Much like my last (and similar) post, I have a circuit that I’d like some feedback and/or any suggestions the community may have. It’s based on the popular “Passive Warp” circuit design: I’ve seen it here and here and I think there are other designers who have made similar designs using the input-through-rgb leds-and-capacitor model.
I’ve built a few with different numbers of LEDs and it works well; adds interest to patches and looks nice as it glows on the panel. I got to thinking, though, that adding an op amp buffer into the mix could be interesting. Here’s what I put together in KiCad:
I built it on a breadboard and I came across something interesting that I can’t explain: when nothing is plugged into the input the circuit self-oscillates. My best guess is that it’s coming from the unity buffer on the input, but that’s just a guess. Any thoughts on why that could be? It’s actually a welcome surprise, though, tbh.
I’ve got the power filtering on the op amp and on the power input in the circuit snippet above, but I left of the actual power header with reverse power protection via diodes out since I could get a cleaner snippet this way. That being said, any and all ideas and suggestions are welcome. I’ll eventually put this up on GitHub so I’ll make sure to follow up here with those links.
I can see how if they’re much different forward voltages you’d get much different currents in them, and you have no way to individually adjust the brightness of each; but if the LEDs are the same what problem does it cause?
EDIT: These apparently are RGB LEDs with internal ICs to flash them in which case my presumptions are probably wrong, you can’t think of these like regular LEDs.
As for the input, another thing you can do is add a 100k resistor from pin 3 to ground. That establishes a fixed 100k input resistance, which is good practice.
The capacitor: Is it intended for AC coupling? I can’t see any other reason for it. But it won’t do that in this circuit because it needs a resistor to ground after it. In the passive version the downstream module’s input resistance (100k, remember?) is counted upon to supply that. But here it’s connected only to an op amp with no path to ground, hence no AC coupling.
Thinking about this some more, I don’t understand how it does anything at all. The voltage at the first op amp output isn’t affected by the LEDs, not unless they’re trying to draw more current than the op amp can supply. Maybe they are, since there are four of them. In which case the output voltage would sag when the current draw goes too high, and you’re depending on that sag to get an interesting output. But aside from that I’d think the output would have to be the same as the input.
This was something I wasn’t aware of until now, so good to know. Did a little reading, but @analogoutput asks an good question (as usual) re the two pin RGB LED. I’ll see if I can find out more about what’s in the the bulbs themselves.
I’d assumed it was AC coupling, but clearly I need to do more reading on this.
Thanks for the input @eric and Rich. Much appreciated
I was wondering if that was the case but didn’t feel like slogging through the videos to find out. Makes a good deal more sense that way. And if there were one resistor per LED… well, it might still do something if they were in addition to rather than instead of the common resistor, with the output connection between the common resistor and the individual ones. But presumably not the same thing.
Ah. Good catch! I think I threw that circuit together to quickly I’m gonna go back to the drawing board with these suggestions and see what I come up with.
