Adding a sound gate to a feedback pedal.

[HeavyFog]

Went to revisit my first ever pedal build today. It was a feedback looper made from the beavis audio layout below.


For anyone who has never used one you put pedals in its fx loop and it feeds the sound in a feedback loop, which makes some crazy droning noises.

After playing with it for a bit (before it broke) i found that by putting a gated fuzz like my scrambler or a gated fuzz face in the loop it wouldn't make a drone and it would only make noise when i played, hence the gated part of the fuzz pedal blocking the sound until i played.

What i'd like to do is add a dedicated sound gate (not a noise gate) right after the return jack so i can get things really weird while still being able to make noise only when i play. What i'd like to do is have a sound gate in place so i can select if i want a drone or not. Haven't been able to find any suitable schematics yet for what i'm looking to do so my thought now is to take an lpb-1 or SHO boost and poorly bias it to get a gated affect i can adjust with a trim pot so i can dial it in just right. Would this work for this purpose?

Any other ideas?

[ashcat_lt]

The simple way to do it is to put a pair of diodes in series with the signal, but even with Shottkeys you'd be chopping out more than half a volt in the middle of your signal.  So put a gain stage ahead of it and complementary attenuation after.  If you want to be able to dial it in, give it variable gain and/or attenuation - preferably both tied together some way.  It's not the best way to do a noise gate, but it's a bit easier than most envelope-based options.  Works like the gated fuzz, but a bit more predictable and a lot more symmetrical.

[HeavyFog]

Sounds like a good idea! Il give it a try next time i get a chance. I've got a small pile of 1N400x diodes so il give those a try. Any diode recommendation? I might just stick an SHO or something equally as simple after it and add a master volume to make up for losses.

[GibsonGM]

The simple way to do it is to put a pair of diodes in series with the signal, but even with Shottkeys you'd be chopping out more than half a volt in the middle of your signal. 

I suggest you look up a Schottky diode from your favorite electronics supplier.   The lower the forward voltage (Vf), the better, which is what Ashcat was talking about. 

Typical small signal diodes Vf ranges from about .3 to .5v...you can find a Schottky that would have a Vf as low as about .15V.  This is better for keeping your signal levels up, and not raising the noise floor...

Edit:  Also, the diodes must be in parallel with each other (like clipping diodes) or you'll only pass one half of your signal...I am not sure what the clipping threshold would be, tho....where they'll go from operating in their linear region to non-linear (distortion)...something you'll have to play with by controlling signal level input and recovery gain.

[HeavyFog]

IL take a look through the shokkys that small beat offers anIL pick out 2 pairs each pair with a different forward voltage for different levels of gate maybe il skip adding a boost to it. In fact the pedal could maybe benefit from a volume drop since it's so insanely loud.

[GibsonGM]

Yeah, this is the fun of design   I personally would PLAN to have a buffer in front, and a boost behind, just a dual opamp (TL072).

BUT if it's only for your one pedal, to be tacked on inside, then your idea seems cool.  Who knows, you might get away with NOTHING more than the diodes!

Let us know how it works.

[ashcat_lt]

My idea was this:

Once you pick a diode, you've got a set Vf to work with.  Can't change it.  You can switch to other diodes, but those are fairly coarse steps, and I tend to think that even Shottkys are likely to be too big for most things.

You can, however, vary the gain of an amplifier preceding those diodes.  I would just use a non-inverting opamp to do that.  The bigger the gain, the smaller the diode looks. 

That's likely to make the signal too big on the other side, so I'm not sure you'd need an output boost, and in fact you might want to attenuate instead.  An inverting opamp stage could let you do both with one knob, but you'd want to invert again before the output because polarity makes a big difference in how the feedback is going to go.  Course, for exactly that reason it might be cool to make that "re-inversion" switchable.

Edit - In fact (!) the knob on the (first) inverting output amp could replace the feedback knob altogether.  This whole thing would basically go where that knob is, and then any time you had feedback engaged, it would have selectable polarity AND a known out-Z no matter what pedal is in the loop.

Course then we start to want to control the whole thing a bit more, and end up with an input buffer and some actual mixing amps and I think I figured out how to do it (minus the gate diodes which IS a pretty neat idea or the polarity switch which is also a good idea) in six opamp stages (and four knobs) once, but I'm not sure that scrap of paper is around anymore.

[HeavyFog]

I like the sound of both ideas and it neither would be too hard to implement since there's already a 9v jack for the led. Since im putting an order in from smal bear soon il throw in a few shottky diodes to give it a try. If the volume drop is tolerable il keep just the diodes but if not il have to try something else to control the madness. I think il add an expression pedal jack to it aswell so that the feedback can be controlled via expression pedal or maybe even an ehx 8 step. So many possibilities! IL report back with a layout in a bit when i get the parts to try it out.

[HeavyFog]

UPDATE

I tried a few diodes in parallel to see what would happen, BAT-41s did nothing at all with both pointing the same way and nothing either with one reversed. BAT46s on the other hand also failed to gate, but did succeed in making the oscillations even crazier and somehow even less predictable. Every time i flicked the toggle to add in the diodes the noise would change in pitch, volume, intensity, and frequency, so i decided to leave it in and call it the "crazy switch". This was for a friend and i think he's going to enjoy it. Added an exp jack as well that takes the place of the pot when plugged in so the feedback can be altered without reaching down for it. Might revisit the idea sometime but for now i'm happy with the way this came out.

[GibsonGM]

I didn't think it would be so easy    Something is floating around in my memory, a ham radio kind of circuit...that uses a Ge diode as a reference, and turns off some active device below .3V....gain > comparator-type thing, recovery stage...I can't come up with it though.  Peak limiter or 'silencer' (squelch) kind of thing.

It's probably going to be easiest to research the most simple gate you can find than to reinvent it, I think!