Joe Ballantyne fusor V2. (Low cost fusor.)
-
- Posts: 557
- Joined: Tue Jul 20, 2010 4:08 pm
- Real name: Joe Ballantyne
- Location: Redmond, WA
Re: Joe Ballantyne fusor V2. (Low cost fusor.)
So during the last month or so, I fired the fusor up a couple of times, and also spent some time working on a low cost ammeter.
I found that the ignition rod feedthrough that I thought was leak free, was only leak free initially.
I left the fusor under vacuum for a month or so, and when I decided to fire it up, the vacuum level was a lot higher than I expected.
Furthermore, it didn't pull down as low as it had when the feedthrough was leak free, so I ran a 1 minute pressure rise leak test after it got down to 8 microns or so, and indeed, the fusor was leaking at about 1 micron a second or so.
So the ignition rod feedthrough after sitting under vacuum for a few weeks, started leaking as badly or worse than the spark plug feedthrough.
IMO that means there is not a whole lot of reason to use it, instead of the spark plug, as the ignition rod costs about $50 instead of $15. Certainly not really worth the additional expense, given that it only gives you an extra 3-5kV in upper bound cathode voltage levels over the sparkplug.
Sparkplug will go to 22kV or so, ignition rod will do 25kV or so, and they both leak eventually.
Joe.
I found that the ignition rod feedthrough that I thought was leak free, was only leak free initially.
I left the fusor under vacuum for a month or so, and when I decided to fire it up, the vacuum level was a lot higher than I expected.
Furthermore, it didn't pull down as low as it had when the feedthrough was leak free, so I ran a 1 minute pressure rise leak test after it got down to 8 microns or so, and indeed, the fusor was leaking at about 1 micron a second or so.
So the ignition rod feedthrough after sitting under vacuum for a few weeks, started leaking as badly or worse than the spark plug feedthrough.
IMO that means there is not a whole lot of reason to use it, instead of the spark plug, as the ignition rod costs about $50 instead of $15. Certainly not really worth the additional expense, given that it only gives you an extra 3-5kV in upper bound cathode voltage levels over the sparkplug.
Sparkplug will go to 22kV or so, ignition rod will do 25kV or so, and they both leak eventually.
Joe.
Last edited by JoeBallantyne on Mon Sep 16, 2024 10:17 pm, edited 1 time in total.
-
- Posts: 557
- Joined: Tue Jul 20, 2010 4:08 pm
- Real name: Joe Ballantyne
- Location: Redmond, WA
Re: Joe Ballantyne fusor V2. (Low cost fusor.)
I wanted to add a low cost mA meter kit to my selection of kits for the low cost fusor so I spent some time working on that.
Originally on my version of the low cost fusor, I was using a Sensitive Research 0-30mA meter that I bought for about $45.
It works great, but the scale was labeled in 3mA increments, which was a little odd, and it turns out that the 30mA range really just means that its kinda hard to read, as the current really never goes above 10mA or so with the precipitator supply, and most of the time is down at less than 3mA.
So, I bought a couple Simpson 25mA meters, and put one of them in a project box, and it worked fine, and had numbered scale markings every 5mA which is a bit more conventional, but again it was a bit hard to use due to typical precipitator currents being 3mA or less. Also, you generally can't find analog Simpson meters on eBay for less than $25 or so, and they also all tend to be one off listings. So they aren't the optimal choice for a kit either.
So then I figured I would pick a range that worked well with the precipitator, and decided on 15mA. Because the precipitator will limit its output before it hits 15mA, and that will give much better resolution at the low end. 10mA would work also, but I figure that sometimes you do get more than 10mA output (like 11-12mA) and I wanted to minimize the risk of above range current damaging the meter. So I decided on 15mA as a reasonable compromise.
I found some inexpensive 15mA meters on eBay, and bought a number of them, as well as some project boxes that the meters fit in very nicely.
I expected the meters I bought to work just like the Simpson meters I purchased, where you just put them in the box, connect them up, and they work.
But NO.
When I tried an air plasma run with the new meter, and had the precipitator at its lowest output voltage, and ramped the pressure up, I fully expected the meter to register the expected 6mA or so that the supply puts out when the plasma lights off in that configuration. Instead the meter instantly pegged as soon as the plasma lit.
Turns out the meters I bought, have a 15mA scale, but do not have the proper shunt resistance in them, and are actually 1mA full scale meters.
I measured the 15mA analog meter resistance as 47ohms on the 2000ohm range of a Harbor Freight multimeter. When I did so, the meter showed a mA reading of 0.5mA on the face - so about 1/30th of full scale - which would be a nominal 33uA current for a full scale 1mA meter. I put one of the 50uA meters I use in my HV meter kit in series, while making the measurement again, and measured a current of 25uA - lower because the 50uA meter raised the resistance of the circuit from 47 ohms up to more than 1kohms. So indeed, the meter reads full scale at 1mA current not 15mA.
