Tube Screamer mix knob.. once again!

[xorophone]

Hello! I'm currently planning a tube screamer build using the tonepad layout (http://tonepad.com/project.asp?id=1). I want to add a dry/wet mix/blend pot and there's already a few threads about that here, but I just can't seem to find a clear solution. I found this article: http://www.muzique.com/news/tube-screamer-mix-control/. I can't really wrap my head around it though. Wouldn't the mix-pot just reduce the amount of signal that reaches the diodes, and therefore reduce the amount of clipping? I guess I'm wrong about that, so would you recommend this solution? Will the clean and distorted signals be gain matched? And would the image below be the correct way to implement it in the TS circuit?



I've also looked at a few other alternatives and one solution would be to implement a Buff 'N Blend circuit (https://www.aronnelson.com/gallery/main.php/v/BarcodeVeroLayouts/album120/buffnblendPCB.gif.html), but the problem is that the dry and wet signals won't be gain matched (and I also don't know how the phase relation between the signals will be.)

Since a lot of people use the tube screamer to boost the input signal of an amplifier, I think it's very important that the two signals are at least somewhat gain matched. Do you know of any clear schematics or layouts that show how to do this?

Thank you!

[marcelomd]

Hi,

In a non-inverting opamp with diodes on the feedback loop, the signal is amplified and, eventually, clipped/limited to gain=1 every time the output goes past the diodes forward voltage threshold. Jack's mix knob works by increasing this threshold, so more of the clean signal passes through without distorting. I would call it "increasing the headroom" more than "mix". Whatever passes the threshold is amplified by the new resistor ratio, so it distorts/compresses less. Still not what I would call "mix". ToMAHto, toMAYto and all that.

I have a Darkglass B3K bass overdrive that has the clean side fixed at unity gain. Works well enough, so I wouldn't worry about matching gains. If you reeeeeally want to, you could use a simple booster for the clean side and control both sides with a dual gang pot.

The TS does not invert phase, so a simple mixer will do the trick.

Voodoo Labs Sparkle drive is a TS with a clean blend IIRC, so you could look into it. I don't know how Doodoo Labs Sprinkle Drive is related, but it also has a clean blend that you can study.

[tekhna]

Hi,

In a non-inverting opamp with diodes on the feedback loop, the signal is amplified and, eventually, clipped/limited to gain=1 every time the output goes past the diodes forward voltage threshold. Jack's mix knob works by increasing this threshold, so more of the clean signal passes through without distorting. I would call it "increasing the headroom" more than "mix". Whatever passes the threshold is amplified by the new resistor ratio, so it distorts/compresses less. Still not what I would call "mix". ToMAHto, toMAYto and all that.

I have a Darkglass B3K bass overdrive that has the clean side fixed at unity gain. Works well enough, so I wouldn't worry about matching gains. If you reeeeeally want to, you could use a simple booster for the clean side and control both sides with a dual gang pot.

The TS does not invert phase, so a simple mixer will do the trick.

Voodoo Labs Sparkle drive is a TS with a clean blend IIRC, so you could look into it. I don't know how Doodoo Labs Sprinkle Drive is related, but it also has a clean blend that you can study.

Yeah, the Voodoo Labs Sparkle Drive is what you want to look at. I'd love it if someone could actually pull that part of the schematic out for general use--I'm not skilled enough, unfortunately. There are some tagboard clean blends available too.

[marcelomd]

EDIT: Duh...

[marcelomd]

Hi,
I had 5 min to spare so I sketched something. Nothing special and you can probably improve it. I did not put any values. GAIN A and GAIN B can be a dual gang potentiometer.


EDIT1: Messed the tone sction. Should be ok now.

[ElectricDruid]

Yeah, the Voodoo Labs Sparkle Drive is what you want to look at. I'd love it if someone could actually pull that part of the schematic out for general use--I'm not skilled enough, unfortunately. There are some tagboard clean blends available too.

The Voodoo Labs Sparkle Drive clean blend is a basic inverting op-amp mixer circuit. The two inputs (clean and overdrive) are given different input resistor values because the clean channel is boosted, but it doesn't have to be. A dual pot is used so that as you mix more of one channel, you mix less of the other. You could do something very similar with a single gang pot if you used the typical "pan/balance" arrangement instead.

HTH,
Tom

[marcelomd]

Voodoo Labs Sparkle drive is a TS with a clean blend IIRC, so you could look into it. I don't know how Doodoo Labs Sprinkle Drive is related, but it also has a clean blend that you can study.

Same idea is discussed here. Both schematics available: https://www.diystompboxes.com/smfforum/index.php?topic=81705.0

[xorophone]

Thanks to everyone for the great replies! It's definitely starting to clear up for me now.

Hi,
I had 5 min to spare so I sketched something. Nothing special and you can probably improve it. I did not put any values. GAIN A and GAIN B can be a dual gang potentiometer.

