I built a standard feedback loop pedal, and decided to add a low pass filter to attempt to tame the ear-destroying high frequencies that sometimes happen, and I've run into a problem, maybe someone can help.



Initially I did the RC calculations and decided that 2KΩ and 470nF capacitor would do, but when I tried it, no feedback at all. I realised that the 500KΩ feedback level pot was being added in series so it was a 502KΩ resistor instead, which is making it filter all audible frequencies.

I figure that I've got to make the values for R&C in the filter such that it makes the 500KΩ pot negligible but I can't work out viable values that work and still give me a good LP filter for audio.

Is this possible, or does the 500KΩ pot really screw it all up? Is there some other way to add a filter that in my noobie-ness I don't know?

Hm... Here's something to try:

On the TONE pot: Move the top of the tone cap to the wiper, move pin 1 to the input side, leave pin 3 as is. That should help to control your feedback signal through the filter cap to ground.

Try that and let us know what happens.

The cap goes *after* the resistor (implied in Rob's answer).

Trying to control both gain and low-pass in one leg with resistors may not have a good answer. Better if we _knew_ the source could drive low impedances; but in typical guitar-cord work the source may not like low-Z loads.

Sorry for the delay in getting back to this, but I've not had the time to sit and play around with the circuit for a while.

I tried the suggestion above, but it didn't seem to work, so in the end I just added a switch so the signal goes through either the feedback pot, or the filter circuit, which obviously clears up the problems I was having.

Thanks for your responses though, now I just need to implement the blend control and I'll be happy with it.

The best way to take treble in a feedback loop is in the feedback loop - like on one of the pedals in the loop.  I really don't think you're going to find a good way to do this passively because there are too many moving parts.

One of the big issues here is that as you head toward more feedback, your source Z (the output of a pedal) starts to get really close to your load Z (also the output of a pedal) and then the value of the LPF resistance becomes critical to the action of the whole system.  Very small changes in that value will make rather large changes in the overall feedback level AND the treble response.

I'd like to point out that if you're actually getting much feedback, that original feedback pot must be set to a value much less than 500K.  In fact, I think mine is only 100K, and I still have to turn up more than half before I get much action.

Frankly, this passive feedback loop is far less than ideal for all kinds of reasons.  The right way to do it requires like six buffers, but would make your task here a lot easier.

But anyway,  you might have better results with a shelving filter like in a guitar tone control, where your filter pot doesn't change the series resistance through the loop.  It'll still be tough to choose good part values, but will make the whole thing a touch more predictable.

Another quick tip:  If you cut the track of the post at the CCW lug, it acts to switch the feedback off when it's turned all the way down and leaves with a basic TB looper.  Some folks install a seperate switch for that, but either way it means you can get away with a smaller pot for better control of the actually interesting stuff and still get rid of it when you don't want it.

QuoteSome folks install a seperate switch for that

Yeah, I have a both a momentary and a latching footswitch wired in parallel to control whether the feedback is on or off (probably overkill, but I'm just experimenting here...)

QuoteFrankly, this passive feedback loop is far less than ideal for all kinds of reasons.  The right way to do it requires like six buffers, but would make your task here a lot easier.

Ideal is in the ear of the beholder and it's currently sounding ok to me. But yeah, from my complete beginners point of view, as I learn more about how it could be improved upon I'm realising there should probably be some buffers around the place.

I'm thinking there should be a high impedance buffer at both the input and return jacks because the pedals connected to those are expecting a high impedance. I think there also needs to be some sort of buffer before the send jack where the clean signal coming from the input jack is mixing with the feedback from the return jack.

Where else are you thinking, or am I completely off base again?

Yes, we definitely want to isolate the pedal iutputs from one another.  One of the things about this is that as you turn up the feedback you also end up mixing the clean sound back in with what's coming out of the loop.  Sometimes that's cool, but sometimes it's not what you want, and it can be tough to actually dial in a usable mix.  That starts to want more pots (my plan has 4, but I think we could live with 2) but it also needs more buffers/active mixers.  You need a buffer between the input buffer and the send mixer, two buffers on the return (one to the send mixer and one to the out mixer), and the send and out mixers themselves.  That's six.  I had it worked out where all the polarities ended up working out, but I don't remember for sure how that worked.  :/