CW, plasma, Paschen curve

Current images of fusor efforts, components, etc. Try to continuously update from your name, a current photo using edit function. Title post with your name once only. Change image and text as needed. See first posting for details.
Gunnar Mein
Posts: 40
Joined: Sat May 29, 2021 12:24 pm
Real name: Gunnar Mein
Location: Kirland, WA

CW, plasma, Paschen curve

Post by Gunnar Mein »

Here are some recent pictures from our project.

First, the repaired Cockcroft-Walton multiplier. It normally lives in oil in a fish tank. It is our second CW (and this is Gavin's third board design), and designed to deliver 10 mA at 35 kV. We had originally used polystyrene-film capacitors for it, but we managed to lightly toast 2 of them just before COVID. When we figured this out in late April of this year, we also found they are no longer made, as polystyrene film was not a profitable product for anyone. We were able to find 2 Mylar-paper capacitors with the same specs, and "sort of" work them into the existing board. The colored zip ties in the pic are only to keep track of the capacitors during the repair. (Edit: Since people are pointing it out, we are aware that the diode templates on the board point the wrong way. The soldered diodes, however, point correctly. Thanks!)

Image

Next up, our best milestone so far - minutes of plasma. We can pretty much reproduce this at will now. No, we are not applying for the plasma club yet, since we are lacking a high-side current measurement device (working on it). Sorry for our limited view port - we used to have a nice one, but that's where the turbo pump is now, so all we have left is this tiny window. But we do have an acrylic bottom on the chamber. and might mount a camera underneath.

Image

Lastly, a plot of V over P of our traversal of the Paschen curve from 865 and 2090 V, made by our mentor Dr. Whitmer from our log files with Mathematica.

Image

Some more overall pictures are coming tomorrow.
Last edited by Gunnar Mein on Mon May 31, 2021 12:43 am, edited 1 time in total.
User avatar
Joe Gayo
Posts: 404
Joined: Sun Jan 06, 2019 9:34 pm
Real name: Joe Gayo
Location: USA

Re: CW, plasma, Paschen curve

Post by Joe Gayo »

"High-side current measurement"
Although you could float a battery-powered or analog meter that would make data collection challenging (maybe wireless, but unnecessarily complicated) since this will be at the cathode potential. My suggestion is to interrupt the return path of your multiplier with a 1kohm resistor and measure the voltage across it. This will be a low-side, ground-referenced measurement of effectively the current to the cathode.
User avatar
Joe Gayo
Posts: 404
Joined: Sun Jan 06, 2019 9:34 pm
Real name: Joe Gayo
Location: USA

Re: CW, plasma, Paschen curve

Post by Joe Gayo »

On another note, multiplying 60Hz AC from an NST and expecting 34kV with film/ceramic capacitors and supporting milliamperes of current is not feasible. Have you used LTSpice to simulate or looked up the multiplier equations that account for frequency?
Last edited by Joe Gayo on Mon May 31, 2021 12:59 am, edited 1 time in total.
Gunnar Mein
Posts: 40
Joined: Sat May 29, 2021 12:24 pm
Real name: Gunnar Mein
Location: Kirland, WA

Re: CW, plasma, Paschen curve

Post by Gunnar Mein »

Thanks Joe, will look into all of these comments. I am not the HV guy, I can only pretend. Yes, LTSpice was done at the time ... 2 years ago ... of course that student has since graduated, I will ask our mentor to confirm. By "return path", do you mean the middle (ground-level) connection on the input side?
Gunnar Mein
Posts: 40
Joined: Sat May 29, 2021 12:24 pm
Real name: Gunnar Mein
Location: Kirland, WA

Re: CW, plasma, Paschen curve

Post by Gunnar Mein »

The current plan for the current measurement is to measure the actual output, keep an Arduino Nano floating at high potential to ADC it (gets 5V through induction loops separated by the fish tank wall), and relay this to the "ground-side" electronics through an optical cable.
User avatar
Joe Gayo
Posts: 404
Joined: Sun Jan 06, 2019 9:34 pm
Real name: Joe Gayo
Location: USA

Re: CW, plasma, Paschen curve

Post by Joe Gayo »

Yes. The green cable.