So the proper shunt resistance for that meter to properly read 15mA full scale when a total of 1mA is flowing through the meter, and the remaining 14mA is flowing through the shunt resistance is:
1mA*47ohms = 14mA*Xohms
X = 47/14 = 3.357ohms.
This of course is assuming that the 47ohms I measured with my cheap Harbor Freight multimeter is accurate.
So I am now waiting for the approximately correct 3.3ohm 1% resistors I bought on Amazon to show up, so I can make the meters I bought actually useable. 3.3 ohms is about 1.7% too low, so the actual current will typically be about 2% higher than what the shunted meters read.
Of course there was no indication on the eBay listing that these meters were actually 1mA full scale meters and required a shunt resistor. On the box each meter comes in however it does say 1mA DC in addition to 15mA Gamma. These 15mA analog meters were originally made for Gamma High Voltage Research, Inc, as that is printed on the face of the meter itself.
Oh well. Live and learn.
Joe.
Originally on my version of the low cost fusor, I was using a Sensitive Research 0-30mA meter that I bought for about $45.
It works great, but the scale was labeled in 3mA increments, which was a little odd, and it turns out that the 30mA range really just means that its kinda hard to read, as the current really never goes above 10mA or so with the precipitator supply, and most of the time is down at less than 3mA.
So, I bought a couple Simpson 25mA meters, and put one of them in a project box, and it worked fine, and had numbered scale markings every 5mA which is a bit more conventional, but again it was a bit hard to use due to typical precipitator currents being 3mA or less. Also, you generally can't find analog Simpson meters on eBay for less than $25 or so, and they also all tend to be one off listings. So they aren't the optimal choice for a kit either.
So then I figured I would pick a range that worked well with the precipitator, and decided on 15mA. Because the precipitator will limit its output before it hits 15mA, and that will give much better resolution at the low end. 10mA would work also, but I figure that sometimes you do get more than 10mA output (like 11-12mA) and I wanted to minimize the risk of above range current damaging the meter. So I decided on 15mA as a reasonable compromise.
I found some inexpensive 15mA meters on eBay, and bought a number of them, as well as some project boxes that the meters fit in very nicely.
I expected the meters I bought to work just like the Simpson meters I purchased, where you just put them in the box, connect them up, and they work.
But NO.
When I tried an air plasma run with the new meter, and had the precipitator at its lowest output voltage, and ramped the pressure up, I fully expected the meter to register the expected 6mA or so that the supply puts out when the plasma lights off in that configuration. Instead the meter instantly pegged as soon as the plasma lit.
Turns out the meters I bought, have a 15mA scale, but do not have the proper shunt resistance in them, and are actually 1mA full scale meters.
I measured the 15mA analog meter resistance as 47ohms on the 2000ohm range of a Harbor Freight multimeter. When I did so, the meter showed a mA reading of 0.5mA on the face - so about 1/30th of full scale - which would be a nominal 33uA current for a full scale 1mA meter. I put one of the 50uA meters I use in my HV meter kit in series, while making the measurement again, and measured a current of 25uA - lower because the 50uA meter raised the resistance of the circuit from 47 ohms up to more than 1kohms. So indeed, the meter reads full scale at 1mA current not 15mA.
So the proper shunt resistance for that meter to properly read 15mA full scale when a total of 1mA is flowing through the meter, and the remaining 14mA is flowing through the shunt resistance is:
1mA*47ohms = 14mA*Xohms
X = 47/14 = 3.357ohms.
This of course is assuming that the 47ohms I measured with my cheap Harbor Freight multimeter is accurate.
So I am now waiting for the approximately correct 3.3ohm 1% resistors I bought on Amazon to show up, so I can make the meters I bought actually useable. 3.3 ohms is about 1.7% too low, so the actual current will typically be about 2% higher than what the shunted meters read.
Of course there was no indication on the eBay listing that these meters were actually 1mA full scale meters and required a shunt resistor. On the box each meter comes in however it does say 1mA DC in addition to 15mA Gamma. These 15mA analog meters were originally made for Gamma High Voltage Research, Inc, as that is printed on the face of the meter itself.
Oh well. Live and learn.
Joe.
-
- Posts: 557
- Joined: Tue Jul 20, 2010 4:08 pm
- Real name: Joe Ballantyne
- Location: Redmond, WA
Re: Joe Ballantyne fusor V2. (Low cost fusor.)
Ok, so my 3.3ohm 1% resistors came in today. I clamped one between the two input terminals on the back of the meter, and then retested the meter with the fusor.
