Xray PSU block failure

Things at the limits.

Xray PSU block failure

Postby johnf » Wed Aug 18, 2010 10:05 pm

While testing our home made PSU block I slipped with the scope probe on the control electronics sending full audio power ie clipping severly to the multiplier stack transformer. The DMMs monitoring showed just over 240kV when there was an awful thump from the tank and the positive supply fell to 125 volts while the neg supply was at the correct 65kV. and now the audio amp is frying itself so out of the oil it came.

test fail1.jpg


Closer inspection showed all caps and diodes OK but a very small dark spec on a part of the positive supply transformer secondary

test_fail2.jpg


The winding should be around 40H but it shows a few mH --a shorted turn or two

so now I have to rewind this secondary
johnf
 
Posts: 433
Joined: Sun Aug 08, 2010 3:51 pm
Location: Wellington New Zealand

Re: Xray PSU block failure

Postby Doug Coulter » Thu Aug 19, 2010 7:52 pm

40 Henries? That must be one heck of a lot of turns and the core must be iron? What F were you driving that at? I guess I'm presuming that theres some multiplication on that after the transformer? 240kv is a pretty wow kind of voltage output! I have a way to get to 125kv here (at some decent power) and boy, do you want to be careful with that -- seems everything under about a gram picks up a charge and flies around the lab, huge long sparks -- scary with the stored energy in the CW stack too. I had a spark jump off a FT, away from the big grounded tank, fly halfway across the room, make a U turn just before it hit my grounded milling machine, then come back and hit the tank -- only a few inches away from the end of the FT! Glad I was on the other side of the tank from that action!

We got ourselves stuck with some big X ray supplies that self resonate around 2khz (which is probably about how fast the diodes are anyway) but it takes around 10a to get 120v across them at no load anyway. Seems they cheaped out on the amount of iron in the core (small cross section). They seem to have real low leakage inductance anyway -- seems they were rated for about 200 ma at +/- 80kv or so. I've never seen anything like that out of them, at no load the cores are hard saturated at about half that at any frequency.

At 60hz, forget it (they saturate at 6v or so), they must have used something HF in the controller, which we didn't get -- we just got two xfrmrs and volt doublers in a big tank full of PCB oil (yuck -- but all cleaned up and safely disposed of now). They don't even like 2khz sq waves, as I found with my H-bridge drivers -- I had to add some series inductance or the fast rise of the square waves made things even worse (probably diode turn off times). So I wound up using that audio amp in bridge mode and sine waves for when I use them at all, not so great but usable for some things. Seems these were meant for full load only, so the cores don't saturate so easily. A learning experience.
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: Xray PSU block failure

Postby johnf » Thu Aug 19, 2010 10:24 pm

So the secondary coil for the positive supply seems to be the only casualty.
the coil former had some nasty black bits on it that I cut off
zapped coil former.jpg


then of course 2 couple of hours of hand winding

coil rewinding.jpg


two transformers on the one set of :U: cores one each for the negative and positive X-ray tube supplies. each transformer feeds the centre of a 18 stage multiplier ie 9 each way from the tramsformer input.
Max output volts from transformer is supposed to be 7kV giving each psu 126kV minus a bit for diode drops etc.

PS cores are gapped to 0.7H on the secondary side with the primary having 29uH which at 31kHz = approx 5 ohms
johnf
 
Posts: 433
Joined: Sun Aug 08, 2010 3:51 pm
Location: Wellington New Zealand

Re: Xray PSU block failure

Postby Doug Coulter » Fri Aug 20, 2010 9:12 am

Cute little coil winder. I have a version of that with a motor and a footpedal I sometimes use, but I also built a jig for my lathe I use a lot more (running slow, of course).
Someday I plan to make the cam jig for the lathe so I can do that "universal" winding thing for lower stray capacity. I use the lathe all the way down to #40 wire, with my hand on the
wire feed to avoid sudden tugs on the wire that would break it --kind of like the arm on an old RR tape deck to help with constant tension, same idea. In low gear, the lathe starts and stops slowly, making it easy, and when I goof, I can make it go backwards to unwind the goof part and do over. Of course, for single layers, it's nice to use the lathe threading settings to get "just so" wire spacing -- pretty much whatever you want there. As long as you keep a little tension on the wire, it will hold till whatever glue/wax/shellac/glyptal you use to keep it right gets hard.
Very nice for RF kinds of things.

