[sdiy] LM741 substitute

[David G. Dixon]

> >> It depends on the circuit. Some saw->tri waveshaper circuits rely on
> >> the crappy slew rate of the 741 to mask the switching glitch.
> >
> > A 20 pF (or so) cap across the feedback loop would be the preferred
> method,
> > I'd think.
> 
> I tried that in spice simulation, but it didn't work very well. Using slow
> slewrate on the other hand worked almost perfectly.

I guess that's why I've only ever built triangle-core VCOs (which typically
give glitch-free triangles), and done the tri-saw conversion.  I've found
that the little glitch in the middle of the saw is very handily dealt with
by putting a 100pF cap across the back-facing diode on the
polarity-switching JFET.

I've also found that, when it comes to itty-bitty caps and their effects,
Multisim (a dressed-up version of SPICE) gives at best only an indication of
the what the effect will be.  I always have to experiment with cap values
and a scope to achieve the best results.