Nathan Marshall - Fusor Progress

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.
User avatar
Nathan Marshall
Posts: 53
Joined: Wed May 08, 2019 8:13 pm
Real name: Nathan Marshall

Re: Nathan Marshall - Fusor Progress

Post by Nathan Marshall »

Hello Mark,

Thank you! The inner grid is about 1.25” in diameter and the chamber is a 6” conflat cylinder, so it’s inner diameter is about 4”.
User avatar
Nathan Marshall
Posts: 53
Joined: Wed May 08, 2019 8:13 pm
Real name: Nathan Marshall

Re: Nathan Marshall - Fusor Progress

Post by Nathan Marshall »

I finished building a ~30kv high voltage power supply, and I got it in an enclosure with oil yesterday. The high voltage is supplied by a very large flyback-style transformer with a huge secondary winding and a roomy air gap space between the primary and secondary. The primary is wound with 12 gauge solid copper wire. The output is fed into a full wave voltage tripler made with 30kv 820pf doorknob capacitors and 30kv 100ma diodes. The output is taken from the enclosure using 40kv wire. The output also goes to four 100 megohm MOX resistors under oil for voltage measurement. The output of these resistors is fed through a 100 uA meter, giving me a maximum voltage measurement of 40kv. The primary of the flyback is fed with a cheap "1000 W" eBay ZVS driver (I don't believe this purported power rating, but I won't be coming anywhere near this level of power). The DC power for the ZVS driver comes from two 15V power transformers in series, rectified with a beefy bridge rectifier and smoothed with a large capacitor. Using a variac, this gives me a heavy-duty variable 0-35 VDC supply for the ZVS driver. Some initial testing last night with just my mechanical pump running showed a current just over 60ua on the meter, or ~25kv. X-rays were pouring out of the viewport. I wear a lead radiology apron as a precaution to keep the rads off my nads, but where I was standing there was little to no radiation above background. I just put a 10 ohm 20 watt resistor from the fusor shell to the supply so I'll try to get a current measurement later today. I plan on making a nice enclosure for all the power circuitry soon.

I rigged up a Raspberry Pi single board computer with a cheap $15 arducam camera. After some fiddling, I got things to the point where I can remotely desktop into the RPi with my laptop and view what the camera sees through WiFi. This allows me to view the fusor grid from anywhere and with no wired connection to the camera.
Attachments
IMG_5610.png
IMG_5610.png (153.13 KiB) Viewed 3925 times
IMG_5611.png
IMG_5611.png (150.07 KiB) Viewed 3925 times
IMG_5612.png
IMG_5612.png (142.85 KiB) Viewed 3925 times
IMG_5613.png
IMG_5613.png (93.56 KiB) Viewed 3925 times
User avatar
Mark Rowley
Posts: 909
Joined: Sun Dec 29, 2002 12:20 am
Real name: Mark Rowley
Location: Sacramento California
Contact:

Re: Nathan Marshall - Fusor Progress

Post by Mark Rowley »

Excellent work Nathan. I'm hoping for the best possible results as this hobby in particular needs to get past the perception of reliance on ultra costly or impossible to locate power supplies.

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

Re: Nathan Marshall - Fusor Progress

Post by Richard Hull »

Nathan, You are admitted to the Plasma club in the fusioneer listings. Great work.

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
Nathan Marshall
Posts: 53
Joined: Wed May 08, 2019 8:13 pm
Real name: Nathan Marshall

Re: Nathan Marshall - Fusor Progress

Post by Nathan Marshall »

Mark:
Thank you, I hope I can get some fusion results soon! I was also interested to see if I could build an entry-level fusion power supply that was not XRT-based. I think I'll get one of those 30kv precipitator supplies as well and give that a shot.

Richard:
Thank you!! Now onward to neutrons!


I played around a bit more with the supply. I got a steady dull glow from the tungsten at 60-70 uA on the voltage measurement meter and 15-30 mV across the 10 ohm resistor which translates to ~25-28kV at 1.5-3 mA. I should be able to do fusion with this! One issue: my stainless steel input feedthroughs on the lid of the HV supply got really hot... hot enough to start melting the plastic. It started to bulge and I turned off the power before it melted through. The copper input wires themselves were barely warm. I think the higher resistance of the stainless steel really becomes and issue at these switching frequencies. I'll ditch the feedthroughs and just run the copper wires through the holes like I did for the HV output wire.

I found that my arducam was out of focus. Once I focused it, the image quality drastically improved! I just thought the camera was having issues in the low light, but nope, it was just out of focus. Great performance for $15!