BINGO! Worked just as expected. With an air plasma running and the precipitator at its lowest voltage, the meter ranged from about 5.8mA up to 6.5mA depending on the pressure. At 90 microns it was about 6.5mA, which is pretty much exactly what the other meters read.
I am very pleased with this meter, now that it actually reads the correct range. It is a pretty big meter at 3.75" x 2.9", and the 15mA range means it is very easy to see the 0.5mA markers - as there is about 1/10th of an inch between each one. So reading 1, 2 or 3mA off the meter is a breeze.
It proved really easy to vary the pressure to stabilize the current at 1, 2 or 3mA.
I like this meter much better than the 30mA or 25mA meters I was using before.
Here is the BOM.
(1) 15mA Gamma High Voltage Research ammeter (w/1mA full scale actual range)
$13 https://www.ebay.com/itm/401590275001
(1) LeMotech ABS project box (115 x 90 x 55mm)
$8 https://www.amazon.com/dp/B092HS9CMR
(1) 3.3ohm 1% 1/4W meter shunt resistor (Chanzon 100pc)
$5 https://www.amazon.com/dp/B08QRVK9WF
(2) 20" Alligator clip leads (from a pack of 50)
$15 https://www.amazon.com/gp/product/B06ZXSCLDH
I will be adding this meter as a kit available from my store in the for sale section of this site. Again, sold at cost plus nominal shipping charges.
Note that I bought a number of meters and project boxes for significantly less than the above single item prices, so the total cost of the kit including shipping should be about $20.
Here are a few shots of the meter in its project box.
Joe.
BINGO! Worked just as expected. With an air plasma running and the precipitator at its lowest voltage, the meter ranged from about 5.8mA up to 6.5mA depending on the pressure. At 90 microns it was about 6.5mA, which is pretty much exactly what the other meters read.
I am very pleased with this meter, now that it actually reads the correct range. It is a pretty big meter at 3.75" x 2.9", and the 15mA range means it is very easy to see the 0.5mA markers - as there is about 1/10th of an inch between each one. So reading 1, 2 or 3mA off the meter is a breeze.
It proved really easy to vary the pressure to stabilize the current at 1, 2 or 3mA.
I like this meter much better than the 30mA or 25mA meters I was using before.
Here is the BOM.
(1) 15mA Gamma High Voltage Research ammeter (w/1mA full scale actual range)
$13 https://www.ebay.com/itm/401590275001
(1) LeMotech ABS project box (115 x 90 x 55mm)
$8 https://www.amazon.com/dp/B092HS9CMR
(1) 3.3ohm 1% 1/4W meter shunt resistor (Chanzon 100pc)
$5 https://www.amazon.com/dp/B08QRVK9WF
(2) 20" Alligator clip leads (from a pack of 50)
$15 https://www.amazon.com/gp/product/B06ZXSCLDH
I will be adding this meter as a kit available from my store in the for sale section of this site. Again, sold at cost plus nominal shipping charges.
Note that I bought a number of meters and project boxes for significantly less than the above single item prices, so the total cost of the kit including shipping should be about $20.
Here are a few shots of the meter in its project box.
Joe.
- Dennis P Brown
- Posts: 3667
- Joined: Sun May 20, 2012 10:46 am
- Real name: Dennis Brown
Re: Joe Ballantyne fusor V2. (Low cost fusor.)
Nice find!
Ignorance is what we all experience until we make an effort to learn
-
- Posts: 557
- Joined: Tue Jul 20, 2010 4:08 pm
- Real name: Joe Ballantyne
- Location: Redmond, WA
Re: Joe Ballantyne fusor V2. (Low cost fusor.)
So I finally got real sick of dealing with sucky feedthroughs, and decided to put a REAL feedthrough on this low cost fusor.
I had a ~40kV feedthrough that I bought some time back from Joe Gayo that had a KF40 connector on it, so I added another shaker ball SS cathode to it using a 2mm to 3mm shaft coupler, and slapped it on the low cost fusor using a KF40 to KF50 conical adapter. I don't remember if I bought the feedthrough from Joe on eBay or at HEAS last year. I think it was at HEAS, but I'm not 100% sure. I think I paid something like $150 to $200 for the feedthrough, and I know I got a good deal, because IIRC Joe wasn't real happy with the price...
But he agreed to it. Reluctantly.
After connecting this actual feedthrough to the fusor, I pumped the system down, and guess what!?
NO LEAKS.
Finally.
When the chamber got down to 9.5 microns, I ran a 1 minute pressure rise leak test, and the pressure rose to just 16 microns in a minute. The chamber is currently down at 6.9 microns. I just ran a second 1 minute pressure rise test, and the pressure went up to 8.9 microns. So about 1 micron every 30 seconds. Likely essentially all due to outgassing, not any leaks.
