Author |
Message |
SineHacker
Joined: Mar 09, 2010 Posts: 99 Location: United Kingdom
|
Posted: Tue Mar 09, 2010 9:17 am Post subject:
No-input mixing desk (simulation) **mentalist alert!** Subject description: Creating bizarre oscillators with feedback loops? anyone? |
|
|
You may or may not be familiar with the no-input mixing desk technique, sorta pioneered by toshimaru nakamura:
http://www.furious.com/perfect/toshimarunakamura.html (if you're interested)
basically by taking outputs of a mixing desk and wiring them back into the channel inputs, you create feedback loops (though that may be bad terminology), an oscillation which I think is generated by a number of steps of amplification / preamplification. But I'm not sure. - I love this technique, the sonic possibilities you can reach with some simple tuning (by controlling gain and eq etc) are extensive and I have played the mixing desk for a number of improv performances.
I have been trying to build a standalone battery-powered handheld device that contains this method within a simplified interface for my own personal abuse (I'm a bit obsessed with having to build this device actually!) but I have been unable to actually achieve any kind of oscillation so far which is down to my limited understanding of electronics and the mixing desk, so I'm wondering if anyone can shed any light on this...
I guess I have / will have a number of questions which arise from this, so rather than me write an essay, perhaps I should just try to show how i think things should be working and perhaps people can point out the errors from there.
My understanding of the mixing desk (stripped down to one channel, line input and probably wrong!) is as follows:
input > line level amplification > input gain (variable) > fader/volume > output
wiring the output to the input in this setup should create a feedback oscillation as far as I understand (I have omitted eq/pan because I know they aren't required for oscillations to take place from using simple powered mixers, however there may be some kind of fixed internal eq to create some kind of sound character in the instances I have fiddled with). But building this on the breadboard doesn't do anything (I'm using pretty cheap amps like the lm386)
I won't write any more now though, I may have missed some bits out but it would be good to here some initial thoughts, whatever they are, and then start generating some dialogue. well done for reading this far!
-- aid |
|
Back to top
|
|
|
blue hell
Site Admin
Joined: Apr 03, 2004 Posts: 24075 Location: The Netherlands, Enschede
Audio files: 277
G2 patch files: 320
|
Posted: Tue Mar 09, 2010 9:24 am Post subject:
|
|
|
You'll have oscillation when the loop gain is high enough ... so probably you just need more amplification. _________________ Jan
also .. could someone please turn down the thermostat a bit.
|
|
Back to top
|
|
|
Sebo
Joined: Apr 27, 2007 Posts: 564 Location: Argentina
|
Posted: Tue Mar 09, 2010 9:42 am Post subject:
|
|
|
Hi:
Be sure you are using a non-inverting configuration, you need positive feedback
to achieve what you want. _________________ Sebo
---------------------------------------
My Music:
https://www.facebook.com/cosaquitos/ |
|
Back to top
|
|
|
SineHacker
Joined: Mar 09, 2010 Posts: 99 Location: United Kingdom
|
Posted: Tue Mar 09, 2010 10:15 am Post subject:
|
|
|
Blue Hell wrote: | You'll have oscillation when the loop gain is high enough ... so probably you just need more amplification. |
this is what I thought and this may be one of the problems, I'm guessing that if there isn't enough gain there will be no oscillation whatsoever (so it won't show on the oscilloscope) and I know from playing with desks that increasing gain drops the oscillation frequency - so I'm guessing that too much gain would cause an inaudible oscillation, though I would have been able to see it.
- Sebo that's very interesting, hadn't heard of positive/negative feedback before, I'm looking into it now hopefully this will help! |
|
Back to top
|
|
|
SineHacker
Joined: Mar 09, 2010 Posts: 99 Location: United Kingdom
|
Posted: Tue Mar 09, 2010 10:17 am Post subject:
|
|
|
i'm loving the second sentence of the wiki page on positive feedback:
"A system undergoing positive feedback is unstable, that is, it will tend to spiral out of control as the effect amplifies itself."
thats what i need |
|
Back to top
|
|
|
Boogdish
Joined: Sep 21, 2009 Posts: 122 Location: Bloomington, IN
|
Posted: Sat Mar 13, 2010 10:31 pm Post subject:
|
|
|
I love no input mixer stuff. I used to put this cheap chorus in the loop and the modulations from it would cause pitch to change randomly, so it was like my mixer was making music by itself. I felt like I really got to know mixer better doing that sort of stuff, and it helped me when I was actually using it to mix later on.
