- Edited
There is a model of the Crybaby wah from CCRMA but it has one major fault:
Because it's based on measurement data it's pretty much stuck as a model of the GCB-95 voicing with no directly intuitive way to tweak it to other voicings.
I sat down and worked out the transfer function and tweaked it in gnu octave and compared to LTSpice simulations to find out what parts of the circuit needed to be rigorously represented. As such it is somewhat simplified, but with some physically-informed factors to tweak some of the more subtle parameters to match circuit simulation. Here is a frequency response graph of the V-847 voicing:
Anybody who has spent time studying this circuit will identify the set of frequency response curves as signature shapes for this type of circuit. I feel like I got a reasonably faithful representation of this circuit.
The model I have created computes digital filter coefficients directly from resistor and capacitor values so you can look up any voicing (mod) on the internet for this type of circuit and enter resistors and capacitors and come up with something that sounds reasonably close.
I also added a softened clipping function on the output so exceeding the +/-1.0 limits doesn't produce such brittle hard clipping, then added some of the distortion back into the filter to emulate inductor saturation (not very rigorous, but it does bring in some of the "quack" when you whack the strings).
Anyway, here is the source, as well as program to analyze the frequency responses and impulse responses over a stepped range of pot position values (which was used to generate the graph posted above):
http://cackleberrypines.net/transmogrifox/src/bela/inductor_wah_C_src/
Videos to come, but for now if anybody wants a "real" classic wah-wah, there it is.