How to check fault in transformer?

Things at the limits.

How to check fault in transformer?

Postby chrismb » Sun May 22, 2011 3:42 am

On Friday I did a booboo. I have a 600V/600mA old fashioned linear electrophoresis supply, in which both sides of the output are isolated. I caused the outputs to float high and they probably [can't tell for sure - no part of the system is intentionally grounded] ended up going 5kV to ground.

(There's a lesson in there for deciding when to tie to ground and when not, I guess, but it was an experiment and if I knew exactly what was going to happen then it wouldn't be!! :? )

Well, a transformer-type humming started up and then the fuse blew. I was hoping it was a short to ground of a small part, but oddly, after desoldering each part in turn to trace, the fault appears to be in the isolation transformer itself.

What is odd is that there is no short to ground in it, but it seems to be a short in the secondary. The resistances all look OK on the terminals so I suppose what has happened is that a couple of turns into the secondary coil something has fused making an alternative 'few-turn' coil causing very high current internal load that I can't see.

Does this make sense? And perhaps it just gave up (being 40 years old) and was unrelated to terminal float?
chrismb
 
Posts: 620
Joined: Thu Aug 05, 2010 6:32 pm

Re: How to check fault in transformer?

Postby Doug Coulter » Sun May 22, 2011 9:43 am

I've had this happen for a variety of reasons. You probably have a shorted turn or few. You can test for that various ways, but you already know the answer -- the thing is toast, stick a fork in it, it's done. Usually can't be repaired without heroic effort.
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: How to check fault in transformer?

Postby chrismb » Sun May 22, 2011 10:49 am

OK.. so decided to get on with substituting the transformer, and the result is very nice.. looks like it has always been there. It is a slightly higher rating but otherwise exactly the same spec and is a 'new' isolation transformer, so happy [enough] I did that job anyway.

So the basic layout is variac -> feeds isolation transformer -> feeds bridge rectifier, then two 'double' caps via an inductor. [I think - please let me know if it looks like I misunderstood that little transformer thing with two inputs!!!).

When I disconnected the bridge rectifier from the transformer it still hummed badly, so am satisfied the transformer was toast and needed replacing.

But after replacing the transformer, the inductor is also humming alot now. [I have disconnected the smooting stage from the output lines, so only the smoothing stage attached to transformer.] Problem is, I'm not sure what 'normal' humming is, but I don't think it is right. (Still no measurable shorts to ground on the inductor.)

Question is... what is this secondary fault most likely to be following a transformer short; is it a fault with the rectifier [could one side have gone down?] or a capacitor internal breakdown, or a short resistor. I don't know what fault can happen to inductors that would make it hum?

If it could be anything, I might as well replace all the parts. :roll:

I'm only thankful enough I have an ancient linear supply that I have a half-chance of understanding!

smoothing_board.gif
"Gutted"


out_old_in_new.gif
Out with the ancient, in with the new!


new_inst.gif
New Installation, with circuit of smoothing board.
chrismb
 
Posts: 620
Joined: Thu Aug 05, 2010 6:32 pm

Re: How to check fault in transformer?

Postby Doug Coulter » Sun May 22, 2011 10:58 am

If your inductor is humming, it's an indication of too much AC current going through it. Likely a shorted diode (easy to check usually, they rarely fail in a mode that makes them into a low voltage zener), or a shorted capacitor.
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: How to check fault in transformer?

Postby chrismb » Sun May 22, 2011 11:32 am

Yeah, sounds like a plan to check out the rectifier.

....[later..] yup, rectifier gone. Good call, Doug.

Sigh. I'm a bit annoyed because I think [thinking back] the signs were that that went first, then the transformer which I might've saved if I powered off quicker. Oh well... such is the game we play doing this stuff! I daren't even think of the more expensive things that can go wrong! :?

I'm wondering if I should change the capacitors as well. Seem a bit underrated - they are curious german things with what looks like two (47uF) capacitors build into each case?! Each have '350V' on with two + terminals connected together for each can and the two cans are in serial. Seems a bit low on overhead, 700V, for 600V output, but I'm wondering if this isn't just the way the German's used to do their over-engineering in the 70's - when nominal 350V includes overhead and means 500V (!).
chrismb
 
Posts: 620
Joined: Thu Aug 05, 2010 6:32 pm

Re: How to check fault in transformer?

Postby Doug Coulter » Sun May 22, 2011 2:27 pm

If old, likely the caps have lost some capacity and are getting ready to open up. It's only recently that 450v electrolytics haven't commanded a super high price premium. Electrolytics "like" to run near their rated voltage. If let sit, or run at lower voltages, the capacity tends to increase, or they tend to need "forming" to run at spec after that. Matshusta (panasonic) make the best E caps going for cheap, it's all I use for quite some years now.
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: How to check fault in transformer?

Postby chrismb » Mon May 23, 2011 4:46 pm

Changed the bridge but decided to leave the caps, but now thinking about changing those also....

Looking up the [old] spec of the bridge that was in there, 420Vrms/600Vrm and 1.6 A. Blimey! For a 600V/600mA PSU! They ran close to the wind in those days [1971], eh?!!

Well, I got hold of some IXYS 54A 900V parts for next to nothing and fitted one of those. No worries about engineering margin any more!!
chrismb
 
Posts: 620
Joined: Thu Aug 05, 2010 6:32 pm

Re: How to check fault in transformer?

Postby Doug Coulter » Mon May 23, 2011 5:58 pm

It may not have been so close to the wind, as the Germans using a more conservative version of "specsmanship" in those days.....they got into marketing spin later than the rest of us, and might just have rated their parts more conservatively than we -- taking actual use cases into account better than some jerk trying to say his line of semis is so much better than the others.
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: How to check fault in transformer?

Postby johnf » Mon May 23, 2011 6:45 pm

Okay you want to check a transformer for possible shorted turns

easy
use acheap RCL meter to measure the inductance this should vary as the square of the turns ratios and the inductance drops like a stone with a shorted turn or two
johnf
 
Posts: 433
Joined: Sun Aug 08, 2010 3:51 pm
Location: Wellington New Zealand

Re: How to check fault in transformer?

Postby Doug Coulter » Mon May 23, 2011 6:59 pm

Since that's the title, John is totally on-target here ;) So even though Chris' problem is solved, the info payload for other cases is now here. Cool :D

I do that some (bridge reactance/impedance meter), but more often just look at quiescent current at rated input, or in extreme cases, hang a cap on there, drive it from an oscillator with about 100 ohm to 1k series R, and look at the resonance curve. There isn't one if there's a shorted turn, more or less, and if the transformer or inductor is good (assuming I've chosen a cap and F that aren't stupid) it's very much obvious on the scope on a frequency sweep -- this isn't a case of looking in the noise for the information, it's big. Either technique is best - depending on what's easiest to lash up on one's lab.

Due to hysteresis in some core types, things look a little different at low level than at high (or at different frequencies where eddy currents kick in big on laminated steel). It's easy to learn that "fudge factor" and get good diagnostics despite it.

Blowing fuses at no load -- that's a gimmee.
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