Ceramic Beads breaking down?

This forum is for specialized infomation important to the construction and safe operation of the high voltage electrical supplies and related circuitry needed for fusor operation.
User avatar
Anze A Ursic
Posts: 157
Joined: Thu Mar 10, 2022 9:28 pm
Real name: Anze A Ursic

Ceramic Beads breaking down?

Post by Anze A Ursic »

So I've rebuilt my fusor and preliminary testing showed excellent results, I was able to get plasmas up to 12kV on the first run after pumping down for several days.

However, every once in a while I noticed an orange "spark" in the bottom corner of the camera and it kept getting worse until I shut down the run and tried to ignite a plasma again only to only see these erratic sparks happening in the same place until a bigger flash happened, at which point I stopped everything. I captured the frame of the flash and it was this:
arcing.PNG
I have two tungsten electrodes wired in parallel going to a Spellman DF3 power supply. The tungsten electrodes are connected to 40kV HV feedthroughs and are insulated by ceramic beads / cylinders as can be seen below with the problematic part circled. I apply 12kV and slowly raise the pressure by letting in Deuterium until a plasma strikes but now it seems the plasma does not even strike where it's supposed to (in the middle of the two electrodes), but there is just some sort of arcing / shorting in that region nonstop.
ceramic.PNG
My question is, what is going on here? Am I breaking down the ceramic electrically? My friend said that there is a voltage difference between the beads that is causing arcing. It's also odd because only one electrode's insulation does this.

Any help appreciated. I wish there was a way to fix this without opening the chamber but I assume there won't be...

Anze
Matt_Gibson
Posts: 505
Joined: Thu Sep 23, 2021 10:36 am
Real name: Matt Gibson

Re: Ceramic Beads breaking down?

Post by Matt_Gibson »

What’s the purpose of those ceramic beads?

-Matt
User avatar
Anze A Ursic
Posts: 157
Joined: Thu Mar 10, 2022 9:28 pm
Real name: Anze A Ursic

Re: Ceramic Beads breaking down?

Post by Anze A Ursic »

Shielding - preventing arcing between wires carrying the HV and the chamber walls.
Matt_Gibson
Posts: 505
Joined: Thu Sep 23, 2021 10:36 am
Real name: Matt Gibson

Re: Ceramic Beads breaking down?

Post by Matt_Gibson »

Gotcha. I think that ceramic (alumina?) usually winds up acrcing and is removed by most of us. I don’t use any other than the bit that sticks out on my feed through and don’t have any issues with insulation up and over 60kV.

-Matt
User avatar
Liam David
Posts: 518
Joined: Sat Jan 25, 2014 5:30 pm
Real name: Liam David
Location: PPPL

Re: Ceramic Beads breaking down?

Post by Liam David »

It could be any number of things, such as triple junction field enhancement, dirty (metalized/metal-streaked) alumina, local outgassing, quasi-static charge buildup, and/or some geometry effect. I would remove the alumina beads, and while you have the chamber open, round as many sharp edges and corners as you can.
User avatar
Anze A Ursic
Posts: 157
Joined: Thu Mar 10, 2022 9:28 pm
Real name: Anze A Ursic

Re: Ceramic Beads breaking down?

Post by Anze A Ursic »

I see.... I have to open the chamber up again. Darn.

So out of curiosity, are people usually not worried about arcing inside the chamber, from the wires to the chamber walls? Especially with sharp corners. I would assume that would be an issue intuitively, although from the Paschen curve I suppose not. Do most people on this forum just not shield their cables inside the chamber?

Anze
Matt_Gibson
Posts: 505
Joined: Thu Sep 23, 2021 10:36 am
Real name: Matt Gibson

Re: Ceramic Beads breaking down?

Post by Matt_Gibson »

From what I’ve seen, most of us go bare and don’t have any issues.

-Matt
User avatar
Liam David
Posts: 518
Joined: Sat Jan 25, 2014 5:30 pm
Real name: Liam David
Location: PPPL

Re: Ceramic Beads breaking down?

