Some unexpected breakdown voltage measurements

For posts specifically relating to fusor design, construction, and operation.
Post Reply
User avatar
Maciek Szymanski
Posts: 193
Joined: Wed Nov 20, 2019 2:31 pm
Real name: Maciek Szymański
Location: Warsaw, Poland

Some unexpected breakdown voltage measurements

Post by Maciek Szymanski »

I’ve finally moved from building to experimenting with my garage fusor. (The vacuum system build is here: viewtopic.php?f=10&t=13708 and the power supply here: viewtopic.php?f=11&t=13873). As I mentioned earlier I’m interested in ionization mechanisms in fusors I’ve done some preliminary measurements to gather some reference data and mostly important to gain some experience with my setup and measurement system. To some extent the results has surprised me.

The experimental setup
The setup is fairly classical and straightforward. The fusor chamber is a grounded 100x100 mm stainless cylinder acting as an anode. The cylinder is closed from top by a 95mm glass viewport and on the bottom directly connected to the port of the vacuum system. On the periphery of the cylinder are 4 KF ports spaced by 90º. Three are 25mm and one 50mm. The KF50 port is used to connect the HV feed through with the cathode assembly. On one of the KF25 ports there is mounded the needle valve for pressure regulation. As the filling gas the environmental air was used.
The cathode is connected to the negative pole of the variable 5KV DC power supply. A variable bank of high wattage resistors may be connected in series with the supply and the fusor. The supply has built in voltage and current meters as well as provisions for connecting external measurement equipment. For the experiments I’ve used the two channel Siemens C1012 chart recorder to record independently the voltage and current.


8A1F7538-36E0-4788-8DF7-A20FFF84D567.jpeg
Overall view of the setup.


AD05B09A-90E7-459B-909A-EC28F06078C1.jpeg
Detail of the chamber assembly. The regulation valve is fitted with a micrometric stop.


F40540A2-0595-4368-9B4E-306BD10AF8A8.jpeg
Alternatively the stop may be replaced with a dial indicator. It has proven a reliable means of repeatable adjusting of the desired pressure.


25F4459F-5699-4172-A1C5-5A5416B733B8.jpeg
The cathode assembly. The stalk is insulated with the glass tubes. The cathode is constructed from individual stainless steel wires crimped into the brass holder.


5A6A4E13-F06C-42F0-8262-16E73CCD758D.jpeg
Overall diagram of the circuit. RB is the internal primary ballast resistor, R0 is the additional HV side ballast.

Pressure measurement
For the current measurements I’ve used only the rotary vacuum pump and the low vacuum pirani gauge (down to 1e-3 Torr) for the measurements. As the pirani gauge is sensitive to the gas temperature and tends indicate higher pressures during the discharge all pressure readouts were taken with the HV switched off and the system cooled down. Following are the photographs of the discharge at various pressure levels.


A98D7F4B-9028-4B8B-9A78-A6BC1E3233AE.jpeg
The discharge at 1.2e-2 Torr in air. The glass luminescence due to electron bombardment clearly visible.


752E1BB7-3264-46A5-BBCB-E555B90EC665.jpeg
The discharge at 4.5e-2 Torr in air.


E46990BA-F929-419F-B3F4-E9ABD8CDF465.jpeg
The discharge at 3e-1 Torr in air.


Voltage - current characteristics
This was a straightforward measurement. At the pressure of 1e-2 Torr (the lowest pressure achievable with the rotary pump only) the power supply was adjusted in 10% steps (the variac setting) up to 80% and then down to 0%. At each step the voltage and current was recorded by the chart recorder for 5 seconds.


27C90078-A162-4288-BB9B-DDE72C3140AB.jpeg
The recording of the U-I characteristics at 1e-2 Torr. Red line - voltage (57 V/cm), green line - current (full scale 10V, 0.2V/mA). Please note, that recording is from right to left!



The actual values were calculated as the mean of the values at the start and at the end of each step. Here the results were not surprising. The resistance was getting lower at higher discharge currents and the typical glow discharge hysteresis of the U-I curve has been obtained.


FE552ABA-EC29-404C-8472-3E7AABC79274.jpeg
The current-voltage curve. Pressure 1e-2 Torr, primary ballast RB=68Ω, secondary ballast R0=0Ω. Percent values indicate the PSU variac setting, arrows - direction of voltage change.

