Saturation control (Dist +)

[matopotato]

I am building a Distortion + based project, with some suggestions from Jack Orman (http://www.muzique.com/lab/sat.htm and other pages close by)
And I liked the sound of example 2 and 3 on that page, but with lower R (270R or similar) and higher C (47nF).
Both are nice and I tried them in series as well as in parallel.
There is probably an obvious choice here, but I would appreciate if someone can suggest and possibly explain why one is better than the other. I plan to have it always on.





Then I am also considering the last suggestion on that page. It also sounds nice, but here I plan to add an SPDT to have it on or off.
Also using 47nF instead of 1nF suggestion. Will that 47nF (or what might be better) have some unwanted impact on the circuit?



Thanks in advance

[antonis]

I tried them in series as well as in parallel.

Not enogh data..
Plz elaborate..

Will that 47nF (or what might be better) have some unwanted impact on the circuit?

Other than more bass clipped, no..
(the bigger the cap value the more close to short between wiper and pot lower lug, for any given frequency..)

[Mark Hammer]

Sometimes, you just have to see a circuit drawn a different way for it to make more sense.  Here's the first circuit, redrawn.  Note that many different diodes-to-ground clipping circuits will feed the diodes with a 10k resistance.  From what I understand, the behaviour of the diodes will depend on the current one is feeding them (though I know little about what the specific relationship is), so the 10k provides a limit on current, and the way the pot is wired reduces that series resistance. 

Perhaps some smart folks here may have something to say about what changes to that current-limiting resistance may imply when it comes to different types of diodes, like Schottky or germanium.  I mention that, because any attempt to implement a "saturation" control in conjunction with diode-switching may require some adjustments.

[ElectricDruid]

My comment with this would be that the worst that can happen is that you lose a lot of output level. Like if you make a cap to big and finish up filtering off all the treble and most of the rest of the signal too, or the reverse, and you lose all the bass and most of the midrange too.

You won't blow anything up. So experiment. There isn't a "best" cipping circuit. There's only the one that makes you most happy. Try different diodes, different resistances, different caps.

[Mark Hammer]

Tom's right that you won't "blow anything up".  At the same time, there is a question of what sorts of mods yield the most audible and usable changes.

The Dist+ uses the ground leg of a non-inverting op-amp to vary gain.  This is in contrast to a TS-9 that uses the feedback loop to vary gain.  When using the ground leg, in conjunction with a cap, increasing the gain simultaneously results in a drop in bass.  My own pet theory is that the designer/s of the Dist+ may not have intended it as such, but found a perk in that approach because it did not amplify 60hz hum quite as much (hum-rejecting pickups and layouts being less common during those years).  It also made the added harmonic content from the clipping more obvious.  When the feedback resistance is used (in conjunction with a feedback cap) to vary gain, bandwidth at the high end is reduced as gain is increased.  This contributes to the TS-type drives sounding "warmer".

Lower-frequency content from the guitar is generally of much greater amplitude.  If one retains bottom end in the gain stage, you "push" the clipping diodes harder.  This is the reason for Ibanez rolling off bottom below 720hz in the TS, so as to provide a more consistent clipping across all strings and the fretboard.  Increasing bass in the Dist+ gain  stage will get you more "grind".  Indeed, I suspect that varying the bass content will not only get you a different tonal balance, but likely achieve a more noticeable change in the resulting clipping - even with the same diodes - than  a saturation control will.  That's not to take anything away from a saturation control.  But if you want a Dist+ that does not look like the %^&*pit of a 747, consider other approaches beyond futzing around with diodes.  They're not quite as magical as the internet would have us believe.

[EBK]

But if you want a Dist+ that does not look like the %^&*pit of a 747, consider other approaches beyond futzing around with diodes. 

Because of the weird, nonsensical nature of the auto-censor rules, I didn't read that as "cockpit", but rather as "shitpit".  My very tired brain thought, well, I suppose airplane toilets have to flush into something. 

