BBD Bias Voltage - Noise Filter?

[MarshallPlexi]

Hey guys!

I'm working on an SCH-1 clone using a MN3007 running at 15V and using a microprocessor to send the clock signal to the BBD through a CD4047 VCO. It's working great and does all that I need it to do and can do it very quietly. What I am noticing is that when I adjust the BBD bias, I have to get the PIN 3 voltage to 3.85V and the Chorus is DEAD quiet and sounds great but with substantially diminished clean headroom as evidenced by the oscilloscope.

When I bias the BBD to around half travel on the trimmer I get a PIN 3 voltage of 6.7V and as expected I get much more clean signal through the pedal again using the oscialloscope as my guide.

For reference I am using 1kHz/1V-PP signal for testing.

Why does all the extra noise go away when the BBD is biased RIGHT at cutoff? I mean it's super tweaky. So much so that the 100k trimmer on the SCH-1 schematic can just barely be touched to find the sweet spot.

There must be something I'm missing. Anything I can do to bias the BBD hotter and get more clean signal through it while noise cancelling?

Thanks!

[MarshallPlexi]

[anotherjim]

I kind of don't like C12 in that circuit. Its polarity bias & therefore leakage is dependant on how the preceding filter Q2 emitter bias voltage sits compared to the BBD bias voltage. If you now have the R2/R3 bias divider nearer 7v or 8v, C1 will most likely be reverse polarity. A 1uF film capacitor could be fitted if it's not too big for the board, but I can't help. wondering if 100nF couldn't work instead - or you could just reverse C12 polarity to cope with the difference.

[MarshallPlexi]

the 1uF cap I have is a ceramic film capacitor.

I wonder if running the MN3007 at +15V could be part of the problem.