193D2EF8-31AE-4BD1-BFBD-F777A4DC2D03.jpeg
The resistance-current curve. Pressure 1e-2 Torr, primary ballast RB=68Ω, secondary ballast R0=0Ω. Arrows indicate - direction of voltage change.


Breakdown voltage vs pressure
In the next step I’ve made measurement of the breakdown voltage at various pressures. First the desired pressure was adjusted with the needle valve with the PSU turned off. Then the voltage was slowly raised while recording the voltage and current. The current was raised to few mA then the voltage lowered back to 0. The breakdown point was very clearly visible in the current recording and usually also pronounced in the voltage curve as a small drop or a short flat segment.



9F9FD57C-9B35-4955-A0CA-0027E7120EA2.jpeg
Typical breakdown voltage recording. Cathode 35/0.8, pressure 1.2e-2 Torr. Red line - voltage (40 V/cm), green line - current (0.2 mA/cm). Recording direction from right to left. Also note, that the current pen lags about 5mm behind the voltage pen. The current raise point is very clearly visible and coincident with the voltage drop.

The measurements were made for two cathodes. Both were constructed from three rings of stainless steel wire spaced at 120º. The diameter of the rings was 35mm. Firs cathode was constructed from 0.8mm wire and the second from 0.6mm. For the 0.8mm I’ve repeated the measurements twice in few days and got the same results. The 0.6mm cathode was measured only once, so probably I should repeat it.


2CDD28AE-8A24-4411-B16F-0FED10F01A19.jpeg
The breakdown voltage vs pressure. Continuous line for 0.8 mm wire, dashed for 0.6 mm.

The first curve for 0.8mm cathode looks quite ok for me. It is quite smooth, steep on the low pressure side and seems like it is approaching Paschen’s minimum somewhere on the high pressure side. But the one for 0.6mm seems quite surprising. First of all I would expect the thinner wire to have LOWER breakdown voltage due to higher E - field strength. And for lower voltages it apears to have higher breakdown voltage. And the second surprise is the unexpected drop somewhere around 7e-2 Torr to less than 10V! Any ideas?
“Begin at the beginning," the King said, very gravely, "and go on till you come to the end: then stop.” ― Lewis Carroll, Alice in Wonderland
User avatar
Richard Hull
Moderator
Posts: 14992
Joined: Fri Jun 15, 2001 9:44 am
Real name: Richard Hull

Re: Some unexpected breakdown voltage measurements

Post by Richard Hull »

Could some sort of unseen heating of the .6 wire have an effect not evidenced in the .8?? What was the current draw for each grid wire? The heating at glow start or breakdown would be different based on the joule heating, but one would have to know the volts and current for each grid. You did note that the breakdown voltage was greater for the .6mm. Heat differentials would have more electron emission at higher joule energy in the .6 wire as it could not conduct away heat as fast as the .8, even if they have the same joule energy applied. Just a thought.

You are doing fabulous work and learning a lot about operation and plasma hysteresis. All of this becomes a lot more tricky at high startup energies in the >20kv applied area. I spent 2 years using only air in a bell jar system that ultimately would hit 15kv. The bell jar is especially instructive in that with a wire geodesic anode, I would find the break-over starting voltage to a glow then have to lower the voltage to extinguish the glow mode. I would then bring the voltage back up to almost at break over into glow mode. I would then take a 5/8-inch 14mm diameter Teflon rod 14 inches long and rub it with rabbit fur, charging the rod. A sudden push of the charged rod at the bell jar would break the system into glow mode. This amused me and showed how critical the electrostatic field is in the system. This could not, in theory, happen in a solid metal sealed fusor anode. ( Internal Field shielded)

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
Maciek Szymanski
Posts: 193
Joined: Wed Nov 20, 2019 2:31 pm
Real name: Maciek Szymański
Location: Warsaw, Poland

Re: Some unexpected breakdown voltage measurements

Post by Maciek Szymanski »