Back on topic, a lot of people have made a lot of money off of "magic" diodes, but I agree, real magic often hides elsewhere, such as filtering.

[matopotato]

Apologies for not being very clear on what I was asking.
Will try to make up for it.

First though, I am fully aware of experimentation is the key in designing and finding a sound and tone you like. And I do that. In fact my very first DIY kit, I couldn't help but add a modification.
I am quite new to electric guitars (16 mts since my first) although I began with acoustic in '76 (ish). Since I went electric and asked "What are pedals?" a whole world opened up and has progressed via diy kits to own inventions or rather extensions of some known circuits (Fuzz Faze, Wampler's Distortion YT clip how to gradually build things up and now Dist +).
Since I am still a bit new, I have not yet reached the "nirvana of subtle", so I notice I prefer effects where I really can tell they're on and doing their thing. It wasn't until 2 moths ago that I appreciated Compressor for the first time, having not heard or understood what they did to anything.
So, when I experiment I notice after some time as my "subtle" gets trained a bit better how things sound.
That is why I wanted to ask for some advice to avoid making something that in a few months time will sound "bad" when it could have been "good" and I just drop the build.

Here is my attempt at redrawing the Dist + with marked In and Out as reference to Jack Orman's examples.


Here is what I meant by having R and C in parallel




And this is my meaning of them being in series





And my question is more or less, over time what will work best?
I thought from other builds that series would work best, but it sounded as if parallel was preferable.
But I have no idea if either could result in "Well, with Humbuckers you'll find that..." or "If you have a low Vf diode pair in there, it's not gonna work out well with ..."

I hope this makes my post a bit more clear? (@Antonis)

Then adding the 47nF to the "other side" of the Saturation control, I was wondering if it could interfere, but I got the reply that it will not, just cut bass.
Jack Orman suggested 1nF for "some sheen or enhancement to the sound", but 47nF might be too much. I will try out with a bit lower since I do not want to kill all the bass. But have it more of an EQ switch or Hum/SC switch

[matopotato]

Sometimes, you just have to see a circuit drawn a different way for it to make more sense.  Here's the first circuit, redrawn.  Note that many different diodes-to-ground clipping circuits will feed the diodes with a 10k resistance.  From what I understand, the behaviour of the diodes will depend on the current one is feeding them (though I know little about what the specific relationship is), so the 10k provides a limit on current, and the way the pot is wired reduces that series resistance. 

Perhaps some smart folks here may have something to say about what changes to that current-limiting resistance may imply when it comes to different types of diodes, like Schottky or germanium.  I mention that, because any attempt to implement a "saturation" control in conjunction with diode-switching may require some adjustments.

Many thanks!
The out goes from the variable part of the SAT control though. (No idea if that matters)

I am actually experimenting with different diodes that I have picked up from 70's TV sets and radios, 90's TV set and a few phones from around that time.
And my first idea is to add them using a Rotary switch, so it will be a Dist+ with many diode pairs.

I will also add an alternative "Crossover" control (without the saturation one) and have 1+1 diode pair and another 1+2 pair (4148) to get a feel for the difference.
So there will be many diodes, and all might not do anything innaresting, but I guess it is where I am in my learning process.
I have not considered much of adjustments in the circuit, but more of hoping the pots will help me out.

[matopotato]

My comment with this would be that the worst that can happen is that you lose a lot of output level. Like if you make a cap to big and finish up filtering off all the treble and most of the rest of the signal too, or the reverse, and you lose all the bass and most of the midrange too.

You won't blow anything up. So experiment. There isn't a "best" cipping circuit. There's only the one that makes you most happy. Try different diodes, different resistances, different caps.

