GGG Stereo Tremolo-Mods?!

[the recluse]

I have tried searching for some relevant info on this project but it's been eluding me.  If the info I'm looking for is already on here somewhere, I'd appreciate a point in the right direction.

Anyway, I bought the PCB for Dean Hazelwanter's stereo tremolo (an updated version of the Anderton tremolo) from GGG about six months ago.  Since I built it, I have periodically pulled it apart to tinker with it.  Overall, I liked the sound of it, but there was a huge level difference between the effect switched in, and out.   I made the changes suggested in another post and increased R9 + 10 to 6k8 and R3 +4 to 220k.  This has helped some, although there is still a somewhat noticeable level change.  The other shortcoming I have found is that if the input level is above a certain threshold, it overwhelms the effect.

In the documents posted on GGG, there is a schematic, and parts layouts, but there really isn't an explanation as to how the circuit works.  What I am curious about is if certain resistor values could be tweaked and or replaced with pots/trimmers to allow for more adjustment, or if this is more or less an as is type of circuit, and the tweakability is limited by the design.  To boil this down to some simple questions:

1. In this circuit, is there anything else to be done to bring the effect level up to unity gain?

2. Is there a way to make the input of the effect a little more robust, so that an overdrive/fuzz/distortion in front of it won't overwhelm the effect (or is the solution to this simply one of effect placement)?

3. This isn't addressed above in the post, but related anyway.  I built the pedal to run on +/- power from two 9V batteries.  Is is possible to wire this up to a Boss Style DC adapter jack so that it could be run from a power supply like the Gator G-Bus-8 or a Pedal Power 2?  Is there a diagram of this somewhere?  In the Electronic Projects for Musicians book, on of the projects details how to wire a stereo  1/4" jack to run bipolar power, but is that adaptable to the typical 2.1mm jack?

P.S. the schematic for the tremolo is here: http://www.generalguitargadgets.com/diagrams/catrmsc.gif

[Mark Hammer]

The unit should not be distorting...at all.  Certainly the optoisolator should not and will not be a source of distortion.  The op-amps are not subject to anything that an op-amp in any other sort of pedal circuit is subject to.  If yo do the calculations, the gain factor in stage 1 is x1, and the gain factor in stage 2 never exceeds x5.  Hardly pushing the op-amps past their limits.

That being said, because of the 4049-based LFO circuit and the LED, the + side of the power supply draws more current than the - side, and it is easy for the user to have a pair of batteries in there that have ended up draining asymmetrically.  Ideally, you want to run this one from an external power source for optimal performance.  I'd explain how to do it, but I have to make my kid's lunch before I rush off to catch a bus for work.

The thing about tremolo is that in comparing effect to bypass levels, you are comparing true constant level with perceived average level.  That perceived average level will depend on the tremolo rate and depth you have it set to.  If the speed is slow and the depth high, then the quiet parts stick out more, and there is a perceived need to turn the level up.  As it happens with that particular pedal's design, if the depth is set to minimum, it actually increases the gain of the second op-amp and you get a volume boost no matter what the speed is (some, myself included, might consider that a kind of "easter egg" hidden in the design - a free booster!).

Consequently, although the original build article has the output volume as a trimpot, you probably want to have it as a panel-mount pot.  In concert with this, adding a bit of gain to that output stage isn't a bad idea if you have the capability to adjust the output level.  The feedback resistor in the second op-amp is normally a 4k7 unit.  Go with 10k-15k for a bit of a boost that should cover equal perceived loudness at any speed/depth.

[the recluse]

Thanks!

That certainly helps me make some sense of this thing.  Just to clarify, in referring to putting an overdrive in front of the trem "overwhelming" the effect, the tremolo is not distorting or clipping, rather, when I play with a TS808 (modded clone, of course) in front of the trem, and the volume of the 808 is above a certain point, I don't hear the tremolo effect until the note decays, or is played more softly. 

I am trying to find a way to explain it more clearly, it's almost as if the tremolo effect only works on signals within a certain amplitude range, and if the signal exceeds that, I can't hear the effect until the input signal falls back into the range.

Does that make sense?

[Mark Hammer]

Your explanation makes sense, though why it should happen doesn't make sense to me at the moment. 

You will note that the circuit uses two consecutive inverting stages.  One of the properties of the inverting stage is that if the input resistance goes higher than the feedback resistance, you can achieve gains less than x1 (i.e., attenuation).  That can work to your advantage here.  If you increase R1, that will attenuate the signal coming out of pin 1, mimicking the net result of turning down you overdrive pedal.  If you have the extra gain in U1b, you can compensate for that turn-down-the-input adjustment atthe output of the tremolo.

Why is that a good thing?  Because it means you can run the 808 at full tilt to do whatever it is you like to the amp, and compensate within the tremolo.

Try replacing R1 with a 68k fixed resistor and a 100k trimpot.  With the trimpot set to zero ohms, U1A has a gain of x1.47.  With the trimpot set to full resistance (100k), U1A will have a gain of x.6 (i.e., 40% lower/attenuation).  Given that Tube Screamers tend not to have humungous output levels, that ought to be enough.

Let us know what happens.  It is an interesting phenomenon you describe.

[the recluse]

Mark,

I really appreciate you sharing your expertise.  Thanks again.

I'll play around with it some more and see what happens.  Maybe the first thing to do is give it a run with fresh batteries.