Not really a surprise there. When you use actual real vacuum feedthroughs spec'ed for reasonable voltage levels, they GASP, tend to just WORK.
I plan on running an air plasma on this cathode shortly, but will need to tweak the shaker ball cathode a bit before I raise the voltage much, because I didn't get it very well centered on the KF50 central chamber axis. One side of the cathode is a bit too close to the chamber wall for comfort, since my plan is to run it up as high as the precipitator supply will go. (40kV) I want to have the cathode properly centered in the chamber before I do that.
Joe.
I had a ~40kV feedthrough that I bought some time back from Joe Gayo that had a KF40 connector on it, so I added another shaker ball SS cathode to it using a 2mm to 3mm shaft coupler, and slapped it on the low cost fusor using a KF40 to KF50 conical adapter. I don't remember if I bought the feedthrough from Joe on eBay or at HEAS last year. I think it was at HEAS, but I'm not 100% sure. I think I paid something like $150 to $200 for the feedthrough, and I know I got a good deal, because IIRC Joe wasn't real happy with the price...

But he agreed to it. Reluctantly.
After connecting this actual feedthrough to the fusor, I pumped the system down, and guess what!?
NO LEAKS.
Finally.
When the chamber got down to 9.5 microns, I ran a 1 minute pressure rise leak test, and the pressure rose to just 16 microns in a minute. The chamber is currently down at 6.9 microns. I just ran a second 1 minute pressure rise test, and the pressure went up to 8.9 microns. So about 1 micron every 30 seconds. Likely essentially all due to outgassing, not any leaks.
Not really a surprise there. When you use actual real vacuum feedthroughs spec'ed for reasonable voltage levels, they GASP, tend to just WORK.
I plan on running an air plasma on this cathode shortly, but will need to tweak the shaker ball cathode a bit before I raise the voltage much, because I didn't get it very well centered on the KF50 central chamber axis. One side of the cathode is a bit too close to the chamber wall for comfort, since my plan is to run it up as high as the precipitator supply will go. (40kV) I want to have the cathode properly centered in the chamber before I do that.
Joe.
-
- Posts: 557
- Joined: Tue Jul 20, 2010 4:08 pm
- Real name: Joe Ballantyne
- Location: Redmond, WA
Re: Joe Ballantyne fusor V2. (Low cost fusor.)
Ran the low cost fusor a few times recently, and learned some things.
1) A real vacuum feedthrough makes it much easier to pull a lower vacuum on a fusor. After staying under vacuum for a couple months, and getting pulled down to 5 microns every couple of days for a week or so, the 1 minute pressure rise on the fusor fell to 0.3 microns per minute, which is the best I have ever gotten. (I am not using turbo pumps or diffusion pumps or baking out to pull to very low pressures, so outgassing is just higher. But I was happy with the 0.3 microns per minute.)
2) I decided to try using the Specstar 9.8 pump with the low cost fusor instead of the Kozyvacu, to see how it would work. Turns out it works pretty well. It actually pulls the chamber down pretty fast - it gets down to 6-8 microns within a couple of minutes. Which is MUCH faster than the Kozyvacu. Currently pulls down to a minimum of about 4.2 microns after 6 minutes or so, and then the pressure slowly rises as the pump gets hot. Running the Specstar pump on the fusor for several hours one afternoon did seem to improve its ultimate vacuum a bit, as it now tops out about 5.8 microns instead of ~7 microns or so before. The total hours on the Specstar now is still very low, probably less than 24 hours of total runtime. It might have just needed some breaking in to improve the ultimate vacuum. I am going to keep an eye on it over time, to see if it improves further.
3) This feedthrough, although WAY BETTER vacuum wise, still has some issues, as it is quite long, and it turns out that the long center conductor when attached to my shaker ball feedthrough with the brass coupler, has an almost resonant oscillation of the cathode from side to side. (I have the feedthrough horizonal, not vertical, so I can easily see through the KF50 viewport on the reducing tee that is the main chamber.) An oscillating cathode is BAD. Really bad. As if it oscillates too much it makes sparks... which it did... and killed one of my 901p gauges. The 901p control/display board was fine, but the gauge suddenly started reading a pressure up at 7e4 torr, and the pressure reading was independent of the actual pressure in the chamber. Swapping in a different 901p gauge fixed the issue. So it was definitely the gauge, not the little control board.
