VCLFO ic or easy circuit?

[donalddokken]

I have been playing with this circuit (the Moog Source LFO at https://urekarm.tripod.com/synth/srcvclfo.pdf) since last week and I'm wondering if there is a way to produce pulse/saw waveforms with it. I removed the ground on Pin 4 of the 13700 and replaced it with a 100k potentiometer set at +/-12v on either end. This very nearly produces the results I am looking for, but it seems to create some kind of strange DC offsets that I can't explain. Has anyone experimented with modifying this circuit a bit? Maybe there is a suggestion for how to do this better?

[guest]

try replacing the 12k at pin5 of the 072 with 1k instead. then its a matter of restricting the range of your trimmer at pin4 of the 13700 to +/-100mV. there are a few other ways to do it, but if that works for you, it would be the easiest.

[EDIT] i just realized that will reduce the amplitude. disconnect the 12k from ground, and insert a 1k between the 12k and ground. then route that juncture to pin4 of the 13700. this way pin5 of the 072 still gets the +/-1V signal, but the 13700 only gets +/-100mV.

[donalddokken]

Thank you for the suggestion - I'm still grasping at these ideas, so let me see if I have that right...
I've added the 1k resistor, I believe where you suggested. This does produce a sawtooth waveform at the Triangle output. I am not sure, however, where to add the +/- 12v pot - and whether I should constraint it to +/- 1v?

[guest]

looks good. you want a signal to go +/-100mV at pin4 of the LM13700. maybe the best way to do this, is to put a 100k pot between +/-12V, so the wiper can go between the extremes of voltage. and then use a 100k and 1k to make a voltage divider connected to pin4, to know the +/-12V down to +/-120mV.

the one donwside to this way of making a variable pulse width circuit, is that the frequency will change a bit as you adjust the pulse width.

[donalddokken]

Hmm..I wouldn't mind the frequency changing a little bit, but for some reason, this is pinning the voltage at either rail. Maybe I need to turn the gain up on the op amp?

[guest]

you mean its not oscillating anymore? or that the output has increased in amplitude to go to the rails?

[donalddokken]

It is no longer oscillating.. the output simply pins to either rail, based on the trimmer at Pin 4 on the 13700.

[guest]

what are the voltages on pins 3 and 4 of the 13700? maybe the trimmer isnt getting the pin4 voltage in the right range, and its always higher/lower than the pin3 voltage. you can also try increasing the value of that 1k at pin3 and see if it just needs more signal.

[donalddokken]

Well, the trimmer is the only thing being input to Pin 4 - and I was feeding it with +/- 12v. The voltage on Pin 3 changes, depending on different factors - it seems to be at 0v with the 1k resistor and produces an oscillation without a resistor to ground.

Changing the 1k resistor to a 100k resistor allows me to produce oscillation and wave-shaping with a +/- 1v input on the trim. This works also in the original circuit, having gone back and tested again. The issue at this level is that the frequency seems unstable to some extent: it "jitters" a bit, but maybe that's just a feature.

If you have additional ideas about it, I'd be curious to understand more - but this is probably close enough for my purposes, so I just want to thank you again for helping me get to this point.

[guest]

with this circuit, the frequency will vary with the pulse shaping, but it should generally be stable if you dont move the pot.

if you had the wiper of the pot going directly to pin4, that would explain why it was saturating. the usable range with the 1k in place is +/-100mV, and if the pot is sweeping over +/-12V, then you wont be able to position it finely enough to get it into that range. for this reason, you need to put a 100k between the wiper and pin4, and then put a 1k between pin4 and ground. if you do this, then you can probably go back to the 1k at pin3 and it should work fine.

[donalddokken]

I have the 100k resistor in serial with the wiper at pin 4 and I've been experimenting with different values to ground. I'm seeing that that jitter still, though. Do you have any idea what causes that? I'm wondering if it's something that might go away once everything is in an enclosure.

[guest]

what if you put a small capacitor on pin4, like 10nF or something, does that help? also, what does the jitter look like? it is in a wobble of the signal? a variation in amplitude of the signal? a difference in time between zero crossings?

[donalddokken]

The capacitor does seem to help a bit, but there is still some jitter at the more extreme poles of the shaping. It's not an urgent problem, but something that I'm just curious about.

The jitter is essentially the waveform moving slightly left or right on the scope, a very small distance, with each oscillation. There doesn't seem to be any change in amplitude - I would assume this does entail a difference of time in zero crossing, though.

[guest]

are you triggering off the square wave or the triangle wave? the square wave will be a better test, as it has sharp transitions. im curious as well as to what it is. without the pot, the signal is very strong at the 13700 input, and it just saturates, so it cant vary much, its just full on. with the pot, there is a small signal at the input, and a bit of noise at the input will cause variable current out. is it possible that the powersupply is fluctuating a little bit, and this is getting coupled into the input? this can happen either through the pot circuit to pin4 or comparator circuit to pin3. the former you can just filter out with larger capacitors, but the latter is a bit more difficult, and might require some zener diodes or something (or a more stable powersupply).

[donalddokken]

Locking to the square wave further improves the reading, so it's down to a pretty negligible amount of jitter at this point. I experimented with multiple power supplies also, which didn't seem to have much effect. In any event, thanks for the help! I learned a lot in this process.

[guest]

great to hear that!