Wow, thank you so much for this! I've been looking at the sparkle and sprinkle drive schematics and especially on the sprinkle drive there are some differences from the tonepad layout. If I for example build the sparkle drive, will there be any audible difference from the tonepad TS layout? I think I'm going to add a switch to be able to flick between the 808 and TS9 values, but if the circuit doesn't sound like the original in the first place, that feels like it's pretty much pointless.

I found some slight variations from the tonepad schematic on your schematic. If I choose to follow your schematic, will I be able to completely replace the distortion part of the circuit with the tonepad layout or did you customize it to make it work with the mix pot? I guess I'd just "tap out" after the 1k resistor at the output of IC1b at the tonepad layout, right?

Also, what's the purpose of U1_D on your circuit? And any ideas which values I should choose for R8, R9, R15 and the mix knob on your circuit? And last but not least, does anyone have any suggestions for how the circuit could be improved? (considering it was designed in 5 minutes  )

Sorry for all the questions, but once again a big thanks to everyone for the great explanations and replies.

[marcelomd]

Hi,

Thanks! I'm just paying it back after learning so much here.

I just got the first TS schematic I could find and built from there, no changes other than the clean booster and mixer. It is just a starting point, anyway. The only difference I see between the Tonepad schematic and the one I used is an extra diode. Some people like it, some don't. My suggestion is to assemble the circuit on a breadboard and experiment a bit.

The Sparkle Drive probably sounds a bit different than the Sprinkle Drive. And both different than this quick sketch, but I'm sure they are in the same ballpark. I wouldn't worry about getting it 100% equal to the originals. Otherwise you wouldn't be trying to mod the circuit in the first place, right? That's is the fun of DIY for me =)

Right, R14 in my schematic is equivalent to the 1k after IC1b, but without the decoupling 1u cap (because both sides are referenced to VA). You may need a bigger value there, so it won't clip the input of the mixer.

U1D is the "mixer". In quotes because the mixing is done passively by R14, R15 and the potentiometer. This opamp is just a high impedance input (you could use a JFET, but this way you can use two duals or one quad opamp). The advantage is the low parts count. The disadvantage is that there will be crosstalk between sides. Which is not a problem in a dirtbox (for me).

R8 and R15. Same as the other side (4k7 and 51k) if you want 100% gain matching. Tweak to taste.

Improvements, suggestions, things I would like someone to check:
- Use some filters on the clean side;
- Tonestack for the clean side;
- Fixed gain for clean side;
- Place the level control in the feedback loop of the mixer;
- If I reference the level pot to VA, can I cut C9, C6, R15?
- Do I need decoupling caps before the mixer pot?
- Mixer options:
  - Passive (as is);
  - Single pot to inverting input;
  - Equal power panner;
  - Level control for each side;
- Any of the standard mods (clipping diodes, mids controls, tonestack, etc.);

Have fun =)

[xorophone]

The only difference I see between the Tonepad schematic and the one I used is an extra diode. Some people like it, some don't. My suggestion is to assemble the circuit on a breadboard and experiment a bit.

Right, R14 in my schematic is equivalent to the 1k after IC1b, but without the decoupling 1u cap (because both sides are referenced to VA). You may need a bigger value there, so it won't clip the input of the mixer.

I was mainly referring to this decoupling capacitor. I'll exclude it if it's not needed, but will it hurt in any way if it's left there?

U1D is the "mixer". In quotes because the mixing is done passively by R14, R15 and the potentiometer. This opamp is just a high impedance input (you could use a JFET, but this way you can use two duals or one quad opamp). The advantage is the low parts count. The disadvantage is that there will be crosstalk between sides. Which is not a problem in a dirtbox (for me).

Does crosstalk mean that the wet signal might spill into the clean signal? If so, wouldn't this happen even at the amplifier stages (U1_A and U1_B on your schematic)? And could it be eliminated simply by using two separate 4558-ICs for the wet and dry signals respectively? Maybe I could use one 4558 for U1_B and U1_C and then another one for U1_A and U1_D, because dry signal spilling into the distorted signal isn't a problem, but if it's the other way around it might become a problem (at least if you decide to only let dry signal through and use the pedal as a clean boost.)

R8 and R15. Same as the other side (4k7 and 51k) if you want 100% gain matching. Tweak to taste.

I guess I'll have to do some breadboarding. My setup is very inconvenient so I was hoping I wouldn't have to do it, but those expectations are of course unrealistic. I don't have access to my equipment in a few days now though, but I'll make sure to try it out as soon as possible and post the results here. It would be nice to be able to post a working schematic, since I've had some trouble finding it myself. Oh and by the way, I don't have any 4558 chips at home. Could I try it with for example a tl072 for the time being and expect similar results? At first I just want to make sure the gain is matched on both sides, and the ratio should be the same even if I use another chip, right?

