w3c
 
Welcome, Guest. Please login or register.
Did you miss your activation email?

 
   Home   Help Search Blog Login Register  
Pages: [1]
  Print  
Author Topic: Coaxing delay pedals into self oscillation  (Read 6397 times)
expanoncolin
Administrator
phpBB Member


View Profile WWW
« on: December 28, 2004, 11:08:42 PM »

There are other ways to get that thing to self oscillate.  Please note that this info is more generally for less-programmed delays (IE, analog delays, not the DL4).

First off, what gets the thing repeating and finally oscillating is an actual feedback loop connected from the output to the input of the the delay creating chip in the pedal. The signal goes through some amplifiers, filters, expansion, and compression... the repeat potentiometer determines how much through all of this process the echoed, repeated signal gets attentuated. If you're lucky, there might be another resistor elsewhere in series with the repeat potentiometer... look at the traces on the PCB. Then, but lowering the resistance, you'd get self oscillation. It's hard to explain without looking at the circuit itself.

You can also do a more "brute" method by just making a whole new feedback loop in the pedal-by creating a simple feedback loop from the input of the circuit to the output. Here, instead of connecting the input and output of the delay chip with a carefully tweaked compression, expansion, and attentuation circuit, you're just creating a damn feedback loop. It will work, but it will be harder to control... what you are going to want to do is solder a wire to the input jack's tip conductor, then to one end of a potentiometer, then the center lug of the potentiometer to the tip conductor of the output jack... Find the resistance where the pedal starts to self oscillate comfortably when the repeat pot is about 3-4oclock. Use a 100k pot to start. Measure that resistance, get a potentiometer of close value, and just solder the thing together... wham, you've got a new feedback loop in series with the other one that will coax it into oscillation. Note that if you disconnect the input with the output still plugged in, it will go NUTS because (basically, even though it is more complex) you have 1 signal, just the feedback, going back into the circuit, instead of 2, the guitar and the feedback.

I did the first mod to an SAD-1 a while back and some other delay... I've done the feedback loop mod to make an underclocked chorus self oscilllate along with a couple other times to other, nonecho effects.

Hope some of that was useful.

-Colin
Logged

The best way to learn is to experiment.  Try it first, then learn from what went wrong.

http://www.eaced.com
http://www.experimentalistsanonymous.com
Simonetta
phpBB Junior Member

View Profile
« Reply #1 on: August 26, 2005, 08:46:08 AM »

I got a cheap analog delay recently on eBay, a Rocktek ADR-02.  Inside is a MN3206 2048-stage BBD delay chip.  The input signal to the BBD is biased by a potentiometer.  
  Setting the bias voltage low causes the device to feedback oscillate at short delay levels.  Setting the bias at about halfway between V+ and Ground stabilizes the circuit completely.
Logged

Beauty is the most subtle form of intelligence.
megadan
phpBB Junior Member

View Profile
« Reply #2 on: April 01, 2006, 05:45:41 PM »

Very interesting... do you think this would work on a flanger pedal? (a DOD FX-82B to be precise)?

I'm building a feedback looper right now, but I never thought of adding it to existing pedals...
Logged

expanoncolin
Administrator
phpBB Member


View Profile WWW
« Reply #3 on: April 13, 2006, 06:25:33 PM »

Quote from: "megadan"
Very interesting... do you think this would work on a flanger pedal? (a DOD FX-82B to be precise)?

I'm building a feedback looper right now, but I never thought of adding it to existing pedals...

The second method pretty much definitely would, give it a try!

That's interesting, simonetta...  it could be that you are actually making the bias better, thus providing more amplification, causing oscillation.  Or it could be something else :confused:

-Colin
Logged

The best way to learn is to experiment.  Try it first, then learn from what went wrong.

http://www.eaced.com
http://www.experimentalistsanonymous.com
Pages: [1]
  Print  
 
Jump to:  

Powered by SMF 1.1.12 | SMF © 2006-2009, Simple Machines LLC