'
Attachments
IMG_5617.png
IMG_5617.png (65.12 KiB) Viewed 3924 times
IMG_5616.png
IMG_5616.png (105.07 KiB) Viewed 3924 times
IMG_5615.png
IMG_5615.png (142.63 KiB) Viewed 3924 times
IMG_5614.png
IMG_5614.png (138.75 KiB) Viewed 3924 times
User avatar
Dennis P Brown
Posts: 3160
Joined: Sun May 20, 2012 10:46 am
Real name: Dennis Brown

Re: Nathan Marshall - Fusor Progress

Post by Dennis P Brown »

Can't tell you how impressed I am with your home-built power supply. That is really an excellent, quality build.

Your total power (under a 100 watts) is on the low side so you will need a good detector system. However, using your smaller volume fusor (higher pressure deuterium), that power should, as you said, be sufficent to get detectable fusion.

Keep posting about the supply, too; maybe a circuit diagram would be nice. I'd be especially interested in the power supply for your ZVS. If you haven't already done this, maybe some details on how you made your 'flyback'.
User avatar
Nathan Marshall
Posts: 53
Joined: Wed May 08, 2019 8:13 pm
Real name: Nathan Marshall

Re: Nathan Marshall - Fusor Progress

Post by Nathan Marshall »

Dennis:
Thank you for the compliments! I will draw up a schematic for the whole supply in the near future. I didn't realize that very few on this forum have gone for the ZVS/multiplier approach so I will try to document my results with it as thoroughly as possible.

I got rid of the stainless steel feedthroughs for the primary coil input and just pulled the copper wires through the holes. Then I soldered the output from the ZVS to them. This is performing way better! The wires just get a bit warm to the touch. I didn't expect the stainless steel to perform so poorly, but this is good thing to remember for the future. I am easily getting currents of 4 mA at 25-30 kV! The grid glows brightly and there are many hot pixels from x-rays showing up. After about 5-10 minutes of running, I turned everything off and observed that the outside of the chamber was too hot to touch! So I don't think I'd be able to handle much more input power anyways.

The main issue that I'm having now is a very infrequent spark happening around the connection of the HV supply to the feedthrough. It is very low power, almost like a static buildup spark. So it is definitely not arcing from the supply. It happens every 20-30 seconds when I get over 20 kV. The problem is that it shuts off my harbor freight voltmeter I'm using for current measurement! I tried coating the feedthrough connection with a lot of hot glue which makes the spark even less frequent, but it still happens. Any suggestions? In one of the attached pictures you'll see the voltage just under 30 kV (~75 uA on the meter), but unfortunately the current meter has been shut off due to the spark.

Edit: just pushed the system a little harder and registered a stable 6 mA at 28 kV!
Attachments
IMG_5619.jpg
IMG_5621.jpg
IMG_5623.jpg
User avatar
Nathan Marshall
Posts: 53
Joined: Wed May 08, 2019 8:13 pm
Real name: Nathan Marshall

Re: Nathan Marshall - Fusor Progress

Post by Nathan Marshall »

I've drawn up a schematic for my current circuit configuration.
Attachments
HV Power Supply.png
User avatar
Dennis P Brown
Posts: 3160
Joined: Sun May 20, 2012 10:46 am
Real name: Dennis Brown

Re: Nathan Marshall - Fusor Progress

Post by Dennis P Brown »

Thank you for taking the time/effort to provide the circuit diagram. I was wondering, what are the current/voltage values for the ZVS power supply diodes?
Also, what is the current out put/rating of the ZVS supply x-formers?

You know, if you keep up these excellent posts, despite my very sore back/arms from so much contrustion on my house, I'll have to mix and pour half a ton of concrete to finish the floor of my (still) future laboratory facility? I'm getting too impatient after seeing this work! ;)

Thanks
User avatar
Nathan Marshall
Posts: 53
Joined: Wed May 08, 2019 8:13 pm
Real name: Nathan Marshall

Re: Nathan Marshall - Fusor Progress

Post by Nathan Marshall »

Thanks, Dennis! Perhaps I should stick a recommendation in the schematic for the bridge rectifier but I left it blank since there are so many options. I’m using a cheap MDQ 100A 1600V rectifier brick unit I had sitting around. It has screw terminal inputs/outputs. It cost about $10. Simple as that! I’m unsure of the current rating of the AC power transformers... they were salvaged from a dumpster. They are about the size and weight of a microwave oven transformer and output 15 VAC.

