[sdiy] LM358 datasheet VCO on spilt rails

[nicolas]

That datasheet circuit has been designed around an assymetrical supply with the CV referenced to the negative supply.  If you have a symmetrical supply available you would be better to go with a more conventional approach which really just means rearranging those four resistors and the transistor on the CV input side.  For symmetrical supply and CV referenced to 0V that resistor network would be:
 - 50k between the CV-in and the transistor collector.
 - 50k between the transistor collector and the -ve in of the integrator op-amp.
 - 100k between the CV-in and the +ve in of the integrator op-amp.
 - 50k between the +ve in of the integrator op-amp and 0V.
 - transitor emitter connected to 0V.

Nicolas


--- On Tue, 7/2/12, James Patchell <patchell at cox.net> wrote:

> From: James Patchell <patchell at cox.net>
> Subject: Re: [sdiy] LM358 datasheet VCO on spilt rails
> To: synth-diy at dropmix.xs4all.nl
> Received: Tuesday, 7 February, 2012, 9:33 AM
> Yes, you can run it on split rails,
> BUT, and it is a big BUT...
> 
> Because of the nature of the LM358/LM324, you may find that
> this causes more problems that it may solve.  The
> common mode input range of the LM358 includes ground, and
> the designer of this VCO took advantage of this fact. 
> Also, you will note, that the control voltage range is from
> 0 to 2 * VCC - <some amount>.  If you were to run
> it from +/- 15 V, well, you can see the problem.
> 
> I used this VCO in the first synth I built, which was the
> circuits for the Paia 2700.  I did not like the VCO
> circuit that was in the magazine (Paia later made some major
> improvements on it), and when I built it discovered it
> worked a lot better.  The supply rails in t6he 2700
> were +18, +9 and -9 volts.  I ran the VCO on 0 to +9,
> and with the 18 volt rail, I was able to provide the VCO
> with all the voltage it needed to go through its tuning
> range.
> 
>     My recommendation is to use it in the single
> supply mode.  If need be, use level shifters to get the
> signal swings (using op-amps running on split supply rails)
> to get what you need.
> 
>     Oh, yes, and this VCO is a true VCO. 
> That, of course, has it's own problems.  This VCO will
> never be anything more than a linear control VCO.
> 
>     -Jim
> 
> On 2/6/2012 7:31 AM, Dave Kendall wrote:
> > Hi All.
> > 
> > Wondering if any of you have tried the VCO on page 1 of
> the National semiconductor LM358 datasheet, and if it could
> run on a spilt rail instead of a single rail.
> > http://www.hebertech.com/view/1448-national-semiconductor-lm358.pdf
> > As it has V+ /2 going to the + input of the right hand
> opamp, could that be replaced by GND, and all GND signals
> connected to -V?
> > 
> > Am most interested in using it as a simple voltage
> controlled TRI wave LFO.
> > 
> > Any thoughts anyone?
> > 
> > cheers,
> > Dave
> > 
> > _______________________________________________
> > 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
>