[sdiy] VCO that goes SHARP in highest octaves??

Neil Harper metadata at gmx.com
Tue May 10 06:07:59 CEST 2022

so this one is pretty strange to me...

i breadboarded a few of Thomas Henry's VCO-1 (i'm sure that design is
familiar to most, if not it is here:
https://www.birthofasynth.com/Thomas_Henry/Pages/VCO-1.html) with 3080
replaced with a TI 13700 that worked worked exactly as intended, i was
able to adjust the scale tracking and both needed a little bit of
high-frequency compensation as expected - with C8 initially being about
10-20cents flat. no problem. built a few thru-hole pcbs, worked exactly
the same.

my most recent project was to design a SMT version using the exact same
schematic - but with SMT equivalents. I received 5 boards back and all
of them go SHARP at C8 (10-15 cents) without any hf-compensation.

what's going on?

my first suspicion was the hf-compensation trimpot never 'went to zero',
so I removed the trimpot completely and the diode in the Rossom
Compensation circuit - no change.

as another test, I removed C1 ("C1 is a speed-up capacitor; its duty is
to drain away any accumulated charge from the base of Q2. This permits
the transistor to change states more rapidly"). the oscillator required
re calibration, but C8 was now only 2-4 cents SHARP. i feel like this
isn't the solution though, and that i was just crippling the
high-frequency linearity of the oscillator to compensation for another
non-linearity somewhere. but i can't think of anything that could be
different going from breadboard to TH PCB and then to SMT PCB.

/// Neil Harper
/// Every Wave is New Until it Breaks

More information about the Synth-diy mailing list