Old X ray supply with a *real* transformer

Things at the limits.

Old X ray supply with a *real* transformer

Postby Doug Coulter » Fri Mar 29, 2013 5:06 pm

I'm taking apart a very old X ray power supply (WWII?) that BillF found us. This had been hit so hard with forklifts and so on, the case was so dented it just wouldn't come apart. Heck, that just doesn't happen on my watch...Time to get brutal. After screw-jacking the case off as far as we could, trying to drill out and pull the dents - not enough...time to get real.
http://youtu.be/-gdPzZRNaFA

It took grinding off two edges to spring this thing far enough to make the innards come out. Whew, that was a lot of sparks and stink, but I didn't hurt anything other than the outer box doing it. If I was crazier, I'd fix the dents and weld it back up, but I'm not - I'll find something else to put the supply and oil into.
DeadBox.jpg
Take that, you 70 lb piece of junk!

And so - we now have the innards. One of the edison base tubes was internally broken - but I don't plan to use any of them. Anyone need big old tube HV rectifiers? 3 are good.
XrayInnards.jpg
And, the mystery is revealed.


We're not going to need all those filament transformers etc, so apart it comes. I'm just going to use the transformers, and some diodes I bought from JerryB (another member here). Sadly, I can't get full volts - the transformer is two windings, in series, but the CT has to be near ground...there's not enough insulation to series them up. Still...we'll see what we can get out of this. Maybe I'll make some nice real HV caps and a voltage doubler.
WheatChaff.jpg
Just the part I want, thanks.


This time I won't be so dumb as to try real volts/power on this before getting it back under real, fresh, oil. The stuff in there - yuck, rancid and high sulfur. Sickening smell. More when I make some progress...
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 chrismb » Sat Mar 30, 2013 4:02 am

Doug Coulter wrote:This time I won't be so dumb as to try real volts/power on this before getting it back under real, fresh, oil. The stuff in there - yuck, rancid and high sulfur. Sickening smell. More when I make some progress...


Gee, Doug.... does this stuff contain PCBs? I think byphenyls smell 'nice' (!) but don't know what they would smell like as they break down. Probably stink of chlorine? Be careful, fella....
chrismb
 
Posts: 620
Joined: Thu Aug 05, 2010 6:32 pm

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

Postby Doug Coulter » Sat Mar 30, 2013 12:31 pm

Got some testing done today. Looks like these were meant for 240v input. All the transformers seem good, and the big boy has a 169::1 turns ratio between the primary and either secondary. With a one stage full wave doubler (4 diodes, 3 caps), I calculate 57kv for 120v input, or 114 for 240. Zowie! #8 wire for the primary makes me think those tube ratings of 125kv and 300 ma are real...I ordered 3 100kv .06uf maxwell caps off ebay - it's too darn many joules (and much higher quality than needed), but it's what I could find. Now in the hunt for some really horsey 125kv or so diodes. This is going to be a real monster. As luck would have it, we have cables and receptacles that match that will let it all be under oil with a cable sticking out, so that end, no problems. Obviously we will have to put some ballast R right there in the tank to avoid arcs with all those joules, we don't want flying molten metal in the lab.

Edit:
I've had this junk oil all over me, and yes, I'm sensitive to PCBs - no reaction other than the sick smell. My best guess is that this is so old it predates them(!). It also passes the burn test - burns yellow like normal oil, not greenish like PCBs do. It's a good thing, this has been a pretty messy undertaking. I'll replace the oil with fresh mineral oil from some vendor I suppose.
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 » Tue Apr 09, 2013 4:37 pm

Got some cool new parts today. Looks like I'll be giving some real positive feedback to ebay sellers on these. Imagine - maxwell caps, .06 at 100kv, 6nh series L - $75 each.
I paid a bit more for the variac (20 amps, 240v) - about $300, which is about 1/3 of what everyone else wanted - and this one is for all intents, new and perfect.
HVParts.jpg
Might be a trick to fit all this in the box with the right clearances...but I won't need much oil either.


