Fusor Plasma Extinguishing

For posts specifically relating to fusor design, construction, and operation.
Post Reply
User avatar
Anze A Ursic
Posts: 157
Joined: Thu Mar 10, 2022 9:28 pm
Real name: Anze A Ursic

Fusor Plasma Extinguishing

Post by Anze A Ursic »

Hello,

I've recently rebuilt my fusor due to some issues I kept encountering, but I still have one problem that I have yet to solve. Specifically, my fusor plasma just extinguishes for no reason. It's there one second and it dies the next. Currently, I am running 99.9% pure Nitrogen to see plasma formation and get to star mode / close to the 25kV goal before wasting my lab's Deuterium. However, whenever I get a plasma, it at some point dies for no reason. I see no spike in current or voltage or quick change in pressure, although the latter is difficult to detect on my remote interface.

Some specs:

1. Gas: N2 going through a regulator that reduces pressure down to about 30 PSI, going to a MKS MFC that I control with a 0-5VDC power supply.
2. Pressure: I start with lowest pressure, around 2E-5 Torr, then throttle the turbo to get around 9E-5 to 1E-4 Torr. That way, I only need about ~3 SCCM of N2 flow to bring pressure high enough to ignite plasma.
3. Power supply; Spellman DF3, with a 18" 100kOhm ballast resistor in series with the fusor. The HV feedthroughs are rated for 40 or 50kV and have been tested up to that before.

Since the minimum voltage on the DF3 is -8.5kV, and there is a slight delay between throttling the MFC and pressure dropping, I start at -12kV at the 9E-5 to 1E-4 Torr range and start opening the MFC to let in N2 and raise the pressure. At around 4.2E-3 I get a plasma, but this is quite violent since I am starting at -12kV, so current jumps to 20mA for a second before I slowly pull down the voltage to roughly 9-9.5kV. Here, current is anywhere from 5.5-6.5mA. I then have a way to raise or lower the voltage using up and down arrows on my keyboard (I run a Labview script to control everything). The resolution is 100V, so up key is +100V and down key is -100V. So once current stabilizes at the 9.5kV range, I start slowly dropping pressure and raising voltage, maintaining current at some normal level. I have tried many different current levels at which I control current, from 4mA to 7mA, the same thing always happens. Yesterday I was able to get to 13kV until this happened but today I could not clear 10.5kV, except one run where I tried running at around 7mA where I got to 11.6kV. I doubt it is arcing because I've seen arcing in this chamber before and it would be seen from the viewport. Further, the power supply has an arc warning thing and also, there is no jump in current when it happens. I monitor current and voltage on a real-life time plot in Labview and it just immediately jumps to 0.

I am genuinely curious as to why this would be happening. There are no major instabilities really and it's not like it can't handle the voltage, I start at a higher voltage at 12kV and I've gone up to 13kV.... For a representation, I have uploaded this video. It's not very clear but you can easily see how at one point there is a plasma and then it just disappears. The increase / decrease in brightness is me going up and down with the arrow keys, increasing or decreasing the voltage to match the current to what I need.

https://www.youtube.com/shorts/FR_M7U6MpQA

Any info appreciated.

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

Re: Fusor Plasma Extinguishing

Post by Liam David »

My first guess is that you're dealing with the left side of the Pashen curve, where it's almost vertical, and that the pressure changes just enough to extinguish things. What change in the pressure can you resolve? Mentioning that "the latter is difficult to detect on my remote interface" immediately gives a red flag.

Since you have a very large chamber, the plasma current density at 10 mA (order of magnitude) is much lower than that in almost all fusors I've seen. You could probably support at least 100 mA if not several times more, given a large enough power supply, of course. I have not reasoned through how the current density affects such instabilities and so cannot give any causal link, but it does seem plausible that low current density glow discharges are inherently less stable. I would decrease the size of your cathode to something on the order of a couple of inches.
User avatar
Anze A Ursic
Posts: 157
Joined: Thu Mar 10, 2022 9:28 pm
Real name: Anze A Ursic

Re: Fusor Plasma Extinguishing

Post by Anze A Ursic »

Changing the cathode size now is not doable, but I could certainly raise current. The power supply is rated for -60kV and 80mA. The reason why I did not go higher yet is simply because the wires carrying the voltage to the cathodes start glowing. This wasn't an issue before, when they were insulated, but as one of my previous posts showed, the high voltage would break down the insulation, so I had to get rid of it. They also have plasma around them which makes sense but I've noticed the ionization region shift towards the inside of my cathodes as the pressure drops. So I just have to get to a good pressure at a high enough voltage and then I think I'll be able to crank the current higher.
Frank Sanns
Site Admin
Posts: 2119
Joined: Fri Jun 14, 2002 2:26 pm
Real name: Frank Sanns

Re: Fusor Plasma Extinguishing

Post by Frank Sanns »

The voltage, current, pressure, and fill gas are not independent of each other. Change one and the others will change.

It is not possible to say that you want a particular voltage and current at a given pressure. It just doesn't work that way.