Thanks! Point taken.
I was not so concerned (anymore) about blowing up. I just don't like to put in a lot of effort, test and try and box up only to find a week later that it sounds kind of bad and it turns out I could have fixed/understood it sooner if I just asked.
But I agree that experimentation is key.
Then I do not always hear the difference between distortion and distortion. I've seen clips from the famous YT pedal shows/reviewers where I honestly don't really hear why a certain combination or setting is so bad. But I also realize that given some training to my ears, I probably will later on.
Pedalism is an acquired taste I suppose.

I built a pedal based on this Wampler video https://www.youtube.com/watch?v=F-WHBgIowmU and added several diodes in the feedback loop to sort of see what happened.
Some OA74 1+1, AA114/9 2+2 and 1N60 2+2 and another rotor for 1N60 1 + x with x from 1-6. Total diode overkill, but a bit fun.
The OA74 just grows on me, and the 1+ x kind of peaks at 2 or 3.
I also added EQ switches for a few Cap variations both where the feedback loop hits ground and also at the tone control. And now I realize I should have combined those two switches into one.

[matopotato]

Tom's right that you won't "blow anything up".  At the same time, there is a question of what sorts of mods yield the most audible and usable changes.

The Dist+ uses the ground leg of a non-inverting op-amp to vary gain.  This is in contrast to a TS-9 that uses the feedback loop to vary gain.  When using the ground leg, in conjunction with a cap, increasing the gain simultaneously results in a drop in bass.  My own pet theory is that the designer/s of the Dist+ may not have intended it as such, but found a perk in that approach because it did not amplify 60hz hum quite as much (hum-rejecting pickups and layouts being less common during those years).  It also made the added harmonic content from the clipping more obvious.  When the feedback resistance is used (in conjunction with a feedback cap) to vary gain, bandwidth at the high end is reduced as gain is increased.  This contributes to the TS-type drives sounding "warmer".

Lower-frequency content from the guitar is generally of much greater amplitude.  If one retains bottom end in the gain stage, you "push" the clipping diodes harder.  This is the reason for Ibanez rolling off bottom below 720hz in the TS, so as to provide a more consistent clipping across all strings and the fretboard.  Increasing bass in the Dist+ gain  stage will get you more "grind".  Indeed, I suspect that varying the bass content will not only get you a different tonal balance, but likely achieve a more noticeable change in the resulting clipping - even with the same diodes - than  a saturation control will.  That's not to take anything away from a saturation control.  But if you want a Dist+ that does not look like the %^&*pit of a 747, consider other approaches beyond futzing around with diodes.  They're not quite as magical as the internet would have us believe.

Again, thanks!
So I guess it would be back to the drawing board again.
Is there some obvious way to add bass control while keeping the other mods?
I kind of liked Jack Orman's way of suggesting the Saturation control, Warp control and crossover examples on the AMZ page, so I tried the all (almost) and felt that the Saturation I described above plus one crossover setup would be cool to have.
I have been quite focused on the diodes though:
4148, 1N60, Red+Yellow LED, 1N5817 and these two: the red red black striped at ~500-550 mV from Ericsson Dialog phone and the yellow green black too but at 220mV



I hope I will grow out of the diode 747 "control room" eventually and get deeper in to what does make a difference and in what ways.

[matopotato]

But if you want a Dist+ that does not look like the %^&*pit of a 747, consider other approaches beyond futzing around with diodes. 

Because of the weird, nonsensical nature of the auto-censor rules, I didn't read that as "cockpit", but rather as "shitpit".  My very tired brain thought, well, I suppose airplane toilets have to flush into something. 

Back on topic, a lot of people have made a lot of money off of "magic" diodes, but I agree, real magic often hides elsewhere, such as filtering.

Thanks, yes it is a learning curve, but a pleasant one. I have tried to explain to people why it is more fun to build your own stuff, but I can't properly put my finger on it...

