Rebote 3, shielded cables, LFO circuits

[Dwubyd]

Hi peeps,

I had a lot of trouble/challenges building a Rebote 3 pedal. Finally got it done though. Some notes:

1. Some of the pots worked the wrong way - counter clockwise was "higher". I just swapped the outer two wires on each offending pot to fix this.
2. Had problems with noise - particularly LFO noise when the "Rate" pot was fastest. I used shielded wires to and from the PCB which seemed to help, although it did not completely cure it.

Questions:

I also seem to have noise issues with other circuits with LFOs. Is it good practice to used shielded wires to and from the PCB for any pedals with an LFO?

Modulation pedals seem to be trickier than gain pedals. Is there merit to this point of view?

[Mark Hammer]

The standard 2-opamp LFO begins with a square wave, which it then turns into a triangle.  This tends to create a brief spike on the power line, as the square wave makes a sudden grab for current, unless precautions are taken.  One of the standard precautions is to use a low-power op-amp chip, like an LM358, TL062, or TL022, whose "sudden grab" for current will be negligible, compared to many other dual op-amps.  I'm sure there are others, but these 3 are quite frequently found in modulation pedals as the "engine" of the LFO.  I realize Francisco's schematic shows a TL072, but any of the three I mentioned here would likely provide an improvement in ticking noise.

[Dwubyd]

Well, that is easy to try, so I get a couple in my next parts order.

Thanks.

[anotherjim]

Just to note that shielded cable, in this case, works best if only the PCB end of the shield is connected to circuit 0v. Don't use the shield as a ground connection wire, just treat it as a shielding wrap around the signal carrying wire.
It ought to be good enough to simply twist plain wires going to each pot together. This gives basic noise cancellation since induced currents are constantly reversing phase for every twist.

[ElectricDruid]

Modulation pedals seem to be trickier than gain pedals. Is there merit to this point of view?

I'd say yes. Very high gain pedals have their own set of challenges, but LFO ticking is a perennial problem with the typical Schmitt-integrator LFO design (and that's the "standard LFO", if there is such a thing).

[amptramp]

I have always found it odd that the LFO design of choice for most delay pedals is the Schmitt trigger / integrator design which is prone to noise spikes whereas the usual tremolo is a phase-shift oscillator that doesn't seem to have the same problems.  My Univox and Garnet amps both use phase-shift oscillators and there seems to be no trouble with that.  They are also a bit simpler with one gain stage although the gain must exceed 29 for it to work.  In both amps, this is done with a single transistor.  Frequency ranges are similar.

[Mark Hammer]

Boss came up with a brilliant anti-tick strategy, detailed here: http://www.moosapotamus.net/files/stompboxology-mo-tremlo.pdf

[ElectricDruid]

I have always found it odd that the LFO design of choice for most delay pedals is the Schmitt trigger / integrator design which is prone to noise spikes whereas the usual tremolo is a phase-shift oscillator that doesn't seem to have the same problems.  My Univox and Garnet amps both use phase-shift oscillators and there seems to be no trouble with that.  They are also a bit simpler with one gain stage although the gain must exceed 29 for it to work.  In both amps, this is done with a single transistor.  Frequency ranges are similar.

One reason might be that you can't get triangle waves from a phase shift oscillator (or not easily, at least). For chorus and some other applications, a triangle wave is a better option than a sine. But for tremolo, that "throb" that you get from a sine can't be beaten!