EBS Octabass (roughly OC-2 clone). Difference between 74HC74 and CD4013? Buzzing

[BluffChill]

I've been breadboarding the EBS Octabass which is roughly a clone of the Boss OC-2, without the superfluous 2nd octave. The schematic is almost exactly the same as the OC-2 but with a few improvements like a buffered VREF (using 1/4 of a TL064), a separate divider to give the other TL064 a roughly isolated supply (VTRIG) and some different filtering, as well as a slight change to the JFET implementation (330k on the gate and no resistance to GND from gate or drain). https://www.freestompboxes.org/download/file.php?id=22797&sid=9f394cfe7fb9627a8637df279711799a

It seems to be mostly working - I get a big fat signal out of U7, and then a big fat square wave out of U3 Q (the first flip flop) and a big octave down square wave out of the second flip flop.

However, something seems to go wrong after that, as the output gives a very very quiet signal which is dominated by a huge fuzzy un-octaved signal which you can hear even when the output is unplugged.

It's also worth noting I've switched the 74HC74 for a CD4013, with all the relevant pinout changes. Can anyone tell what I might be doing wrong to be getting this fuzzy signal (even when I disconnect the output from the circuit?)

[Rob Strand]

Some time back forum member bushidov had a fizzing issue on one of those Octavers.   During the course of debugging one thing which was picked up was that the sense of either the clock or the set/reset signals was important.   Boss's fundamental detector (the rectifier etc.) aims to track the signal well.   However, it became clear to me at the time that if the sense of the flip-flop inputs was not correct it undermined the whole fundamental detector.    The 4013 has some specific behaviours when both S and R are active.

Off hand I think bushidov's unit had a 74HC14 and the sense of inputs was different.

[BluffChill]

Thanks, yes I did notice that on the 4013 the Set/Reset have to be pulled low, whereas on the 74HC74 they have to be pulled high. But still, no dice...

[Rob Strand]

Thanks, yes I did notice that on the 4013 the Set/Reset have to be pulled low, whereas on the 74HC74 they have to be pulled high. But still, no dice...

It's not the pull-up/pull-down resistors themselves.   The opamps are not open collector they drive the logic input high and low.  Those resistors help the opamp swing but only in one direction.   

The problem is the *direction* the opamp outputs that feed logic gates move when you get a peak that's important.  You would need to flip the sense of the output of the opamp.     The conceptual solution would be to put logic inverters between the opamp outputs and the logic gates.     The next step is to achieve the same effect but without the inverters by flipping the sense of the opamp outputs that drive the logic gates.

FWIW, bushidov's problem was not resolved.  There was a lot of debugging to get to that point and I think he had enough!

EDIT:

So as a first pass perhaps try leaving the pull/pull downs as they were but swapping the +/- inputs of each opamp.
For example, with the Boss schematic,
https://www.hobby-hour.com/electronics/s/oc2-octave.php

Swap pins 2 and 3 of IC2, then swap pins 2 and 3 of IC3.   The outputs of the opamps should idle in the opposite state.

Having said that, don't forget the 4013 has some specific behaviour when both R and S are active.   The 74HC14 might not do the same thing and I don't know if it matters.   If it matters I'm not sure how to get around it, it's a long time since I looked at this.