[matopotato]

Here is the full schematic.
I don't have any good program, and not very versed in drawing, so there is bound to be some flaws and tons of improvements.
But it should show my plan.
There will be OUTPUT (vol), DISTORTION (Gain), DPDT for Rotary vs Crossover, Rotary x 8 slots which will engage SATURATION and SPDT for bass cut, the CROSSOVER path with a DPDT for high pass/low pass and SPDT for Symmetrical vs Asymmetrical clipping.
4 knobs, for switches and 1 rotary.
I also plan to cut out holes for 3mm LEDs so they will shine up the eyes of some artwork.
If I name if "Jack of all" (in honor of course), the Jack card juggling stuff might be the "art".
And yes, I am aware that it is quite overloaded, but I kind of like the idea of having many options. That is what I usually end up pursuing in most walks of my life. (And it is not always desirable)



Full credit to R.G Keen schematics of Dist + and Jack Orman for the selected mods. No part is my own original.

[Mark Hammer]

Is there some obvious way to add bass control while keeping the other mods?

Yes.  The .047uf cap on the ground leg, where the Gain pot is, sets the bass rolloff to 720hz when the Gain is maxed.  Lowering that rolloff to include more bass simply involves increasing the value of that cap.  A value of 0.1uf lowers that 720hz to around 340hz, and 0.22uf drops it down to just over 150hz.  Bear in mind that this is at maximum gain, and that rolloff point gets lower when the gain is at less than max.  So, for instance, even with a .047uf cap, when the Gain pot is at 100k (not the full 500k), the rolloff is around 40hz.

One strategy that some designs use, including the Proco Rat and several Tube Screamer derivatives, is to provide two ground legs, one that sets the gain for the entire guitar signal, and another that sets the gain for content above some particular frequency.  Let's say we add a resistor and cap in parallel with the 4k7 and .047.  Let's make them 12k and 0.1uf.   At max gain (i.e., Gain pot is at 0 ohms), the rolloff in that path will be 132hz; so reasonably flat across the entire guitar spectrum.  But because we've used 12k, that maximum gain will be 84x, instead of the 218x applied to stuff above 720hz.  So, you keep some bass, but still give more push to the mids and highs.  That secondary ground leg still conforms to the same rules as the main one; namely that as the gain pot resistance is reduced that rolloff moves downward.

[antonis]

I hope this makes my post a bit more clear? (@Antonis)

Clear enough..

P.S.
Of couse, you realize we're now talking for totally different R/C filters configuration..
(LPF Vs HPF with only the corner frequency common between them..)

[matopotato]

Is there some obvious way to add bass control while keeping the other mods?

Yes.  The .047uf cap on the ground leg, where the Gain pot is, sets the bass rolloff to 720hz when the Gain is maxed.  Lowering that rolloff to include more bass simply involves increasing the value of that cap.  A value of 0.1uf lowers that 720hz to around 340hz, and 0.22uf drops it down to just over 150hz.  Bear in mind that this is at maximum gain, and that rolloff point gets lower when the gain is at less than max.  So, for instance, even with a .047uf cap, when the Gain pot is at 100k (not the full 500k), the rolloff is around 40hz.

One strategy that some designs use, including the Proco Rat and several Tube Screamer derivatives, is to provide two ground legs, one that sets the gain for the entire guitar signal, and another that sets the gain for content above some particular frequency.  Let's say we add a resistor and cap in parallel with the 4k7 and .047.  Let's make them 12k and 0.1uf.   At max gain (i.e., Gain pot is at 0 ohms), the rolloff in that path will be 132hz; so reasonably flat across the entire guitar spectrum.  But because we've used 12k, that maximum gain will be 84x, instead of the 218x applied to stuff above 720hz.  So, you keep some bass, but still give more push to the mids and highs.  That secondary ground leg still conforms to the same rules as the main one; namely that as the gain pot resistance is reduced that rolloff moves downward.

Again, many thanks!
Gotta try. Project delayed, but I feel I will learn from it and hopefully understand a bit more what you meant by not focusing only on diodes.
Two ground legs, does that mean 2 parallel ways to the gain pot, so a new path from IC to Gain pot?