4) After centering the cathode in the chamber side to side (it was previously angled slightly and closer to one side than the other), and adding a piece of 49mm diameter 2mm thick sacrificial glass in front of the viewport, I was able to ramp the voltage up to 30kV without any arcs. However as the voltage got up to about 30kV the cathode would suddenly start oscillating side to side even when I wasn't making the fusor move in any way. Because these oscillations are basically at a resonance, they take a long time to die down.
5) When the voltage is high, and the pressure is at 5 microns or so, sometimes the fusor will just light up with plasma. This happened first, when I went up to 35kV and just let it sit there. I was hoping it would just stay dark and not arc, but once I get up to 30kV or thereabouts, and just let the voltage sit there, it seems the fusor will eventually fire off a plasma. When this happens it generally starts the cathode oscillating. At one point it was so bad, that it touched the sides, and then there were little purple sparks between the cathode and the wall of the fusor a few times, and after that episode the 901p gauge was dead.
6) Arcs/sparks are BAD. They can kill electronics that are attached to the fusor. The less sparks that fly when you run a fusor, the happier and longer lived will be your electronics that you use with it.
I have ordered a couple of MPF KF40 30kV feedthroughs - which are significantly shorter than the feedthrough I am currently using, and so will hopefully eliminate this oscillation. Cathodes need to stay mostly put, especially in small chambers, so having a stiff central conductor that doesn't move much is likely beneficial.
The precipitator supply (number 2) seemed to survive this episode of arcing/sparks OK, even though the 901p gauge did not.
So, at this point the low cost fusor is on precipitator #2, and 901p gauge #2. Hopefully the new feedthroughs will eliminate cathode oscillations. We shall see.
Joe.
1) A real vacuum feedthrough makes it much easier to pull a lower vacuum on a fusor. After staying under vacuum for a couple months, and getting pulled down to 5 microns every couple of days for a week or so, the 1 minute pressure rise on the fusor fell to 0.3 microns per minute, which is the best I have ever gotten. (I am not using turbo pumps or diffusion pumps or baking out to pull to very low pressures, so outgassing is just higher. But I was happy with the 0.3 microns per minute.)
2) I decided to try using the Specstar 9.8 pump with the low cost fusor instead of the Kozyvacu, to see how it would work. Turns out it works pretty well. It actually pulls the chamber down pretty fast - it gets down to 6-8 microns within a couple of minutes. Which is MUCH faster than the Kozyvacu. Currently pulls down to a minimum of about 4.2 microns after 6 minutes or so, and then the pressure slowly rises as the pump gets hot. Running the Specstar pump on the fusor for several hours one afternoon did seem to improve its ultimate vacuum a bit, as it now tops out about 5.8 microns instead of ~7 microns or so before. The total hours on the Specstar now is still very low, probably less than 24 hours of total runtime. It might have just needed some breaking in to improve the ultimate vacuum. I am going to keep an eye on it over time, to see if it improves further.
3) This feedthrough, although WAY BETTER vacuum wise, still has some issues, as it is quite long, and it turns out that the long center conductor when attached to my shaker ball feedthrough with the brass coupler, has an almost resonant oscillation of the cathode from side to side. (I have the feedthrough horizonal, not vertical, so I can easily see through the KF50 viewport on the reducing tee that is the main chamber.) An oscillating cathode is BAD. Really bad. As if it oscillates too much it makes sparks... which it did... and killed one of my 901p gauges. The 901p control/display board was fine, but the gauge suddenly started reading a pressure up at 7e4 torr, and the pressure reading was independent of the actual pressure in the chamber. Swapping in a different 901p gauge fixed the issue. So it was definitely the gauge, not the little control board.
4) After centering the cathode in the chamber side to side (it was previously angled slightly and closer to one side than the other), and adding a piece of 49mm diameter 2mm thick sacrificial glass in front of the viewport, I was able to ramp the voltage up to 30kV without any arcs. However as the voltage got up to about 30kV the cathode would suddenly start oscillating side to side even when I wasn't making the fusor move in any way. Because these oscillations are basically at a resonance, they take a long time to die down.
5) When the voltage is high, and the pressure is at 5 microns or so, sometimes the fusor will just light up with plasma. This happened first, when I went up to 35kV and just let it sit there. I was hoping it would just stay dark and not arc, but once I get up to 30kV or thereabouts, and just let the voltage sit there, it seems the fusor will eventually fire off a plasma. When this happens it generally starts the cathode oscillating. At one point it was so bad, that it touched the sides, and then there were little purple sparks between the cathode and the wall of the fusor a few times, and after that episode the 901p gauge was dead.
6) Arcs/sparks are BAD. They can kill electronics that are attached to the fusor. The less sparks that fly when you run a fusor, the happier and longer lived will be your electronics that you use with it.