I don't suppose that the cathode temperature is an issue here. The measurements were made with the grid cold and after breakdown I have not increased the current beyond 2 mA. I'm suspecting surface (deposits etc) of the wire. The 0.6 cathode has seen less discharges than 0.8, but anyway before the measurements I've kept it for some time orange hot in the discharge to get rid of any dirt. Or it is a measurement system fault. Anyway I've to repeat the measurements with more points in the suspected pressure region.
And yes - I would like to have a bell jar system if only for the looks of it. Unfortunately the pyrex bell jars are prohibitively expensive in my country. But I keep my eye on the surplus bell jars with the rotary pump once used in school demonstrations, but they are quite rare an in most cases the bell jar is missing or broken.
“Begin at the beginning," the King said, very gravely, "and go on till you come to the end: then stop.” ― Lewis Carroll, Alice in Wonderland
User avatar
Richard Hull
Moderator
Posts: 14992
Joined: Fri Jun 15, 2001 9:44 am
Real name: Richard Hull

Re: Some unexpected breakdown voltage measurements

Post by Richard Hull »

By keeping the grid at orange heat you are doing what all neon sign makers do. It is called "Bombarding". It cleans and conditions electrodes to run in a gas environment. When you remove the electrode from vacuum, if it is out for a long period it may need a bit of bombardment again, but it is never as bad as the first pass at bombardment with raw new grid material.

As you are able to pour in more energy at higher voltages and higher currents, thing will get a bit more hectic. Your slow process is to be applauded. Few people stop on their way to fusion, long enough to smell the roses using air and studying the plasma characteristics in gases.

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
Rich Feldman
Posts: 1470
Joined: Mon Dec 21, 2009 6:59 pm
Real name: Rich Feldman
Location: Santa Clara County, CA, USA

Re: Some unexpected breakdown voltage measurements

Post by Rich Feldman »

Applause for using a paper chart recorder, and hand-drawn graphs!

My high school physics teacher taught that when "manually" reading instruments in the lab,
it's a good idea to chart the point immediately, instead of just write a number in a notebook.
That way, transcription blunders are caught right away,
and unexpected shapes in the curve can immediately be confirmed and invite new data points.

Sometimes I get to practice that even at work, when an instrument needs to be read by eye.
Type reading into spreadsheet program, set up so it immediately appears in an XY chart.
In many cases 10 points are enough to tell the story, and 100 or 1000 computer-recorded points don't add much except glitz to the presentation.
OTOH, the 100 or 1000 point curve may reveal a wiggle that would otherwise have been overlooked. Sometimes that's led to historically important discoveries.
All models are wrong; some models are useful. -- George Box
User avatar
Maciek Szymanski
Posts: 193
Joined: Wed Nov 20, 2019 2:31 pm
Real name: Maciek Szymański
Location: Warsaw, Poland

Re: Some unexpected breakdown voltage measurements

Post by Maciek Szymanski »

I really hate to go trough many levels of options and sub-options to get the right looking chart in a computer program. By hand I can add any marks and annotations with just one tool :-) And I can have a really big screen area to view all the data at once:

29D7A216-DAA0-41B7-8D54-E853108235C2.jpeg

The surprising glitch in the breakdown voltage curve was probably due to some outgassing of the cathode despite the conditioning I’ve done. Today I have made another series of measurements and got a more expected curve. But still for the low pressures the thinner grid has higher breakdown voltage.


41BE3F3D-3CC3-4D5A-9E39-6D0A9159E256.jpeg
Breakdown voltage vs pressure (ambient air). Continuous line - 0.6mm wire cathode, measurement on March 17th. Dashed line - 0.6mm wire cathode, measurement on March 14th. Dashed - dotted line 0.8mm wire cathode.

Anyway the kink in the 2e-2 to 3e-2 Torr seem to be repeatable for both wire diameters.
“Begin at the beginning," the King said, very gravely, "and go on till you come to the end: then stop.” ― Lewis Carroll, Alice in Wonderland
User avatar
Maciek Szymanski
Posts: 193
Joined: Wed Nov 20, 2019 2:31 pm
Real name: Maciek Szymański
Location: Warsaw, Poland

Re: Some unexpected breakdown voltage measurements

Post by Maciek Szymanski »

Things are getting more interesting. I've made a solid aluminum cathode. I will make the drawing later, but the cathode is a ball of 35mm diameter with five tapered holes at right angles. The hole diameter at the ball center is 10mm and the taper angle 15º.


85F2C674-C91B-4705-A06D-DCBB892DF9BA.jpeg

From the first measurements the cathode tends to have much more "soft" brekdown characteristics. What is interesting at higher pressures the breakdown occurs at extremaly low potential. At lower pressures it it much more gradual compared to the wire cathode and allows operation at much higher voltages and currents. I've done the measurements for the discharge U-I characteristics, but I've to process the data.


