imaging results. Good luck!

[John Taylor]

Thanks Chris for the reply. I really brought this up more as something to think about than something for an in-depth study. I ran across a paper sometime back talking about thermal emission of ion impacted insulators and thought that it was interesting. While I don't believe that a gridded IEC device will ever be efficient enough for power generation, it fun to try to fine tune it to see what can be accomplished.

I often learn more from pondering such things and then studying enough to put together an experiment (even if it ends up just being a mind experiment).

John

[Frank Sanns]

As you get farther from the inner grid region, the volume goes up with the radius cubed. Not much of the ion path is taken up by the stalk. If I recall, Carl Willis even ran some fusion without the insulator even present and achieved good results.

Frank Sanns

[Dustin]

There was a poster at 2011 IEC conference on this subject , Dan posted a link which now appears to be broken.
It was about voltage devision on the stalk to alter the field gradient to be more normal to the rest of the chamber. (Thereby not altering stable recirculation and reducing stalk collisions)

I have discussed this before here;
viewtopic.php?f=6&t=2797&hilit=stalk#p16975

Insulating the stalk doesn't change the electric field from the insulated wire very much as the dielectric constant of the insulator >1 compared to vacuum =1 and the voltage devision across the differing dielectrics in series is inversely proportional to the dielectric constant.
Once the surface charges however, who knows at what potential it sits at.
Steve.