I have ordered a couple of MPF KF40 30kV feedthroughs - which are significantly shorter than the feedthrough I am currently using, and so will hopefully eliminate this oscillation. Cathodes need to stay mostly put, especially in small chambers, so having a stiff central conductor that doesn't move much is likely beneficial.
The precipitator supply (number 2) seemed to survive this episode of arcing/sparks OK, even though the 901p gauge did not.
So, at this point the low cost fusor is on precipitator #2, and 901p gauge #2. Hopefully the new feedthroughs will eliminate cathode oscillations. We shall see.
Joe.
-
- Posts: 120
- Joined: Fri Nov 25, 2022 9:25 am
- Real name: Ryan Ginter
Re: Joe Ballantyne fusor V2. (Low cost fusor.)
Issues such as this were exactly the reason I chose to go with a 6" sphere for my first fusor. An oscillation that is detrimental to a small cross becomes insignificant at a larger scale. I have many ideas I've been tossing around in my head for future fusors I'd like to make, and will eventually be tackling smaller form factors myself.
There's no doubt that compact fusors are the ideal neutron source for activation, as they allow you to irradiate samples in a higher neutron flux. That being said, I think they can be quite challenging for beginners who also intend to attempt high power input.
Your work is truly beneficial to all those who follow after you. I'm eager to see the results which will hopefully enable anyone regardless of income to achieve good results.
There's no doubt that compact fusors are the ideal neutron source for activation, as they allow you to irradiate samples in a higher neutron flux. That being said, I think they can be quite challenging for beginners who also intend to attempt high power input.
Your work is truly beneficial to all those who follow after you. I'm eager to see the results which will hopefully enable anyone regardless of income to achieve good results.
Last edited by Ryan Ginter on Wed Jan 22, 2025 10:11 pm, edited 1 time in total.
- Rich Gorski
- Posts: 360
- Joined: Mon Aug 01, 2022 4:34 pm
- Real name: Rich Gorski
- Location: Illinois
Re: Joe Ballantyne fusor V2. (Low cost fusor.)
Joe,
Interesting report about the oscillating cathode. I wonder if this could be related to the precipitator supply and its poorly regulated output. The supply voltage/current gets into some kind of resonance with the cathode that is triggered by an arc or a plasma ignition. It might be interesting to see if the oscillations stop with a different better regulated supply at the same voltage and current.
Rich G.
Interesting report about the oscillating cathode. I wonder if this could be related to the precipitator supply and its poorly regulated output. The supply voltage/current gets into some kind of resonance with the cathode that is triggered by an arc or a plasma ignition. It might be interesting to see if the oscillations stop with a different better regulated supply at the same voltage and current.
Rich G.
-
- Posts: 120
- Joined: Fri Nov 25, 2022 9:25 am
- Real name: Ryan Ginter
Re: Joe Ballantyne fusor V2. (Low cost fusor.)
Rich,
That's an interesting thought. You could attempt to record the grid oscillation in slo-mo with a smartphone camera and stopwatch in-frame to determine if the resonant movement is at 60 Hz. If so, that would likely suggest the issue is due to voltage drop via poor filter capacitance.
It could also simply be due to a slightly off-center feedthrough. With high ion current and low rigidity, the side of the grid closest to the wall may undergo more force than the opposite side, causing it to move off-center in the other direction, only to repeat the movement back and forth.
Regardless, a thicker stalk on the feed through is beneficial, both for reducing motion and the electrostatic field strength at its surface.
That's an interesting thought. You could attempt to record the grid oscillation in slo-mo with a smartphone camera and stopwatch in-frame to determine if the resonant movement is at 60 Hz. If so, that would likely suggest the issue is due to voltage drop via poor filter capacitance.
It could also simply be due to a slightly off-center feedthrough. With high ion current and low rigidity, the side of the grid closest to the wall may undergo more force than the opposite side, causing it to move off-center in the other direction, only to repeat the movement back and forth.
Regardless, a thicker stalk on the feed through is beneficial, both for reducing motion and the electrostatic field strength at its surface.
- Rich Gorski
- Posts: 360
- Joined: Mon Aug 01, 2022 4:34 pm
- Real name: Rich Gorski
- Location: Illinois
Re: Joe Ballantyne fusor V2. (Low cost fusor.)
Ryan,
The precipitator supply is a high frequency flyback type so its output is tens of kHz and it has only rectified output. No filtering. However if it is a 60 Hz frequency oscillation it could be related to the 60Hz VAC input powering the supply.
My thought was that the oscillation might be a loading down of the supply when a plasma ignites. As the voltage builds on the stalk electrostatic forces increase on the stalk. At some point a plasma strikes. This then loads down the supply extinguishing the plasma and relaxes the forces on the stalk. Then the cycle repeats this oscillation. A high speed camera (smart phone ?) might catch the plasma cycling on and off. The oscillations would be dependant on several factors such as voltage applied, length of the stalk, weight of the cathode grid, plasma current etc.
