I managed my first completely hands-off fusor runs last Friday and Saturday. Finally!
Chad Ramey was here for the Saturday run (we were working on his fusor too, which he brought along). The Friday night run produced activation of 1560 cpm on silver, the Saturday one some 780 or so, in the first 10 sec of counting after the run. Runs were timed at 5 minutes, and time to get silver to counter was roughly 20 seconds (the ten second counter free runs, so the first real count may have to wait for a new period to begin -- we allow one count period to verify background first as well).
Power was 51kv DC (pure) with current limit set at 12 ma, but never or very rarely reached that, bouncing from about 4 to about 7 ma mostly. I had a second grid in the big part (14" by 26") of the tank as a supplementary ionizer, current limiting and running about 10kv or so, at a few ma DC (on its own supply). This gives better ion/neutral ratio and seems to make gas pressure less critical (lots less). The gage (PKR 251) was reading 1.6 e-2 mbar the whole time for both runs, a decent amount less than I can run with just the little ion source (about 1.9 in that case) or without any source (about 2.2 in that case). The gage is known to read high on D, and higher yet if the D is ionized, so those are kind of relative numbers. Anyone who actually does fusors knows that the fusor itself is more sensitive than most any gage....I trust it more on a relative basis than my gage, anyway. The fusor was blinking and pulsing some during both runs, not real stable, but stable enough. There is some evidence here that it makes more neutrons doing that than stable...
During the runs, a faraday probe far from the action (an unused HV feedthrough) read in the 250 volt range with a 100 meg load to ground, negative polarity.
We are now logging most of the run data to a MySQL database for later examination, so we now can tell how long it took the silver to get to the counter, which is left on the whole time -- you can see the radiation from the fusor, then it goes to background, then the silver counts, all timestamped in the database. We are working to add more things to this log so it will be more valuable to "data mine" later -- in fact, it's been the main thing we've been working on for the last couple of months.
So, soon, I'll be posting pretty plots with all the parameters on them!
Key to this was the new flow control system, which is still manual, but very stable now -- it would be easy to automate at this point. What we do is let in some D to help the turbo spin down to the "standby" speed I preset, then open a small solenoid valve between the turbo and forepump in pulses (slap a microswitch) till the gas pressure is just right, and it simply stays that way for a long time after that. I sometimes overshoot a bit and have to let in more gas, then try again, but a couple of tries at most gets it done. The runs were done with no gas flowing in at all. I normally keep this system at high vacuum (~4e-8 mbar) when not in use, so it stays pretty clean in there. With the turbo valved off from the forepump, it simply doesn't move any gas, but tends to keep purifying it, as it has a higher compression ratio for all gases not hydrogen isotopes. We got the little solenoid valve from a leak detector we stripped for vacuum parts.
I *think* I saw a line at mass 6 post-run on the first run, very close to the e-14 noise floor of the mass spectrometer, but to use that at all, I have to pump out most of the gas first, so who knows....it sure looked real, and I've seen it before, not just this time, but I'd have been a lot happier if it had been 10x the noise floor instead of ~2x -- hard to pump out just the right amount of gas to get into the mass spec "happy" range, the turbo is real quick. I will have to perfect that technique some more, I guess.
I'd never see a line at 3 because there's always one at 3 anyway after a run -- HD is always there in my systems from water and D both being there, and a tiny amount more from something really 3 (like 3He or T) wouldn't show reliably. But my "6" is right on the floor if I don't run the fusor first, and off it if I do. It's the same for "4" as for three -- if you have D in there, you have 4.
What a relief -- when we started this it was like trying to fly a helicopter upside down while blindfolded and trying to run an RC airplane with the other hand, all at once, with tactile feedback through duct tape, and an angry wife screaming in your ear and throwing ashtrays. Did I mention having to balance that pencil on my nose the whole time? Boy, that was HARD.
This is now so stable that when we wanted to test Chad's and my old B10 tubes later in the day with his new Ludlum electronics unit, all I did was slap the HV on switch again and bingo -- neutrons.
Amazing those old things still work, one I bought in the late 60's and didn't store carefully (rust) was still fine!
Now things are stable enough so measurements can really mean something beyond "am I having a good day as fusor pilot" which of course is the goal. This is a beginning, not an end!
A picture of the setup (but not the same other conditions) is here:
http://www.youtube.com/user/DCFusor#p/a/u/2/gs8mWLjO5Yc
In that video, the gas pressure was changing, and I was driving the second grid with AC. |