Good luck with the concrete, that sounds like quite the project... but having a nice dedicated lab space will be worth it!
User avatar
Mark Rowley
Posts: 909
Joined: Sun Dec 29, 2002 12:20 am
Real name: Mark Rowley
Location: Sacramento California
Contact:

Re: Nathan Marshall - Fusor Progress

Post by Mark Rowley »

Keep it up Nathan, love the progress and Can-Do attitude.
I had limited success with the HF meters past 25kV. Exact same issues you and Jim have described and that's with copious shielding and grounding. Ultimately I opted for a nice Triplett analog meter which quickly solved the problem.

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

Re: Nathan Marshall - Fusor Progress

Post by Richard Hull »

Analog metering will never display the precision of a digital, but they will be around a lot longer and do good enough in the HV realm. No one here needs to know 36kv to better than +/- 200 volts. Harbor Freight meters are not good in high RF or EMP environments.... Few digitals are. It is extremely tough to kill an analog VTVM or VOM. (unless you drop it on a concrete floor - jeweled movement) Digitals will usually survive such a drop.

Choose all metering based on where and under what conditions it is built for. Hamfests are the best place to pick up a venerable old VTVM or lots of specific analog meter movements that you can turn into anything you wish. I have so many meter movements purchased over my lifetime that I have them in several large, cataloged boxes stacked up in the lab from 0-20ua to 5 amps. I tend not to buy voltmeters unless they are 0-30kv or the like, scaled, and have 50ua or 100ua movements.

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
Nathan Marshall
Posts: 53
Joined: Wed May 08, 2019 8:13 pm
Real name: Nathan Marshall

Re: Nathan Marshall - Fusor Progress

Post by Nathan Marshall »

Mark and Richard, you reminded me that I have an old analog meter that my grandfather gave to me years ago! I almost forgot about it. It has a 250 mV range so it is perfect for this application. I hooked it up and now even when the static EMPs occur there is no effect. The power supply is running great. The furthest I pushed it was 6mA at just over 30 kV. My Ludlum 44-9 probe starts to pick up x-rays even back where I am standing at that kind of power. So now I need to hook up the deuterium tank and regulator, and get some reliable neutron detection working.
Attachments
IMG_5637.png
IMG_5636.png
User avatar
Jim Kovalchick
Posts: 717
Joined: Wed Apr 13, 2011 8:00 pm
Real name:

Re: Nathan Marshall - Fusor Progress

Post by Jim Kovalchick »

Nathan, I am certain you will be in the neutron club soon. I admire your careful work, especially the power supply.

Stay safe.

Jim K
Victor Gonzalez
Posts: 28
Joined: Tue Jul 16, 2019 10:16 pm
Real name: Victor Gonzalez

Re: Nathan Marshall - Fusor Progress

Post by Victor Gonzalez »

Do you have a reason to use these 2 capacitors?

I have almost the same setup for a PMT (I used an lm317 for voltage control), and the voltage output with and without these 2 capacitors are the same.
Attachments
IMG-20200501-WA0046.jpg
IMG-20200501-WA0046.jpg (33.54 KiB) Viewed 7359 times
User avatar
Nathan Marshall
Posts: 53
Joined: Wed May 08, 2019 8:13 pm
Real name: Nathan Marshall

Re: Nathan Marshall - Fusor Progress

Post by Nathan Marshall »

Victor: I was under the impression that the way to make a negative voltage multiplier was simply to reverse the diodes of a positive voltage multiplier. I haven't read any source that recommends removing those capacitors when making a negative multiplier so I never thought to. I'd have to do some simulations or experiments to decide if they really contribute greatly to the function.

Since my last post, I got everything ready for fusion. I did some tests of pumping down the chamber with the diffusion pump, throttling off the pump with a butterfly valve, and admitting air through my flow restriction setup to see if I could get stable plasmas and pressures. I found that my grid was too large. I was getting crazy negative resistance which resulted in wild oscillations of current and voltage. So I ditched my grid and made one that was a bit less than 1 inch in diameter and with two rings instead of three. This works far better. I was able to tune the pressure and get stable plasmas at ~30kV and 2-5mA. I did some hydrogen runs with hydrogen generated by an old PEM cell I was given in 6th grade. I used Mark Rowley's syringe method for storing the gas. The main reason I did this was to convince myself that I wasn't sucking gas in at high rates which would waste deuterium. Lastly, I hooked up the deuterium tank to the gas line with a used Harris 9200-50 H2/CH3 regulator from eBay.