Complex but interesting.
Rich G.
The precipitator supply is a high frequency flyback type so its output is tens of kHz and it has only rectified output. No filtering. However if it is a 60 Hz frequency oscillation it could be related to the 60Hz VAC input powering the supply.
My thought was that the oscillation might be a loading down of the supply when a plasma ignites. As the voltage builds on the stalk electrostatic forces increase on the stalk. At some point a plasma strikes. This then loads down the supply extinguishing the plasma and relaxes the forces on the stalk. Then the cycle repeats this oscillation. A high speed camera (smart phone ?) might catch the plasma cycling on and off. The oscillations would be dependant on several factors such as voltage applied, length of the stalk, weight of the cathode grid, plasma current etc.
Complex but interesting.
Rich G.
-
- Posts: 120
- Joined: Fri Nov 25, 2022 9:25 am
- Real name: Ryan Ginter
Re: Joe Ballantyne fusor V2. (Low cost fusor.)
That's what I meant by filter capacitance (on the 120 VAC to DC switching supply side). Having an improperly rated capacitor for a given load will result in a voltage ripple superimposed over the supply output, causing it to load down as you've described.
- Rich Feldman
- Posts: 1559
- Joined: Mon Dec 21, 2009 6:59 pm
- Real name: Rich Feldman
- Location: Santa Clara County, CA, USA
Re: Joe Ballantyne fusor V2. (Low cost fusor.)
Never would have thought of mechanical oscillation of grid at a resonant frequency, sustained by electrostatic forces or (?) temperature gradients in the metal.
Couldn't oscillation of fusor current be monitored safely with an oscilloscope connected across the 3 ohm low-side milliammeter? In my experience, the inductive aspect of sense resistors won't matter much below 1 MHz (for single digit ohm, single digit watt values). Be mindful of peak currents, and voltage across sense resistor, from HV spark events. The 'scope could be far away from fusor (e.g. 20 feet) and get undistorted view if you use an ordinary coaxial cable instead of high-impedance probe. If signal risetimes are short enough that transmission line reflections matter, set scope input termination mode to 50 ohms, or use a BNC in-line 50 ohm terminator at scope input.
PPS output has voltage ripple at 2x the transformer frequency (as shown by Finn Hammer in 2017 post) and, I guess, at 2x the powerline frequency because of bridge rectifier charging its DC bus capacitor. Will know more after my 3rd DIY load resistor bank is finished.
Couldn't oscillation of fusor current be monitored safely with an oscilloscope connected across the 3 ohm low-side milliammeter? In my experience, the inductive aspect of sense resistors won't matter much below 1 MHz (for single digit ohm, single digit watt values). Be mindful of peak currents, and voltage across sense resistor, from HV spark events. The 'scope could be far away from fusor (e.g. 20 feet) and get undistorted view if you use an ordinary coaxial cable instead of high-impedance probe. If signal risetimes are short enough that transmission line reflections matter, set scope input termination mode to 50 ohms, or use a BNC in-line 50 ohm terminator at scope input.
PPS output has voltage ripple at 2x the transformer frequency (as shown by Finn Hammer in 2017 post) and, I guess, at 2x the powerline frequency because of bridge rectifier charging its DC bus capacitor. Will know more after my 3rd DIY load resistor bank is finished.
All models are wrong; some models are useful. -- George Box
- Richard Hull
- Moderator
- Posts: 15459
- Joined: Fri Jun 15, 2001 9:44 am
- Real name: Richard Hull
Re: Joe Ballantyne fusor V2. (Low cost fusor.)
Joe created a flurry of replies on his recent post. I gave up on tiny fusors when I tried the 2.75 cross. It blew out my Turbo controller.
The suggestion that such small fusors in high voltage, high power, high neutron numbers might need plus ultra components is a viable thought. However, a larger device with plus-ultra components would be demanded for the 2-3 mega mark for a fusor as the super HV needed for future work demands.
It is impossible for me to conceive of an oscillating cathode as everything I have ever built had a very rigidly attached cathode to the HV stalk feed through! My supply has alway operated of 60 hz power, supplying 120 hz fullwave DC totally unfiltered. No vibration noticed.
For big neutron numbers, a full 1or2-inch clearance of the cathode from the positive, grounded shell or wall of the fusor vessel would be ideal.
I still hold that the sphere holds the most uniform electrostatic field with a spherical cathode. The sharp edges of the cross re-entrant arms will create an unequal field regardless of cathode geometry.