63D0E4BD-4178-457A-AE68-5915FB7FA5C1.jpeg
The discharge 3/21.03.2021. Pressure 1.5e-2 Torr (air) voltage 1,1kV current >100mA, inernal balst RB=22 Ω. Visible traces of electron induced glass luminescence.


8BA32F75-8F95-4C37-AE42-9778818016FA.jpeg
Breakdown voltage vs pressure (air). Continuous line - solid Al cathode Φ35mm. Dashed line - voltage at which the voltage-current caracteristics linearizes. Dotted dashed line - wire catode, inox Φ35mm/Φ0.8mm (reference). For pressures below 3e-2 Torr the breakdown for the Al cathode is not abrupt like for the wire one, but the current rises slowly until the "linearization" potential is reached.


C14547ED-EE9E-4495-A85D-600A66BF22C0.jpeg
Current vs voltage in the breakdown region. Pressure 1.2e-2 Torr (air). Continuous line - solid Al cathode Φ35mm (interpolated form dischachge 5/21.03.2021 at 2e-2 Torr and 4/21.03.2021 at 1.5e-2 Torr). Dashed line -solid Al cathode Φ35mm (discharge 8/14.03.2021. Please note the gradual increase of the current for the solid cathode.
“Begin at the beginning," the King said, very gravely, "and go on till you come to the end: then stop.” ― Lewis Carroll, Alice in Wonderland
User avatar
Maciek Szymanski
Posts: 193
Joined: Wed Nov 20, 2019 2:31 pm
Real name: Maciek Szymański
Location: Warsaw, Poland

Re: Some unexpected breakdown voltage measurements

Post by Maciek Szymanski »

I've made the comparison of the voltage-current characteristics of the solid and wire cathodes.


6588B498-491F-48FE-997E-E6835D4DE6AF.jpeg
Discharge current vs voltage. Air at 1e-2 Torr. Solid line - solid aluminum cathode Φ35mm. Dashed line - inox wire cathode Φ35mm/Φ0.8. Secondary side ballast R0=0Ω, primary side ballast RB=68 Ω.
Last edited by Maciek Szymanski on Sat Apr 03, 2021 8:17 am, edited 1 time in total.
“Begin at the beginning," the King said, very gravely, "and go on till you come to the end: then stop.” ― Lewis Carroll, Alice in Wonderland
User avatar
Richard Hull
Moderator
Posts: 14992
Joined: Fri Jun 15, 2001 9:44 am
Real name: Richard Hull

Re: Some unexpected breakdown voltage measurements

Post by Richard Hull »

Good curves, but an awful large amount of current at 10 microns with only 2000 volts. Interesting...

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
Maciek Szymanski
Posts: 193
Joined: Wed Nov 20, 2019 2:31 pm
Real name: Maciek Szymański
Location: Warsaw, Poland

Re: Some unexpected breakdown voltage measurements

Post by Maciek Szymanski »

The voltage was measured without the secondary side ballast, so it it just the voltage across the fusor. But the figures are similar to what I got with the demo fuosr at work. Both with the NST supply as well as with 50kV Glassman. At pressures above 10 microns it was impossible to bring the voltoge to more than 1-2 kV.
It would be interesting to put together the data from the forum and do some comparisons. I think there is a lot of measurements made by various people over the years spread through many threads. It is some work, but maybe worth the effort.
“Begin at the beginning," the King said, very gravely, "and go on till you come to the end: then stop.” ― Lewis Carroll, Alice in Wonderland
User avatar
Richard Hull
Moderator
Posts: 14992
Joined: Fri Jun 15, 2001 9:44 am
Real name: Richard Hull

Re: Some unexpected breakdown voltage measurements

Post by Richard Hull »

Good thought, but a lot of those reports have one or more critical measurements missing. Volts and current but no pressure. Volts and pressure but no current, etc.. My ballast is in the oil in my X-ray tank and thus all of my measurements of voltage have been after the ballast and the current in the grounded leg of the X-ray xfrmr. Thus, when I report fusion at 41kv and 10ma, it is the real deal. The DC voltage internally before my 62kohm ballast is 41 + (1e10-2 X 6.2e10 4) or 41.62kv.

It is a pity, but many fusor folks over the last 20 years have not always availed themselves of full or highly accurate instrumentation. Thus the useful union of useful data is limited and problematic.

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)”