Extend freq audio xfmr

I'll take a stab at it..... Allison's Simply Dippy?

Don

I usually figure I've done a good job with a transformer design if the leakage inductance is under 1% of the magnetizing inductance. It's been a couple of years, but it seems like that one had a much higher ratio, with the magnetizing inductance well up into the 10s of milliHenry;. All turns are effectively multi-filar, with no insulation other than the 'varnish' between the primary and secondary. The main point I was trying to make is that audio distribution transformers, when not designed as if they were 50/60 Hz power transformers rated for high average power and requiring lots of insulation, can have much better performance. According to my investigations, most of the popular distribution transformer designs do not take advantage of this fact. Paul Mathews.

"Paul Mathews"

L mag is in the Henries, for a toroidal design.

So that absurd 20mH figure was OK by you after all ?

....... Phil

That makes good sense. Somewhere in this thread, I thought you were holding these transformers up as something state of the art, close to the ultimate, not just very functional and better than most. Fair enough.

The O.P. seemed interested in finding transformers that perform well at frequencies (possibly much) higher than 25 kHz, not just in the audio spectrum. Are you proposing he use these? I was just thinking about where the first limitations for say, (pulling a number out of my ass) 100 kHz use would occur with your design.

Those transformers are designed to be much better than those commonly used for PA distribution without being priced radically higher, and they are. For crazy Phil's benefit, I dragged one out and measured its magnatizing inductance: 7.8 Henry at 1 kHz. Leakage inductance, as measured on the secondary side for the 70.7V (100 watt) tap, was 13 microHenry. That's really pretty good, if I do say so myself. These are also not autotransformers, which makes the job of reducing leakage inductance more difficult. However, any transformer designed to transfer power is a compromise, so choose the best compromise for your app. Returning to the OP's quest for extended HF response, there have been several good suggestions. If he sticks with a single transformer, crazy Phil is absolutely right to say that leakage inductance is the main limiting factor, but it results in a simple rolloff for at least an octave or 2, and core saturation is not a problem at high frequencies, so some sort of pre-emphasis can be used to compensate. Paul Mathews

No, none of that! Rather, "pedantic ass".

Get it right next time.

"whitless ASS"

- f*ck off .

....... Phil