Re: A nice little open-source data acquisition package
Posted: Wed May 14, 2014 6:41 pmYeah, I had all too much trouble with office politics, believing it all was or should be a meritocracy (which would have put me at the top). Nope, doesn't work like that. Making every problem harder, you have to come up with an "everybody believes they win" solution...even if it takes crazy patience for your idea, rated lousy at the meeting, becomes the boss' idea 6 months later and then you get to do it. It's why I started my own firm, consulting, and was only willing to deal with customers at or very near CEO level themselves. Less BS all around. And it was a meritocracy internally. Hard to do - scum rises to the top if left alone, you have to actively keep the politicians at bay or fire them, and that's hard because they specialize at looking good without being good, to the boss (me, in that case). Fah, glad I'm retired.
I'm trying to put together a system that needs all that hyper accurate timing to tease apart cause and effect in the fusor. Can't just take data at 2.5 ghz continuously while looking for oddities that might only happen 3-4 times in 5 minutes - that's an overwheming amount of data. But lookee here - correlated within one instrument (and soon across all my data aq).
Blue trace - from antenna picking up emi at onset of a discharge. Yellow trace - neutron detector firing. Long scale at top, zoom on bottom. Nice! And I can grab this with the computer over USB, or if I buy it (likely) ethernet interface. You can see a discharge start, then the main event produce neutrons when the other grid lights off and makes it's own EMI burst....
I can't believe no one uploaded the update I put on this thread recently.. If's the first one that gives multidimensional plots off this log format for one thing. EG the reason for the entire thing to exist at all, so you can mine your data as you please, with arbitrary mapping and presets for all axes and unlimited stored mappings if you like. Stuff you'd normally be into a few grand for.
I'm trying to put together a system that needs all that hyper accurate timing to tease apart cause and effect in the fusor. Can't just take data at 2.5 ghz continuously while looking for oddities that might only happen 3-4 times in 5 minutes - that's an overwheming amount of data. But lookee here - correlated within one instrument (and soon across all my data aq).
Blue trace - from antenna picking up emi at onset of a discharge. Yellow trace - neutron detector firing. Long scale at top, zoom on bottom. Nice! And I can grab this with the computer over USB, or if I buy it (likely) ethernet interface. You can see a discharge start, then the main event produce neutrons when the other grid lights off and makes it's own EMI burst....
I can't believe no one uploaded the update I put on this thread recently.. If's the first one that gives multidimensional plots off this log format for one thing. EG the reason for the entire thing to exist at all, so you can mine your data as you please, with arbitrary mapping and presets for all axes and unlimited stored mappings if you like. Stuff you'd normally be into a few grand for.