Could you draw us a schematic of what you've got breadboarded right now? That would help us to troubleshoot why it's not working for you. |
|
Back to top
|
|
|
SineHacker
Joined: Mar 09, 2010 Posts: 99 Location: United Kingdom
|
Posted: Sun Mar 14, 2010 7:06 am Post subject:
|
|
|
I'm going to try some new stuff, just going back to the drawing board, I met someone who used to work for this company: http://www.ddaconsoles.com/ some of their mixing desk series are discontinued and they've put up engineering info and detailed schematics - the S series looks particularly good, simple design and good performers.
so i'm thinking I can adapt a relevant schematic from one of these - I will post up results and schematics over the next couple of weeks.
boogdish - my previous attempts have really just been adapting really simple mixer schematics from the web, usually controlled by one transistor, I'm guessing this won't create enough gain to really achieve anything. I will look through my notes to see if I tried anything different (I've been away from the breadboard for a while so the memory dims) |
|
Back to top
|
|
|
electri-fire
Joined: Jul 26, 2006 Posts: 536 Location: Dordrecht NL
Audio files: 4
G2 patch files: 4
|
Posted: Mon Mar 15, 2010 1:39 pm Post subject:
|
|
|
SineHacker wrote: | Blue Hell wrote: | You'll have oscillation when the loop gain is high enough ... so probably you just need more amplification. |
this is what I thought and this may be one of the problems, I'm guessing that if there isn't enough gain there will be no oscillation whatsoever (so it won't show on the oscilloscope) and I know from playing with desks that increasing gain drops the oscillation frequency - so I'm guessing that too much gain would cause an inaudible oscillation, though I would have been able to see it.
|
Ermm... wouldn't just the tiniest bit of gain be enough? Because your gets amplified a little, gets fed back to the input, gets amplified more, etc. etc.
Oops, there is no signal is there? Hehe, maybe you do need more gain. Or cheaper opamps with a bit more noise.
But probably you DO want equalising. I've mixed stuff on a mixer with effects sends, but instead of using the effect returns I was giving the effects their own input. So I had input, but no input worked nice as well.
Anyway, the moral of the story: the eq settings were a nice and more responsive way to tune the oscillations to a desired frequency.
Using amp feedback without eq may oscillate at frequencies above hearing limit.
Easy to try out, adding a capacitor to ground from the output should make a difference. Use a not too small capacitor to make sure it works. |
|
Back to top
|
|
|
SineHacker
Joined: Mar 09, 2010 Posts: 99 Location: United Kingdom
|
Posted: Tue Mar 16, 2010 3:26 am Post subject:
|
|
|
could anyone recommend a nasty, noisy op-amp to work with?
yes I do agree, eq does make for some fantastic sounds, (I've been rocking out on the yamaha MG124C) - I think I'm going to add a bunch of nice momentary buttons to switch between capacitor values for a sort-of filtering effect. |
|
Back to top
|
|
|
electri-fire
Joined: Jul 26, 2006 Posts: 536 Location: Dordrecht NL
Audio files: 4
G2 patch files: 4
|
Posted: Tue Mar 16, 2010 9:43 am Post subject:
Opamp for No-input mixer |
|
|
I used an old Dynacord mixer that has the Ibanez tubescreamer opamps, JRC4558D.
http://en.wikipedia.org/wiki/Ibanez_Tube_Screamer
I've used a Behringer mixer in a similar way, and while it has six aux sends, expanding the amount of effects you can feed into each other, the Dynacord does have a more organic feel.
So, the JRC4558D, while not especially noisy, does deliver the Oomph. |
|
Back to top
|
|
|
jksuperstar
Joined: Aug 20, 2004 Posts: 2503 Location: Denver
Audio files: 1
G2 patch files: 18
|
Posted: Tue Mar 16, 2010 10:24 am Post subject:
|
|
|
This is an old dub technique, although effects are also used in the feedback path. Use a delay in the return path, and you get a tuned feedback that resonates, which is how Eddie Kramer came up with chorusing, phasing, and flanging while working with Hendrix and tapes in the studio. |
|
Back to top
|
|
|
electri-fire
Joined: Jul 26, 2006 Posts: 536 Location: Dordrecht NL
Audio files: 4
G2 patch files: 4
|
Posted: Tue Mar 16, 2010 10:49 am Post subject:
Subject description: Slew Rate |
|
|
So, what makes a nice response?
Come to think of it, it's the opamp's slew rate. You want feedback to be sluggish. Not the kind that kicks in instantly, "Pweep!" , giving you tinnitus the rest of the week if you're lucky, giving you a permanent dip in your frequency spectrum if not.
What you want is the feedback to slowly build up, "Hmmmpffwweee" , then during the "eeee" you decide you want another pitch, turn the knobs and go "eeeeeooooohh" .
http://en.wikipedia.org/wiki/Slew_rate
Your best bet when looking for a circuit might be to study distortion/ overdrive stomp boxes. You want those that are said to be dynamic, or "responsive" to input level.