Post by Liam David »

Unfortunately, the issue of vacuum breakdown is much more complicated than a quick glance at Paschen's law. It can give you some idea at moderate voltages and in smooth geometries, but surface contamination, finish, complex geometry, dielectrics, etc... make it much less useful, especially in the pressure regime of fusors. The question of when and where to use ceramic insulators isn't straightforward and would take considerable effort to do justice, but in your case, I think it would be best to remove them.
User avatar
Anze A Ursic
Posts: 157
Joined: Thu Mar 10, 2022 9:28 pm
Real name: Anze A Ursic

Re: Ceramic Beads breaking down?

Post by Anze A Ursic »

I'm mostly worried about the tungsten wire connecting to the HV input pin, because the connector has a few sharp edges. I could try and sand it off, but has anyone ever use Capton tape for shielding purposes? Or is that another bad idea?
User avatar
Jim Kovalchick
Posts: 717
Joined: Wed Apr 13, 2011 8:00 pm
Real name:

Re: Ceramic Beads breaking down?

Post by Jim Kovalchick »

My experience with Kapton tape is that grid heating is enough to cause it to break down.
Matt_Gibson
Posts: 505
Joined: Thu Sep 23, 2021 10:36 am
Real name: Matt Gibson

Re: Ceramic Beads breaking down?

Post by Matt_Gibson »

Don’t use any sort of tape inside the chamber. Maybe you could take a few pictures of your cathode?
User avatar
Anze A Ursic
Posts: 157
Joined: Thu Mar 10, 2022 9:28 pm
Real name: Anze A Ursic

Re: Ceramic Beads breaking down?

Post by Anze A Ursic »

Got it, no tape. One other thing I was thinking about is that the HV feedthrough has a 1 inch diameter, so what if I just put a 1" inner diameter glass / ceramic tube around the actual ceramic part of the feedthrough, so that it basically hangs off it, while not touching the HV wire at all? At least for a few inches, until the wire is more distant from the chamber walls.

Something like this. The picture sucks because I made it in Paint a few minutes ago.
shielding option.png
Anze
Matt_Gibson
Posts: 505
Joined: Thu Sep 23, 2021 10:36 am
Real name: Matt Gibson

Re: Ceramic Beads breaking down?

Post by Matt_Gibson »

To be honest, I wouldn’t add anything until I was having trouble. I think that less is more to worry about when it comes time to trouble shoot. For now, try removing those beads and going bare conductor. Watch for arcing to see where the trouble spots are. For me, I don’t have any arcing up to 63kV (I haven’t pushed beyond that).

-Matt
JoeBallantyne
Posts: 326
Joined: Tue Jul 20, 2010 4:08 pm
Real name: Joe Ballantyne
Location: Redmond, WA

Re: Ceramic Beads breaking down?

Post by JoeBallantyne »

What is that very large and beefy structure in the middle of your chamber with ~12 vertical rods attached to it? (6 each on an x and y axis if the vertical rods are considered as being oriented in the same direction as the z axis.)

It is not clear, but it looks like that structure may or may not be isolated from the chamber itself.

Is that huge structure your cathode? If not, and if it is not isolated from your chamber, you should remove it. Unless you want your anode to look like that.

Why do you have two wires coming off your feedthrough?

Most cathodes are made with a much less beefy structure, and are either cylindrical, or spherical, and are hollow (in at least 1 dimension if not all 3).

It looks to me like that structure was perhaps just in the chamber you bought and was there to fulfill the original purpose of the chamber.

If so, why haven't you removed it?

Joe.
User avatar
Anze A Ursic
Posts: 157
Joined: Thu Mar 10, 2022 9:28 pm
Real name: Anze A Ursic

Re: Ceramic Beads breaking down?

Post by Anze A Ursic »

No, the big structure is actually my cathode setup. It used to be a bit more complicated which is why it's so big. But it's completely separated from the chamber so those metal endplates are not connected to anything (ground or HV).

