Page 1 of 2

Jan Ruge Fusion Attempt @ Chaos Darmstadt

Posted: Sat Dec 02, 2017 2:39 am
by Jan Ruge

First of all, as I'm quite new here, pleas let me know if this is the wrong section :)

This is a brief introduction to my Fusor project or static neutron source to be precise (shooting deuterium ions on a metal plate).
I try to keep this up to date as we make any progress.

The vacuum system is based on an oil diffusion pump, I picked up on Ebay a couple of years ago, a TV-203 to be precise.
I ended up building a compact module with all the necessary parts to run the pump, as water cooling system a KF-40 inlet flange and an additional transformer to power up filaments for example.
The inlent flange was the main problem with that design as the pump should originally be mounted to a flat surface, what was a problem as I want to use the cheap standard KF vacuum fittings.
I managed to connect a ISO-F/KF40 adapter, that had aprox the right size using some metal parts, proper clamps, an O ring and some grease.
The first picture shows the first run of the diffusion pump creating a good vacuum as tons of xrays are generated, the second one shows the final version of the setup.
The current high voltage supply is a flyback transformer with a ZVS driver with a voltage multiplier which is encapsulated in epoxy to avoid corona discharge.
In the end this seems to be a bad idea, as after the epoxy cured, the glass started to crack.
The cascade worked for a couple of years, but now there is some arcing on one of the hv feedthroughs for the AC.
This problem might be fixable, but on the long term I need to build a new one.
The ion gun is mostly based on an anode ray tube, a glass tube that allow a gas discharge with a small channel to let the ions pass.
First, I tried to tighten the assembly using vacuum grease and rubber o-rings, but there was too much air leaking inside (I was not that accurate in drilling the holes) so I ended up fixing the problem using epoxy.
There is a small needle valve as used in CO2 applications for the gas inlet.
The following tests looked very promising as the ion ray were clearly visible, even though the vacuum was not ideal.
I was not able to open the needle valve even a bit, as there were too much air leaking inside the assembly.
This was most likely due to dirty seals, but we are currently debugging that problem.
The pre acceleration voltage for the ion gun was generated by using a cheap 'High Voltage Taser Module' for tasers.
It turned out that this module has one output grounded what can be problematic depending on the configuration (a friend was zapped a couple of times and the lab power supply went crazy).
This problem was fixed by using the built in transformer that has no reference to ground.
By applying a the acceleration voltage, there was a strong gas discharge what is not useful for that design as it slows down the ions and introduce much current.
Even though that was not intended, but this also seems to be a good sputtering device, so the inner of the glass tube was quickly plated with copper an zink during the tests.

Re: bolek42 Fusion Attempt @ Chaos Darmstadt

Posted: Sat Dec 02, 2017 5:56 am
by ian_krase
Cool! It's always nice to see simple ion guns.

You will want to change your name to your actual name. Rules are the rules.

Re: bolek42 Fusion Attempt @ Chaos Darmstadt

Posted: Sat Dec 02, 2017 7:18 am
by Richard Hull
Yes we do not allow online pseudo-names only real names. You needed to introduce yourself in the introduction forum. Posts with no user real name can suffer deletion.. update your profile.

Richard Hull

Re: bolek42 Fusion Attempt @ Chaos Darmstadt

Posted: Sat Dec 02, 2017 10:26 am
by Jan Ruge
Okay, I will do asap

Re: bolek42 Fusion Attempt @ Chaos Darmstadt

Posted: Sat Dec 02, 2017 10:56 am
by Dennis P Brown
Very good approach and excellent design; add deuterium and Ti target (to load Deuterium) as well as neutron detector and you should be able to have detectable fusion.

Be aware such a gun (depending on your voltage), can be an x-ray generator. While Al isn't too bad, might be a good idea for you to check the x-ray output with a Geiger counter or similar detector. The issue tends to be back-streaming electrons; a suppressor (a bias voltage at your target) can reduce that threat.

Re: Jan Ruge Fusion Attempt @ Chaos Darmstadt

Posted: Fri Dec 15, 2017 1:56 am
by Jan Ruge
Thanks for the positive response and also for the tip wit the suppressor grid :). Sadly the ion gun is sealed with epoxy but I keep that in mind, especially for the target.

A Geiger counter is always with us (Gamma Scout Alert) but to be honest I don't trust the readings that much. It detects gamma radiation above 30keV. But it might not detect anything, if the voltage lower than that (voltage drop due to high current). Even if there is a massive amount radiation. Fluorescenting glass makes me always nervous. So if possible I try to use a delayed shutter and leave the room.

