GuitarPCB ParaMix

[JuneauMike]

Any one built this pedal. I've got three PCBs to play with and wanted to try my hand at modifying it to a three channel parallel signal blender (two effects, one clean). I think the first one will be stock. But the second will hopefully be modified with some offboard components.

Curious as to what I'm getting into with this one.

 

[zippofan]

Looks like a cool circuit, though I haven't built it, or many GuitarPCB circuits. I have used their 3PDT breakout boards though.

 

[Digital Larry]

What's the actual question? To add more FX channels duplicate the circuit sections outlined in yellow. Attachment points to the existing circuit circled in blue.

https://ibb.co/BzHvxkG

 

[JuneauMike]

What's the actual question? To add more FX channels duplicate the circuit sections outlined in yellow. Attachment points to the existing circuit circled in blue.

https://ibb.co/BzHvxkG

No, the question is what recommendations for modifying the PCB to accomplish this. My back of the envelope thought was to just break the connection at C2 and R4.

I'd jumper a clean channel from the junction of C2 and the first IC subdivision, adding an offboard circuit that duplicates C7-R9. That would drop in at the junction between R9 and R10.

R4 would become an output jack to capacitor and the tail end of the second effects circuit.

The second effects circuit send would also be offboard and would be at the junction of C2 and the first IC subdivision. (This uses a TL074). Roughly like below.





It gets a little funky at the second IC subdivision since I need to get a phase switch and resistor between pin 3 and the socket. The other thing I'm really worried about is gain. Each of the effects return channels uses a portion of the IC as a recovery stage. The clean channel only has the initial buffer at the start. I breadboarded the first section of this where the signal splits three ways and it seems to give a strong enough signal on all three, but that's without all the rest of the circuit goodness bearing down on it.

Wondering if there's a way to squeeze a little bit gain out of the clean channel, and/or reduce the incoming gain from the two effects signals to give a good balance to begin mixing the signals. This is essentially what I'm working towards.

 

[Digital Larry]

You need to put C3 and R10 between the clean buffer op amp output and the (-) input of IC1B. (I see you already did that). You can adjust the gain of that by adjusting R10. The gain is -R11/R10. R10 lower, gain higher.

https://www.electronics-tutorials.ws/opamp/opamp_2.html

 

[JuneauMike]

You need to put C3 and R10 between the clean buffer op amp output and the (-) input of IC1B. (I see you already did that). You can adjust the gain of that by adjusting R10. The gain is -R11/R10. R10 lower, gain higher.

I'm really out of my depth here, but would it make much sense to use trim pots at R9, R10 and the R from the new incoming clean channel? Wouldn't that let me adjust the output of each channel separately before they are summed?

And I see your pic, it makes sense and is probably a cleaner way to skin the cat. Just build an entire send and return channel on a Vero board. I'm assuming it would add a TL071 to the circuit and I'd have to tap power from the power rail. But yea, that would get it there.

 

[Digital Larry]

I'm really out of my depth here, but would it make much sense to use trim pots at R9, R10 and the R from the new incoming clean channel? Wouldn't that let me adjust the output of each channel separately before they are summed?

You already have level pots on each channel. I'd just tack in various resistors there (probably in the 1k to 10k range) to set the maximum gain. You can always turn it down from there. Of course you can add trimpots wherever you like but I don't think it's warranted here.

And I see your pic, it makes sense and is probably a cleaner way to skin the cat. Just build an entire send and return channel on a Vero board. I'm assuming it would add a TL071 to the circuit and I'd have to tap power from the power rail. But yea, that would get it there

Yeah, I don't know about how to cut up the existing board to accomplish your goal. Sounds like real work to sketch that out.

 

[JuneauMike]

Yeah, I don't know about how to cut up the existing board to accomplish your goal. Sounds like real work to sketch that out.

Well, it's largely been done already, right? If I take your suggestion then it's just a matter of flying a wire out from the input buffer and then fly another wire in between R10 and R11. In between those two wires is a vero board with the second effects channel on it, both send and return. The clean channel stays the same. That's kinda it.

The advantage of your idea is that I sidestep having to shoehorn a phase switch and resistor between IC pin 3 and the rest of the circuit. Hmm.

And at that point there's a recovery stage on each of the channels ( by adding a TL071 to the vero board). And I could add a trim pot to each of the Negative Feedback resistors in the inverted Opamp stage. (He's doing that to the main channel in the original schematic). That might give just a little tweakability, which is all I'm looking for from gain. (And to state the obvious, "level" is not "gain.")

Thoughts?

 

[Digital Larry]

Well, it's largely been done already, right?

I'll take your word for it, I'm just focusing on the schematic. If you can make the wiring match the schematic you should be OK.

