[sdiy] Time fer a new string: Low Pass Gate Update

Grant Richter grichter at asapnet.net
Fri Jun 7 06:09:58 CEST 2002

> I received today from a modular
> manufacturer I will leave nameless unless he wishes to be credited (which I
> would surely like to do).

We all watched too many spy shows while growing up...

> I suspected was I wasn't giving the vactrol enough juice to sufficiently
> raise the resistance in the filter to allow passage of the higher harmonics
> of my gated signal.  I suspected this due to the fact that the voltage
> limits of my system for both AC and DC were marketably different from those
> in Donny B's box.

The Buchla envelopes run at +15 volts. So you are going to have to massage
the circuit to get the same sweep range out of a +10 or +5 volt envelope.
You can decrease the value of the resistor from the envelope input to the
summing node of the op-amp.
> If you look at the schemo of a 292C or D, you'll notice a 3.9 volt zener
> limiting the low voltage going into the vactrol's LED supply.  It was
> suggested to me that this is where the problem may lie and I went over to my
> local electronics shack to find zener's of a lower value.

The zener kicks in when the voltage drop across the resistor exceeds the
zener voltage. It supplies extra current to the Vactrol LED when the input
is above a certain level. The zener is out of circuit for lower envelope
voltages. The combination of the two form a two segment non-linear function.
Presumably to approximate an exponential response.

By lowering the 4.7K value, you increased the high end of the first linear
segment. Lowering the zener voltage would bring the second segment in
sooner. But after thinking more about it, the better method is to lower the
resistor value to the envelope input, so the two segments are driven
correctly by the lower peak envelope voltage.

> Not only did I gain high end, I stiffened up the attack response of the gate
> (which some of you mentioned before) AND I raised the signal level coming
> out of the gate as a whole, which before was easily 10 DB lower than my
> normal system level.

With the gate control fully on, the circuit should show a gain of 1 (0dB).
When the feedback from the output stage is connected back to the center tap
of the Vactrol in LP mode, this adds 3 dB gain. So the resistor at the input
is switched in, providing a -3 dB attenuator and bringing the overall gain
back to 0 dB. Otherwise there is a noticeable increase in volume when you
switch to LP mode.

More information about the Synth-diy mailing list