[sdiy] Compensating output level for Q

Andrew Simper andy at cytomic.com
Mon Jun 3 11:08:38 CEST 2013

Hi Nick,

On 1 June 2013 20:23, Nicholas Keller <niroke at tampabay.rr.com> wrote:
> Andy,
>   Thank you for the very detailed response.

You're welcome!

> That link you included is a wonderful resource!   I might take a few hints from that, although the CEM3320 has on-chip resonance, so it's not quite the same circuit.
I meant to ask about that actually, the circuit you have shown, at a
glance, doesn't look like it will self oscillate, it only has a gain
of 1 where you need 4.

> Also I'm not trying to toss out the original character of the Cat SRM.  I'd like it to still sound like a Cat, but have the option of other filter modes and a more usable resonance, as right now it is very difficult to adjust the Q while playing live or recording because of the jump in output level.  I'm afraid that the Hi-Q/Low-Fc bass it produces will be sacrificed once the Q is tamed.  So an ideal solution might allow me to turn off/bypass whatever control mod I install when I want the extreme low-end.
> I'm going to try the diode clipper between the Q pot and R166 first.

Yep that looks like a great place to try putting the back to back
diodes, hopefully that will bring the resonance under control enough
without damping it too much to change the tone.

> My first idea for creating BP was to cascade HP into LP, but when I learned that BP could be tapped right from the circuit, I decided to try that instead.  However, with the difference in level between the BP and the other tap points, perhaps it is easier to just make a cascade BP and avoid having to install op-amps to balance all the signals going to the VCA.  Also, I really didn't notice that much difference between the 12dB and 24dB modes, so perhaps trying to provide all modes is not necessary.  I will consider limiting the mode options to fewer modes.  I realize I will still need to invert something for the HP4 Q to work. I was hoping to avoid a 3pole rotary switch, but I might need to use one after all (input to second stage, selected output to VCA, inverter bypass)

One way to get around having to boost the band pass mode with extra
gain is to leave it alone, but instead cut the level of the low and
high pass outputs post switch with higher resistor values and then
lower R194 so you boost the levels back up again to where they were.


More information about the Synth-diy mailing list