I plan to use the two 50k 200w resistors as a ballast - under the oil. The big 30k will be in series with the transformer CT to help protect the diodes should the main ballast not be enough - I just don't need all this thing can possibly give. Diodes rated at 110ma continuous, 10a peak, 150kv or so. Transformer, who knows, but calculates out to 117kv out of a full wave doubler...and the input wires are #6...zowie! I expect the variac to be the "weak link" if there is one, since my inverter for this will happily put out 4kw continuous (doesn't even make the fan run) or about 10 times that for 10-20 minutes...

Since the pic was so shabby (should have used a tripod), I thought I'd try a movie off my new Nexus 10 tablet...Not that it's a ton better, but at least the motion gives more angles. It's not as high rez as my "real" vidcam, but then there's this tradeoff - the real vidcam takes over 150 mb/minute to upload...this one has lower rez and is already in the right format for youtube. http://youtu.be/1kxlb365yNw



Looks like a real challenge to fit this into the box I selected at Lowes with decent spacing for the volts involved, even under oil. The upside is that if I can make it fit - I won't need much oil at all to fill in the empty space, since there won't be much of that at all. Most of the tricky stuff will go elsewhere in a main control box that will be remote from the supply - the supply will be placed right next to the HV input for the fusor, of course, and I don't like to be back there when its running - I like to be in front and behind the lead shielding. Which I'll have to add a lot more of at 100kv - it was already marginal at 50kv. I may even have to plug up the deliberate leak I left in the back to measure things, since there is a lot of scattering of X rays off the air, the ceiling and so on as is. It's only going to be worse at higher energies I suppose, though there is a chance that what I measure in front includes things like 2.2 mev X rays from neutron capture in all the moderators around and things like that - and I simply don't have the tons of lead required to stop all those. Luckily, this is getting to be mundane enough to consider operating it remotely with some automation, so I can put it someplace I'm far away from during a run!

The reason behind all this is what I saw in the parametric sweep plots I took earlier. It looked like a sweet spot for Q was developing from about 40kv on up, increasing in current (and Q, and output) as the volts went up in very good agreement with the rules for space charge limits on focus I've never been able to convince the people at fusor.net are real and well worked out long ago for electrons - a couple constants have to change for the different e/m ratio is all...nice to have the theory agree with what I measure!
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 » Tue Apr 09, 2013 9:54 pm

Sure hope this is right (or close) for the mineral oil I intend to use as "transformer oil". Else I'm going to have problems.
Breakdown.gif
From Kohl
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 Jerry » Thu Apr 11, 2013 1:59 am

Need to nip this one in the bud...

Jerry
 
Posts: 573
Joined: Sun Jul 18, 2010 12:07 am
Location: Beaverton, OR

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

Postby Doug Coulter » Thu Apr 11, 2013 10:16 am

Funny one, Jerry - I suppose I probably fall closer to group 2 than anything :oops: . But in my own defense, if you saw my floor plan and how tight it is - I actually couldn't get to a place to shoot that long skinny pile from the side in the normal orientation, so I went with what didn't require me to move all that very heavy stuff that's still coated with stinky oil.
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 » Tue May 07, 2013 12:43 pm

Every time I order from McMaster, they amaze me. While this stuff wasn't super cheap, it's the right stuff. Boron nitride rod to replace the rapidly degrading quartz in my new hv feed-through, a drill to bore it (special, long, self centering), OFHC copper rod to push through the middle - ordered last night, here at lunch. Gheez.
I also got some bigger rectifiers from a Chinese outfit that actually ships on time at decent prices. When I find the link, I'll update our vendors list - these guys are great, and it's all the HV stuff you can't find and dream of scoring. I just got 10 of those 80kv/.2amp guys, am getting a gig's worth of 200m 5W resistors from the same place, and yes, they sell all sizes and ratings of capacitors, diodes, and resistors for HV. Pretty neat.
But here's the pic - might have been eye-candy for some, it was for me. It doesn't pay to look too closely to my Tig welds on the brackets I built to hold that huge transformer off it's windings in the tub - but they're all in compression and almost not needed at all. Good enough.
NewHVParts.jpg
New stuff to help make the big supply....
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 » Sun May 26, 2013 1:32 pm