Question -- how does that system handle getting from one "pie" to the next without having an insulation problem like you have if you just come off the top of the first one to the bottom of the second (which means a wire alongside the whole second pi to be arced to from the top of the second pie). That one has been an issue here...with the HF stuff having so many volts/turn.

Nice to know there's someone else here who winds their own coils...not common, but a very useful skill :) I've been lucky to collect a large amount of magnet wire in nearly all sizes there are, so I go for it. With the price of copper these days -- it's an investment!
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: Xray PSU block failure

Postby Bill Fain » Sat Aug 21, 2010 10:12 am

Hi, Could you run the feed wire through a pot of coil/corona dope or shellac as you are winding the coil to increase breakdown voltage? Would the trouble and mess outweigh any benefits? -bill
User avatar
Bill Fain
 
Posts: 90
Joined: Sat Jul 17, 2010 10:23 am

Re: Xray PSU block failure

Postby johnf » Sat Aug 21, 2010 2:08 pm

Bill
yes it might help but Shell Dialla has very good specs with regards volts per mil.
other than getting a bit oily I can pull the transformers out change primary turns, change from pimaries in series to primaries in parallel etc very quickly. also under oil there is no corona loss, hence ozone, hence no corrosion of the metal bits
johnf
 
Posts: 433
Joined: Sun Aug 08, 2010 3:51 pm
Location: Wellington New Zealand

Re: Xray PSU block failure

Postby Bill Fain » Sat Aug 21, 2010 2:57 pm

John, Hi. When I initially read the post I thought "240kv, Wow! that's a lot". I guess somewhere deep in my head I knew that it would be nearly impossible to get that kind of voltage without oil (with the dimensions you are using now), but conveniently forgot about it when I posted. Sorry about that. -bill
User avatar
Bill Fain
 
Posts: 90
Joined: Sat Jul 17, 2010 10:23 am

Re: Xray PSU block failure

Postby lutzhoffman » Mon Aug 23, 2010 3:15 pm

Hello:

Just a comment: If your voltage surges way above the X-ray tube rating, then it can arc across the vacuum space between the cathode and anode. With a small amount of stored energy, the tube can often survive. I am just mentioning this because creating what is in essence a superfast switch at the output of the HV power supply could factor into the failure analysis, HV pulse energy can be a complicated animal. Very impressive setup none the less, some pretty cool home engineering.

In the past I fried a perfectly fine x-ray setup by simply not loading the tube correctly, so with a to low, or even maybe a no load, condition the HV shot up and fried my VM. Lesson learned well, I never repeated that mistake. On a related note: In the medical field I have seen tube failures due to rotating anode driving circuit failure. When a 10-50kW electrom beam hits a square mm of anode, even W cannot stand the heat. With one partial rotar failure, the tube anode had a seried of pits which "recorded" the HV 3 phase pulse pattern. The tube was of course ruined due to the conductive W coating which was now deposited on the inside of the tube envelope.
lutzhoffman
 
Posts: 31
Joined: Thu Jul 15, 2010 1:57 am

Re: Xray PSU block failure

Postby Doug Coulter » Tue Aug 24, 2010 10:53 am

Lutz (and all), the topic you bring up here inspired me to share a bit more of my library on this, here.
I will also soon add one on high volt, high power switches, same idea -- we'll try to keep it organized, sorta-kinda. Looks like we'll want to start a thread/topic here about HV transformers, a subject dear to at least me, Lutz, Tyler, and of course John...

So if anyone has more on those two topics - get them up there with links. I will also work on getting some good literature on HV transformer design and start a thread on that along with some of the attempts here (many of which might as well go into "it almost worked" instead).
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 8 guests