Welcome to Solid State Guitar Amp Forum | DIY Guitar Amplifiers. Please login or sign up.

March 28, 2024, 11:06:05 AM

Login with username, password and session length

Recent Posts

 

why won't this damn feedbacker work?

Started by thomasds, October 01, 2012, 08:17:23 AM

Previous topic - Next topic

thomasds

i've wanted to make a feedback looper without true bypass channel. soldered this together but must've made a mistake. it works a bit as dry/wet mixer but not as a feedback looper... anyone an idea why?

Roly

It may be brilliantly obvious to everyone else what the signal flow is here, but I'm finding it a bit obscure.


A potential difficulty is that the "Send" output and (what I presume is the Fx) "Out" will both be very low impedance driving points, and simply fading them together using a 500k pot (which will be very much higher than the two source impedances/resistances) means that these two outputs will dominate the circuit and try to dominate each other.

Assuming the signal flow goes;

Send -> In -> {Fx} -> Out -> Return

...and by "feedback looper" you mean you want to feed some of the Fx output back into its input...

You need a resistor, say around 1k ohm, between "Send" and everything else, and I think your 500k pot is way too high in value and you should try something more like 10k ohms.  That will then act like a basic mixer of the signal coming out of the Send and the signal coming out of the Fx Out, both back in to the Fx In.

I hope this guesswork based on assumptions helps
If you say theory and practice don't agree you haven't applied enough theory.

thomasds

Well, it's the other way around: in is guitar-in, out is to amp (or fx-chain) en the effect is between the send-return. i don't think that this makes a lot of difference in your reasoning...

and yes, that's what i mean by feedback looper...

but i'll give it a try using a smaller pot. found that value though in FB-looper with true bypass and in that one, there is no resistor at the "in"... is that one necessary?
http://diyshoegazer.tumblr.com/post/3042080186/diy-guide-to-feedback-loop-pedals

Roly

I'm still a little confused by your terminology because "Send" would normally be an output from an amplifier, but here you seem to mean sending from your new device.

In (guitar) -> Send {in>Fx>out} Return -> Out (to amp)

(I'm also assuming that there is a left-right ground path between the sockets through a metal box or similar; you show two top-bottom earth paths that are not linked.)

The reason I suggested the resistor is because a driven output actually looks like a short circuit to signals coming from anywhere else.  If you are using an active guitar (one with a battery in it) or coming via any sort of stomp box, then the "guitar" signal source will look like a short circuit to signals coming via your pot.

What you seem to need is some form of basic two channel mixer at the input of your Fx that will mix the fresh signal with the Fx output being recirculated, that way the recirculated and fresh inputs will have some isolation from each other.

For example, a passive resitive mixer;



An active mixer using a FET;



Google "simple audio mixer" for more.
If you say theory and practice don't agree you haven't applied enough theory.

J M Fahey

Dear Roly, this isn't what you think (or I would have thought either):
This is an accessory for a couple pedals which *may* benefit from it (most do not).
In delay pedals, there is a "Feedback" pot which allows some of the delayed signal back into the input to provide "repeats".
They are usually designed to have a "reasonable" range.
This "Looper" is a very crude, unbuffered pot which connects the pedal output to input and allows for crazy high feedback, getting all sorts of peaky responses, lots of hiss, oscillation, you name it.
If that rocks your boat, so be it.
For a free sample of what you can get (ugh!):
http://youtu.be/rZyYDlmr0_o
By its very crudeness, there's absolutely no guarantee that a different guitar, pedal, cable or Moon phase will allow you to repeat the same effect twice.

Dear thomasds: build it *exactly* as suggested in the Beavis Audio schematic. No guarantee of anything, as I mentioned above, but it still is the best chance.

Roly

Uh huh.  This reminds me of the sort of thing I was doing with a reel-to-reel sound-on-sound back when I was ... ah ... (well never mind).

I must say I was wondering what had happened to the "feedback/recycle" control on his delay, thought perhaps it didn't have one.

I notice in the vid that (apart from it being fairly uncontrollable and making some bloody awful noises) the pot on the outboard box is very "touchy", so yeah it looks very "device dependent".

Still think some form of crude mixer, even passive, would give better and more dependable results than trusting to whatever device impedances turn up.

The Bevis loopers don't actually control the signal level around the loop, only coming out of the loop.

http://www.beavisaudio.com/projects/Looper-Switcher/index.htm
If you say theory and practice don't agree you haven't applied enough theory.

J M Fahey

Agree.
A device with proper *input* and *output* buffers, plus an input mixer, coupled to a feedback pot (which now should work properly) would be the "real stuff".
Somebody might design and post such a design in the interest of the hobbyist; not me because at present I'm with more work than I can handle.
The Beavis one is very simple and does work, sort of, but is hard to set up and probably won't work at all with many pedals.

Roly

@thomasds - are you still with us?  Are you up for something a wee bit more complicated, a bit of circuit development, that may end up delivering the "looper" goods for more people?   8|
If you say theory and practice don't agree you haven't applied enough theory.