Then came a major disappointment. I was going for my first fusion attempt this afternoon. I had everything set up and ready to go: diffusion pump running, BD-PND bubble detector set up near the chamber, and my power supply ready to rumble. When I opened the valve on my deuterium tank, the high pressure gauge didn't even budge... My heart sank. I obtained this tank over six years ago with the help of my high school chemistry teacher. That was my freshman year of high school, and I was naive enough to think I would be able to achieve fusion then. I didn't realize the time and money that it really took. It is a Sigma-Aldrich tank and was originally pressurized to 25 atm, and has a volume of 459mL. I wonder if over the years it slowly leaked since deuterium is such a light gas. To my surprise, when I turned the knob on the regulator I was able to get some PSI on the low pressure side. So, I went ahead with my test. I got a beautiful red 25kV 5mA plasma at 30 microns for just a minute before the gauge on the low pressure side went back to zero. Any thoughts on what went wrong here? Did the tank just leak over time? Or did I have some sort of leak? I imagine a leak of that rate would have been quite audible, and the fact that the high pressure gauge never even budged leads me to think it was basically empty to begin with.

Plasma without deuterium:
05-07-2020_10-59-12resize.png
05-07-2020_10-59-12resize.png (221.38 KiB) Viewed 3828 times
Plasma with deuterium before it ran out:
05-07-2020_15-45-23resize.png
05-07-2020_15-45-23resize.png (208.89 KiB) Viewed 3827 times
User avatar
Jim Kovalchick
Posts: 717
Joined: Wed Apr 13, 2011 8:00 pm
Real name:

Re: Nathan Marshall - Fusor Progress

Post by Jim Kovalchick »

Nathan,
Maybe your regulator has an obstruction. You should check it on another bottle or at least inspect the internals of the high side before assuming your bottle is empty. Also, cycle your bottle valve open and closed and open again.

I am curious about your short run. Did your plasma flash at the pressure you expected for your voltage. If it flashed at a lower pressure then it was just air in the line.

Good luck. You will be 'there' soon.

Jim K
User avatar
Nathan Marshall
Posts: 53
Joined: Wed May 08, 2019 8:13 pm
Real name: Nathan Marshall

Re: Nathan Marshall - Fusor Progress

Post by Nathan Marshall »

Jim,
I did try opening and closing the tank a few times but to no avail. The pressure required in the chamber was higher than my runs with air. With air I was at about 5 microns on the baratron and for this short run I needed about 30 microns to get a similar plasma wattage. I'll take another look at the regulator like you suggested, but from what I could tell it appeared clean and in good condition. The fact that I got a little pressure on the low pressure side before it went away makes me think the bottle still had a bit of gas in it but then ran out. Thanks for the reassurance, I will soldier on! As one of the grad students I work with says: "It's always a fight."
User avatar
Dennis P Brown
Posts: 3160
Joined: Sun May 20, 2012 10:46 am
Real name: Dennis Brown

Re: Nathan Marshall - Fusor Progress

Post by Dennis P Brown »

Sorry to hear; regulators aren't something that last - I've had some fail and a tank empties. But unless your primary valve (independent of the regulator) wasn't fully closed, the tank should have remained filled. Deuterium can't really diffuse through the lecture bottle (I've had one for over 6 years (in use) and it is fine but I always carefully and fully close the main valve fully tight. I will say, the residue gas in the regulator does leak away over a few days. Those small (read cheap) regulators can't be trusted to be the sole seal on the tank.)

I understand only too well getting ready to make one's first real fusion run only to take a hit. For me, my detector counter that zapped by the detector's HV supply. It is frustrating but you'll make it - your so very close. and have shown you can handle these issues.
User avatar
Nathan Marshall
Posts: 53
Joined: Wed May 08, 2019 8:13 pm
Real name: Nathan Marshall

Re: Nathan Marshall - Fusor Progress

Post by Nathan Marshall »

Dennis,
Well then perhaps it was my worst fear that occurred: the tank was full to begin with and something wrong with the regulator caused the whole bottle to leak in just the few minutes that I had the tank valve open... I should have closed it the moment that the high pressure gauge didn't budge. Hindsight is 20/20 and I think I was too blinded by the desire to go for fusion. Also if something is wrong with regulator, it once again goes to show that deal used buys on eBay usually lead to more frustration and money spent than just buying new to begin with. I still think that if the whole bottle leaked I would have heard some sort of hissing.
User avatar
Richard Hull
Moderator
Posts: 14992
Joined: Fri Jun 15, 2001 9:44 am
Real name: Richard Hull

Re: Nathan Marshall - Fusor Progress

Post by Richard Hull »

I always open the tank valve jsut barely to watch the low bressure gauge go to 4 PSI into my CO2 cartridge tank and immediately turn of the main tank valve. The little cartridge at 4 PSI will run the fusor for 10-20 minutes. I have never nor will I ever leave a main tank valve open regulator or no regulator attached. It stays hard off always until I need to let more into the little catridge at 4 PSI.

