Need help understanding diode pre-distortion circuit

[living sounds]

This is the voice summing/output circuit of a vintage polysynth.

What I do understand is the TL081 summing amp, the CA3080 OTA used for volume control, including it's biasing arrangement as well as the output buffer/driver and peripheral components.

What I do not really understand is what the internet calls the "pre-distortion" circuit based on a matched diode IC (CA3019). My guess is, that it's sort of a crude diode bridge limiter combined with a clipper (two addition diodes) to prevent overloading the OTA.

Can anyone elaborate why it would be done exactly this way? And would it be possible to substitute modern diodes (probably doing some matching)?

Thanks!

 

[JohnRoberts]

OTAs like the 3080 basically use an input LTP (long tail pair) differential operated open loop. These have usable linearity down below about 25mV AC and typical audio applications pad down the input audio level to stay relatively linear. That circuit looks like it it is driving the audio into a diode junction and then feeding that crude/y logged audio signal into the OTA's input LTP.

I don't really see the benefit of this, perhaps they liked how it sounds.

JR

 

[living sounds]

Yes, they did like how it sounds. Are closely matched, "ultra fast, low capacitance" diodes a necessity here?

 

[living sounds]

This seems to have been the inspiration for the circuit:

 

[JohnRoberts]

look up a data sheet for CA3280 OTA. It has linearizing diodes integrated right into the silicon.

JR