Feedback

[Magnus Danielson]

From: Don Tillman <don at till.com>
Subject: Re: Feedback
Date: Sat, 1 Jul 2000 19:33:00 -0700 (PDT)

>    Date: Sun, 02 Jul 2000 01:45:32 +0200
>    From: Magnus Danielson <cfmd at swipnet.se>
> 
>    I just wanted to comment some on this feedback. For some of you it
>    migth not be obvious on how it work, it's a kind of non-linear thing
>    so let's make a model just for the sake of it.
> 
> [...]
> Wow!
> 
>    So, where this little exercise enougth enligthening to understand why
>    negative feedback in high gain op-amps may not be so beneficial to
>    distorsion?
> 
> Yes indeed.

Good.

> Ten years ago I hacked together a program to perform an FFT on the
> output of any arbitrary transfer function with feedback so I could
> watch how the harmonic distortion products changed with the amount of
> feedback.  The results were enlightening.  Really enlightening.

I take it that you ran a sine input through some model and then just FFTed it?
Thats a good start I beleive. But then, FFT has never really said the full
story (I could elaborate rather extensively on that).

> Sure enough, if you take a transfer function with simple harmonic
> distortion products and add feedback, the total quantity of distortion
> goes down but lots of new harmonics get created, and thus the
> character of the distortion moves to higher harmonic content.  And
> that translates to more intermodulation effects.
> 
> My original program was written in Lisp, so it wouldn't be very
> practical to run assuming I could find it.  Some day I'll put together
> a Java applet to do this.  It's a great demo.

Java - YUCK!

Well, if you make it work stable it migth be interesting....

I guess that one could hammer down a few lines of Matlab/Octave code that does
the whole thing.

Cheers,
Magnus