maestro fsh-1 latching, stairstepping

[vacuumdust]

Anyone else have this symptom with the maestro fsh-1 build.  It seems like using a 2n3904 in the "noise" position in this circuit provides a nice random sample and hold, all be it subtle.  But i've tried a couple other transistors that result in much better intensity/depth of the sample and hold but at the price of stairstepping, or latching on to a frequency, thus screwing up any sort of vibe you had going.  The pn2222a and mpsa13 are the 2 transistors i've used.  Beautiful sounds but then eventually the latching starts...when that happens, just a slight nudge of the speed control snaps it out but of course that ruins your tempo.  And I've messed with the trimmers until the cows come home.  If anyone has tackled this problem or even attempted it i'm all ears. THANKS

[liquids]

I'm not personally familiar with the latching and 'stairstepping' issues, but this topic mentioned something quite similar: http://www.diystompboxes.com/smfforum/index.php?topic=24142.msg155750#msg155750

Likewise, when I first got mine 'working' I had it creating a filter of some kind that was not moving...DMM the control voltage (CV) confirmed that what should have been a continually variable voltage was static...hence the 'stuck' filter. 

I know say you are messing with the trimmers, but more so, are you using your DMM to measure voltages on some of the fets/transistors/chip pins, or only turning them trimmers by ear?  Even If you've got it 'working,' the DMM reading voltages on a handful of places helps in many ways...and can help understand the sounds you are hearing.  I wrote about my findings at the end of this altogether useful thread about the FSH-1:

I posted some of my finding http://www.diystompboxes.com/smfforum/index.php?topic=69381.0