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: 3160
Joined: Sun May 20, 2012 10:46 am
Real name: Dennis Brown

Re: Nathan Marshall - Fusor Progress

Post by Dennis P Brown »

As for hearing the leak through a failed regulator; yes, you should have. It appears the tank emptied before that, which can only happen if the tank's main valve isn't fully closed or its 'pressure relief' valve failed at some point (my lecture bottle has one of those relief valves - don't know if all do. The main valve does not seal that off.)

No matter. Your next concern is getting deuterium so generating your own is the most straight forward, I'd think; and a number of people have done that here.
User avatar
Nathan Marshall
Posts: 53
Joined: Wed May 08, 2019 8:13 pm
Real name: Nathan Marshall

Re: Nathan Marshall - Fusor Progress

Post by Nathan Marshall »

I do have 10mL of heavy water on hand, so I can generate some deuterium with the PEM cell and use a syringe to store it. I got the gas line set up for that now this morning. I just need to dry out the PEM cell from the water that was in it before I put in the heavy water. Also I will do some tests to make sure my connections with the syringe are air-tight and see if I get flow rates in the 1mL per minute range. But at the moment I have a quantum take-home final exam to finish so I will prioritize that.
User avatar
Nathan Marshall
Posts: 53
Joined: Wed May 08, 2019 8:13 pm
Real name: Nathan Marshall

Re: Nathan Marshall - Fusor Progress

Post by Nathan Marshall »

I generated some hydrogen, stored it in a 60mL syringe, and fed it into the fusor. The fusor was sucking in gas at quite a high rate, so after throttling off the diffusion pump a bit I was able to taper off the gas consumption. Even after these adjustments it was still quite a bit higher than 1mL/min. Attached is a picture of the plasma. This time I found that to get a plasma in the ~25kV 2-5mA range I needed a pressure on the baratron of about 16 microns. So this is quite a bit lower than what I got with the deuterium from the tank. I'm guessing this points to the lesser purity and potential water contamination of the generated hydrogen. I'll keep playing around with this a bit more and then consider filling the PEM cell with some of the heavy water that I have in my element/chemical collection.

H2_run_05-08-2020_17-24-26_resize.png
H2_run_05-08-2020_17-24-26_resize.png (118.28 KiB) Viewed 3586 times
Edit:
So I did another hydrogen run this evening after some more work on the quantum exam. Feeling pretty good about my answers now. This time I was able to establish a stable 1.3mL per minute flow rate into the chamber with stable 3-5mA 25-30kV plasmas at ~12 microns on the baratron! After the hydrogen run, I brought the chamber to the same 12 micron pressure with just air as a control test. As expected, the electrical characteristics were completely different! At 5mA the voltage was <8kV. Below is a picture comparison of the two plasmas.

Hydrogen at 12 microns, 28kV 4mA:
H2_run2_05-08-2020_20-10-10_resize.JPG
H2_run2_05-08-2020_20-10-10_resize.JPG (28.26 KiB) Viewed 3586 times
Air at 12 microns, 7kV 5mA:
H2_run2_05-08-2020_20-24-23_resize.JPG
H2_run2_05-08-2020_20-24-23_resize.JPG (28.63 KiB) Viewed 3586 times
Cai Arcos
Posts: 163
Joined: Fri Oct 12, 2018 9:30 am
Real name: Cai Arcos
Location: Barcelona, Spain.

Re: Nathan Marshall - Fusor Progress

Post by Cai Arcos »

Nathan:

First of all, I want to express my gratitude for all the details provided about a superb build, especially the affordable power supply.
I myself intent to imitate your efforts using a transformer I have found from a tube based CRT TV. The only problem are the capacitors themselves. The only item that I found on Ebay that is affordable on my budget is a set of 100pF 20kV caps. Even if I could connect two in parallel, that would still be 200pF, substantially lower than your own values. I understand the principle of operation behind a voltage multiplier, and thus know that the necessary capacitance depends on the frequency. From your own values, I assume that the frequency must indeed be high, however because I don't know the frequency of operation (which I suppose would also vary with the transformer itself) I can't perform the necessary calculations myself in order to check if those capacitors are adequate.
I would highly appreciate if you could use your experience and observations to guide me.

Thank you very much:
Cai
Post Reply

Return to “Images du Jour”