I got some time today to take a look at my old VCO build. Couldn’t figure out what the hell was wrong with it, so i just built it again. Rather than just be boring, i also attempted to add the other features that were not included in the original stripboard design. Things like soft / hard sync and FM modulation.
I dont really have the capacity to check that this works right as is, i did my best given the diagram in the datasheet and some example schematics on Electric Druid. All i can say is the pulse is now working, when it was not before. This uses the cheapo alfa 3340. Although i saw in the datasheet it wanted a 51k pulldown resistor, it works just fine with the 10k so i just left that since its probably more available.
The new features are pretty basic. I probably need to add resistors to the syncs as well, but i have no idea what voltages are going to go in there, and no idea what to do there. As always any guidance here is appreciated.
Just thought i would share since i was just trying to fix it for myself, but documenting it was probably better than just letting it go to waste, even if there are problems.
The AS3340 example circuit uses a higher pulse out pulldown, and the onboard zener is a bit different so you can use a slightly lower Vee series resistor, but it works just fine with the CEM3340 values.
The sync inputs expect brief spikes, and the example schematics just use 1 nF capacitors on the way in, to turn pulses into spikes:
EDIT: The 1222 performance VCO follows the example schematics, and routes the sync input via a 1 nF capacitor (C4).
EDIT: I just noticed that the CEM3340 data sheet has a section on sync on page 6, with a couple of alternative sync circuits, and some discussion about voltages.
Just a heads up folks, this is still a work in progress. @d42kn355 pointed out a few errors already, and im noticing some noise on my pulse. Ill try some different values for filter caps on the pulse, and mark to put traces on the filter caps.
Good news is i think the FM works, I have taken one output, and i see the frequency changes when i trace the modulated input.
It would be nice if i could get the curve to fall off faster, but in my experimenting, it just messed with the signal too much. Listening to it, it sounds just fine. You could probably even skip this step.
Edit v2.1:
Changed where the PULSE OUT is routed from. Where it was, the RC filter was being applied to the signal, instead of just dumping the fuzz. Now, its just nice and square! There is a bit of noise, but not nearly as much.
This weekend i found some time to build up two more of these. Still in straight up gubbins mode, but functional testing was a success! Since i have to do some sizing and shits to fit the stuff through the panels im just postponing that until later in the week.
In v2.1, it looks like you have two 10k pull down resistors in parallel on the pulse output of the 3340, is that intentional?
The way you connected the additional RC makes it act more like an additional load on the opamp in parallel with the main output, than like a filter. As long as you are not overloading the output of the opamp, its output should not change much based on the loads it drives.
Hi all!
Is there a particular reason for using 2 TL072 instead of 1 TL074? I assume Sam’s original schematics uses 1 TL072 for the simple version, so you need to add a second one to upgrade. But do you see any reason not to use a TL074 in the first place, if you want to build the whole circuit?
Cheers!
Not in this circuit. Unless you plan to occasionally power it up backwards, perhaps, in which case you may fry only one of them if you’re lucky
(there are some cases where you may want separate ICs to minimize cross-talk, and I intentionally use two TL072s in my MS20 filter builds so I have the option of switching to rail-to-rail versions for the one in the filter path, but here it shouldn’t matter at all)
Thanks for your fast replies guys!
This is odd: the black wire in Y2 in this schematics is connected to nothing. I assume it should be connected to ground, right?
@Caustic was experimenting with pulse output filtering here so either that is the “could probably even skip this step” bit or it’s indeed supposed to go to ground.
(not convinced it’s needed, but haven’t looked too closely at the pulse outputs in my builds)
!