I know how most people's setups look, but this is a very large chamber and it's difficult to make a cathode directly on the HV feedthroughs, so I use the ceramic rods to hold it in place and the cathodes connect to the HV feedthrough.
User avatar
Anze A Ursic
Posts: 157
Joined: Thu Mar 10, 2022 9:28 pm
Real name: Anze A Ursic

Re: Ceramic Beads breaking down?

Post by Anze A Ursic »

Joe, I have computer problems so I can't post any pics on here unless I'm on campus, but if you wanna see the geometry producing plasma to see how the geometry looks, check my thread here, I posted a few pics.

viewtopic.php?t=14462&start=60
JoeBallantyne
Posts: 326
Joined: Tue Jul 20, 2010 4:08 pm
Real name: Joe Ballantyne
Location: Redmond, WA

Re: Ceramic Beads breaking down?

Post by JoeBallantyne »

So the vertical rods look like they might be insulators. Is that the case? They look like they are glued to the metal end pieces on the top and bottom.

The pictures in the other thread were not helpful. But I saved the ceramic.png picture from this thread and then zoomed in on it to see better what was going on.

It appears that your cathode is just 2 wire loops held in place by the inner 4 vertical rods, which appear like they might actually be insulators.

Is that correct?

Is the feedthrough on the side of the chamber, or the top or bottom?

Joe.
User avatar
Anze A Ursic
Posts: 157
Joined: Thu Mar 10, 2022 9:28 pm
Real name: Anze A Ursic

Re: Ceramic Beads breaking down?

Post by Anze A Ursic »

Yes, high voltage rated ceramic rods are held in place vertically and the electrodes are attached to that. It was the only way I could figure out how to make this design, although it was originally a bit more complicated, which is why the structure is somewhat unnecessarily big. This is a crude picture of how everything works. The steel endplates holding the ceramic rods in place are separated by about 2" of ceramic standoffs from the actual chamber (the two blue legs on the bottom of my device in this picture.)

The HV feedthrough is on the side of the chamber, it is 4 large HV feedthroughs rated for 40kV.

Anze
structure picture.png
Matt_Gibson
Posts: 505
Joined: Thu Sep 23, 2021 10:36 am
Real name: Matt Gibson

Re: Ceramic Beads breaking down?

Post by Matt_Gibson »

You might want to try something a little more “tried and true” for now.

Here is what I went with as a start…It’s stainless steel tubing clamping together my tungsten wire loops and pushed up the feed through stalk/wire. You could probably use stainless steel tubing to do the same with yours…
F279191C-A9B4-4506-B4C7-09C8BEF1B8A7.jpeg
6161D1E2-9DF8-4E30-A59C-9B866F5757B5.jpeg
User avatar
Anze A Ursic
Posts: 157
Joined: Thu Mar 10, 2022 9:28 pm
Real name: Anze A Ursic

Re: Ceramic Beads breaking down?

Post by Anze A Ursic »

Yeah I know that's how most people build their fusors and I wish I could too but I'm working within the confines of a university and this is all I have. I am not complaining, I have access to extremely cool equipment, but the chamber itself is not geared towards fusors... I have to do with what I have. Do you have any further advice for me before I open the chamber tomorrow and close it up again after removing the beads? I'll try and keep the wires as far away from the walls as possible.

Anze
Matt_Gibson
Posts: 505
Joined: Thu Sep 23, 2021 10:36 am
Real name: Matt Gibson

Re: Ceramic Beads breaking down?

Post by Matt_Gibson »

Don’t overthink it. Remove those beads and avoid sharp or rough edges and don’t kink those wires. I think you’ll be fine with clearance. You have a lot more than I do.

-Matt
JoeBallantyne
Posts: 326
Joined: Tue Jul 20, 2010 4:08 pm
Real name: Joe Ballantyne
Location: Redmond, WA

Re: Ceramic Beads breaking down?