Yes, I have hit the 1.3 mega mark with my 6" sphere at my best point of operation, but never applied more that the max my X-ray transformer can give and that is about 45 KV. The operation is always smooth, even and reliable.
Just my experience and thoughts based on a long time playing with fusion.
Finally, Joe has made it clear his fusor noted in the above is a bottom of the line, inexpensive fusor for the neophyte or noobie on a limited budget seeking to be "in-it-to-win-it". I did my first fusion with no turbo or diff-pump at 28kv in 1999. It is possible!
Richard Hull
The suggestion that such small fusors in high voltage, high power, high neutron numbers might need plus ultra components is a viable thought. However, a larger device with plus-ultra components would be demanded for the 2-3 mega mark for a fusor as the super HV needed for future work demands.
It is impossible for me to conceive of an oscillating cathode as everything I have ever built had a very rigidly attached cathode to the HV stalk feed through! My supply has alway operated of 60 hz power, supplying 120 hz fullwave DC totally unfiltered. No vibration noticed.
For big neutron numbers, a full 1or2-inch clearance of the cathode from the positive, grounded shell or wall of the fusor vessel would be ideal.
I still hold that the sphere holds the most uniform electrostatic field with a spherical cathode. The sharp edges of the cross re-entrant arms will create an unequal field regardless of cathode geometry.
Yes, I have hit the 1.3 mega mark with my 6" sphere at my best point of operation, but never applied more that the max my X-ray transformer can give and that is about 45 KV. The operation is always smooth, even and reliable.
Just my experience and thoughts based on a long time playing with fusion.
Finally, Joe has made it clear his fusor noted in the above is a bottom of the line, inexpensive fusor for the neophyte or noobie on a limited budget seeking to be "in-it-to-win-it". I did my first fusion with no turbo or diff-pump at 28kv in 1999. It is possible!
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
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
- Rich Gorski
- Posts: 360
- Joined: Mon Aug 01, 2022 4:34 pm
- Real name: Rich Gorski
- Location: Illinois
Re: Joe Ballantyne fusor V2. (Low cost fusor.)
Maybe we all forgot the more likely culprit causing the oscillations. The vibrations from the fore pump transmitted through the table and or plumbing. Easy thing to test. Move the fore pump to its own table. Any perceived connections to electrostatic forces are probably just coincidences.
Rich G.
Rich G.
- Rich Feldman
- Posts: 1559
- Joined: Mon Dec 21, 2009 6:59 pm
- Real name: Rich Feldman
- Location: Santa Clara County, CA, USA
Re: Joe Ballantyne fusor V2. (Low cost fusor.)
Good point, Rich.
I would bet that JB's shaker balls (and other Meiro spiral balls) have mechanical resonant frequencies much, much lower than balls made from 2 or more wire rings that are rigidly connected at every crossing. Is that a cost that offsets the elegance of single-wire jointless grid?
If grid is manually rung like a bell, can't the dominant frequency be measured with a musical instrument-tuning app? Ought to be plenty of sensitivity if the microphone is touching the grid's attachment stem. Or adapt an electric-guitar pickup or old fashioned telephone pickup coil. Or optical sensor with LED and detector on opposite sides of a gap.
I would bet that JB's shaker balls (and other Meiro spiral balls) have mechanical resonant frequencies much, much lower than balls made from 2 or more wire rings that are rigidly connected at every crossing. Is that a cost that offsets the elegance of single-wire jointless grid?
If grid is manually rung like a bell, can't the dominant frequency be measured with a musical instrument-tuning app? Ought to be plenty of sensitivity if the microphone is touching the grid's attachment stem. Or adapt an electric-guitar pickup or old fashioned telephone pickup coil. Or optical sensor with LED and detector on opposite sides of a gap.
All models are wrong; some models are useful. -- George Box
-
- Posts: 45
- Joined: Fri Aug 18, 2023 3:45 pm
- Real name: Cinar Kagan
Re: Joe Ballantyne fusor V2. (Low cost fusor.)
The mechanically induced resonance might actually induce plasma as well. The inequialized fields that form due to oscillation might develop to high enough intensity to locally ionize gas to the point of one plasma channel. As the grid gets further away the ionized channel might sustain and even grow as it reaches homogenous fields. After that point any change in field intensity might cause a part of the ionized chamber to develop an even more ionized region. This highly ionized region might cause sparks due to high current conduction.
Cinar Kagan
Cinar Kagan
-
- Posts: 120
- Joined: Fri Nov 25, 2022 9:25 am
- Real name: Ryan Ginter
Re: Joe Ballantyne fusor V2. (Low cost fusor.)
If this discussion is to continue any further it should be moved elsewhere. Let's leave Joe's post to its intended purpose, helping others get into fusion on the cheap.