Single buss keyboard question

[jordroid]

Howdy,

I have just built a single buss keyboard controller, using the MFOS single buss 1V/oct keyboard controller schematic.

It tracks well and the gate and trigger sections work fine, but i have an issue with the voltage at the 1v/oct outputs changing over time. With an LF444 for IC-3 the voltage will slowly, but noticeably rise over time, so if i leave a note sustaining it will go sharp. If i press the same key again it will return to the correct voltage, and then start rising again.

It is a slow rise and if i am playing fast and not sustaining notes for more than five seconds or so it is not noticeable.

I have rolled a few different ICs in the IC-3 position, and some make it better, some make it worse and some make the voltage *fall* over time. None seemed perfect.

I assume this is an issue with the sample and hold section, with the capacitor C19 continuing to charge slowly when it should just be holding steady. Does anyone have any experience with this or suggestions? The fact that trying different ICs in the IC3 position changes the speed at which it happens makes me suspect that end of things, but it could also be an issue with the FETs Q2 and Q3 i suppose.

I did build it on perfboard, but i try to keep things neat around sample and hold sections, and i added plenty of power rail decoupling *i think*. C19 is a polystyrene type.

Anyone chase down this kind of thing before? Or general suggestions for keeping sample and hold circuits stable?

thanks,

jordan

[jordroid]

In case it's useful to anyone, i got the most improvement by increasing the size of C19. I hoped that if the cap was larger, it would take longer to discharge or charge, and the voltage would change more slowly. I feel like i got it "acceptable" with C19 at 0.47uF. I guess it's possible that i damaged the polystyrene cap in there originally whilst soldering it, but since it's not bothering me anymore i think i'll call it good.

cheers,

jordan

[Mongo1]

Hi,

I just noticed your post - I'm planning on building one of these boards in the near future, so I'm quite interested in what you're seeing out there.

This is a really odd symptom. Generally when you deal with Sample and Hold circuits the problem you see is leakage, where the voltage drops over time. It sounds like you've got the opposite problem.

The only reason I can think of that pitch would rise like that would be if the FET isn't turning on all the way. If the FET has too much resistance in the main channel, you might see what you're talking about. Did you try to replace those?

Gary

[jordroid]

Yeah, it was weird, depending on the opamp it would either rise or fall. I'm guessing input impedance, bias currents and stuff like that that i don't fully understand are at play. I didn't try replacing the FETs, and i never really felt satisfied that i knew what was going on.

It was a gift for a friend and i needed to get it done in a hurry, so i settled for "pretty good". It was a funky little hand held mini keyboard thingy, halfway between a real instrument and a toy, so i didn't sweat it too much. By pretty good, i mean it really was quite good, i'm fairly picky about intonation stuff and i was satisfied.

Sorry i don't have more info about it, i think it's a pretty nice circuit, the gate/trigger section is quite reliable even with the super funky mini switches i used, and it scales to 1v/oct quite easily (by ear). I built mine on perfboard, and it's possible i didn't follow best practices and had reduced performance because of it, or maybe i made some mistake or fried some component...

thanks,

jordan