Post by JoeBallantyne »

If I were you, I would build a standard spherical cathode (at least 3 rings of wire separated by 120 degrees each) and attach it to just one of your feedthrough wires. Remove the second feedthrough wire.

I would remove all of the ceramic wire coverings so that the feedthrough wire is bare, and I would not use that massive complicated cathode structure you have built at all.

I would just remove it completely from the chamber.

It is basically a lot of structure, much of which may be floating to some unknown potential, and will just block ions and electrons from circulating - because they are going to hit all the currently unused parts of the structure. (8 vertical unused ceramic rods and the floating metal structure on the top and bottom of those rods that is holding them in place.)

Making the innards of your fusor as simple as possible will make it easier for you to be successful at what you want to do.

If you have to support the cathode somehow, use a single ceramic rod from either the top or bottom of the chamber to hold a spherical wire cathode in place.

My 2 cents.

Joe.
JoeBallantyne
Posts: 326
Joined: Tue Jul 20, 2010 4:08 pm
Real name: Joe Ballantyne
Location: Redmond, WA

Re: Ceramic Beads breaking down?

Post by JoeBallantyne »

Also, you indicated that it took days to pump down the chamber. That seems a bit long.

Are you sure you don't have leaks?

If the pump sizes you are using are reasonably matched to the chamber with good conductance, it shouldn't take more than an hour or 2 at the most to pump your chamber down to the micron range I would think.

If you have leaks, or your pumps are tiny compared to the chamber, or the conductance is low (think tiny straw sucking on a huge volume) then it would possibly take days.

Joe.
User avatar
Anze A Ursic
Posts: 157
Joined: Thu Mar 10, 2022 9:28 pm
Real name: Anze A Ursic

Re: Ceramic Beads breaking down?

Post by Anze A Ursic »

By pumping down days, I mean getting rid of contaminants. Once I open the chamber, and mind you, the chamber is large, I pump down but every day it gets exponentially faster to reach the E-5 Torr range. I have a large viton flange on the top because it's the best solution we could think of, but the outgassing isn't bad, it goes from 2E-5 to 1E-2 torr in about 10 minutes and the next day, roughly after 17h I am still in the ~9E-1 Torr. Just today I checked the pressure (I pumped down to 1.5E-2 with just the mechanical pump on Sunday), and it was 2.8 Torr after 49h and 40 minutes. Ideal? No. Good enough? Sure! :-)
User avatar
Anze A Ursic
Posts: 157
Joined: Thu Mar 10, 2022 9:28 pm
Real name: Anze A Ursic

Re: Ceramic Beads breaking down?

Post by Anze A Ursic »

Hello again everyone,

Sorry to revive an old(er) thread, but I had to leave the fusor work alone and focus on my MS thesis. Anyway, I am still gnawing over possible solutions to this problem. Without insulation, my primary area of ionization shifted to the HV carrying wires and there was more plasma around those than in the actual center as it should be. Those ceramic beads that broke down were in fact weakened somehow (should never have reused them...), so I will try this setup one more time and see where I get. I was getting up to 13kV+ last time.

If it doesn't work I agree that making my structure less complicated is a good way to go. With other issues I have with this setup, including the chamber being massive and leaky, it's the only sane thing to do.

However, I am slightly confused as to how I would be able to build a "standard" spherical fusor with my HV feedthroughs. Basically, from what I gather, most people use a single HV feedthrough somewhere in the middle of their chamber and their cathode is directly attached to that. In my case though, as you can see in the figure below, the HV feedthroughs are on the side and are "indented", so the wall of the chamber is very close to them. If I build a spherical fusor, and just attach it to one of these feeds, how would that work? I would have uneven electric fields and the cathode would be quite close to the chamber, which makes me worried. I drew a picture showing how a spherical fusor would look in my chamber.

Does anyone have any advice?

Much appreciated!

Anze
fusor idea.jpg
chamber.jpg
Post Reply

Return to “High Voltage - Fusor Input Power (& FAQs)”