I had asked about the active lowpass filter used to filter harmonics out of the fundamental extractor part (and also in the synthesized octave down to minimize artifacts), and how to change the cutoff frequency, STM was good enough to post this link:
That topology corresponds to a third order lowpass filter. Here is an online calculator that would allow you to choose cutoff frequency to your needs:
Just make sure you select a third order response like 3rd order butterworth and you're done.
Also, Zero the Hero corrected a mislabeled pin, thanks, and had some good questions:
Sorry for the newbee questions, but i'm trying to divide by blocks this schematic and I do not understand how the lower octave is generated.
The guitar signal enters in the effect and after the first non-inverting op-amp it get splitted in three parts:
- the direct level goes straight to the output buffer Q2 via pot R59
- IC1 (pin 12, 13, 14) forms a lowpass filter, then R10 / R32 / C20 another first order lowpass filter
then the filtered signal goes splitted and squared (o rectified?)
into both positive and negative half-wave by IC2. What this part of the circuit actually does?
You have it right, so far.
I suppose that IC2 (pin 5/6/7) and IC2 (pin 8/9/10 correct this!) are rectifiers that with C22 and C23 become envelope followers which drive alternatively the filtered signal applied on non-inverting pins of IC2 (pin 1/2/3) and IC2 (pin 12/13/14)
I don't understand this section very well either, but I think this is the clever part that makes the OC-2 track notes a little better than other analog octavers. The filtered half wave rectified signals drive opamps set up as comparators, notice one half wave goes to the non-inverting input of IC2 pin 12, and the other half wave signal goes to the inverting
input, IC2 pin 2. The half waves are compared to the signal from R10/R32. This part is mysterious to me, I would be very interested to hear a better analysis.
CORRECT ME PLEASE! all this mess drives the CD4013.
I don't quite get the how and why, but the two comparators put out short square wave pulses. One side pulses when the signal goes positive, and the other when it goes negative. These pulses alternately pulse the set, and reset pins on the first 4013. The output of that section of the 4013 is an even duty cycle square wave, at the same frequency as the instrument. I'm guessing it is an attempt at hysteresis, to prevent glitches when the fundamental extractor would otherwise trigger on little overtones rather than the fundamental. This square wave from the first section of the 4013 gets divided again to be an octave down by the next, second section of 4013, just like other octave dividers that use flip flops.
I do not understand how IC1 (pin 5/6/7) works and how the divided signal is mixed. The output of the CD4013 is applied to Q8 which is configured as a variable resistor (right?) more than an amplifier. Then I get lost...
Q8 is used here as an electronic spst switch.
Q8 varies the amount of straight signal applied to the non-inverting pin of IC1 (5/6/7), which normally is applied through resistors of the same value on both inverting and non-inverting pins. Since CD4013 returns a square wave whose frequency is half the original frequency, is it correct to say that IC1 (4/5/6) has gain >1 only for an amount of time equal to half the input frequency (actual the inverse of the half of the input frequency)? Thus, the divided signal actually modulates the straight signal rather than being mixed with it.
Q8 is used as a switch, to change the phase at the output (pin 7), at a frequency an octave down from the fundamental. The Ge diode adds a DC offset when it switches phase. This Stompboxology article
explains it better than I can, skip to page 14 for the concept at work here. This Geo article
does to, except for what the Ge diode does.
This can't be correct... and explaining everything in English is a torture...
Your English is excellent, thanks for getting me thinking about this, some time I will hook up the scope the fundamental extractor part we are interested in, to see how it works. Might make a good extractor to graft onto other synthesizing effects, like Escobedo's PWM, or simple square wave shaper, or Blue box