And to state the obvious, "level" is not "gain."

Well, not so fast. "Level" is certainly "gain" although it starts at 0 dB and goes down from there. So if you want a 20 dB gain boost (factor of 10 voltage gain IIRC) then you make your Rf/Ri = 10. Also note that the load impedance on the wiper of the level pot is going to be Ri. So if you have a 500K level pot, it's going to drop off really fast if the load is 1K. The high pass corner set by the coupling caps will also change as Ri changes. To keep the fc the same for Ri =10k as Ri = 1k then you'd multiply the coupling cap by that same ratio (to keep R x C constant).

All of these things interact and it's not possible for me to give you a really simple way to understand it that doesn't just cut to the heart of the matter.

I'd start by figuring out the max gain you want in each leg to determine the Ri. From there we can calculate the appropriate values for pots and coupling caps.

Also I don't see a need for phase reverse switches in all 3 legs. I'm somewhat questioning the need at all, depending on what kinds of effects you plan on using there. What kinds of effects are you planning on using?

 

[JuneauMike]

I'll take your word for it, I'm just focusing on the schematic. If you can make the wiring match the schematic you should be OK.


Well, not so fast. "Level" is certainly "gain" although it starts at 0 dB and goes down from there. So if you want a 20 dB gain boost (factor of 10 voltage gain IIRC) then you make your Rf/Ri = 10. Also note that the load impedance on the wiper of the level pot is going to be Ri. So if you have a 500K level pot, it's going to drop off really fast if the load is 1K. The high pass corner set by the coupling caps will also change as Ri changes. To keep the fc the same for Ri =10k as Ri = 1k then you'd multiply the coupling cap by that same ratio (to keep R x C constant).

All of these things interact and it's not possible for me to give you a really simple way to understand it that doesn't just cut to the heart of the matter.

I'd start by figuring out the max gain you want in each leg to determine the Ri. From there we can calculate the appropriate values for pots and coupling caps.

Also I don't see a need for phase reverse switches in all 3 legs. I'm somewhat questioning the need at all, depending on what kinds of effects you plan on using there. What kinds of effects are you planning on using?

Thanks. Yeah, the reverse phase option would only be on the effects channels, not the clean channel. Two SPDT switches only. And the reason would be to have that option if a random pedal has phase issues.

Respectfully, gain magnifies the potential power of a signal output (and unfortunately line noise as well), while level only attenuates what is already there. The furthest travel of a volume pedal or pot is as loud as a signal can get (assuming we're talking in terms of dB, or loudness in general). Everything below 100 percent of that pot's rotation is being attenuated to ground. If the pedal makes the signal louder, then that's the gain structure in the circuit doing it. The level pot is only compensating for that on the output, not introducing any new potential to the equation. But maybe we're getting tripped up by the semantics.

I think in this case, the baseline gain would come from the clean channel. I guess I could start the math there if its important. But as far as gain adjustments I suspect I'm talking about here, they would be miniscule and really for just correcting the input to the level pots, if necessary. Or more likely to roughly calibrate the output of the three signals and then not touch.

As I said in the OP, I bought three of these PCBs and have already socketed one of them completely. So I could see a lot of fun here playing with the math you are suggesting. Thanks, I really appreciate the input. And I like your idea better than mine. I think my idea is a product of staring at a problem for too long.
.

 

[Digital Larry]

Thanks. Yeah, the reverse phase option would only be on the effects channels, not the clean channel. Two SPDT switches only. And the reason would be to have that option if a random pedal has phase issues.

Respectfully, gain magnifies the potential power of a signal output (and unfortunately line noise as well), while level only attenuates what is already there.
.

It's fine. I do agree with you, but over time I have merely come to accept that "gain of -20 dB (0.1)" is a valid concept. Also, thanks for being respectful. I try but don't always succeed. I'll follow this thread but feel free to @mention me if you want me to comment on something and I'm not picking up on it.

 

[JuneauMike]

It's fine. I do agree with you, but over time I have merely come to accept that "gain of -20 dB (0.1)" is a valid concept. Also, thanks for being respectful. I try but don't always succeed. I'll follow this thread but feel free to @mention me if you want me to comment on something and I'm not picking up on it.

Thanks a lot for your input. I've been stuck on how to approach this for a while and I think it can be much more simple than I've made it. Your observation has definitely inspired me to look at it from another way.

The big thing I was worried about was splitting the buffered signal three ways, but after breadboarding it I think it will be fine. The "gain" trimmer pots is maybe a comfort thing for me in case things aren't fine. But I'm really looking at them to just tune the individual circuits to each other as a baseline.

But I'm not that versed on this stuff to begin with, and math terrifies me. I appreciate the input.