Orman's diode blender

[Ben Lyman]

An LPB-1 with Jack Orman's OD/clean blender added on. Pretty cool in it's most basic form but I wish there was a little more separation between OD and clean plus a little more crunch in the OD half. Some more high end might be nice, maybe a tone stack or something. Any thoughts?

[Kipper4]

Couldn't resist googling up so this is where your coming from partly

http://www.muzique.com/lab/sat2.htm

What input cap are you using for C1?
Where do you want the high end ?

What taper are you using for the blend?

What diodes are you using?

[Ben Lyman]

oops! I forgot to click "show value" on all my components! 
Sorry, please stand by while I repost the schemo.

[Fast Pistoleros]

dude, very nice link man..you guys .quit posting so much good stuff, I dont have any time lololl 

[Kipper4]

Where do you want more high end Ben ?

[Ben Lyman]

I think maybe some more highs in the clean half would be good

[Kipper4]

Maybe try a high pass filter in place of r6 r8
Cap in place of r6 (simple rc filter) start small and move up.
And maybe a 100k pot in place of r8.
Now it's tunable filter.
There will be some loss of signal I'd of thought. On the clean side.

Let us know how it goes.

[Ben Lyman]

Thanks kipper. Replacing R6 with a cap brightened it up but also let a lot of OD bleed thru so I put the cap between R6 and junction of R8/R9. Then of course the OD was a little to dark so I changed C1 to a 10n. Pretty nice balance now with a 100n after R6. Still tons of volume boost if needed.  Haven't yet tried any different diodes/LEDs. Here's the current set up, starting with bypassed then the clean side and turning the pot towards the OD side:

[R.G.]

a little more crunch in the OD half. Some more high end might be nice, maybe a tone stack or something. Any thoughts?

Scale the resistors all down so there's less resistance between the amp output and the diodes.

[Ben Lyman]

Scale the resistors all down so there's less resistance between the amp output and the diodes.

ok, thanks RG. Do you mean reduce *ONLY R7* value (10k to the diodes) or reduce *R7 AND R6* by equal amounts?
I tried a 100k pot as a variable resistor between output and diodes which gave some more versatility but I'd rather avoid it if possible in a build.

[R.G.]

Sorry - it depends.   

What's going on here is that the two diode are being driven by the transistor's output through R7. R7 limits the amount of current that can get into them, and that limits how "saturated" they can clip. More current, harder apparent clipping. Notice also that the diodes are mismatched. One is silicon, one is germanium, presumably to get some asymmetry in clipping. You could also change the 1N914 to a 1N34a and get more clipping. Changes the nature of the clipping a bit. Or put a resistance in series with one 1N34A, etc.

The "blending is a result of the wiper on the 100K picking off various mixes of the two outputs, the resistive one on the top, and the clipped one on the bottom. In such a simple pan circuit (that's what it is, at heart), you never get entire all of one or the other, and the "blending" in the middle depends on the various resistances.

The blend pot has to be about 10x the impedances at each end to not affect the signals at the two ends very much, and it has to be relatively lightly loaded (that 100K pot on the output) to help with the not-too-heavy loading and with the relative abruptness of the mix as you get away from each end. It takes a lot of tinkering to get those right.

That tinkering works against you when you try to lower the impedance driving the diodes. So you may have to scale them all, and that puts pressure on the ability of the transistor to drive this, given that it has a 10K output impedance (that collector resistor) itself. So - try some stuff.

The blend-between-two circuits, variants of the "Big Muff Tone Control" and wherever it came from, is, as our UK brothers say, cheap and cheerful. It's not perfect, but it sure is cheap, and if that's enough, great. Tinker the resistors a bit, see how it sounds.

I nearly always get sucked into making things do what I wanted them to do as I first envisioned the sound. So I'd probably drop back to a dual opamp in one of several variations to get better control. But the complexity does creep up, the price of better control.

[Ben Lyman]

Thanks RG, that was a very helpful post, much appreciated 

...Notice also that the diodes are mismatched. One is silicon, one is germanium...

...it has a 10K output impedance (that collector resistor) itself....

...I'd probably drop back to a dual opamp in one of several variations to get better control...

The mismatched diodes are just what I had sitting on my bench when I decided to do this, I will try some more experiments.
I did change the collector resistor to 18k just because I got out the DMM and 18k got the collector right at 4.45v  hmm... wonder if that was a good thing or bad.
I really wanted to keep this super cheap and easy but I have also considered trying it with an opamp, I want to believe it would be super cool with a OD250 or Dist+ type pedal.

[R.G.]

This is all fairly fresh on my mind, as we just revamped the product line for at The Company I Work For last year, as we added a clean blend to the distortion pedals.  I tried all the simple ways, and finally wound up doing it with the dual opamp panner for better control. But that's a fundamentally different set of requirements.

There's another blend dimension you might want to try. Back in the 60s, they often did distortion and then added the dry signal back in, but out of phase. This had the effect of tending to cancel out the remaining fundamentals in the distorted signal, leaving much more of the distortion products, but without adding even more high-order harmonics.

[Ben Lyman]

This is sounding really good to me right now. I think I'm gonna box it up. I added a switch to the 100n input and a 3.3n which turns it into a very cool treble booster with blendable grit.
I wonder if I should put a 1M across the switch in case it pops? 
I also changed the blend pot to 500k to get a little more separation (I think?) and adjusted to clean side 100k to ground down to a 56k, now there is a pretty smooth level throughout rotation from clean to crunch. 500k pot for master volume seems important to keep that even blend, not sure why and I hope I didn't mess everything up.
I've tried a bunch of diodes and LED's but the original sounds best to me, I will use sockets anyway.

[R.G.]

Yeah, put a 1M across your switch to keep it from popping.

And if it does what you like, you're good.

[Ben Lyman]

Thanks RG. I Just played around with it thru my tweed princeton with a strat. I replaced the 3n3 with a 2n2 and now it sounds EXACTLY like a vintage Dallas-Arbiter Rangemaster! (Tongue planted firmly in cheek) But seriously, it does sound really good, I can get a lot of great sounds from it with one toggle switch and one blender knob.

[karbomusic]

From a player point of view, I build in an opamp blend for most any circuit I can (that I build) except for fuzz faces. I've always thought that blending should be the norm.

[Ben Lyman]

I've always thought that blending should be the norm.

+1

It's done and I'm really diggin' it. Some pix & a vid:

[Kipper4]

Ben I like it