"Dual" overdrive..

[antonis]

Just finished (with a lot of helpfull notices from a previous post) an overdrive circuit with dual distorted signals (hard & soft)..



Practically, it's my first designed effect so I'd like (and be appriciated of..) ANY comments/suggestions/notices/proposes about the circuit..

(propably I made a lot of "mistakes" and have also a lot of doubts about many individual designs i.e.: the gain of the first stage was designed for a single singal -  should I double it because of the splitted one..?? or: are the 2 saturation pots a good idea or just an overkill..?? or: have the different band zones in the gain stages a noticable result..?? or or or anything else..)

P.S.
Please, be gentle to an aged first timmer...

[seedlings]

Nice!  I like this idea... and it would be pretty easy to channel switch one or the other or both.  How does it sound?

I'd say that the bias for Q1 is a kind of heavy load on the guitar, less than 100k.  Might help to multiply R1 and R2 by a factor of 10.  Also, what is the purpose of C9 and C10?

Keep up the good work.

CHAD

[antonis]

How does it sound?

I'm not sure...
(I haven't constructed yet..)  

I'd say that the bias for Q1 is a kind of heavy load on the guitar, less than 100k.  Might help to multiply R1 and R2 by a factor of 10.  

Not a bad idea...
(but I think a little guitar "loading" should sound more "all time classic" fuzz..)

what is the purpose of C9 and C10?

In theory, they "soften" the corners of the waveform and "mellowing"  out the clipped high end...


P.S.
Maybe I should try the "soft" clipping in the opamp feedback section...

[seedlings]

All good design points.  And now for the breadboard sound samples coming soon, right??? 

CHAD

[PBE6]

One thought, you may want to insert a buffer after the blend pot. The tone stack will load the blend pot fairly heavily, which will make the pot much less effective until you get to the end of the rotation. Adding a buffer in-between should fix that.

[PBE6]

One other thought, just before the blend pot you may want to include a small trimpot (1k? 5k?) in line with the signal being clipped by the LED's. This signal could be twice the voltage (or more!) of the signal being clipped by the 1N914 diodes. The series trimpot can be adjusted to cut that signal back a bit for more predictable blending.

[blackieNYC]

Assuming you are breadboarding this, I would want to be blending two really different distortions. One idea would be to work a HP filter into one and a LP filter into the other.  Maybe fixed, with just a switch.  Might give you clearer lows if the lows only go to the LED clipper.
Don't get me wrong, diode clippers like the MXR D+ might be my favorite, but you might consider blending one of your circuits with a transistor based fuzz.  Or something really wacky that has great character but might not stand on its own as well.

[PRR]

> The tone stack will load the blend pot fairly heavily,

I think this was thrashed in another thread??

The 10K blend pot wiper impedance is about 10K/4 or 2.5K.

The tonestack starts with 56K and won't go much lower.

2.5K<<56K.

[PBE6]

Hmmm, I haven't seen that thread but it sounds like something I should read. 😳

I based my conjecture on a quick analysis on the source voltages, the pot value and an assumed load, but didn't actually check what the proposed load was. Where does the 10k/4 come from though?

But bottom line antonis, looks like no buffer is needed.

EDIT: Ah, the maximum impedance of the 10k pot occurs when both legs have half the resistance, and the legs are taken in parallel so R = 1/(1/(10k/2) + 1/(10k/2)) = 5k/2 = 2.5k.

[antonis]

Assuming you are breadboarding this, I would want to be blending two really different distortions. One idea would be to work a HP filter into one and a LP filter into the other.  Maybe fixed, with just a switch.  Might give you clearer lows if the lows only go to the LED clipper.

Unless I've got something wrong (my English are realy bad..), I think that it is "covered" by the 2 different bandpass zones at the Hard & Soft gain stages (1,6kHz - 8kHz & 720Hz - 16kHz respectively..).
I'm not sure if we are talking about the same case...

Don't get me wrong, diode clippers like the MXR D+ might be my favorite, but you might consider blending one of your circuits with a transistor based fuzz.  Or something really wacky that has great character but might not stand on its own as well.

I don't get you wrong as long as you don't get me into mind troubles..
(honestly, your idea is excellent !!! and I promise to do it in my next project - thanks a lot..!!)

[antonis]

> The tone stack will load the blend pot fairly heavily,
I think this was thrashed in another thread??
The 10K blend pot wiper impedance is about 10K/4 or 2.5K.
The tonestack starts with 56K and won't go much lower.
2.5K<<56K.

Oh yes and you are included in <with a lot of helpfull notices from a previous post> Paul..!!!

I know that you "insist" for R11 (slope resistor) to be 100k but I'd like to try a lower value to make the tone appear more "midrangey"..
(although with 100k should have about the half tone sucking than with the 56k - propably I will include sockets in the trial PCB..)

[antonis]

Another question..

If I mix two different "kinds" of distortion (like one from a BJT amplifier and another from JFET OpAmp) with the same polarity (both inversed) will be any kind of "mismatching" (like some kind of frequency drift between the two signals or anything else)..