"Iain M Churches" wrote in message


"Stewart Pinkerton" wrote in message
...


Given a flat open-loop response, NFB affects
all harmonics equally. What cancels even
harmonics is push-pull operation, not NFB,

If you had taken the trouble to read carefully what I wrote,
instead of frothing with expletives, you would have realised
that I was referring to the feedback loop from the secondary
of a push pull output stage.

So what?

The NFB greatly reduces the even harmonics, but not the
odd.

Not unless the transformer is wildly nonlinear.

The odor of male bovine feces comes to mind for some odd reason.

Come on over and have a look at the spectrum analyser.
I have a HP3581A. Even you cannot argue with that:-)

Actually, one can if it is compared to modern equipment.

The HP 3851A cost about $8K US in its heyday, but can be easily be found in
the used equipment market for less than a tenth of that. This is nature's
way of saying it is obsolete. It's no doubt as suitable for the purpose at
hand as it was on the first day of its life, but any bragging rights
associate with its ownership have long since dissipated.

According to the HP 3851A tech manual before me, this analyzer has only
about 80 dB dynamic range. A $100 PC audio interface with appropriate
analytical software (some of which is freeware) can exceed its performance
handily.

BTW, here are my calling-cards: www.pcavtech.com and www.pcabx.com .