So not full out fuzz, more like overdrive, for blues or jazz players. |
|
Back to top
|
|
|
slabman
Joined: Sep 01, 2005 Posts: 102 Location: UK
|
Posted: Fri Mar 19, 2010 8:24 am Post subject:
Escobedo |
|
|
So, you could conceivably build a no input mixer where each of the input channels was a different one of Tim Escobedo's fuzz circuit snippets. I'm, not sure the planet would survive the onslaught though. |
|
Back to top
|
|
|
electri-fire
Joined: Jul 26, 2006 Posts: 536 Location: Dordrecht NL
Audio files: 4
G2 patch files: 4
|
Posted: Sat Mar 20, 2010 11:07 am Post subject:
Re: Escobedo |
|
|
slabman wrote: | So, you could conceivably build a no input mixer where each of the input channels was a different one of Tim Escobedo's fuzz circuit snippets. |
Hey, slabman, what a great idea. Indeed you could. I love the Escobedo designs that much that I saved all schematics. His site if down a lot.
The one that comes to mind as being a perfect contender would be Escobedo's "Bronx Cheer".
Why?
Because it already has a built-in filter that responds to input level. And the diodes limit slewrate, I think the feedback path could be used with many transistors or opamps.
So you could mod it, the essential part of the circuit would be the diodes and capacitors in the feedback path.
Disclaimer: I havn't tried it. But I have built the Bronx Cheer, and like it a lot. I THINK it'd do nice as a no-input mixer channel. |
|
Back to top
|
|
|
slabman
Joined: Sep 01, 2005 Posts: 102 Location: UK
|
Posted: Mon Mar 22, 2010 2:42 pm Post subject:
|
|
|
Good thought, electri-fire. The Bronx Cheer looks as though it would get seriously chaotic in a feedback loop - which is a good thing in this application. |
|
Back to top
|
|
|
elmood
Joined: Sep 05, 2009 Posts: 22 Location: Toronto
|
Posted: Sun Apr 18, 2010 11:19 am Post subject:
|
|
|
You don't need very much gain... just a bit more than 1!
One thing you might like (which I used to play around with quite a lot) is to include some delays in your system. Being able to selectively route some signals through delays can create some slowly building rhythms. They also act to change the frequency response quite dramatically, especially if you use short delays. (delays are the basis of digital IIR filters) |
|
Back to top
|
|
|
electri-fire
Joined: Jul 26, 2006 Posts: 536 Location: Dordrecht NL
Audio files: 4
G2 patch files: 4
|
Posted: Mon Apr 19, 2010 9:05 am Post subject:
|
|
|
elmood wrote: | You don't need very much gain... just a bit more than 1! |
Indeed! Also being a guitarplayer, and guitar feedback requiring more gain and distortion, I overlooked this technical detail. Come to think of it, too much gain would actually not be that user freindly. A gain slightly over unity gain would be sufficient.
You get feedback already with gains slightly below unity. And delays are cool, the more the merrier. Short delaytimes have filter properties leading to Karplus-Strong type physical modelling tones. Then you send these to longer delays. And reloop those through other delays making counter rhythms. And reloop those. Etc. Etc. Etc. Etc. Etc. Etc. Etc. Etc. Etc. Etc. |
|
Back to top
|
|
|
electri-fire
Joined: Jul 26, 2006 Posts: 536 Location: Dordrecht NL
Audio files: 4
G2 patch files: 4
|
Posted: Mon Apr 19, 2010 9:10 am Post subject:
|
|
|
Hrmpff...
My Etc.'s with well placed spaces between them illustrating multiple delay polyrhythms has the spaces deleted between them for some strange reason. |
|
Back to top
|
|
|
electri-fire
Joined: Jul 26, 2006 Posts: 536 Location: Dordrecht NL
Audio files: 4
G2 patch files: 4
|
Posted: Mon Apr 19, 2010 9:19 am Post subject:
|
|
|
Testing. (inserting loads of spaces after this) Testing. (inserting loads of spaces after this) Testing. (inserting loads of spaces after this) Testing. (inserting loads of spaces after this) Testing. (inserting loads of spaces after this) . |
|
Back to top
|
|
|
electri-fire
Joined: Jul 26, 2006 Posts: 536 Location: Dordrecht NL
Audio files: 4
G2 patch files: 4
|
Posted: Mon Apr 19, 2010 9:20 am Post subject:
|
|
|
Erm... Moderator???? |
|
Back to top
|
|
|
jfox
Radio operator
Joined: Mar 06, 2010 Posts: 208 Location: GMT
Audio files: 26
|
Posted: Mon Apr 19, 2010 9:27 am Post subject:
|
|
|
hi electri-fire,
I think the board [software] removes extra spaces automatically.
|
|
Back to top
|
|
|
blue hell
Site Admin
Joined: Apr 03, 2004 Posts: 24075 Location: The Netherlands, Enschede
Audio files: 277
G2 patch files: 320
|
Posted: Mon Apr 19, 2010 12:29 pm Post subject:
|
|
|
Yup, space will be removed ... if not by the board your browser will do it ... unless you use the [code] tags which will make blockquotes. _________________ Jan
also .. could someone please turn down the thermostat a bit.
|
|
Back to top
|
|
|
|