Fusor MK 0.5 design

For posts specifically relating to fusor design, construction, and operation.
Post Reply
Joseph Jerkins
Posts: 12
Joined: Sun Oct 25, 2015 12:10 am
Real name: Joseph Jerkins
Location: San Francisco, CA

Fusor MK 0.5 design

Post by Joseph Jerkins »

After about two years of working on and off on a fusion capable fusor (after a thoroughly underwhelming demo fusor), I have completed design plans for a fusion capable fusor. I have been using Solidworks to CAD a model of my fusor down to each screw and keep track of sources for almost all parts so that when I finally order everything I can put it together and be comfortable in the knowledge that everything fits and I haven't ordered the incorrect sizes, which will hopefully ease the difficulty of troubleshooting things and getting it operational. Before I move on to the step of beginning construction however, I would like to bounce the design off the fusor forum to get feedback and fix errors that I have inevitably made along the way. Hopefully having a full model (apart from some trivial wires for communication that I haven't added quite yet to connect things) will make presenting my design a bit easier (forgive any formatting mistakes, I have not posted much on this forum).

I have broken my design up into a few systems, namely the chamber/vacuum, deuterium production/handling, power supply, and measurement/control systems. Going through each of these:

Chamber

The chamber is currently based around a piece of equipment I managed to find from a local scrap dealer in decent shape, and is one of the items I am least sure about keeping. I am not sure what it was used for in the past, but it has thick enough walls to withstand vacuum and CF flanges on 3 of the 4 ports (Two are 8" CF flanges, one is a 10" CF flange) so it seemed a suitable candidate for a cheap base. It has a dent (visible in the picture, on the section near the left flange) that I am worried about, and the knife edges on the CF flanges are probably worn down.
Image
Everything else on the chamber has not been ordered, but the full chamber is connected as shown in the link below (image wasn't correct scale in post):
Exploded view.

Going around the ports:

Top is for deuterium inlet and pressure sensor. The chamber base I currently have is missing a CF flange at the top, so a sheet of 1/8" SS is welded on. The vacuum gauge is attached with a 1/8" NPT fitting welded on, although if reaching deep vacuum with NPT is too difficult then I will have to use the CF or ISO versions instead (would love advice on this). In the center, some thin steel tubing is welded on and wound to stabilize the pressure and to act as a conduit from the fitting that leads to the deuterium handling system. This fitting is currently 1/4" VCR, but is entirely dependent on what connections the mass flow controller I end up getting has. The MFC is one of the few components that I have not chosen a specific model for, just because it will need to be bought off of ebay and there is too much variation to be able to choose a specific model prior to buying one with how often auctions end. Chamber top.

Left (or front in the final assembly) is the view port. Pretty self-explanatory, it is an 8" to 4.5" reducing flange with a 4.5" CF glass viewing port from lesker attached to the other side. Chamber view port.

Bottom is vacuum outlet. This has a 10" to 4.5" CF adapter connected to a 4.5" CF to NW50 adapter. I couldn't find a 10"CF to NW50 adapter, or I would have done this in one step. This connects to a throttle valve, which is modeled after MDC's part 306002 (I managed to find a cheap one on ebay or wouldn't consider this due to price). Again, because I may not be able to find a valve with similar specs when it comes to buying parts, I may have to change the 4.5" adapter to something else that interfaces with available valves. The throttle valve connects to a diffusion pump, in this case modeled after the agilent AX-65. The diffusion pump has an NW16 flange output that goes to a roughing pump. Chamber vacuum system.

Finally, the right is the electrical feedthrough. This is just an 8" to 2.75" CF adapter connected to MDC part 9442011, a feedthrough rated for 40kv 3A. Electrical feedthrough

Deuterium

The deuterium system consists of an electrolyzer, drying tube, storage tank, solenoids, and mass flow controller. The electrolyzer and drying tube are arranged as shown in this image. The image is a bit hard to see on the white background, but the setup is very similar to Andrew Seltzman's electrolyzer with a glass tube in the center of a larger polycarbonate tube. There is a ring of platinum wire around the outside of the glass tube and a wire on the inside for hydrogen production. The apparatus is connected via swagelok tube fittings to a drying tube stuffed with calcium chloride or some such desiccant. This is then connected to a solenoid, which is connected to a two port 0.5 gallon air tank with 0.25" NPT fittings. The other end is connected to another solenoid via 37deg brass flare fittings and copper tubing, which then leads to the MFC and ultimately the chamber as shown in these images. The system is meant to operate so that deuterium can be made shortly prior to operation from heavy water, then used from the tank without needing to carefully monitor the electrolyzer's pressure or rate of reaction. First a vacuum is pulled with the MFC solenoid open and the electrolyzer solenoid closed, to evacuate the tank of air. Next, the MFC solenoid is closed and the electrolyzer solenoid is opened. This will pull in any hydrogen present in the electrolyzer into the tank until pressure reaches approximately atmospheric pressure. To avoid pulling all the heavy water in, the solenoid will likely have to be opened for short periods of time. If this sounds ridiculous (I'm a little unsure of this method), a needle valve would make a suitable replacement for the electrolyzer solenoid. When the fusor is operated, the electrolyzer solenoid/needle valve is closed and the MFC solenoid is opened to allow deuterium into the chamber. Assuming storage of 0.5 gallons deuterium and a flow rate of 4sccm (a typical flow rate), this supply should last for many hours.

