Re: Custom transformer



On Sun, 21 Dec 2008 20:25:30 +0100, David Bourgeois wrote:

Hi,

I just built the ESR meter described at
http://ludens.cl/Electron/esr/esr.html and wired a small 400:20
transformer. As expected, it doesn't have the theoretical 20:1 ratio at
50kHz. As I'm building that meter to repair my oscilloscope, I can't
-yet- look at the square wave but using a true rms multimeter, I have a
ratio of 36:1.

Would you have any pointer where I could learn on construction of real
transformers and their loss? I've read that an iron core would have too
much loss at 50kHz, I just would like to understand why. My
transformer's core is made of these "E" shape stacked metal sheets so I
guess it's an iron core.

Thanks,
David

I don't know if there is one unified source -- there's a bit of black art
to the whole magnetics thing. The ARRL has some good publications if you
want to build transformers at RF, I would expect that a good book on
switching power supplies would devote a good bit of text to magnetics,
but I don't know if there's One True Book.

In your case, yes, stacked steel laminations count as "iron core". You
have a better chance of the thing working well if you took the core from
an audio transformer rather than a power line transformer, but AFAIK
laminated steel cores aren't the good choice for switching supplies --
the problem is eddy currents, which get worse the thicker the laminations
are (who eddy is, and why his currents cause problems, is a mystery that
you may be able to clear up with some Googling).

The number of turns you are using is alarming -- one of the sources of
loss in a transformer is inter-winding capacitance; you may be running
afoul of this. Checking the voltage ratio with a sweep source, and doing
some leakage inductance and parasitic loss measurements, may illuminate
some problems without resort to an O-scope.

But your best bet is probably to just chuck the thing and get some
ferrite E-cores, and rewind the transformer. See if you can find the
Amidon website and puzzle through what it has to say on the subject; you
may find what you need there (and be able to buy some cores).

--
Tim Wescott
Control systems and communications consulting
http://www.wescottdesign.com

Need to learn how to apply control theory in your embedded system?
"Applied Control Theory for Embedded Systems" by Tim Wescott
Elsevier/Newnes, http://www.wescottdesign.com/actfes/actfes.html
.