Any thoughts on the "series" vs "parallel" over by the Saturation?

[matopotato]

I hope this makes my post a bit more clear? (@Antonis)

Clear enough..

P.S.
Of couse, you realize we're now talking for totally different R/C filters configuration..
(LPF Vs HPF with only the corner frequency common between them..)

Thanks
Well, I hope so...
The alternatives over by the gain you mean?
There is also sort of LPF/HPF over at the "Crossover" part

[antonis]

Well, I hope so...
The alternatives over by the gain you mean?
There is also sort of LPF/HPF over at the "Crossover" part

OK, let's make more simple..



(A): 270Ω is in parallel with 47nF (and their equivalent resistance is in parallel with pot X)
(B): 270Ω is in seires with 47nF (and their equivalent resistance is in parallel with pot X)
The higher the signal frequency the lower the 47nF impedance (capacitive reactane) hence the lower their parallel equivalent resistance (A) and the lower their series equivalent resistance (B)..

You can see that by increasing 47nF cap value (or frequency or both), we can effectivelly by-pass pot part X in case A where set a minimum of 270Ω in parallel with X in case B..

e.g. At 12.5kHz, there is a resistance of 135Ω in parallel with X (case A) and a resistance of 540Ω in parallel with X (case B)..
(no need for transfer functions mess, I hope..)

[matopotato]

Well, I hope so...
The alternatives over by the gain you mean?
There is also sort of LPF/HPF over at the "Crossover" part

OK, let's make more simple..



(A): 270Ω is in parallel with 47nF (and their equivalent resistance is in parallel with pot X)
(B): 270Ω is in seires with 47nF (and their equivalent resistance is in parallel with pot X)
The higher the signal frequency the lower the 47nF impedance (capacitive reactane) hence the lower their parallel equivalent resistance (A) and the lower their series equivalent resistance (B)..

You can see that by increasing 47nF cap value (or frequency or both), we can effectivelly by-pass pot part X in case A where set a minimum of 270Ω in parallel with X in case B..

e.g. At 12.5kHz, there is a resistance of 135Ω in parallel with X (case A) and a resistance of 540Ω in parallel with X (case B)..
(no need for transfer functions mess, I hope..)

Thanks,
I will read through a few more times, but I feel I am nearing some current level of understanding.
Seems I am pushing to a point where the pot won't matter much and it all boils down to impedance.
So I will try to get a few values that have some effect probably to a lesser extent than what I have now "just to make it feel a lot" and where the pot still matters as I kind of like the effect it makes. The crossover pot feels similar to when I was biasing transistors, although this is probably way off electronically. I mean there is a sweet spot rather than a "more vs less" scenario as usually the case with the other pots. IMHO.

[antonis]

Seems I am pushing to a point where the pot won't matter much and it all boils down to impedance.

Pot DOES matter but in conjunction with R/C wiring..

To have an visual sense, a graphic depiction of both A & B transfer functions is needed..
(maybe a good willing guy who loves maths or sims does bother to utilize it..)

[matopotato]

Seems I am pushing to a point where the pot won't matter much and it all boils down to impedance.

Pot DOES matter but in conjunction with R/C wiring..

To have an visual sense, a graphic depiction of both A & B transfer functions is needed..
(maybe a good willing guy who loves maths or sims does bother to utilize it..)

Ok, good to know.
I am not sure how to make such graphs though. I also do not have more than a DMM...
I was thinking more if I push the R and C even more, then the effect of the variable R becomes less.
Anyway, I will experiment a bit more with the values in A and B to see what is maybe a bit more optimal and not too extreme, so that it contributes instead of becoming a sort of spectacle.
I will also try out the suggestions for the Gain - IC part to see how Bass is affected. Maybe another SPDT. I mean if you already are flying a 747, then one tiny extra switch will not ruin the party...