Compensating multi-stage feedback (was: RE: all tranny vca+ )

[Harry Bissell]

(sound cue: Austin Powers...)

Oh YEAH baby....

This is true.  For instance crossover distortion is very noticible...
because it is level independant.  And the "soft clipping" that (here comes
the "T" word...) tubes provide
is sought after by guitarists. I even own two such amps....

And if you allow the listener to A/B compare, at very close time intervals
they 'can' tell
a difference between two sources that differ in distortion.

But I don't usually hear any "op-amp" distortion unless the circuit is
poorly designed or
driven beyond its limits... slew rate limiting distortion isn't there
unless you exceed the slew rate. Did you take too much gain in one
stage... or choose an unsuitable device... and of
course the dreaded "phase reversal" distortion sounds like PURE SH!T...
but that again is a misdesign or misapplication problem.

OnceI built a device said to generate "even-order" harmonics at specified
levels. I have NEVER heard a worse sounding distortion in my LIFE !!! But
I bet that was at 1% or
more...

Transient Intermodulation Distortion is everyone's favorite (not!). But I
bet it takes a hell of a speaker system to reproduce it at levels that the
mean (average) listener can detect... and most of them do not own such a
speaker system.

I don't like nails on a chalkboard any more than the next guy... I'm just
saying that nails-on-a-chalkboard, at 80db down, in the presence of a
window fan running... just
is not an issue.  If YOU hear distortion and it is a problem, fix it. But
if we make up un-real world tests to uncover misbehavior... then we have
another problem.

How much high-end "audio-phile" gear requires hand selection of components
to find
the "sweet" sounding one.  At least the opamp-negative feedback is
repeatable. Try putting 12AX7 (or equivalent) in the front end of a
Mesa-Boogie guitar amp (very high gain...) and you will probably find you
need to throw away 75% of them.

For synth use... waveform buffering, modest gain, etc... the opamp stamds
up very nicely and can be used even by a "newbie-phile"

BTW thanks for the torch-job.  Gets the blood boiling a little bit...

NEXT ???

H^) harry

Don Tillman wrote:

>    Date: Sat, 01 Jul 2000 16:44:21 -0400
>    From: Harry Bissell <harrybissell at prodigy.net>
>
>    I think (flamesuit ON) very few of the people in the world can hear
>    the distortion in an audio system unless it is GROSS.  Usually the
>    speakers are far worse than anything else in the audio path!
>
> [Blowtorch on...  whahishhhhhhhhhhh...]
>
> Your argument assumes that all distortions of a certain measured
> amount have roughly the same effect on the listening experience,
> regardless of the process that causes the distortion.  You know that's
> not true; 0.001% of one type of distortion could be especially awful
> while 5% of another type of distortion might not be a problem, or in
> some instances might actually be an improvement.
>
> The process that creates opamp distortions is nothing like the process
> that creates speaker distortions, so their characteristics will be
> completely different.  Personally I'd place opamp distortion in the
> nails-on-a-blackboard 0.001% catagory.
>
> [blowtorch off :-)]
>
>   -- Don
>
> --
> Don Tillman
> Palo Alto, California, USA
> don at till.com
> http://www.till.com