Hair trigger Uglyface

[nocentelli]

I've just breadboarded the uglyface exactly as per Escobedo's original (i.e. no LFO) with all specified parts, but the runaway oscillation is far too easy to trigger; That is to say, unless the threshold is fully down, the oscillation is constant. Also, even with the threshold set so the oscillation is not there, pushing the frequency pot past 12 o' clock brings it back in. The sensitivity pot has a similar effect. Additionally, once the S/O starts, I have to "reset" the pedal by dialling back the freq+sens pots WAY back to stop it. I know it's a beast of a noise-maker of a circuit, but I've built it before (with LFO on vero by Blanik) and I don't remember it working like this. My recollection is that the threshold pot only triggered oscillation at halfway, and there was plenty of scope for setting a nice gated sound with no oscillation across most of the frequency settings.

It's doubly puzzling since I am using the same 386, 7555 and optocoupler (a VTL5C3) that I used for my first one. Could the oscillations be worsened by the breadboard?  Is there another fix to make the threshold pot more effective?

[edit - I tried doubling the "upper limit" resistor on lug3 of the threshold pot to 47k, and it helped quite a bit. I then noticed that the vero layout i must have used before (link - http://www.frigidworld.com/rich/uglyface_lfo_blanik.jpg) has a 56k instead of Escobedo's 22k in that position. ]