GGG pt - 80 feedback resistor question

[postagerock]

i am building a general guitar gadgit model pt-80 for my friend.  i am noticing that when the repeats are set higher, it falls into feedback quite quickly.  all op-amp values are correct.  i am assuming this is fixable, however i am not sure which resistor to tinker with. can someone let me know, and also offer advice as to what value may be relevant to change to. thanks much.

[BAARON]

This is absolutely normal for this design, so that you Can create feedback.  There are a few things you can do, which can help limit the feedback, but I'm not sure I'd try to get rid of it entirely, because sometimes feedback can be fun.

1. Remove the 22k resistor that connects to pin 12 of the PT2399, the one that's only seperated from the chip by a 0.1µF cap.  Replace it with a 10k and a 12k resistor soldered in series with each other, then connect a 0.015µF cap from the middle of the 10k/12k pair to ground.  This will eliminate a lot of high-frequency oscillation, but will also make your repeats get darker and darker each time they repeat.  This will NOT prevent oscillation/feedback, but will help make it less ear-piercing.

2. See the purple wire on the parts layout diagram that goes from LR3 to the Delay Repeats pot?  If you replace that wire with a resistor, it will limit the level of the signal that the unit sends back to the start of the circuit to create successive echoes.  (Note that this will also lower the maximum volume level of the repeats by the same amount.)  I would suggest starting with a 10k resistor, and see how it sounds.  If you still need to limit it more, make the resistor bigger.  If you can't get enough echoes at the maximum feedback setting, make it smaller.  I would suggest making it small enough that you still CAN get feedback at the maximum setting, if you're so inclined... but only when it's turned up to 9 or 10 on the feedback knob.

[cpm]

pin 12 on the compander is a negative input for a standard opamp inverting mixer, so playing with that 22k value can vary feedback level. You can

also reducing that 1uF feedback cap (around ~100n), may help taming the low end when going into oscillation feedback. This, with the low pass from the previous filters, makes a band-pass so the repeast can float on its natural frequencies without the lows saturating and distorting.