[sdiy] Non-inverting output buffer

[David G. Dixon]

NEVERMIND!

I retract everything I said before.  Multisim was playing tricks on me
(again).  I should have thought it through.  I started over and resimulated,
and of course the 75R resistor has no effect on gain unless the load
impedance is really low, in which case the output will be clipped.

There is still no sensitivity whatsoever to the value of the feedback
resistor.

(No, Harry, the models don't typically show the oscillation unless the opamp
SPICE models are very precise, which they typically are not.)


> whats wrong with... oooh I know I know :^)
> 
> TL07x opamps have a real desire to oscillate on capacitive
> loads, especially cable capacitance. A small resistor is
> all you need to decouple that. Bet the models don't show
> that.
> 
> I find even 10 ohms is enough.
> 
> The series resistance will limit short circuit current.
> 
> With no load, the gain should be closer to one, imho...
> the 75 ohms is a divider with what, 10e7 at least ???
> 
> I just use 10 ohms after a simple divider, the 10 ohms and
> a 100K resistor (input to the next stage) gives like
> .009% error
> 
> but, having a TL07x oscillating rail to rail is NOT a pretty
> thing. I'll take the series resistor, please...
> 
> H^) harry
> 
> ----- Original Message -----
> From: David G. Dixon <dixon at interchange.ubc.ca>
> To: 'Andre Majorel' <aym-htnys at teaser.fr>, synth-diy at dropmix.xs4all.nl
> Sent: Tue, 27 Apr 2010 14:08:24 -0400 (EDT)
> Subject: Re: [sdiy] Non-inverting output buffer
> 
> Andre,
> 
> Based on a Multisim simulation using TL072BCP, I believe the question is
> moot, since the 75R resistor on the opamp output limits the gain to about
> 0.9926 regardless.  A gain of 1.0000 is only achieved when that 75R
> resistor
> is removed.
> 
> With that resistor removed, Multisim simulation suggests that the gain is
> completely insensitive to the feedback resistor value between 0 and 5M
> (the
> highest value pot I could put in) and also to the presence or absence of
> the
> feedback capacitor, regardless of the frequency between 1Hz and 20kHz.  At
> 20kHz, the opamp output lags the input by about 46ns, creating about
> 0.004%
> THD in a sine wave.  This is an artifact of the opamp.
> 
> So what's wrong with a simple follower (R1 = 0, R2 = 0, no cap)?
> 
> > Excuse the basic question, but in a non-inverting output buffer
> > like this one
> >
> >                    R1
> >         +---------^v^v---------+
> >         |                      |
> >         +------||----+         |
> >         |     33pF   |         |
> >         |  |-_       |         |
> >         +--|- -_     |   R2    |
> >            |   _-----+--^v^v---+--o  OUT
> >   IN  o----|+_-          75
> >            |-
> >
> > how do you choose R1 ?
> >
> > The goal is to have a gain of 1.0000 from DC to 20 kHz for loads
> > down to 10 k.
> >
> > --
> > André Majorel http://www.teaser.fr/~amajorel/
> > _______________________________________________
> > Synth-diy mailing list
> > Synth-diy at dropmix.xs4all.nl
> > http://dropmix.xs4all.nl/mailman/listinfo/synth-diy
> 
> _______________________________________________
> Synth-diy mailing list
> Synth-diy at dropmix.xs4all.nl
> http://dropmix.xs4all.nl/mailman/listinfo/synth-diy
> 
> --
> Harry Bissell & Nora Abdullah 4eva