Power supply

My power supply is a ZVS driven flyback run through a rectifier. Now before you go and link the FAQ, I am well aware that this forum does not like flybacks. Often complaints are that they "can't quite supply enough voltage and current" and that most run on quite low powers (350W), well below that required for fusion. With this in mind, I have built a ZVS driver rated for 500V 120A and sourced flybacks rated for 40kv 60mA on amazing1 (These have been mentioned but their posts go unanswered). In the name of determining whether these are capable of measurable fusion, I have decided to pursue this method. I have already built the ZVS driver and have tested it on my variac and a weak flyback taken from a TV, but I have not yet tried a power supply capable of even half the rated power. Given that it is rated for over 1kW, I am confident that this setup will supply enough power for measurable fusion. Image of ZVS setup and fusor. Closer image of ZVS driver.

Measurement/Control

Pressure measurement is carried out by a thermocouple gauge that is attached to the top plate via a 1/8" NPT port. These gauges are plentiful on ebay, my model is based off of vacuum research's VRC-6M 1/8" NPT gauge. The driver/display is currently the corresponding part sold by vacuum research.

Voltage measurement is not in the model, but I have a handful of 10M resistors that will suffice for a voltage divider. Current measurement is carried outwith a 10 ohm resistor between ground and the chamber.

Neutron measurement is one that I am not as comfortable with my solution, but I plan to have a regular SBM-20 tube wrapped in a 0.01" silver sheet and surrounded with HDPE blocks. The geiger tube would be run from an arduino kit such as those offered here or on ebay. I have seen setups in the forum with silver activation schemes where a silver plate is activated by fusion and moved to a beta sensitive device, but when I calculated the penetration of beta radiation with 1.2MeV (Ag 110 decay) and 0.63MeV (Ag 108 decay), it seemed that beta radiation is capable of travelling through lead of this thickness. Thus, it would seem that a system in which neutrons are slowed by the HDPE and activate the silver is measurable with a simple geiger tube inside. My cause for hesitation lies in the fact that others have not used this setup (to my knowledge), and that thinner silver means less neutron interaction and less chance of measuring fusion. Any thoughts on whether this might work are appreciated. I am also curious as to where I should place this apparatus relative to the chamber. Is it fine to just have it sitting next to the fusor on the table as shown?
Fusor with sensors. Geiger tube with one block off

Lastly, my fusor is intended to be remotely operable for fusion purposes. It may require gas generation to be done locally if the needle valve is used (unless a remote needle valve is bought/manufactured), but for actual fusing operation electronics will be remotely operated with one or a few arduino modules and some cheap remote cameras. I have left this portion of my design out due to its mostly non-physical nature, having more to do with programming.

Full view, and another full view.

With all of that explained, I ask for any corrections to false information or suggestions regarding how I might improve/rework my fusor. Many thanks to those on the forum for providing such a valuable learning resource.
User avatar
Dennis P Brown
Posts: 3159
Joined: Sun May 20, 2012 10:46 am
Real name: Dennis Brown

Re: Fusor MK 0.5 design

Post by Dennis P Brown »

Your flyback - rated to 60 ma by that company (and no, their FB will get no were near that on continuous load) so, I do not think that flyback (& ZVS) will make anywhere near the current for even a low output fusor. The ZVS is a strange device (tried a few and not at all happy with them - they really need a stack of boat battery's to really work, in my opinion and as I've seen them actually used) so unless it is made very carefully and you have a supply that can really dump current without dropping the voltage, they often fall far short of their rated power.

Bottom line: absolutely test the flyback's total power now before assuming you have a power supply that can do fusion. Well worth your time to build a simple resistive load and test the flyback/ZVS - would not be surprised if you only got a few ma unless you use some boat batteries or a welding supply ...even then, not convinced the current will be enough. Hope I'm wrong but my experience to date, says not a good supply (that fly back; maybe make your own; not as difficult as one would think.)

Your vacuum gauge isn't enough for fusor work; you really need to measure down below what a thermocouple can do - more like 10^-5 torr; otherwise, how will you know your fusor has been evacuated enough to remove trace gases?

Your neutron counter will most likely not work using activated silver; activated silver needs a rather high neutron flux, as I understand it, and using a flyback, very unlikely you will get anywhere near enough neutrons. Again, hope I'm wrong and wishing you succeed but many have tried these methods and posts to date show no success; as for activation, except for experienced fusors that have high fluxes ...but that isn't your system .

I am lost why you want to add even more complexity by creating a remote operation - not saying that isn't worth doing but it is not needed. Even a super high end fusor is very safe to be around. Unless you have a lot of power above 30 kV, even the x-ray flux isn't a problem for the average steel fusor; again, at least as I understand matters.

