clipping diodes and pot adjustment

[aaronkessman]

i want to add clipping diodes with an adjustment pot to dial in just how much clipping those things cause.

does it matter if i put the diodes first, then the pot to ground or if i put the pot first, then the diodes to ground?

Aaron

[Mark Hammer]

Yes and no.  If there is a fixed resistor in the signal path and then another in series with diodes to ground, it is my understanding that those two resistors will act as a voltage divider to yield a softer clipping.  I'm not so sure it wuld do the same if the diodes and series resistor switched places.  Ton, RG, and "the Petes and Pauls" will probably have something to say about this.

[R.G.]

It matters.

In general two items in series may be in either order with no difference as long as you don't care what happens in the middle. In this case, all you do care about happens in the middle.

In the "standard" connection, the diode pair goes to ground and you take the signal off the diodes. The resistor limits how much Max current the diodes get from the signal source. When the signal is small, no current flows, so the signal is reduced hardly at all by the series resistor. When the diodes start to conduct, the voltage in the middle is clamped to the diode voltage, so the signal is flat-topped at a diode drop, and enough current flows so that the resistor eats up all the rest of the signal voltage.

Result: the signal appears unchanged until it gets bigger than a diode positive or negative, then is clipped there.

If you put the diodes first and the resistor to ground, then the voltage across the resistor is what you see. When the signal is less than a diode drop positive or negative, you see... nothing. The diodes don't conduct significantly yet, so there is no current to flow through the resistor to make a voltage. When hte diodes start to conduct, the signal voltage that is greater than a diode drop appears across the resistor.

Result: the output signal looks like only the tops and bottoms of the signal with two diode drops eaten out of the middle. It's distort-y all right, but not what most people expect.

You might like it different ways. Try both.

[WGTP]

Just when you think you understand things ...

R.G. are you saying the second method works like having the diodes in-line?  Or is that yet a different way to use diodes to distort?

I guess you need 2 sets (I used a 10k pot on my Rat) one each way.

2 more Knobs for the TMK.  

[aaronkessman]

well, im going to do it for my fuzzface right before the volume control. it's easy enough to try both ways. I was thinking that there'd be a difference and that the diodes and then the resistor would be weird.

how about a third option - connecting the diodes between one of the end lugs and wiper of the pot so as you turn the pot "up", there's less resistance between the signal and diodes. of course, there's a resistor in parallel there - how would this affect things?

[WGTP]

On second thought, you could do each diode differently.  One with a pot between the diode and ground and one with the diode to ground and the pot to the signal.  Hmmm  

[aaronkessman]

ok. connected them both ways and they're both pretty cool. they don't add quite as much distortion as I expected, maybe i'll use two in series. They both sound pretty extreme and raunchy in the fuzz face and give it a little more versatility. Would I have it on a 10k pot, but about half the pot's rotation does nothing, so the 5k would work great. thanks for the input guys,

Aaron

[WGTP]

Since the FF is already pretty distorted, it may be hard to hear the actuall affect.  Probably an Op Amp deal like Distortion + would be easier to evaluate, I'll have to try that on my perf board.  Cool idea.  

[aaronkessman]

yeah...figured that's why it's hard to tell. i tried to lower the fuzz control, but that wouldn't make them clip at all. sometimes i wonder why the orignal was made as it was. I get much more use out of my DIY "bias" and "contour" pots (based on the 69 circuit) than the fuzz control which is always maxed out.

Aaron