Stereo EA Tremolo with same LFO

[Guite Lectrique]

Hi,

Is it possible to make the EA Tremolo stereo? I want to have stereo inputs and stereo outputs with the same rate on both (not ping pong, but it could be a nice option). My idea is to have two separate EA trem circuits but controlled by the same LFO, is that possible? I made an attempt at drawing a schematic, could somebody tell me if it's right and, if not, point me where I got it wrong? I thought it would be nice to have separate depth controls.

Thanks!

Gui

[Mark Hammer]

Seems reasonable to me.  Having reciprocal control over the two gain stages (i.e., ping-pong), might be a bit tricky, though, given the form of gain control employed.

But I'll let others more knowledgeable about JFETs, and discrete circuits in general, decide that.

[boogietone]

The more experienced will probably better address this, but I would imagine that, as drawn, the two depth pots would interact - being parallel resistive paths for AC to ground.

Also, off topic, but in the series 10uF, .22uF caps after the first transistor, the 10uF is not needed, IMHO.

[Mark Hammer]

Ordinarily I'd agree with your point, but I think the 120k resistors and 470nf caps take care of that.  Pot interaction should be minimal.

[R.G.]

Ordinarily I'd agree with your point, but I think the 120k resistors and 470nf caps take care of that.  Pot interaction should be minimal.

I think you're right Mark. Little interaction. There may be some loading issues, as you have twice the load on the LFO. Might make it hard to start, but probably not too bad.

[Guite Lectrique]

The schematic is based on the GGG Improved EA Tremolo which I have already built and works pretty well, and my knowledge in electronics is pretty basic. You might be right about the two depth pots interacting, I haven't thought of that. I guess I'll have to try it and if they do interact, I will try using a single Depth pot. I'll keep you informed but it might take a while, since I have to order a few parts.

Thanks for your help!

[boogietone]

That makes sense about the 120k now and the cap prevents interaction of the DC components.

Now a question, given the typical levels in guitar pedals what is a minimum resistor value that is needed to keep such interactions from occurring?

[YouAre]

What if you took the oscillator output/input to one of the depth pots (half the LFO) and put that into a unity gain inverting op amp. That way, you could have that ping pong complimenting trem. It could be a great faux panning pedal.

[PRR]

That will work.

You don't really need two 0.47u caps, one will do.

Loading is no big deal. The LFO has 15K load. With all the positive feedback, this might not be the actual output impedance... but it turns out close enough. If the load is much larger than 15K, it won't notice. If it noticed too much, it would quit. To not-quit, the equal-value CRCRCR oscillator needs gain of 29; a bit more for un-equal CRCRCR. This has gain near 150, it should oscillate robustly even if load were near 15K.

OTOH, the input to the FET is dang near infinite. 1Meg is a handy concept; much larger invites some problems.

We pick the Depth network somewhere well above 15K but less than 1Meg. 122K, or 438K, whatever.

I'm not fond of "stereo" with two knobs. I'm lucky to find one spare hand, and two hands on two knobs never gets the channels balanced. And "slightly unbalanced", IMHO, is liable to be more annoying than musically useful. Since we really have one source, and the load may be very high, I'd lean to one common Depth pot. We can diddle the balance later.

Conceptually we could lower the mono Depth pot value to half. Or 100K, next standard value. It happens the shim resistors come out very standard-value.

For "true" stereo you need matched FETs or some trim. Two JFETs from the same tape may be close-enough the ear won't know the difference. If not, a couple 1Meg trim pots can correct a small mismatch pretty-near. One should be full-up, the other just down enough for balance.

_OR_ you can front-panel these pots to raise your knob-count and have more things to mess with.

I want some gate resistance, so if one gate conducts (I think both do for part of a cycle, but not exactly the same) it does not mess-up the other gate. This may be moot, but also now simple switches will kill either FET, so you can have Left out full and Right out wobbly.



> have that ping pong complimenting trem

Yes, though the inverter increases complexity.

_IF_ you can find a P-type JFET similar to your N-types (may be tuff), there's a really elegant trick. This WILL need a trim-pot cuz matching N and P are unlikely.