There also is not really a minimum current (within reason) to keep the plasma alive. I can run star mode with air or deuterium at just 1 ma. Admittedly, running the grid red hot with more current does stabilize the plasma and allow for a different set of operating parameters but it is not necessary in a small or large chamber.

One thing that I found helpful to keep plasma lit is the hundreds of needle points in my outer grid (lattice grid). You can see the point ionizations there during a Fusor run. This seems to give a little more slope to the extinction curve rather than an abrupt end. This is also true if you run ion guns. None of this is necessary though as a smooth chamber with a grid does pretty well on its own.

As suggested by Liam, start with higher pressure and let the voltage rise at constant current. A dirty chamber will extinct the plasma at a low voltage compared to a clean and conditioned one. Don't expect to hit 50kv and clean star mode on your first start up. I may or may not happen. Often a dirty chamber will extinct around 15kv no matter what you do. Do some glow cleaning and let that voltage rise at constant current and you will get there.
Achiever's madness; when enough is still not enough. ---FS
We have to stop looking at the world through our physical eyes. The universe is NOT what we see. It is the quantum world that is real. The rest is just an electron illusion. ---FS
User avatar
Anze A Ursic
Posts: 157
Joined: Thu Mar 10, 2022 9:28 pm
Real name: Anze A Ursic

Re: Fusor Plasma Extinguishing

Post by Anze A Ursic »

But letting the voltage rise at constant current inherently means I lower the pressure right? Because the second I raise voltage, current jumps, so I lower the pressure to bring it down. And I'll make it glow, but like I said, right now the main ionization region is around the wires carrying current but it starts to shift towards the main cathode grid as pressure lowers.
Also, should I just switch to Deuterium instead of nitrogen? I wanted to run nitrogen to see if there are instabilities and to burn off junk from my device. Or should I try seeing if I can get star mode with nitrogen first?
User avatar
Richard Hull
Moderator
Posts: 14992
Joined: Fri Jun 15, 2001 9:44 am
Real name: Richard Hull

Re: Fusor Plasma Extinguishing

Post by Richard Hull »

You do what you have to do to make it go and stay lit. If that means lowering the pressure then do so. Did you read my recent FAQ on conditioning?
It is complete with two PDFs, the second with complete hard date taken over my conditioning pressure, voltage, current, results....

viewtopic.php?t=14505

Richard Hull
Progress may have been a good thing once, but it just went on too long. - Yogi Berra
Fusion is the energy of the future....and it always will be
The more complex the idea put forward by the poor amateur, the more likely it will never see embodiment
User avatar
Anze A Ursic
Posts: 157
Joined: Thu Mar 10, 2022 9:28 pm
Real name: Anze A Ursic

Re: Fusor Plasma Extinguishing

Post by Anze A Ursic »

Richard, in your paper you state conditioning shouldn't be done if the leak rate is more than 1mTorr/minute. Right now, my leak rate is higher. For example, this is my leak rate if I shut all the valves in my system completely, including the valve to my turbo.

1min: 8.5E-4
2min: 2.2E-3
3min: 4.9E-3
4min: 5.8E-3
5min: 6.7E-3
6min: 7.6E-3
7min: 8.5E-3

I guess at some points the leak rate is higher and at others its lower, but I'm curious if this is still too high?

Anze
User avatar
Richard Hull
Moderator
Posts: 14992
Joined: Fri Jun 15, 2001 9:44 am
Real name: Richard Hull

Re: Fusor Plasma Extinguishing

Post by Richard Hull »

Your leak rate does drop over time. I think if you keep working at it with deuterium you will condition. However with a giant chamber that is very costly in deuterium. Fighting a leak with costly gas is not a real solution unless you have a gas supply that matches your chamber size and money to follow that issue.

As I state, try and hit a well sealed chamber. My chamber size is well within my budget to supply D2 and I run it as in the conditioning paper never more than 3 times a year. That means not a lot of D2 wasting and thus, no real impromptu demos to interested parties especially of individuals.

The first paper notes that as I condition the pressure will rise over each run. Given a fixed leak rate and loading of the walls My D2 to air ratio gets better with each run.

You are an unusual case due solely to the size of your chamber and the leak rate. In short, you are on your own here. This is the very reason that I ditched the fabulous 16" chamber I obtained 10 years ago and never even went forward with it. Images of this chamber I paid $400 for in 2011 are below. Lovely and fantastic, but a bridge too far. A deuterium gobbler. This beauty came with (2) 2.75 viewports, a 4.5" view port, A TC gauge, and a naked ion gauge. Also, a number of the ports already blanked off, and one with a short SS welded flex bellows. I sold it a year after I bought it at a great profit. Click images to enlarge.