Best of luck.
Joseph Jerkins
Posts: 12
Joined: Sun Oct 25, 2015 12:10 am
Real name: Joseph Jerkins
Location: San Francisco, CA

Re: Fusor MK 0.5 design

Post by Joseph Jerkins »

Thank you for the help!

With regards to the flyback, they are indeed quite finicky to get working and require much tweaking to get them working optimally. In a few weeks I will have access to a much nicer power supply than the one I am using now, and I will then be able to test its ability to hold voltage under a load. There is still the issue however, that the energy going into the ZVS driver must be going somewhere. If it can pull well into 1kW, then any inability of the flyback to supply that amount of power should be mainly due to heat losses (if I understand the electronics correctly). There are folks in the neutron club who achieved it with less than 400W such as here (33kv@11ma), so even with less than 50% efficiency on the flyback fusion is measurable with silver activation. Nevertheless, I am not charging into this without realizing that the DC supply to the driver will need to be beefy. My plan is currently to get one or two 24V power supplies capable of at least 500W each and put them in parallel. I have found through some experimentation on my driver that the performance is significantly increased with capacitors over the input to battle any impedance the supply might have causing problems at high frequency. Lastly, if this doesn't work I am happy to go with an xray transformer.

With regards to vacuum measurement, the gauges I have been looking at bottom out at 1 mTorr. Isn't that enough for fusion purposes, when it should be carried out in the range of 1-20 mTorr? I have seen posts on here by users who report fusion with data such as "Pressure: <10mTorr (our gauge bottomed out)". If it is really that important to measure low temperatures to ensure the chamber is as evacuated as possible, then I will have to upgrade.

I mentioned silver activation above, which has been verified on a system at less than 400W. What do think of when you say high power in that case? Richard Hull responded to the above post by saying that the measured rate is fairly low for the power they put in, but that can be explained due to the tube not responding much to radiation. Regardless, if I am unable to measure any activity with the silver scheme (which is also quite cheap, the main reason I am aiming for this first), then I will move on to an identical He3 or BF3 tube instead in the moderator. I figure it's worth spending the 30-50$ on a silver system that could work and has worked in the past for decent fusor power input, then to move up to better tubes if that doesn't work.

Finally, remote operation is something that I am not worried about because I have about three years of experience on a robotics team that uses systems nearly identical to what I would need to run it remotely. I will be constructing this reactor at my school, and the administration has seemed hesitant to have a fusor in the building even with the assurances that radiation and x rays will be at safe levels. Even the simplest system such as some relays controlling power supply and a phone with skype running would suffice, I am not trying for anything overly complex.

Again thank you for the feedback. I really can't say whether a flyback will work, but in the coming weeks I will be testing whether they can work with a system tweaked to support high enough power. If not, I will just move on to an xray transformer.
John Futter
Posts: 1848
Joined: Wed Apr 21, 2004 10:29 pm
Real name: John Futter
Contact:

Re: Fusor MK 0.5 design

Post by John Futter »

Joseph
In general Flybacks are used for less than 100 watts output with a maximum output of 300 watts.
Most supplies like PC supplies are forward converters either voltage or current fed.
Big step up or step down ratios are to be avoided in flybacks so as to keep leakage inductance under control ---it is this that strains the active switching components and snubbing components hence the power limitations
User avatar
Dennis P Brown
Posts: 3159
Joined: Sun May 20, 2012 10:46 am
Real name: Dennis Brown

Re: Fusor MK 0.5 design

Post by Dennis P Brown »

What is important, vacuum wise, is to make sure the chamber is well within 10-5 torr before adding deuterium to raise it back up to 1-10 microns. Trace gases are an issue and will poison the fusion process if they remain. One can hope for the best and ignore that low level measurement but it adds yet another unknown and you have a few of them already.

I hope the flyback works but most people here say otherwise. As for silver activation, maybe it will work but again, another unknown since you will be, at best, at the bottom for power. As for 33 kV at 11 ma, yes, that can do some fusion but it is on the low end.

Again, do try and maybe it will all work. Not saying don't try just be realistic.
User avatar
Richard Hull
Moderator
Posts: 14991
Joined: Fri Jun 15, 2001 9:44 am
Real name: Richard Hull

Re: Fusor MK 0.5 design

Post by Richard Hull »

Remember that fusion may be done at 33kv @11ma, but that is measured actual voltage and current at the moment of fusion success and under a real electrical load. It is not merely claimed by a manufacturer of flyback or inverter/multiplier arrangements. Really a giant difference.

Example: A neon transformer can easily have a nameplate of 15,000 volts @ 60ma. Can it supply a continuous 15,000 volts? Yes, absoltuely. Can it also supply a continuous 60 ma of current? Yes, absolutely. Can it supply both the rated voltage and the rated current at the same time? Never, ever in a million years, under any circumstance. At 60 ma the output voltage will be on the order of 500 volts. At max 15,000 volts the maximum load might be 1 ma.

Believe no claims unless you actually measure that they are true. Most always they are not.

So many here want to fall for the claims of a inverter/regulator and flyback that can fit in the palm of your hand capable of a 1000 watts at ultra high voltage and current. So easy.....So simple....So not true.

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)”