As mentioned a high-side current measurement is unnecessarily complicated.
Gunnar Mein
Posts: 40
Joined: Sat May 29, 2021 12:24 pm
Real name: Gunnar Mein
Location: Kirland, WA

Re: CW, plasma, Paschen curve

Post by Gunnar Mein »

One more thing, Joe - I do remember us testing this board last week with a 3.3 MOhm HV resistor as a load. I will confirm that we really went that low on the load resistor. But anyway, it did fine.
User avatar
Dennis P Brown
Posts: 3159
Joined: Sun May 20, 2012 10:46 am
Real name: Dennis Brown

Re: CW, plasma, Paschen curve

Post by Dennis P Brown »

Nice picture; so, what is the voltage output of the supply transformer? Also, what frequency are you using?
Gunnar Mein
Posts: 40
Joined: Sat May 29, 2021 12:24 pm
Real name: Gunnar Mein
Location: Kirland, WA

Re: CW, plasma, Paschen curve

Post by Gunnar Mein »

This gets input from an NST that produces up to 60 mA at over 15 KV peak-to-peak, 60 Hz, and the CW output under the 10 mA load was something like 32 KV. Without load, it makes it to 34 KV.
Last edited by Gunnar Mein on Mon May 31, 2021 3:35 pm, edited 1 time in total.
User avatar
Richard Hull
Moderator
Posts: 14991
Joined: Fri Jun 15, 2001 9:44 am
Real name: Richard Hull

Re: CW, plasma, Paschen curve

Post by Richard Hull »

No neon sign transformer will ever deliver rated voltage at rated current! Not even close! This is dealt with a a nearly 20 year old FAQ here. I wrote the paper in 2003, but updated it 3 times.

viewtopic.php?f=29&t=10333

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
Gunnar Mein
Posts: 40
Joined: Sat May 29, 2021 12:24 pm
Real name: Gunnar Mein
Location: Kirland, WA

Re: CW, plasma, Paschen curve

Post by Gunnar Mein »

Here is the LTSpice result of our CW under a 3 MOhm load, from 2019

Image
User avatar
Richard Hull
Moderator
Posts: 14991
Joined: Fri Jun 15, 2001 9:44 am
Real name: Richard Hull

Re: CW, plasma, Paschen curve

Post by Richard Hull »

I still worry that you case of the transformer may be floating if you are using the two knobs and a simple voltage doubler. Again read the FAQ I supplied in the prior post. A 15kv transformer's metal case can be lethal if not grounded!! The transformer is set up to be, (left knob 7.5kv... zero volts @ grounded case center tap....7.5kv right knob). Voltage doubling across the knobs and grounding the plus of the doubler to use the negative as hot will leave the metal transformer case at a deadly 7.5kv to any one touching the case!

A few fool-hearted souls have done this but wisely placed the transformer in a thick plastic case where no one could touch the case during operation. One still runs the risk of the secondary to primary insulation breaking down as the core is at 7.5kv. By reading the FAQ specific to the neon transformer you will see the current at voltage is grossly reduced. Add to this fact that the doubler has low capacitance in its storage filters and the current available really plunges even further to almost nothing.

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
Gunnar Mein
Posts: 40
Joined: Sat May 29, 2021 12:24 pm
Real name: Gunnar Mein
Location: Kirland, WA

Re: CW, plasma, Paschen curve

Post by Gunnar Mein »

I will go and read that ... however I can assure you that the center tap of the secondary on the NST is grounded. What gives you the impression that it might not be? Well I guess I can answer that, you doubt that it would output the current we are seeing if it were. (I missed your last post as it ended up on the previous page.)
User avatar
Joe Gayo
Posts: 404
Joined: Sun Jan 06, 2019 9:34 pm
Real name: Joe Gayo
Location: USA

Re: CW, plasma, Paschen curve

Post by Joe Gayo »

@gunnar

Heed the words of Richard about safety.

Also, my comment was about multiple milliamperes, like what will be needed ... 5-10mA ... I still suspect you will see a sizable drop ... ~30%+
Gunnar Mein
Posts: 40
Joined: Sat May 29, 2021 12:24 pm
Real name: Gunnar Mein
Location: Kirland, WA

Re: CW, plasma, Paschen curve

Post by Gunnar Mein »

Image
Gunnar Mein
Posts: 40
Joined: Sat May 29, 2021 12:24 pm
Real name: Gunnar Mein
Location: Kirland, WA

Re: CW, plasma, Paschen curve

Post by Gunnar Mein »

@Joe, 32 KV over 3.3 MOhms = about 10 mA. What am I missing?
User avatar
Richard Hull
Moderator
Posts: 14991
Joined: Fri Jun 15, 2001 9:44 am
Real name: Richard Hull

Re: CW, plasma, Paschen curve

Post by Richard Hull »

I see the transformer is grounded in your image.

Are you floating the two doubler output leads?! If you properly ground the hot positive of the doubler, as you should for safety sake, the doubler might only give negative15kv or above output reference to ground. If you float both leads then you will, indeed, get 30kv or more no load across them but touching either lead will be lethal with reference to ground!

What has saved you might be that no one has touched anything while the power supply system is running and none of the floating leads is within arcing distance to true ground. In this manner, you only have one single lead to insulate and keep away from grounded objects.

Floating HV leads is a bad idea. You only want one single, deadly, hot lead in most all HV applications relative to ground.

