Big NST based HV supply

Things at the limits.

Big NST based HV supply

Postby Doug Coulter » Wed Apr 17, 2013 4:29 pm

Well, while going through the extensive junk box looking for just the right stuff for the big X ray transformer supply, I also found a few things that weren't right for that, but that should make a really nice NST type supply. It's a pretty good junkbox...thanks, BillF. I'm sort of doing this for a few reasons - just to get back in the swing, to give my back burner brain more time on the big guy - it's going to be a tough mech engineering job to fit all that in the oil tub I have and it needs more thought - and the fact that despite claims to the contrary, I've found that I can make significant neutrons with my "second, ionizing grid" with voltages and currents that *should* be well within what this can make - I expect around 20kv, and it's a 60 ma transformer. Yes, we know they are "squishy" due to a lot of series L ballast built in, but I've also found that with a little capacity - you can more or less "tune that out" and not have the severe current limit behavior. So, while going through the junk, I found these massively over-rated (for this) diode stacks - 6 AMPS at 1kv each diode, 41 diodes per board (or 41kv prv) and each diode with a 2w 1 meg and a .002 cap across it. This is almost right to resonate the series L - I might add a bit more to get it just so, but we'll fly with this initially. I have a nice pair of .1uf 10kv diodes for the series caps for the doubler, and 6 .0082 uf 10kv caps for the output filter (two sets of 3 in series parallel for .0123uf or 9 joules). That's on the small side, but it's what I have - and it will be a bit safer than a bigger output filter cap would be. The idea here is that this thing should stand any abuse I can throw at it - and if I make significant neutrons off this, with any grid, I want to debunk the idea that NSTs are only demo grade fusion supplies. Hopefully I'll later have some vids of the thing in action doing the stupid stuff kids (even we adult ones) love, like arcing continuously, burning things up and so on. For fun, I do have a 1uF 30kv maxwell to make the arcs really something. We've played with that one before, and it's about like shooting a pistol indoors when you arc it...

And here's the pic of the thing in progress. The white cap shown on the left will be mounted to the clear side plate (and so will another on the other side) as the series cap for the full wave doubler. The little guys, in two groups of 3 parallel (then in series) will go in the middle between the diode strings, output on top. There will be a small copper plate screwed to the baseplate for the main ground connection for the doubler and output filter. All the plastic was scrounged, some by me, some by Bill - dumpsters and such. We won't worry that the clear stuff is scarred and has some junk printed on it, will we?

NSTSupply.jpg
Yes, I'll put a front/back and top on this later...it's not THAT safe.
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: 3515
Joined: Wed Jul 14, 2010 7:05 pm
Location: Floyd county, VA, USA

Re: Big NST based HV supply

Postby Doug Coulter » Thu Apr 18, 2013 1:42 pm

OK, this is finished other than mounting the front cover - I wanted to test with it off as I calculate some watts going to be dissipated in all those 1 meg resistors across all the diodes. It does indeed make a satisfying bang, as I hope the movie will reproduce, but a direct arc causes the transformer to shut down in self-defense for a few seconds. In a real-life arcing situation (fusor maybe, other cap discharge things) we'll have to use a ballast R or L in series to limit the peak current out of this.

Here's a lousy pic of it almost done.
NSTSupply2.jpg
Bad exposure and shake, sorry...but here it is.


And a movie of an arc test. It seems to survive these fine - but the transformer (a new one with all sorts of protective crap - buried in the encapsulation) shuts off for 2-3 seconds when you do that. It appears the R's across all the diodes do act as a good enough bleeder to not need any more of that, but nothing has gotten hot yet on this - I may not need to do more than mount the sides with a little air inlet at the bottom and the top already has some holes to let hot air out, so this ought to be super-robust in real life use. Just like what an amateur fusor maker would need - it takes awhile to build technique to the point of being able to use a less poof-proof power supply and higher voltages.
http://youtu.be/88ILAqXSIc8



It turns out this movie does NOT reproduce how loud that arc is. It's just below the ear-ringing level, quite loud. Not quite a firecracker at your feet.

Now, this only stores on the order of 9 joules (not counting the multiplier caps here since they are in series with a huge inductance in the transformer). For a real test, we'll fire up one of the big maxwells with this. I have 3 120 uf/10kv ones (that's a very large blam - 6kj! - it will vaporize copper pipe), one 1uf/30kv one (pistol shot, 200 joules) and some .06 uf/100kv ones (which we can't reach with this supply - it only goes to 20kv so won't bother with them except to make sure they are good at all).
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: 3515
Joined: Wed Jul 14, 2010 7:05 pm
Location: Floyd county, VA, USA


Return to High Voltage, High Power

Who is online

Users browsing this forum: No registered users and 5 guests

cron