Runoffgroove Tri-vibe - but with complex LFO-generator??

[Sesh]

Hi everyone!

I've been wanting a vibrato pedal.. I looked into the options, and really dig Fairfield Circuitry Shallow Water and the Earthquaker Aqueduct.

Both are, however, as far as I can tell based on delay and thus introduce "latency" to the playing when engaged and fully wet (the Shallow Water can also works as a chorus, though, and a cool one at that).

And of course, since I'm here, I was interested in seeing what I could DIY.

I found the Runoffgroove Tri-Vibe true vibrato that I assume a lot fo you DIY'ers here are familiar with. Schematic and info here: http://www.runoffgroove.com/tri-vibe.html

It's super simple and sounds great in demos, but it is a little pedestrian if you're not going for that subtle Magnatone 50s/60s kind of vibrato and, like me, want something more experimental.

So my question is this: Is it possible to exchange the sinus-esque LFO in the Tri-Vibe to, say, the Electric Druid LFO (that has tons of different waveforms to choose from) or does the circuit "need" the particular LFO in the stock circuit to function as a true vibrato? I

If it's possible, how would one implement the different LFO? I know it might be way beyond my expertise, but if it is indeed possible, I'm surprised noone has undertaken the implementation (like Drolo has for tremolo and phasers with the Twin Peaks & Liquid Mercury circuits)

[Mark Hammer]

The ROG Tri-vibe isn't really a "vibrato" pedal, but is a sort of Uni-Vibe, employing two range-staggered allpass sections.  Like the actual Uni-Vibe, and indeed any conventional phaser, one can get some pitch-wobble by lifting the dry signal.  Typically, the more stages, the more pronounced the pitch wobble.  Just how much wobble one wants or aims for is a matter of taste.  I've made a Tri-Vibe and it's a pleasant tone, however I wouldn't use it as a vibrato.  The Magnavibe and Magnatone amp it emulates doesn't offer up much more than the Tri-Vibe does.  That's not a terrible thing, but I suspect - without proof, though - that more exotic forms of modulation may not be quite all that differentiable when the modulation and effect itself is rather subtle.

I will say that I made a stereo Magnavibe earlier this year and was really impressed with it.  The stereo Magnatone amps employ a single LFO that countersweeps the single vibrato stage in each channel.  The unit I made used independent LFOs for each channel (basically 2 complete Magnavibe circuits) that were unsynced to each other.  The LFOs would occasionally sync up for a moment and then fall out of sync, or occasionally be complementary and fall out of sync again.  Run in stereo, the effect was strikingly rich, animated, and immersive, with the asynchrony providing none of the annoying going-back-and-forth of the Magnatone version of stereo.  That's the long way of saying that a dual Tri-Vibe - mono in, dual out - might be a truly magical thing, whether run as Uni-Vibe OR vibrato.  No need for fancy schmancy  modulation; just two different things going on at the same time.  It is technically possible to use dual-ganged pot for Rate and Depth, such that both channels can be made faster/slower or wider/shallower via a single knob.

[ElectricDruid]

So my question is this: Is it possible to exchange the sinus-esque LFO in the Tri-Vibe to, say, the Electric Druid LFO (that has tons of different waveforms to choose from) or does the circuit "need" the particular LFO in the stock circuit to function as a true vibrato?

It's possible. The vibrato is created by the changing phase of the signal, so the LFO used to control the changes isn't critical to the effect. The stages in the Tri-Vibe are OTAs (LM13600/13700) which needs a control current. The schematic uses 10K resistors on each control input to turn the modulation signal into a current, so it's pretty basic.

In terms of practicalities, the voltage on the OTA's control input has to be two diode drops above the negative power (so ground in a single supply ciruict like this one). That means a 0-5V LFO like the StompLFO would need offsetting upwards by a volt-and-a-bit. I guess that's what that extra op-amp stage after the LFO Depth pot does in the Trivibe schematic, but it probably won't work directly, since it's set up for an LFO that oscillates around the Vref level (so probably 2 to 3V up and down from 4.5V, giving for example a 2V to 7V range). But if you can find out what output voltage range the tri-vibe LFO is producing and can add gain/offset to the StompLFO to get the same levels, you can drop it straight in.