[marcelomd]

I was mainly referring to this decoupling capacitor. I'll exclude it if it's not needed, but will it hurt in any way if it's left there?

If you leave this cap, you must put one on the clean side and a resistor (1M, maybe) to VA after the mix pot.

Does crosstalk mean that the wet signal might spill into the clean signal? If so, wouldn't this happen even at the amplifier stages (U1_A and U1_B on your schematic)? And could it be eliminated simply by using two separate 4558-ICs for the wet and dry signals respectively? Maybe I could use one 4558 for U1_B and U1_C and then another one for U1_A and U1_D, because dry signal spilling into the distorted signal isn't a problem, but if it's the other way around it might become a problem (at least if you decide to only let dry signal through and use the pedal as a clean boost.)

Maybe crosstalk is not the right term. But yes, the drive side might leak into the mix even if it is set 100% clean. It happens simply because there are just a few resistors between signal sources. A 100k mix pot might be sufficient isolation, but you will have to test. If it is not ok, we can work something else.

Oh and by the way, I don't have any 4558 chips at home. Could I try it with for example a tl072 for the time being and expect similar results? At first I just want to make sure the gain is matched on both sides, and the ratio should be the same even if I use another chip, right?

Any opamp will "work" and most will sound "good enough". TL07x will probably sound close but not 100% equal to the original. Use sockets and feel free to mix and match.

What do you mean by "match gains on both sides". Same level of amplification?

[xorophone]

If you leave this cap, you must put one on the clean side and a resistor (1M, maybe) to VA after the mix pot.

Oh ok. I'll try both, because it would be nice if the dry part of the circuit was 100% compatible with the tonepad layout and could be separated on the schematic. I think it'll help more people in the future if it comes across as a mod for the tonepad tube screamer.

Maybe crosstalk is not the right term. But yes, the drive side might leak into the mix even if it is set 100% clean. It happens simply because there are just a few resistors between signal sources. A 100k mix pot might be sufficient isolation, but you will have to test. If it is not ok, we can work something else.

Thank you for the explanation! I'll play around with it.

Any opamp will "work" and most will sound "good enough". TL07x will probably sound close but not 100% equal to the original. Use sockets and feel free to mix and match.

What do you mean by "match gains on both sides". Same level of amplification?

Yes exactly. I basically just want to make sure I get a good perceived volume ratio between both sides. I'll use 4558 op-amps on the final product, but I want to play around with the circuit before my order arrives.

[ElectricDruid]

The Sparkle Drive probably sounds a bit different than the Sprinkle Drive.

I'd expect they sound at least a bit different. The two schematics that Moosapotamus linked in the other thread:

http://home.kpn.nl/chrisdus/download/VLSD.jpg
https://www.diystompboxes.com/pedals/rstrand/doodoo.gif

Sparkle drive is a very clear TS-Screamer-alike overdrive with clipping diodes in the feedback path of an inverting op-amp. The Sparkle drive is a quite different arrangement, with the clipping diodes to ground after a non-inverting amp, and followed by a variable lowpass filer stage, not the Screamer's tone control. I'd say the Sparkle drive is more like a Proco RAT with a clean channel mixed in than anything. So no, not terribly similar, at least technically.

[marcelomd]

Oh ok. I'll try both, because it would be nice if the dry part of the circuit was 100% compatible with the tonepad layout and could be separated on the schematic. I think it'll help more people in the future if it comes across as a mod for the tonepad tube screamer.

Sure. You can always build something like the ROG's splitter blend, or use a pedal like Boss LS2 Line selector in mix mode.

Yes exactly. I basically just want to make sure I get a good perceived volume ratio between both sides. I'll use 4558 op-amps on the final product, but I want to play around with the circuit before my order arrives.

The way I understand it, after the signal level reaches the diodes' threshold, any increase in gain results only in change of distortion character, not volume (the level is limited by the diodes, right?). Having a slight fixed boost on the clean channel might be enough bring it to a nice level. Then you set the "perceived ratio" with the mix pot.

---

Sparkle drive is a very clear TS-Screamer-alike overdrive with clipping diodes in the feedback path of an inverting op-amp. The Sparkle drive is a quite different arrangement, with the clipping diodes to ground after a non-inverting amp, and followed by a variable lowpass filer stage, not the Screamer's tone control. I'd say the Sparkle drive is more like a Proco RAT with a clean channel mixed in than anything. So no, not terribly similar, at least technically.

You are 100% right. To be completely honest, I have never listened to them, so I went by the similarity of names... mea culpa... hehehe. The point was to illustrate ways to do a simple clean blend, so both should be useful anyway =)

[Groovenut]

You might want to have a look at the Maxon OD820. It's essentially a TS with a clean blend using a dual pot and the Klon style mixer.

[xorophone]

You might want to have a look at the Maxon OD820. It's essentially a TS with a clean blend using a dual pot and the Klon style mixer.

Thank you for the suggestion! I'll take a look at the schematic.