Old X ray supply with a *real* transformer

Things at the limits.

Re: Old X ray supply with a *real* transformer

Postby Doug Coulter » Sat Jun 08, 2013 5:32 pm

Almost done! Tested at 1kv/10ma. The current sense is only giving me about 1mv/ma, which is no where near as loud as I want - I'll look into that. I didn't want to test it higher till I got it under oil, which it now is.
Sadly, I forgot to put the copper screen between the tubs, so I'll have to pull it out and deal with that at some point soon...come up with a siphon to remove the oil to make the inner tub light enough and so on.
In real use, the oil will be higher level, which I'll probably achieve with the moral equivalent of putting in a couple bricks (in this case, HDPE scrap). But, it works! No reason to believe it won't work well at full power at this point, once the oil finds its way back into the transformer windings - this guy is too nice to fry.
HVOil.jpg
Under oil now - be patient and let it get into the transformer.


It takes a fat cable to not worry about arcs internally. I used one I got from Cliff S at Spellman, rated for 160kv or thereabouts, as everything else here is overkill, might as well...Since he got me two, I still have the "better" one for the 160kv Spellman power supply that we got ripped off on at ebay, which I have to fully rebuild - the only working parts are the stack and the transformer in that one right now - they shot it here via cannon or something. So here it is with the cable showing - that's 1" diameter insulation, shielded. "The right stuff". I'll add a copper ring to the end so it fits my resistor mounts I have on the back end of my HV FT.
For this one, the ballast is in the supply - it would have been far too dangerous to have a wire out in air with 950+ joules of filter cap directly on it. That's one heck of a bang in the event of an arc, so I put the ballast under the oil with the rest.
HVCable.jpg
Check this fat cable - it's the good stuff for real HV.
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: Old X ray supply with a *real* transformer

Postby Doug Coulter » Thu Jul 18, 2013 4:59 pm

Just got a little testing in, some more to do, but this appears about ready to go into real service. At 2kv pre-ballast, with a 2k load on it post ballast for about 20 ma, it has about 12% of pure 120hz ripple - not great but not bad. Nothing gets hot.
I now need to figure out the burden I want to put on the current sense winding - one more thing to measure before I button it up, and do one more test on load vs no load volts. I'm looking for say, about 1v/milliamp there, which I can easily divide down later (inside a shield) for data aq - I like nice "loud" signals that swamp ground loop noise in my instrumentation coax.

At the low volts I've been testing at, there's quite a difference between no load and say 10 ma - voltage shoots way up (as measured before the ballast resistor) under no load. This could be accounted for at least in part by the large rated forward drop of the diodes under load - 150v/block, and each diode in the full wave doubler is really two blocks, which is quite a lot of forward drop - it amounts to almost 50% of output at these low voltages. If that's all it is, we are golden. What concerns me is if that's not the main cause and open circuit volts are really almost twice the under-load volts, in which case adjusting for say 80kv under load would result in a very high output (maybe breaking things) if the fusor light goes out, as it does inevitably when running near the low pressure limit where I like to run (highest Q is there). I will have to test that with the real setup, though. I can do that at say 40kv or thereabouts under load, then see what happens when I make it "lights out" - a doubling then wouldn't hurt anything much.

Guys, if this works out as well as it looks like it may - we'll be setting some new output and Q records shortly!
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: Old X ray supply with a *real* transformer

Postby Doug Coulter » Mon Nov 25, 2013 3:12 pm

Almost done!
BigHV.JPG
Almost...


I started getting more than a little nervous about this one. Something about the 1500+ joules stored in caps being around the same energy as a rifle shot (small rifle, but...) while being under a flammable substance. Imagine a few gallons of vaporized oil-air mix sprayed all over the lab and ignited. No more lab...and maybe no more me. So.....Bill was kind enough to have this nice steel box made for me, and what you saw in the pics above is now sealed up inside 1/8" thick steel - with air outside the plastic tubs to provide a bit of "pressure rise-time increase", so hopefully, if this "blows" it won't kill anyone, X rays being a separate problem. The small twisted yellow wires are the current sense leads, the purple/brown wires are the voltage sense (post the 100k ballast resistor), and the speaker wire is the mains input. Think of it as a fuse - it's number 16 while the transformer primary is #6...we won't be needing that level of power for this, and I'd rather have that wire smoke than what otherwise *might* happen.

Obviously, the central wire is HV cable, in this case, rated for 160kv, or a lot more than we'll need (the output filter in the supply is only rated at 100kv. It's full coax, which will help with the EMI, and I will put an end on it that fits inside the current feedthrough hole that's now holding a 225w 50k ballast - turns out it's just the same size as 1" ID copper pipe's OD, so that'll be easy to do, and it will integrate fine. My main concern is that this piece of coax surely has some capacity of its own, and the ballast R is at the other end, so I'll have a bit more stored energy at the grid than just what the feed through has. We'll just have to do some testing to see if that's a bug or a feature (worry is, it's probably a bug).

This is all setting on a table I built with 4 way swivel castor wheels. One rolls this *very* carefully so as not to slosh oil over the sides of the plastic tub in there - they are about 1/4" from totally full. When we put this together, we drained out most of the oil first so as not to have to worry about that one.

Stuff left to do is calibrate the V and I sensing so my data aq gets good numbers, and of course, something closer to full power testing. I might wind up driving this with a nice 3.5kw audio amp I have at about 1khz (we'll have to see how high this monster transformer will go before self-resonance) to reduce ripple, and perhaps make it safer (the audio amp has safties in it too). Then make a real controller. It'll never be as good as a Spellman, but it might come out good enough...I really got spoiled with the Spellman's fine control and perfect current limit with decent bandwidth...but this should serve if I do the rest as best I can.
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

Previous

Return to High Voltage, High Power

Who is online

Users browsing this forum: No registered users and 3 guests

cron