I now also finished my own Geiger counter circuit based on a CCFL driver with a voltage multiplier. The output voltage can be adjusted from 200V-1800V. A voltage divider to turns of the inverter as the correct voltage is reached. Here is a picture of the final assembly. The GM tube is a SBM-19.
The reason, why I build this is, some time ago I bought one of those boron coated neutron detection tubes. It is a SNM9 as discussed here: viewtopic.php?f=13&t=6191 . I have not tested it yet hopefully I get this one running.

I now also have some deuterium-oxide and starting work on the Gas supply. I have one of those 'Kosmos Fuel Cells' and also bought a small pressure sensor. The plan would be to fill a reservoir and turn off the fuel cell if a slight pressure is reached. This way there should be no water sucked into the pump or pressed out of the fuel cell.
As I said before the vacuum system wasn't in use for a while and we had issues with some leaks. We spent a lot of time reassembling and cleaning everything (thanks to Zeri aka Julian Wälde for help). It turned out that the hub of the backing pump was way too complex (10 copper tube seals and several plumbing joints). The copper tube seals are not rated for vacuum use but did a good job for a while. Even though, they tend to turn a bit if some force is applied, so they become leaky after some time. We simply removed all of them.
We now have a KF-25 adapter straight of the pump sealed with teflon tape and vacuum grease. The color of the discharge (backing pump only) now changed from orange to white and the diffusion pump creates high vacuum again. Here the comparison:
- Left leaky backing pump
- Middle fixed backing pump (comparable to leaky backing pump + diffusion)
- Right fixed backing pump + diffusion pump (arcing on the surface of the glass tube)
This time we could also adjust the pressure in the ion gun and control the gas discharge. That was an interesting experiment on its own, to see the direct changes in the discharge. Although I was zapped a couple of times. In addition we generated some x-rays if the pressure was just right. Surprisingly even without the diffusion pump as the ion gun is a good vacuum pump on its own. Julian works on a way for remote control for the needle valve to avoid irradiation. As it was too late we do not have good pics of the assembly running under high vacuum.

Re: Jan Ruge Fusion Attempt @ Chaos Darmstadt

Posted: Fri Dec 15, 2017 3:04 am
by Richard Hull
Good show! I have always said that would-be fusioneers need to spend time alone with demo vacuum systems, to study the way pressure and electrical discharge in gases function. This is a fascinating time of real discovery and gives teeth and backing to one's self-directed reading of theory. One comes to know discharge phenomena over a broad range of pressures and applied voltages.

When I distributed my "fusor an introduction" and "fusor- tips and techniques" video tapes, I spent a lot of time with fusor II and on the fusor introduction tape, I showed and interesting pressure vs. voltage phenomenon. I was able to get the voltage and pressure of air in the bell jar of fusor II adjusted to a critical point of no discharge but near break over. I show my charging of a long Teflon rod with silk and moving it rapidly inward towards the bell jar. The jar flashed brightly due to the huge dv/dt electrostatic gradient external to the bell jar. It self-quenched. Now, jerking the rod back away from the jar, it flashed again due to the rapid change in the local electrostatic field. Cool!

Only "monkeying" around and "dinking" about with a plain air discharge system can do this. A very teachable moment of which there are many when actually stepping aside from or slowing the pace of the rush to do fusion.

Richard Hull

Re: Jan Ruge Fusion Attempt @ Chaos Darmstadt

Posted: Fri Dec 15, 2017 10:48 am
by Dennis P Brown
A few points: you should run any deuterium gas you make via the generator through some type of drying agent. Water vapor will contaminate the D2 and reduce/interfere with your device's ability to achieve high vacuum and might interfere with your fusion rate on the target (this later part is just a guess. For fusors, drying the gas is essential but yours is an ion gun/target fusion device. Very different process.)

A home assembled geiger counter, while not calibrated can give one some type of warning but frankly, it is only a no threat (essentially no counts) or don't know because it gives lots of counts but that does not tell anything about the actual levels.

I have three detector systems - a calibrated geiger counter, a scintillation detector and a dosimeter: all three indicate I have zero x-ray emission (no counts at all except background) from my fusor from any angle I measure. I run 32 kV.

My chamber, however, is very thick steel (about 0.5 cm) and my window is a bit over 2 cm thick. Despite that, I have some lead shielding in line with my location of my electrode in the chamber. I also keep my distance and use a mirror for viewing the plasma. One can also use inexpensive slate instead of Pb if concerned about toxic issues of Pb. However, my main concern is protection from my high voltage since my transformer/voltage is lethal.

