Another Triangle VCO Idea

[Jim Patchell]

Harry Bissell wrote:

> Hi Jim:
>
> doing it with current sinks is a twist on an old National Semi Triangle VCO
> method
> AN74 fig. 17 for one example.   If there is a hidden weenie in this I don't
> know what
> it is... the resistor fed version here does not work super well, but your
> current control
> should fix those problems...
>

    Not familiar with that Ap Note, but, if this is the same triangle VCO that
is included in the LM324 data sheet in the national data book, yeah, the idea
is sortof the same.

    What I am hoping this circuit will give me is the same nice clean triangle
that you get when you use the OTA based current switch, without the temperature
dependence that the OTA gives you (the more you saturate the OTA, the less temp
dependence you have, but then that causes other minor problems).  The FET
switchs work pretty good, but, you get the switching signal coupled into the
triangle (charge coupling), which is pretty easy to compensate for, but it is a
tweak.

    BTW...the very first synth I built used that LM324 VCO.  Yep, it was
linear, but I felt it was a better VCO than the one that was originally used
for the Paia 2720 that apeared in radio electronics.  Sure had a lot of fun
with that stuff.

    -Jim

>
> H^)  harry
>
> sssh... we are hunting 'ground loops'
>

    Squeee....... opps, just stepped on one...

>
> patchell wrote:
>
> >     Well, if this isn't a new idea, well, what can I say (I seriously
> > doubt there are any new ideas).
> >
> >     Anyway.  So far, I have built two different triangle oscilators.
> > The first one used a "ota" current switch (nothing new here).  The
> > second one was similar to one of  Ian Fritz's oscilators that used a
> > pair of fet switches and a current mirror (I used JFETs).  Here is my
> > idea for another one:
> >
> >     The exponential converter would have two identical current sinks.
> > One would go to the input of an opamp integrator (in the usual way), the
> > second will go to a current mirror.  The current mirror would be set up
> > to source twice the current of the exponential converter.  Also, the
> > current mirror would be set up so that it could be turned on and off.
> >     So, when the current mirror is off, the current sink would suck
> > current out of the integrator, making the output voltage ramp up.  When
> > it gets to a threshold, the current mirror would be turned on.  Since it
> > is sourcing twice the current as the sink, the result is that the same
> > amount of current is now being sourced into the integrator, as was being
> > sunk before.  This will cause the voltage to ramp down at the same rate.
> >
> >     Don't know if I will be able to try this this weekend.  If anybody
> > else can follow what I said, I would be interested in the results.  Or,
> > is this just a bad dream?
> >
> > --
> >  -Jim
> > ------------------------------------------------
> > * Visit:http://www.silcom.com/~patchell/
> > *-----------------------------------------------
> > *If you didn't buy a home in Santa Barbara,
> > * You didn't pay enough!
> > ------------------------------------------------