[sdiy] Continuously variable waveshaping (was Behringer Neutron)

[Guy McCusker]

On Sun, Apr 8, 2018 at 6:54 PM, David G Dixon <dixon at mail.ubc.ca> wrote:
> My wavefolder (as found in the Intellijel uFold) uses TL072, and doesn't
> generate "high-frequency hash".  Also, I've achieved what looks a lot like a
> resonant filter output with a sawtooth wave, without a filter.  Given the
> way that this folder works (with diode "turnaround" circuits) I'm not sure
> that slew rate is a significant factor.  The opamp outputs in the folder
> chain are confined to relatively small voltages -- the first one to about 3V
> or less, and the subsequent ones much less.  Other folders may have
> significantly different modes of operation where slew rate is important, but
> not the ones based on diode turnarounds.

There's a bit of talking at cross-purposes here I think. The wave
folder in the CGS29 and also the Serge wave multiplier (middle
section) also does not generate "high frequency hash" and works fine
with a fast op amp. Their design is also similar to your "turnaround"
idea, as it happens.

The "high frequency hash" business concerns a different part of the
CGS29, the Grinder section, which exploits op amp overshoot and
ringing. It makes sense that slew rate is a factor there. ("High
frequency hash" is a direct quote from Ken Stone's description of that
circuit, by the way.)