As you now know, the best vacuum system is one with the fewest couplings; and since some couplings are always required, one can't use cheap methods - KF is the best bet for the cost.

Good luck and glad you posted your results.

Re: Jan Ruge Fusion Attempt @ Chaos Darmstadt

Posted: Wed Oct 10, 2018 4:09 pm
by Jan Ruge
First of all, sorry for the late response, we had to some major upgrades before doing further experiments.

Second, some responses:
- Drying the gas is planned at some point (probably using CaCl), but is currently not implemented. I'll keep that in mind.
- We also have a calibrated Gamma Scout Alert, but this is reserved to verify, that we do not get any dose. So we needed a second one to measure the X-ray emission directly at the device. The Geiger counter now has a proper case and shielding, so its also not easily affected by the high E-field. If someone is interested, I can add further pictures and schematics.
- Currently, we simply leave the room, so we can make use of the ~5-10cm concrete walls, what works quite well for those low energy photons. We also have ~14kG of sheet lead, that will be mounted at some point.

Third, our progress:
After hours of debugging, we found the leak at the ion gun. It was caused by a bad O-ring seal and was fixed using a couple layers of epoxy (Again keep it simple). In addition, we now have complete remote control for the device, what was the most time-consuming part. Our lab power supply for the high voltage had this feature already built in. The hard part was the driver for the ion gun.

For that driver, we had the following requirements:
1. High voltage for gas discharge
2. Remote valve control
3. Galvanically isolated
4. Powered by battery

Here is what we came up with:
The driver runs at 7.3V so can be powered by a two cell LiPo battery. The PWM signal for the servo is generated by an Arduino and passed to the driver via a Toslink cable (glass fiber) to provide insulation. The Toslink transmitter, receiver, and servo are powered by 5V (provided by a 7805) and can be connected almost directly. We only had to invert the signal once. The presence of the PWM signal also switches the high voltage generator, so the driver can be turned on and of remotely. A charge pump (C2,C3,D3,D4) converts any AC component sent via Toslink to a DC signal that switches M3. R6 is a bleeding resistor, that will pull the gate low if no AC signal is present. The high voltage is generated via a ZVS circuit and a rewound switch mode transformer. The mechanical setup is still slightly messy.

We stopped working on the neutron detection tube, as debugging the neutron detector and generator at the same time seems to be a bad idea.
A NaI(Tl) scintillation gamma spectrometer is arriving from Russia. So we have two options to detect fusion reactions:
a) Shield the detector using lead and detecting neutrons reacting with the crystal
b) Measure changes in the gamma spectrum. The maximum energy of the bremsstrahlung should be less than the applied voltage (e.g. max 200keV). Every significant amount above this limit should be generated by fusion reactions.
Both methods should be verifiable with a reference measurement using air or protium.

Fourth, current status:
We finally measured some X-rays from the device, therefore we managed to get some ions >20keV! Even though it's not quite stable. This is our current setup:
(10.05 KiB) Downloaded 2332 times
Using only the ion gun without acceleration voltage, we could measure a beam current of approx 30uA with an aperture of 3.3mm (M4 thread). Here is a video of the device in action. The fizzing sound in the background is no Geiger counter, but the device itself.

We seem to cycle through the following states:
- Valve closed, high vacuum, no discharge, no X-rays
- Opening Valve, discharge in the ion gun, X-rays
- Visible beam in the chamber, sometimes X-rays
- Gas discharge near the HV feedthrough (chamber pressure way too high)

Therefore we assume, that the aperture is too large, so we can't establish a stable working condition. Next step would be to try some different sizes, effectively drilling some holes in screws.

Re: Jan Ruge Fusion Attempt @ Chaos Darmstadt

Posted: Wed Oct 10, 2018 7:24 pm
by Richard Hull
You are moving towards some sort of success. Far more important than mere successful fusion is that you are learning a lot by "the doing" and hands on observations in a slow and careful march towards your goal. When you arrive at your goal, you will "own" a lot of concepts others merely read about and take for granted.

I dare say that you have done and learned more not doing fusion than virtually all of the people on our Neutron club listing who have done successful fusion! They came and worked from a plan, followed instructions, and rolled over the finish line rapidly. They were all about the win than the learning process. Learning can be costly in time, effort and even treasure. However, all things worth claiming as one's own are often hard won.

Operation is a deuterium gas volume to produce fusion is never an easy task. Working with plasmas and controlling them is a learned process.

Richard Hull