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
Gunnar Mein
Posts: 40
Joined: Sat May 29, 2021 12:24 pm
Real name: Gunnar Mein
Location: Kirland, WA

Re: CW, plasma, Paschen curve

Post by Gunnar Mein »

@Richard, you can look at the CW picture at the start of this thread. Yes, only the negative output is used, the left and right outputs are floating. In oil. With nobody closer to the box than 6 ft, and even that is only when we are testing it outside of the normal setup, when nobody is closer than 40 ft, and I can shut down everything manually with an emergency switch outside of the room.
Gunnar Mein
Posts: 40
Joined: Sat May 29, 2021 12:24 pm
Real name: Gunnar Mein
Location: Kirland, WA

Re: CW, plasma, Paschen curve

Post by Gunnar Mein »

Reload of a log file from last week. This was not in the vacuum system, but in the lab with the 3.3 MOhm load resistor in oil. Note: The NST rms measure only one knob, to ground.

Image
Last edited by Gunnar Mein on Mon May 31, 2021 3:36 pm, edited 2 times in total.
User avatar
Joe Gayo
Posts: 404
Joined: Sun Jan 06, 2019 9:34 pm
Real name: Joe Gayo
Location: USA

Re: CW, plasma, Paschen curve

Post by Joe Gayo »

You mentioned you measured a current of 1mA. I agree 32kv across 3.3Meg ~ 10mA. If you actually measured 10mA, then I'm surprised the NST can source that current at the full 15kV p-p.
Gunnar Mein
Posts: 40
Joined: Sat May 29, 2021 12:24 pm
Real name: Gunnar Mein
Location: Kirland, WA

Re: CW, plasma, Paschen curve

Post by Gunnar Mein »

Gotcha. Typo in post. Fixed. Thanks!
User avatar
Richard Hull
Moderator
Posts: 14991
Joined: Fri Jun 15, 2001 9:44 am
Real name: Richard Hull

Re: CW, plasma, Paschen curve

Post by Richard Hull »

Yes, floating and safely out of the way in oil the full doubler will function just not much current at the doubled voltage unless lightly loaded. As most fusors are metal shelled, this would not be possible in a real fusor due to all the vacuum plumbing and all attached instrumentation being at real electrical ground.

We tend to think fusion functioning fusor here when talking HV and safety. Safety grounds abound and there is only one viper to worry about.

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
Dennis P Brown
Posts: 3159
Joined: Sun May 20, 2012 10:46 am
Real name: Dennis Brown

Re: CW, plasma, Paschen curve

Post by Dennis P Brown »

A 60 Hz NST will never delivery any significant voltage and current for a CW that drives a plasma; once you hook it up to any real device to generate a plasma, the voltage will drop like a rock and the cap's will delivery very little current. Yes, a scope with its high impedance will show a different story.

None of this matters to join the plasma club! The device will ignite and hold a plasma so it will function for that purpose. Just don't lose sight of the fact that a VM for a real fusor requires a very high frequency driver circuit to deliver real current at a high voltage.
Gunnar Mein
Posts: 40
Joined: Sat May 29, 2021 12:24 pm
Real name: Gunnar Mein
Location: Kirland, WA

Re: CW, plasma, Paschen curve

Post by Gunnar Mein »

@Dennis, I don't know what else to say - we measured 32 KV across a 3.3 MOhm load, that works out to 10 mA for us. There was no high-impedance scope involved, just that load resistor and the actual divider boards we use in real scenarios.

We are going run an LTSpice simulation to convince ourselves that the lowside return on the CW is sufficiently standing in for the highside current. We don't see how it accounts for losses, but we would love not to do the whole optical link to the Arduino and stuff. When that's done, and we have a working current measurement, we'll gather all the evidence to apply.

Today and tomorrow, I am working with 7 students on a couple of improvements - camera under chamber, new 5V offset supply for divider board, maybe shielded cases for some electronics. I will post more pictures.
User avatar
Richard Hull
Moderator
Posts: 14991
Joined: Fri Jun 15, 2001 9:44 am
Real name: Richard Hull

Re: CW, plasma, Paschen curve

Post by Richard Hull »

I went back to your introduction and you noted you have a "beefy" neon sign transformer. It would be great to know the name plate rating of your transformer. A stock non-beefy 15kv neon is rated at 30ma on the name plate. In the past, 15kv neon xfrmrs were made in 60ma and way back, 120ma current ratings. A 60ma version would support HV at a much higher current level and the very rare 120ma transformer would easily supply 20-30ma at nearly full voltage. Now that is beefy! Those ancient 120ma units were used to drive massive neon tube lengths in giant neon signage. Such signage is rare today and those massive hunks of iron have given way to custom designed electronic switching type supplies for modern neon. One day iron, shunted neon transformers will be talked about around campfires referring to the good old days.

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 “Images du Jour”