The final question is - does it sound any good? Mark voiced some doubts. It may be that you'd need to add more stages to get more effect, and that would add noise worries. Feedthrough of the LFO control signals is also another problem as soon as you move away from sine wave modulation. The sine wave conveniently has no harmonics, so even if there is some feedthrough of a 5Hz sine wave, you can't hear it. Once that's a 5Hz square wave (or ramp wave, or S&H random waveform or anything else with sharp edges) then ticking starts to become a potential problem. I added some mild smoothing to the waveforms on the StompLFO for exactly this reason, but when I tried it with a LM13700 filter design, I struggled to get the feedthrough low enough that you couldn't hear any ticking. That's the reason the FilterFX pedal uses a vactrol filter - it was quieter. However, that's my own failing as a designer, and not an immutable law.

Let us know if you give it a try. I'd be curious to hear it.

[Sesh]

Thanks for the replies! I was probably crossing my fingers that it was as easy as a drop-in replacement, haha but the more stages for more effect makes sense.

The idea of the stereo Magnavibe sounds way cool! Hadn't come across that one, but I see there's schematics and layouts and everything. Definitely gonna build that as well.

it sounds like I need to get deeper into electronics theory and breadboarding, heh. THe reason I asked is that I'm conceptualizing an analog "lo-fi tape wobble/wow&flutter emulator" pedal with random pitch wobble, a saturation preamp (perhaps with something similar to the envelope controlled "crackle" function of the SS/BS f*ck Overdrive as it itself is inspired by sound of disintegrating tapes in William Basinkis Disintegration Loops) and a lo-pass/hi-pass filter in one pedal. Real wow and flutter of course isn't too extreme either, so you could probably make do with only a few stages, but the randomness/instability is crucial and it would be fun to open up for extremes for experimentation.

From what I read, the original Magnatone amps used strange varisters to create the vibrato instead of vactrols.

Concepts are easy though, the hard part is the actual designing, and at this point it's way beyond my electronics expertise. I also highly doubt I can do a better job than you!

[DrAlx]

In terms of practicalities, the voltage on the OTA's control input has to be two diode drops above the negative power (so ground in a single supply ciruict like this one). That means a 0-5V LFO like the StompLFO would need offsetting upwards by a volt-and-a-bit. I guess that's what that extra op-amp stage after the LFO Depth pot does in the Trivibe schematic, but it probably won't work directly, since it's set up for an LFO that oscillates around the Vref level (so probably 2 to 3V up and down from 4.5V, giving for example a 2V to 7V range).

The purpose of the final opamp stage is to convert the (approximately) sinusoidal control voltage out of the depth pot into an approximately inverse-sinusoidal control current for the OTA all-pass filters.

For an all pass-stage, you can think of the small phase-shifts it produces as being an approximation to a very small time delay that is proportional to the cap value and inversely proportional to the OTA control current.
A sinusoidal time delay gives a pleasing vibrato.  So to achieve that, the OTAs need a control current that varies like   
    1 / ( A + B sin(t) )
where A is larger than B.

So the point of the final op-amp stage is to give an OTA control current that varies something like this

https://www.wolframalpha.com/input/?i=plot+1%2F%281%2B0.9sin%29

and the Tri-Vibe designers did a good job on approximating that.

[ElectricDruid]

Thanks DrAlx.

[blackieNYC]

I made a stereo Tri Vibe. Stereo or mono input. It's great, but I do wish I had used Druids LFO chip.

[jonny.reckless]

A couple of years back I designed a vibe pedal with a sinusoidal quadrature LFO and used bipolar transistors to do the frequency exponentiation, after being inspired by building a ROG trivibe.

It works pretty well for vibrato, univibe, phasing and points in between but it's not a particularly simple circuit to build. You can get fairly decent vibrato with allpass filters if you get the control right, it has to be an exponential sine for constant phase modulation.

I've often thought replacing the LFO with an MCU would be a neat idea. You could have different wave shapes, tap tempo, sync to a MIDI clock, all kinds of musically useful features.

https://www.diystompboxes.com/smfforum/index.php?topic=119211.msg1146780#msg1146780

https://postimg.cc/5HzsFfYy