Stripboard CGS30 Bandpass Filter
Phwoar, etc. The SH-5 (and the System 100) look so good.
Because it’s a pretty simple circuit, I thought I’d try and do a stripboard version. It’s the first decent sized schematic I’ve tried to put together, it took about a week of scratching my head and checking.
It’s meant to be run off +/-15v, but it works happily off my +/-12v stripboard MFOS wallwart power supply. I didn’t have any 47k resistors for the input buffer op amp so I used 43k resistors instead. The resistor marked 800-ish is originally 4.7k in the SH-5 schematic, but Ken suggested taking that down to 1k-ish to increase the resonance, which is pretty subtle otherwise. I could only get a 47k linear pot for the resonance control, and I was sent a dual 100KA (log) pot instead of a linear, so the frequency jumps very quickly to the high end.
Look at the state of this.
I know, it’s ugly. But it worked first time, so (raises glass) here’s to ugly.
How does it sound? Here’s a square wave from the SH-5, sequenced by the CSQ-100 and sweeping the SH-5 BPF at full resonance.
And this is a pulse wave from an Analogue Systems RS-95, through the stripboard CGS30, sequenced from my dodgy Arduino sequencer.
The hiss in the background is probably a bit of spring reverb, whups.
I prefer the SH-5, it’s got a bit more of a resonant edge than my CGS30, but it does sound pretty similar. There’s not much point in chasing the exact sound given the other variables involved, but I’ll do it anyway. I’m wondering if running it on 15v might help, and I’m using TL072s, I’m wondering if using something like the original 1458s might get me a bit closer.
There’s some more useful commentary on the CGS30 in this synth-diy post – especially like the idea about tapping the output of OA3 (presumably IC2, pin 1?) for a low-pass output.
Anyway. Here’s some SH-5 picture filth.
Update: Sep 2018
Muffwiggler user kragg has had a go a making a more compact layout, their attempt is in this thread here.