Splitting an LFO

[mth5044]

Is it possible to split (and buffer?) an LFO? It is just a voltage swing, not dissimilar to a guitar signal (although more voltage). I'm thinking about having one LFO generator and splitting it up into multiple paths. Then possibly invert them and see what happens. This way all LFO signals can have the same rate instead of setting multiple generators to the same rate. Thoughts? Thanks!

[Labaris]

Have you tried something yet?
You could try this (maybe using opamps instead of JFETs).
I guess it should work

[samhay]

Is it possible to split (and buffer?) an LFO? It is just a voltage swing, not dissimilar to a guitar signal (although more voltage). I'm thinking about having one LFO generator and splitting it up into multiple paths. Then possibly invert them and see what happens. This way all LFO signals can have the same rate instead of setting multiple generators to the same rate. Thoughts? Thanks!

Yes you can. Treat like any other AC signal, but remember that the frequency is significantly lower than a typical guitar or bass signal, so take care with your choice of coupling caps (or DC couple everything).
You may want to look at something like the 'Dimension P' chorus, which uses an inverted LFO etc...

[mth5044]

Great info, thanks guys.

Labaris - I haven't tried anything yet. I was thinking something along the lines of the simple splitter on AMZ, but wasn't sure if it would be that easy.

Samhay - I'll check out the Dimension P, thanks for the heads up.

[R.G.]

Remember that the tricky part of an LFO is not getting the waveform. It's arranging that waveform so it drives whatever is being modulated properly.

In general, this involves making sure the lowest frequencies get through, the peak and valley waveforms are where you want them for the individual modulator, and the DC level the modulator needs is provided. This last involves some thought if you intend to put a modulation depth control in.

Modulators are in general idiosyncratic. Each has its own special needs to work properly. Each one will have a minimum and maximum value of the LFO it can take and work properly. If you use a depth control, you need to figure out *per modulator* what these mins and maxes are, and whether increasing depth leaves the DC level of the LFO from the middle out, bottom up, or top down. If it's middle-out, you can AC couple the incoming LFO, otherwise not, although AC coupling may be difficult because of the low frequencies involved. It's sometimes simpler to DC couple even this case.

"LFO" is a good concept, but it is just a concept. Each modulator will generally need a specially fitted LFO signal, even if the frequencies are the same. The details will matter.

[midwayfair]

Well, here's a famous oscillator with a decoupled LFO:

http://www.home-wrecker.com/eatremolo.html

Going by breadboarding memory, I believe it was possible to split it multiple times.

Hardly the best oscillator in the world, though.

I had some issues trying to decouple a TL072 oscillator.

Don't forget that you can just use a single LED if LDRs are an option in the design. Then you only need one oscillator anyway.

[mth5044]

RG - I hear that. I tried applying a 5V LFO to a VCF with an LM13700. Something about mA, yada yada yada, things aren't as easy as connecting circuit snippet A to B.

midwayfair - thanks for the link. I was thinking about the LED surrounded by LDR's similar to the univibe. Then I suppose it would be as easy as inverting the single with an opamp and driving another LED surrounded by other LDR's. I'd have to think about the placement of LDR's in the circuit/circuits that this could be applied to. Someone varying resistance can still be harder the comprehend than programming a ucontroller 

[midwayfair]

RG - I hear that. I tried applying a 5V LFO to a VCF with an LM13700. Something about mA, yada yada yada, things aren't as easy as connecting circuit snippet A to B.

midwayfair - thanks for the link. I was thinking about the LED surrounded by LDR's similar to the univibe. Then I suppose it would be as easy as inverting the single with an opamp and driving another LED surrounded by other LDR's. I'd have to think about the placement of LDR's in the circuit/circuits that this could be applied to. Someone varying resistance can still be harder the comprehend than programming a ucontroller 

Check out Anderton's stereo trem. It uses a MOSFET chip and generates an inverted signal at one point.

There are other ways to do what you want, including changing what happens on the LDR side.