[sdiy] Inverting Amplifier as a Buffer Output.

David G. Dixon dixon at interchange.ubc.ca
Fri Jun 25 05:45:21 CEST 2010

Harry, in almost any other case, I'd totally agree with you.  However, in
the specific case of pitch-determining CVs, I really must insist that the
"inny" resistor is better.  It doesn't take much offset to create a
noticeable tuning error.

Concerning the layout issue, this is a manifestation of how I do all my
analog layouts: ICs in a line with the rails running underneath, one row of
"free" pads next to the IC pins (where any stability capacitors are placed),
then all resistors, diodes, etc., bent to 0.4" long and arrayed
perpendicular to the IC pins with space for as many as three traces running
in parallel underneath as required, then another "free" row of pads (where
any off-board connectors are placed).  In this way, all my boards are either
2.1" wide or 4" wide (for two rows of ICs).

In the case of "A" (the "outty" resistor), I would require a jumper to
return feedback resistors from the negative input to the opamp output, and
then another 1k resistor in parallel to that jumper to connect to the output
connector at the edge of the board.  In the case of "B" (the "inny"
resistor), that jumper is replaced by the "inny" resistor itself, and the
output is taken directly from the junction of the feedback resistor and the
"inny" resistor, which is naturally located at the outer edge of the board.
Each time I do this, my board loses one 0.4" jumper and gets 0.1" shorter.

> If you have a modular synth (I assume this is the case...) you might have
> multiple outputs, each with an 'outty' resistor. If all your input
> resistances
> are the same (as I said earlier, 100K)... then when you calibrate that
> voltage
> source you will remove the gain error.
> even 10 ohms in series will cure cable capacitance instability... thay
> would be
> .01% error...
> maybe you have a really really really good VCO ???   :^)
> A or B will work. I need to consider the layout like you mentioned...
> maybe "b"
> does lay out better...
> H^) harry
> ----- Original Message -----
> From: David G. Dixon <dixon at interchange.ubc.ca>
> To: 'David G. Dixon' <dixon at interchange.ubc.ca>, 'Oscar Salas'
> <osaiber at yahoo.es>, synth-diy at dropmix.xs4all.nl
> Sent: Thu, 24 Jun 2010 00:11:50 -0400 (EDT)
> Subject: Re: [sdiy] Inverting Amplifier as a Buffer Output.
> > I used to use A all the time, but now I use mostly B, because I found
> that
> > it is a helluva lot easier to lay out on the PCB, with or without the
> 10pF
> > cap (which I always make provision for, whether I use it or not).  I
> have
> > noticed no difference in how my circuits perform.
> >
> > My $0.02 (probably only worth $0.01).
> Oh, and I almost forgot to mention, if you are using it to buffer a
> precision CV signal (like something which will be fed to the 1V/oct input
> of
> a VCO, for example) then definitely use B.  The "outty" resistor in A will
> cause tuning errors by creating a divider with the input resistor at the
> receiving end.  The "inny" resistor in B will not.
> _______________________________________________
> Synth-diy mailing list
> Synth-diy at dropmix.xs4all.nl
> http://dropmix.xs4all.nl/mailman/listinfo/synth-diy
> --
> Harry Bissell & Nora Abdullah 4eva

More information about the Synth-diy mailing list