Snuck in a few minutes of fusor test time and got some interesting results. Not great, just interesting. When I get more, I'll make another thread under Nuclear.
Interesting that moving the setup across the yard seems to have increased the amount of power when there's no gas load....
At any rate, it seems that one doesn't want the max coupling capacitor, or more directly, doesn't want too much time/energy at the positive part of the cycle - it seems to drive the ions into the tank walls, and in fact it makes it harder to stay lit - I had to press the DC ion source into action here. This setup used 3 750pf in parallel, the previous test just one of them. I'm guessing that was closer to correct, and I may be fiddling with doorknob caps in series and parallel to get close, given that I can't just make the waveform I want and amplify it at this point - it's going to have to be created at the point of use by dodges like this (let's hope we don't wind up needing anything too elaborate or pulse forming by saturable reactors).
With more voltage, the neutrons that are produced come out _after_ the negative peak. So frequence has to go up with voltage, not a surprise at all.
This was far enough off that I didn't see any multiple ion bursts going by either faraday probe, but when I did get neutrons, I got more than last time, fat bursts. More on the audible counter too.
I'm having trouble probig the HV correctly. My old B&K type TV probes are way off on the AC content - read too high (while DC reads low due to the scope being 1 meg vs 10). I'm fairly sure that just adding a load capacity won't fix this - as just waving my hand a foot away changes the apparent divide ratio more than 2::1. But if you try to shield that thing, you add more capacity to ground than you'd want, and it'll arc - you need the spacing from the HV end for that reason. I need to get a solution to this so I can feed this info into the rest of the data aq and have some confidence in it.
I may revert to the last setup to get a more direct relative measurement at least for one run, just so I know where I stand.
But it did work "at all" and in fact better than last time. I'm going to need to compute the actual input power to the fusor, as a lot seems to be going into warming up things outside of it - even at 100-200w dc input to the H bridge.
- big scope
Scope trace, nothing calibrated here. Trace one (yellow) is the neutron detector. The sine wave looking magenta trace is a capacitive pickup near the HV feedthrough. Blue is the "near" faraday probe, green is the "far" one on the opposite side of the tank. I couldn't get a clear reading on transit time between them, it's faster than the risetimes. The sudden step in the drive waveform appears to be real, but since it seems to be lit off continuously I'm not sure I have a good explanation for the discontinuity.
Since the trigger is on the neutron detector - and I set that to only trigger on a fat burst, I used another scope to track the waveform and frequency as well as i could:
- Don't think this is showing DC/AC ratio correctly
- DS0013.png (5.17 KiB) Viewed 218526 times
I'll have to make a few tweaks and try more...as always.
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.