Richard Hull
Attachments
P2120002.JPG
P2120001.JPG
Chamber1.jpg
Progress may have been a good thing once, but it just went on too long. - Yogi Berra
Fusion is the energy of the future....and it always will be
The more complex the idea put forward by the poor amateur, the more likely it will never see embodiment
User avatar
Anze A Ursic
Posts: 157
Joined: Thu Mar 10, 2022 9:28 pm
Real name: Anze A Ursic

Re: Fusor Plasma Extinguishing

Post by Anze A Ursic »

Well the other issue is I am leaving the university next wednesday but I'll probably be back in the fall. So I have to decide now whether to stop conditioning now and save a ton of D2 or keep trying to get fusion by next wednesday but as it seems now, that probably won't be possible. Dang it.
User avatar
Anze A Ursic
Posts: 157
Joined: Thu Mar 10, 2022 9:28 pm
Real name: Anze A Ursic

Re: Fusor Plasma Extinguishing

Post by Anze A Ursic »

Richard,

I had a quick question regarding your "Conditioning a leaky fusor" and your "Advanced technical discussion" documents. I understand how wall loading works with deuterium ions, but I am curios what voltage is required to wall load. Meaning, if I am trying to prepare for an actual deuterium fusion run (20kV +), does it make sense to ignite and burn deuterium plasmas in the 15-17kV range days prior to "condition" the chamber, or wall load it? I have recently moved my experiment to a new chamber, which allows me to burn plasmas at 20kV easily for several minutes at a time, and my actual fusion run will be this Friday 10/4/2022 when the safety people will be here. Until then I cannot achieve fusion, but would like to prepare the chamber for it as best I can.

Thanks!

Anze
User avatar
Richard Hull
Moderator
Posts: 14992
Joined: Fri Jun 15, 2001 9:44 am
Real name: Richard Hull

Re: Fusor Plasma Extinguishing

Post by Richard Hull »

No wall will load with deuterium ions. The wall loads with fast neutral atoms of Deuterium. These real atoms of gas are interstitially loaded within the wall by violent collision.

Wall loading in the fusor, as constructed, naturally occurs in the process of normal operation! Loading arises and relies totally on failed fusion! When all deuterons fail to collide and fuse, they typically fail to recirculate due to recombination with an electron. They are now a fast moving neutral atom of deuterium and the direction they were headed when they recombined is fixed. (neutral atoms will not respond to an electric field) Assuming the direction remains fixed, (no reionization in velocity space), 100% will crash into the wall and some percentage will go a micron or fraction thereof to interstitially be bound into the wall's metal lattice.

This is easily proven! Vacuum out a fusor chamber. Valve the chamber off to hold this vacuum, (assume no leaks). Let in 15-50 microns of deuterium gas. Carefully bring the voltage up. You might light off a plasma at 3kv. Keep raising the voltage slowly while monitoring the current so as to not melt the grid. You will slowly note that the pressure in the chamber is dropping and you might easily be running safely at 15kv in 15 minutes. This will continue. The gas atoms are slowly being locked into the walls via bombardment. In early x-ray tubes this was noted as the tube "going hard".

In these early tubes, special gas absorber arms, (palladium), on those tubes locked in the gas and a hard tube failed to perform to specification. A special technician was needed to hook the HV into the metallic collector in this arm. Bombardment heated the absorber to release the gas, thus, re-softening the vacuum, restoring tube to functionality.

Modern hydrogen thyratrons still use this reservoir idea. You must first warm the filament in the tube and at the same time heat the reservoir which "gases" the tube for proper operation.

In the end, all metals absorb to a greater or lesser degree, hydrogen gas. In certain situations this gassing and out gassing can be controlled and used to advantage. Likewise, the horrid condition of hydrogen embrittlement can weaken the metal absorbing it to the point where it is structurally weakened and ruined.

Richard Hull
Progress may have been a good thing once, but it just went on too long. - Yogi Berra
Fusion is the energy of the future....and it always will be
The more complex the idea put forward by the poor amateur, the more likely it will never see embodiment
User avatar
Anze A Ursic
Posts: 157
Joined: Thu Mar 10, 2022 9:28 pm
Real name: Anze A Ursic

Re: Fusor Plasma Extinguishing

Post by Anze A Ursic »

So based on this answer, I assume even sub-fusion voltage plasmas do this, correct? Going further - is it even better to do this in sub-fusion voltage plasmas since in that case 99.99% of deuterons fail to fuse during collision and thus more of them recombine and load the wall? Perhaps my thinking is flawed here.
User avatar
Richard Hull
Moderator
Posts: 14992
Joined: Fri Jun 15, 2001 9:44 am
Real name: Richard Hull

Re: Fusor Plasma Extinguishing

Post by Richard Hull »

One might try this, as I suggested. Vacuum the system out. Valve off the fusor. Pressurize with far more deuterium than a run gas. (50-100 microns). This will not allow high voltage to be achieved without melting the grid, but will bombard the shell with the gas. You know you are loading as the pressure drops. As I noted.

Richard Hull
Progress may have been a good thing once, but it just went on too long. - Yogi Berra
Fusion is the energy of the future....and it always will be
The more complex the idea put forward by the poor amateur, the more likely it will never see embodiment
Post Reply

Return to “Fusor Construction & Operation (& FAQs)”