Fusor gas pressure range for best operation

For Farnsworth type designs.

Fusor gas pressure range for best operation

Postby Doug Coulter » Thu Feb 02, 2017 11:45 am

Well, at least my opinion backed by my observations - which are limited - I haven't had time to try everything possible.

Based on running in a 6" ID sidearm (but with some field access to a larger tank) it seems as though what we want is a situation of pretty long mean free path on average. Say in this case, right around 3-6" MFP in the bulk of the tank. If we get more gas than that, we get far too much scattering to make much fusion (we can't get the stuff going fast enough before it bounces off something), and the system won't let you get to voltage at any current you'd want to put in there - or can. At the high pressure limit, you have an arc welder at some tens of volts and hundreds of amps and it's not going to last long. More likely, your high voltage supply will shut down if it's a good one (Hooray for Spellman!) or let out the expensive magic smoke. Which is why I recommend that you use some arc proof pretty large ballast resistor till you get things right. Here we've used 50-100k ohm 225 watt wirewounds that are a foot long. Usually those live long enough, but sometimes where they wound a few too many turns/inch they die when something in the tank arcs to ground and the power supply caps dump into the resistor.

We tried putting a 4" sleeve around the grid (grounded) to simulate a smaller tank. No good, it took lots more pressure to light off and stay lit, and we got next to no fusion. At the time we were using 1.26" diameter grids with perhaps 2" active length. We've since scaled down both diameter and length to diameter ratio some, which I think is at least partly responsible for our better results. I'll avoid some of my theory here, as that's going to be "prior art" for the patent office (as again, like GPL, I have to own it in order to give it away, but since this isn't software, it's patent turf, not copyright, which is easier and well, free), but there are reasons these ratios matter that do make sense with the standard model.

We have indeed seen now and then some real high neutron outputs from our itty bitty ion grid out in the big tank - maybe a 3/4" grid in a 14" diameter by 26" long tank, with the grid fairly off center - it wasn't intended to be the main show - and with a fairly wimpy power supply, this time a spellman PTV-250N, 40kv at 5.5 ma tops. I think I now know the reasons for this, but as above - more on that later once the legal guys are done - and that indeed might be where we're headed.

Just in case, I also own a big SS tank that's 30" in both directions, but that's going to have to wait for time to make it into a good vacuum tank.

My gauge reads high on deuterium, according to Pfeiffer, it's about 2x high in the range I use. I'll just quote the gauge numbers.

We run from around 1.4e-2 mbar to 2e-2 mbar most times. To hit that lower one - which gives much better performance, we have to use an ion source to keep the main grid drawing current, and we have the end of the main grid level with the end of the sidearm so that field also leaks out into the main tank - but that's not enough, we still need an ion source.

Hopefully I will have some fresh run data which I'll put in the section for that and you can see what I see - we have some pretty nifty data aq going on here, with a nice database for it all, and some realtime plotting stuff that works well (and cameras, and audio). Data mining software is yet to be written in large part. I'm using the mark 1 human eyeball and gut for the moment.

If you have a larger tank, you should be able to run lower pressure and should try it. Don't worry about "compression ratio" - in this setup I've observed that the limiting net collision energy isn't limited by the starting pressure, it's more the volts, space charge issues, and field geometry. Inside the grid you're into viscous flow with short mean free path no matter where you began if you get the rest right. But see Paschen's law and also the field distribution around a cylinder inside another one - diminishing returns would seem to set in pretty quick.

The V/M gradient does matter. You heard that from me, capiche?
voltageofpointcharge.png (9.05 KiB) Viewed 577 times

For those just starting this, you're going to find out that gas control, whether in flow mode as most do it (which wastes a lot of deuterium) or batch, as I do it - which has it's own issues of allowing reaction products to build up (which is good if you want to measure them but bad for DD fusion) - is the hardest part of all this, assuming you can buy vacuum gear, fab some of your own, and buy or make seriously good power supplies - and I do recommend Spellman for those. A disclaimer is that they have helped me along the way (thanks CliffS) but having also tried almost everything else - I'd have needed a lot more help with anything else on the market, and my own switching HV supplies so far haven't cut it nearly as well - and I'm good at this - and sure aren't as handy if you just want to get on with science and not just do a hobby. Gas control should be some other threads, and I think already is, but it's the big one in a lot of ways, and nope, there's really nothing commercial I've found that is totally cool for that. All mass flow controllers are FAR too high in minimum volume with crap resolution, for example. Hand operated fine valves are nice if you can fly a helicopter upside down while blindfolded and don't mind wasting a lot of gas and time...and on and on. Computer controlled insanely small lost-volume solenoid valves are what I do here - I sort of had to make those.
Posting as just me, not as the forum owner. Everything I say is "in my opinion" and YMMV -- which should go for everyone without saying.
User avatar
Doug Coulter
Posts: 3038
Joined: Wed Jul 14, 2010 7:05 pm
Location: Floyd county, VA, USA

Return to Fusors, Farnsworth type

Who is online

Users browsing this forum: No registered users and 1 guest