Different clipping options for DOD 250 / Dist +

[afrogoose]

Hey Everyone,

Thanks to a little help from fellow DIYstompboxers (cheers UKToecutter!) I have built a General Guitar Gadgets GGG Dod 250 Gray clone.  I've never played a real Dod 250 or a Dist + for that matter, and I'm not sure what exactly it's supposed to sound like.  I wired in a three way switch to change the clipping diodes in this pedal and right now I'm trying red LEDs and Ge diodes.  I'm wondering how much gain this pedal is supposed to have.  It doesn't seem to be clipping the signal as much as I expected.  Specifically, the red LEDs don't seem to be clipping the signal at all really.  There isn't much or any difference at all between the LEDs and no clipping diodes at all?  Is this correct?

[sajy_ho]

If I understand correctly, LEDs have higher Vf in comparison to Germanium or Silicon diodes, so they want higher signal level to start clipping. In my TS808 I've used a rotary switch to change between different pair of diodes and just like your build, LEDs didn't clip the signal at all while the other diode did well.

[Mark Hammer]

A pair of red LEDs will generally require a signal of around +/-1.5V in order to clip.  A pair of germanium will only need about +/-250mv, on average, but will yield a much lower maximum output level than LEDs because of that.  The original used a pair of silicon diodes, like 1N914/1N4148, which require about +/-500-600mv to clip.  

Keep in mind that your guitar signal starts out large, but then shrinks pretty quickly.  If the clipping threshold is set high, then even with max gain in the circuit, it will quickly fall below that threshold, such that you'll only get clipping for the first post-pick moments, and little after that.  Setting the clipping threshold low (by means of the type and number of diodes used) will allow for the picked string to exceed the clipping threshold and distort for a longer portion of its lifespan. It will also yield a somewhat more compressed-sounding distortion (i.e., reduced dynamics, and the impression of greater sustain), and a lower maximum output level.

Personally, I think a 250 sounds just fine with a pair of 4148/914 diodes.  Change the .047uf cap connected to the Gain pot to .1uf or even .22uf, and you'll get more oomph to your sound, and a bit more distortion, without having to change the amount of gain.

[TheBigMan]

I find that Ge diodes sound best to my ears in a ground clipper, with as Mark says the loss of overall level.  Back when I used to mod pedals most people wanted LEDs in their DS-1's for example, which I thought sounded much poorer than Ge diodes.  LEDs do sound quite good, again to my ears, in a modded TS.  Mine's got an on-off-on toggle with 1N914/1N4001, diode lift, and 2 LEDs.

[Mark Hammer]

Many people will build their DOD250 or MXR Dist+ clone using a dual op-amp, and many posted layouts are based on use of a dual.  The smart and simple thing to do is to is to use the other op-amp in the chip as a gain recovery stage, such that one gets to take advantage of the clipping properties of Ge diodes, without necessarily sacrificing level to do so.

The old Voodoo Labs Overdrive demonstrates this principle in action.  You can see here that the first op-amp stage is basically very very close to a 250/D+, and then feeds a second op-amp stage with a gain of about 4x, to bring the level back up (or also yield a reasonable punch even when the clipping stage is set for more modest gain).

Just a few details to clear up here, before sending you off to build it:

1) There needs to be a fixed resistor in series with the 500k gain pot to set the maximum gain.  I recommend 2k2 in this instance.
2) The 220pf feedback cap in the clipping stage rolls off above 1.5khz.  Personally, that is my own preference.  But it is not what the DOD or MXR circuits use, so feel free to use something lower in value.  The 220pf in the second stage is simply a smart idea to tame potential hiss and fizz.
3) In the 250 and D+, there is an electrolytic cap on the output of the clipping stage, but there is none here.  That DC blocking is accomplished on "the way out", just before the volume pot.  The circuit works fine as shown (with the addition of the fixed resistor noted in #1).

[petey twofinger]

that sounds really cool mark .

my 250 has leds , i use it a lot, i never put the diodes on . it clips when i dig in , this is for cleaner more OD ish sounds , but i find when i turn the gain past 12 i get enough grit. i run it between 11 and 1 and  have to say i use it too much . unless of course i am going for a more distorted tone but i have been trying to avoid that , there are 2 guitarists and keys , cleaning up a bit has worked well for me . i would like to try that idea that mark posted though .

[Hatredman]

I guess whatever you use, it matters less at what level it clips, and more on how soft it clips.

[ashcat_lt]

I'm wondering how much gain this pedal is supposed to have.

The gain of a non-inverting opamp stage is pretty easy to figure.  In fact, it's exactly the inverse of the voltage divider equation, which is the second most important equation in the game.  Heck, it might even be more important for the things we do around here than Ohm's Law!

Vout = Vin * (Rbottom  /  (Rtop + Rbottom)) 

Remember that.  Then remember that the gain of the opamp is the exact inverse of that.

Look at the schematic, use the above, and answer your own question.

[Hatredman]

In other words

Av = 1 + (Rtop/Rbottom)

Remember the gain is never Unity on non-inverters.

[Mark Hammer]

That wold be where you are incorrect.  If there is no ground leg, the gain is 1.  Anytime you see a unity-gain input buffer where there is no ground leg, but a straight-wire connection from output to the inverting pin, that's a unity-gain non-inverting stage.

But apart from that, if there is ANY ground leg resistance at all, then you are correct.  So, if the feedback resistor is 10k, and the ground leg is 10M, the gain is equal to (10k+10000k)/10000k, which isn't a whole lot bigger than 1, but it isn't unity.

[ashcat_lt]

In other words

Av = 1 + (Rtop/Rbottom)

I don't like that one because it hides the voltage divider and looks like a new equation.

[afrogoose]

Hey, thanks for all the info everyone!  Mark, thank you especially for the tip on where to goose some more gain out of the DOD 250, and thank you for the tips on the Voodoo Labs circuit.  Unfortunately it's not showing right now?  I'm not sure if maybe you took it down?  

I appreciate the responses re: how much gain the pedal is supposed to have, but I guess I mean to say that in a qualitative way rather than a quantitative way.  Meaning, how distorted is it supposed to be?  It's lame because my only point of comparison is comments from DIYstompboxers and watching Youtube videos.  It seems like maybe mine isn't fuzzing out enough.  

Here are two quick vids of the pedal.  Maybe you guys with 250 experience can tell me if this pedal sounds about right?

Ge diodes

LEDs

Thanks!!!

[afrogoose]

Sorry.  Took me a minute to fix those youtube links...

[bloxstompboxes]

Nice grit on the Ge didodes. Never tried the leds in mine. Sounds about right to me though overall.

[Elijah-Baley]

It sounds pretty similar to mine. 
I tried Ge Diodes, but I preferred the Si Diodes stock (a little more volume and and little less smooth than Ge DIodes). I switch it with red led. I tried orange, yellow and green led, but I had not heard differences.
I can choose without clipping with my On/Off/On switch, more volume and less gain than led clipping.

[ashcat_lt]

...more volume and less gain...

No!  Stop that.  It has exactly the same gain, just less clipping.*  I am aware that it's common to use the word "gain" when one means "distortion", but it's WRONG and as we've seen in this thread leads to confusion.  You have the opportunity right now to make a small step in the direction of changing the "common usage" to something closer to correct.  All you have to do stop using the incorrect terms now.  Not trying to specifically call you out, just trying to save us all some headaches down the road.


*Technically, there is more overall gain (input to output) for higher input levels than with the clipping.  That's kinda what clipping is.

[Elijah-Baley]

I'm sorry. You're perfectly right!!! 

Thank you.