stuck on ion guns

[jaaz95]

My design for my fusor calls for 2 ion guns, one at each end. this will act as my gas supply as well so i can avoid cutting more holes in my chamber. this said im having a bit of a problem. Im not sure what type to use. I need it to be simple, fit through a 2.75" flange and be easy enough to build, im fine with something fairly complicated but preferably out of parts that are cheap and easy to get. Does anyone have any ideas where I should start or what kind to use?

[Tyler Christensen]

The easiest would probably be a quick and dirty DC magnetron. I've had fantastic luck with mine. I don't think I've documented it, so I'll throw some pictures up on this post and describe what I've done. If you want to put more time and money into it, I'm sure you could easily make a much better ion source than this one, but it gets the job done.

I've taken a spark plug feedthrough (there is a vinyl tube on the spark plug since sometimes I put oil in for a low pressure run since the voltage can actually exceed that of the NST output due to some strange space charge build up, but no need to worry about oil for, say 10 micron plasma), put a ~2" long ~1/16" stainless rod on the inside end, and put that into an NW50/2.75 flange. Hook that directly to one leg of an NST.

Then, wrap magnet wire that can take 10+ amps for at whatever duty cycle you want to run around the flange 50-100 turns (mine is multi-layered in the picture), and hook this onto a DC supply. I just rectified a step down transformer, put it on a variac, and capacitively filtered the DC bus (absolutely critical to cut back on the space charge effect described earlier). I have 100,000uF filtering, which seemed to be just enough. It really can't take much of any ripple on the magnet.

Permanent magnets could also be used, but I like the simplicity and ease of field adjustment that comes with an electrical magnet.

Now, both the NST variac or the magnet variac can control the current of the ion source. Mine fires well below 1 micron.

Just ran it tonight and it's a dream running the fusor compared to no ion-gun, dial in 10 microns, turn on the gun, turn up the HV PSU, sit back and enjoy, no valve or variac adjustments mid-run required.

[Doug Coulter]

Nice work, Tyler! That's a lot simpler than my fancy one, and if it works, it works.

[Doug Coulter]

Ah, it was late last night, so I forgot a couple of things. I myself have gone to using something similar, but since my fusor's a lot different, my ion generator is a little different too. My (main) fusor runs in a 6" diameter sidearm off a much larger tank - 16" ID and about 30" tall. I stuck another grid out in the main tank, no magnet, but otherwise the same setup - NST to drive the second grid. It can fire at low pressures without the magnet due to the longer paths in the big tank and the Paschen P x D equation.

Now, when I do it this way (rather than say, just putting another DC supply on the ion generator grid) I get a scope shot like pictured below.

The top trace is the 3he tube output, AC coupled (so it misses a pulse or two at the start). The bottom trace is the voltage from the NST terminal on the 2nd grid. Note the pulses on that that go well above ground on the otherwise negative half cycle of the AC, and the rectification on the positive half cycle. This, by the way, burned out the NST eventually -- a positive few kV on the terminal while it was driving negative 7.5 kv into the internal ballast cooks them at some point - internal arcing.

The thing I make most note of here (and I wish I'd saved a better picture, and I will when I get another NST hooked up) is the timing of the neutrons on the tube detector. Since you're the only other guy who has one running like this, it sure would be nice if you could set up something similar and see if that's happening on your rig as well....In other measurements I've taken, I see mainly pulses of current on the main HV more or less time-coincident with those on the NST/2nd grid, and very little power draw the rest of the time. Doing a back of envelope Q computation says this mode is vastly superior to a static running fusor. In my case, I get about the same total neutrons, but a lot less main DC power input -- I've not yet managed to scale this to more neutrons at the same power input, it only seems to happen at the low end of the workable pressure range so far.

You'd never realize this with a BTI - no time resolution. But I've seen this plenty on various time resolving neutron detectors, and I'm pretty sure it's not EMI at this point. Actually, I'm very sure it's not EMI, as silver gets hot and bti's bubble during this mode.

So if you replicate this or not - I think it's worth a try and we'll all know whether what I've seen is a fluke or not. When I just use DC on the 2nd grid, it acts like a normal, stable fusor. My second grid is nothing special, and I don't measure significant neutrons from it alone -- it's just a two loop pear shaped version of a sphere grid like many here use, and it's tiny - about 1/2" OD.

[RobertTubbs]

I'm currently using this setup, something I showed Tyler yesterday, which is just some ceramic magnets around a non resistive spark plug.

It starts up around 2-3 microns, works great, only needs a few kv to run ac/dc, I haven't had an issue with it yet.

RT

[jaaz95]

the main reason i was hoping to use an ion gun was to add gas as i said before but also to use it to sufficiently accelerate the ions that it produces. all these designs are awesome btw, thanks for all the ideas so far, this is awesome.

[Dustin]

Some other ion sources worth looking at are Carls RF ICP ion source.
(great for mono-atomic ions)

viewtopic.php?f=12&t=5005&hilit=rf+ion+source#p32313

Andrew Seltzman Anode layer ion source

viewtopic.php?f=12&t=5021&hilit=anode+layer#p32329

Steve.

[Doug Coulter]

Those are good, and so is this microwave one:
http://www.coultersmithing.com/forums/v ... f=28&t=222

Which has been duplicated in a few other labs now. I adapted the design from a paper in Rev Sci Ins, and added the ECR magnets to that, which improved it a lot.

But all are more complex than the one Tyler describes, and which I'm using a variant of myself. It works well.

The microwave one seems to win the prize for staying "lit" at low pressures, but it seems the basic "ion gage used as ion source" works plenty fine for fusor conditions and is obviously a lot simpler.

I've posted other details on mine here on this board if I can find it in a search, I'll edit and add the link.

Ah, here it is:
viewtopic.php?f=12&t=5022#p33120

[daniel_pflug]

Rob,

Looks like the most simple design I've seen. what do you use as power supply voltage? Current? I've been looking for a simple source that is pretty cheap too. Thanks!