EEEEK! How do I mount these Maxwell caps, which only have one terminal insulated from the world at full rating? After all, in a doubler, both sides are "hot" in varying degrees. I was concerned for a couple reasons. One is the thing tilts when being hoisted, and if just wires were holding them, they might fall off as the oil drains off.
Also, since I'm using an old oil-soaked board (saved out of another HV supply we scrounged) for the base - what if it arcs across the board itself (through the plastic walls of the caps, ruining both)? Gee, it's not like they provide for mounting these by anything other than the terminals (bad move IMO). Soooo...here's my solution - mill out little pockets in some HDPE, and glue the suckers down with oil-proof silicone gasket maker (not caulk - this stuff is special and expensive at the auto parts store). I can then bolt the HDPE down to the board - and use the bolt for a ground point. Since this whole mess sits on top of another piece of HDPE in the bottom of an outer tub, it should be fine. There will be copper mesh screen between the tubs, grounded, to keep me safe - and if an arc blows a hole through the first tub...I still will have the outer one to contain a spill. The amount of energy .06 uf will store - one at 100kv, the other two at 50, is no joke - I'm betting it'll pay to have been this careful. Of course, the ballast R should keep peak currents to the outside world down, but - there's always a but - what if it arcs across itself, and shorts - then I've got nearly a thousand joules that are going somewhere very quick - these caps have only 20 nh of series L and next to zero series R. Better safe than sorry.
MaxwellMount.jpg
Milled out HDPE, work in progress for the other two caps.
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 » Fri May 31, 2013 2:10 pm

Making a little progress on this. Got the base all made, the diodes and caps wired up, now need to drill/tap the top plate for connections to the outside world. Bill found some nice X ray cable - shielded - that will carry the 100kv output to whatever. The plan is to have 100% of the HV stuff under oil (which is why it's so tightly packed - oil isn't cheap) or inside shielded cable insofar as that is possible.

The assembly will go into an HDPE tub from Lowes, which in turn will sit on top of a 1" thick piece of HDPE inside another tube, with copper mesh screen, grounded, between the tubs. The hope is any accident will only hole the inner tub - this is way too many joules, but the caps are what I could find - they are "too good" for this on a number of levels, being Maxwell high energy pulse caps - 20nh or so series L, and virtually zero series R. About 1k joules all together. Zowie! Since the low end of each secondary on the big transformer floats, I decided to copy the original design for current monitoring - you hook those to the outside of a center tapped primary of another transformer, and burden the output of the measurement transformer with a bridge rectifier, capacitor, and resistance to get a floating current signal. I'm putting a 240v spark gap across the highest impedance part of that winding, but probably will take the current signal off at a lower tap - I like loud signals, since they are EMI resistant, and then can be easily divided down for data aq, at that end of things. There will need to be an opamp polarity inverter for the 1 gig ohm voltage sense resistor (made of 5 200 meg 10w resistors in series). But all fhe fancy control box junk can be built while we wait for the air to sneak out, once we get this under oil after some minimal testing at no more than a kv or two.

Here's what it looks like now. I built the cart on wheels to move this around, and make lifting it out of the oil with a winch less work. Learned my lesson after one hernia lifting that pig - that's one huge transformer - a cubic foot of iron and copper. We won't be using it anywhere near its limits, I suppose. But for once - unlike the "generous engineering" used in modern X ray supplies, that run transformers at post-saturation and will only stand very short pulses - this will go all day...if we want.

http://www.youtube.com/watch?v=dcl6Kc6Yigc


Obviously the real fun begins once the oil is soaked back into the transformer and we brutally test it. Stay tuned!
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

Next

Return to High Voltage, High Power

Who is online

Users browsing this forum: No registered users and 3 guests

cron