I accidentally, a filter. Who knew that could happen. Maybe I’m picking something up about that whole electronics stuff after all. One can dream. But here it is and I proudly call it VCF-2, quite original, I know. But I ran out of cool sounding names a while back so I’m going the descriptive route.
I was tinkering with an AS3360 IC. It’s a dual VCA, and to be honest, a pretty good one and very easy to use — stay tuned for an actual VCA using this IC, it cannot not happen. In fact it took me about 10 minutes to breadboard a working VCA, rough, but working just fine. At that point it occurred to me that I should check out what would happen if I’d put it in a filter configuration. And heck.. I got a filter.
Ok. I made it sound far easier than it actually was. I confess that I did spend quite some time afterwards to get the circuit actually working with voltage control and resonance and all that stuff, but a few evenings of tinkering later it did what it should do on my breadboard and it sounded quite ok to my ears.
The 2 sides of the AS3360 function as voltage controlled amplifiers, feeding into 2 integrators. This isn’t unlike many other filter designs out there of state variable filters. Basically you need a way to build a voltage controlled resistor, why wouldn’t the AS3360 be able to do so?
It’s a bit more complex as the VCF-1 circuit in the end, but not by much. The AS3360 expects a voltage control between 0 and 2Volts and, according to the data-sheet will be completely open at 1.7V. This means that there is a fair bit of scaling needed in order to prevent voltages outside the desired range. So, D1 & D2 provides a voltage rectifier, that way there can only be a positive voltages present at the output of U3B. The voltage divider between R17 & R18 scale everything down to the desired range. Additional protection could be build in using zener-diodes and what not, but I don’t think that’s really necessary. If the opamp goes the way of the dodo and starts outputting its rail voltages, the voltage divider will bring it still down enough to not blow up the 3360.
R1 is there to provide a small base voltage, so it’s not really 0 when the filter is completely closed. It need to be there to keep the cutoff control in a useful range. Otherwise a larger part than usable goes into the inaudible range. But, no fear, it still goes way down. You can put a lower value resistor there if you feel like the range is still too wide, or leave it out if you want a wider range.
The filter circuit itself isn’t very difficult either. Both sides of the 3360 are basically configured like a linear VCA. Pretty much like the data sheet stipulates it. They feed into 2 integrators with a 470pF capacitor where the filtering thing happens, so what you put in there has quite a distinct effect on the sound. I avoid ceramic types for this purpose, but feel free to experiment and see what suits best.