DEMO Oomph! Overdrive [with continuously variable clipping symmetry]

This post contains an audio or video demo
Aleph Null

Aleph Null

Well-known member
This circuit started out as an Electra Distortion. I kept tweaking and adding the features on my wish list until I ended up somewhere very different. I think it works best pushing a dirty amp, but it certainly sounds good on it's own as a low to mid-gain overdrive. It's pretty versatile, especially given the low parts count.



The knob placement required using a mix of 16mm and 9mm pots, but I think the aesthetic effect was worth the inconvenience. I attempted a hydro dip finish and failed...so I moved on to a paint splash finish. The perspicacious among you will notice that the PCB is labeled "Electra Supreme"; I had ordered the PCBs before the name came to me.



The soldering around the diodes looks a little sloppy because I tested multiple sets before settling on 1N60s. BAT46, 1N60P, and 1N34A also work well.



An input cap blend acts as a pre-gain Lows control. This feeds what was originally the Electra. The bias is set up so that the Gain can be adjusted without scratching. The real innovation here is the Texture control. This blends continuously between symmetric clipping when fully counterclockwise to highly asymmetric clipping when clockwise. This changes the harmonic content of the overdrive in a way that a normal "mids" eq would not. High Gain and Texture settings start to border on octave fuzz sounds—especially if you push the front end with something else. Next is a make-up gain stage with a presence boost around 2kHz. Boosting the treble after the clipping stage does mean this circuit can get a little noisy in some settings, but the presence boost provides a sizzle that I think makes it worth it. This is followed by a fairly standard passive Highs tone control and then the Volume.

Here's a demo:


As always, I have extra PCBs if anyone is interested in building one. Just DM me.
 

Attachments

  • IMG_1110.JPG
    IMG_1110.JPG
    539.1 KB · Views: 1
Last edited:
Very cool tweaks and tidy design as always!
One question.
Is the input cap blend/low control not neutered by C4 and to a lesser extent C5 and on?
Why not let the low end at least hit the clippers, if not the second stage as well?
 
jwin615 said:
Very cool tweaks and tidy design as always!
One question.
Is the input cap blend/low control not neutered by C4 and to a lesser extent C5 and on?
Why not let the low end at least hit the clippers, if not the second stage as well?
Click to expand...
An RC calculation would suggest that C4 is attenuating bass, but that doesn't actually seem to be the case. With Gain all the way down and Lows all the way up, I'm getting just as much bass response with the pedal engaged as in bypass. The Lows control being at the front does effect clipping, but that's on purpose, as pre-gain bass attenuation tends to help keep things tight. The second stage is for volume recovery and not really for additional clipping. The forward voltages are 200–500mV for all the recommended diodes, so the second stage isn't really getting hit very hard.
 
Aleph Null said:
An RC calculation would suggest that C4 is attenuating bass, but that doesn't actually seem to be the case. With Gain all the way down and Lows all the way up, I'm getting just as much bass response with the pedal engaged as in bypass. The Lows control being at the front does effect clipping, but that's on purpose, as pre-gain bass attenuation tends to help keep things tight. The second stage is for volume recovery and not really for additional clipping. The forward voltages are 200–500mV for all the recommended diodes, so the second stage isn't really getting hit very hard.
Click to expand...
Gotcha. But not really think in terms of an RC filter and more so about the reactance vs frequency relationship of a given value. Just as the 200n input cap allows more bass through, the 100n decoupling cap will restrict it/allow less bass.
Maybe this is all 99% lost using a revLog taper pot on the input caps(I still struggle not mixing up directions of sweeps when looking at a schem). But it would seem like a portion of the sweep would result in little to no change as C4 is blocking some lows that C3 allows to pass before any real harmonics etc are generated.
As neither are part of an RC network, it's only the reactance that is restricting the low end.
Again. that change could all be on the last 5% of the pot and we're also talking fundamentals that are probably below a standard tuned guitar as well...
 
jwin615 said:
Not trying to be a jerk or question design choices. Just curious. A peak under the hood if you will...
Click to expand...
No offense taken!

I always start with theory and then adjust by ear. I designed this one a while ago, so I may be a little fuzzy on some of my thought process, but to my ear 100nF for C3 C4 was big enough. You could up the value to 220nF or even 470nF if you really wanted to be safe. The 220nF and 3.3nF values at the input were chosen to get the best taper and range across the pot. It may be that there's some interaction between C1, C2, and C3...but I can't really characterize it any more scientifically than, "sounds good to me!"
 
Last edited:
Aleph Null said:
Real talk: if this pedal used a more standardized control layout and only required 16mm pots, would it be more appealing to build?
Click to expand...

Not for me.

I like the unusual layout and I try to keep a good stock of 9mm pots for builds like this, PCBs from MBP, elsewhere and 1590A builds — and then you can squeeze in a 9mm pot-mod to a regular build so much easier, too.

Nonetheless, you're sure to sell more PCBs with a conventional layout and 16mm pots (bonus for many here if you lay out the PCB for one of Robert's Drill Templates/Tayda.




Aleph Null said:
...to my ear 100nF for C3 was big enough. You could up the value to 220nF or even 470nF if you really wanted to be safe. ...
Click to expand...
To clarify something from your post #9, you said you thought C3 was okay at 100n, but on the schematic it's 47µ? That's a HOOOJ jump!
 
YourGuitarist said:
It's be interesting to swap a Darlington into the first Q. That texture control is a neat way of varying symmetry.
Click to expand...
If you do, I would recommend omitting R4; that's going to limit the gain.

I chose to limit the gain and use lower forward voltage diodes because I found that hitting Q2 caused the whole thing to get farty and mush out. That could be fun, but it wasn't what I was going for. I got some promising results with mosfets, but you'd have to reduce the bass response in a few places to keep it tight.
 
I mean r4 ain't limiting much I think. Could also be fun to put the bass control in the emitter path. Partial bypass. I'm mainly just throwing ideas out there for fun.

There is such a thing as too much gain.
 
YourGuitarist said:
I mean r4 ain't limiting much I think. Could also be fun to put the bass control in the emitter path. Partial bypass. I'm mainly just throwing ideas out there for fun.

There is such a thing as too much gain.
Click to expand...
Yeah! There are lots of options! I've got a mosfet design I'm working on that used the partial bypass approach. You can also connect a cap from the collector to the positive rail to act as a low pass filter. If you do both, you could get a "mid boost" thing going.
 
You must log in or register to reply here.
Back
Top