idiot alchemist at it again

[boog]

Sooooo.....my newest contraption is an LPB boost (w/ 2N5089 transistor) into a dual op screamer style clipping stage (TL082) and then w/ the tone stack from that Chaos petal.  After all that - 2 yellow LED's for clipping to ground, in parrallel w/ a .047 cap, both in series to a 1k resister to ground (saw this at the BYOC board on a pedal called the bootleg jazzfuzz, really neat pseudo tone/clip mod).  This things loud, compression heavy and a rough crunch distortion.  But the LED's don't light up.  Figuring I'd try and outsmart electricity (which I can't), I connected the 1k pot to the voltage supply.  Still not enough to make the LED's go, but the distortion smoothed out in a nice saturated gainy field.  Why'd that happen?  Not the lighting of the LED's, could care less about that right now, but the change in tone.... if I can figure out how to do pics I'll post the schem...

[Kearns892]

I dont believe the LEDs used for clipping are supposed to get enough voltage to light up, but they are still clipping the signal. Is this what you were asking?

[JKowalski]

Figuring I'd try and outsmart electricity (which I can't), I connected the 1k pot to the voltage supply.  Still not enough to make the LED's go, but the distortion smoothed out in a nice saturated gainy field.

I'm not exactly sure what you mean by this...


Schematic would be helpful, as well as showing on it what the change was.

And clipping LED's typically don't light up in most cases, but they still do their job.

However, depending on the strength of the signal, they might not even be clipping at all. Remove them from the circuit and see if the sound changes dramatically. You need something like a 3-4V P-P signal for most LED's to clip significantly. And if you put a 3-4V P-P signal straight into an amp, thats gonna be LOUD AS HELL. If you use LED's as clipping diodes, you'd probably want to have a gain stage to up the signal voltage, clip it, then a voltage divider to bring the signal back down to line levels.

[boog]

I've not had them light up like they would as a status indicator before, but i have had them "blink;"  I'm more interested on why when i connected my clipping diodes to my voltage supply i got a much smoother sound than when i connected them to ground (which is what i've done in the past)....um, how do i post an image?

[Derringer]

that would effect the threshold voltage of the diodes then ... right?

which would effect the clipping character of the diodes

[JKowalski]

Okay - well, in that case, the LED pointing towards the V+ will do absolutely nothing, as it is reverse biased. The one pointing DOWN from the V+ will be always conducting current to ground. I am assuming that you have something like a 100k resistor on the output of the effect as well, so that limits the current to 90uA, which is definitely not enough to light the LED, and I don't think the signal will be affected much, if at all by this.

Therefore, I think that the reason that you have a smoother sound is because the LED's now can't do anything. If you remove the LEDs from the circuit and listen to the sound, it should probably be the same or very similar  

Download your image to a site like Tinypic.com, get the URL for it, then stick it in [ img] [/img] tags (without the space in front of img)

[boog]

Ah-ha!  Thank you and thank you, JKowalski!  I'm still working on learning what happens in these circuits, so the more I can learn in real applications, the more easily (I hope) this stuff comes to me.  I'll have to try removing the LED's when I get home.  As it stands I just hobble-gobble parts of different circuits I've seen together and hope for the best.  And thank you for the image tidbit as I can show you what I mean.....sorry for the lame descriptions on the schematic, still learning how to use that software as well

[JKowalski]

My pleasure 

And don't be ashamed! Hobble-gobbling circuits together is a great way to learn, and often times a perfectly valid method of designing (especially for guitar effects, just consider the whole basic concept of the traditional hobble-gobble pedalboard) And don't be discouraged if your method fails you - in fact, you should hope for such setbacks. As in this case, they give you a great opportunity to learn very useful bits of information in a non-abstract way.

You should be proud of your motivation to be original, and go beyond cloning. Many people don't actually go beyond that point.


A further note, the LED is current limited by your 1k potentiometer when the wiper is closest to the capacitor (Right now i am talking as though you moved the LED's back to ground). When the wiper is closest to the LED's, it is not. So this means that with the "Smooth" control furthest to the leds you may get some light. Then again, it may still not be enough current simply because the signal is not strong enough for the large voltage drop of the LED's.

[BAARON]

Also notice that the 100k volume pot is acting as a current limiting resistor, as it is placed directly between the diodes and ground.  Also, you have DC voltage sitting on the volume pot, on the output, and on the negative side of the 10µF output cap.  That's probably a bad idea.

Regarding the change in sound:
When the LEDs were connected to ground, they were clipping both halves of the signal as it swung far enough above and below 0v to pass the forward voltage of the diodes (in other words, whenever you played and it put out a loud signal).

Right now, there's no way for those LEDs to clip properly.  One is reverse biased and will never see a large enough guitar signal to turn on (as compared to the 9v it's pointing at), and the other is a constant one-way road from 9v to your signal.  I'm not entirely sure whether or not it's capable of clipping the signal at all, but I doubt it.  If anything, it's clipping one half of the signal, but not the other.

In other words, the reason it sounds different is that when they were attached to GROUND, they were clipping.  Now, they're doing practically nothing, except letting DC voltage reach a spot in the circuit that it shouldn't be touching!

[boog]

Very good everyone and thank you for the encouragement!  One of the things I seem to be having a difficult time in comprehending is exactly how the electricity from the power voltage and that of the input signal interact and cause (or uncause) problems.  So I understand that how the LED's do nothing, including a little bit of what you are saying about the biasing (at least I know what kind of verbage to look up) and that there's actually a resister between the LED and ground in the pot.  But what about what you, BAARON, say about the voltage being on the volume pot.  I understand where it is, but I don't understand what could possibly go wrong ( ) in this instance.  Is it that voltage used in this manner is generally a bad idea, or is it that it's outside a polarized cap?

[BAARON]

The problem is that the voltage on the - side of the output cap (9v minus the forward voltage of the LED = ~7.3v) is higher than the voltage on the + side (about 4.5v, coming off the opamp).  The voltage needs to be higher on the + side of the cap than the - side.  Otherwise, you could kill that electrolytic cap.  That's what the - stripe is there for.

Also, it means that you're applying DC voltage to your patch cable and out to your amp, which is not good.  That can seriously mess up the bias on the first triode of a tube amp, because tube amps don't have input caps to keep out DC.  Output caps are there specifically to keep DC voltage inside the pedal... not outside of it.

[boog]

Thanks for the explanation.  All I've ever seen has been a "just don't do that" warning.  I appreciate the instruction!