Design help. Mixer/Splitter.

[Buffalo Tom]

Here is a schematic for a Boss CE-1 mixer I have designed. The idea is the keep the CE-1 pedal stock but make it work nicely with guitar. Im using the wet output on the CE-1. I don't like the preamp distortion.

The CE-1 is a AC powered device so I wanna keep the loop isolated. I have chosen a 5K pot for the send attenuation, to keep the output impedance low. The CE-1 input Z is only 50K. Any drawbacks with this?

The MET-09 transformer is something I found in a Suhr mixer so quality should be good, its 10K:10K spec. How will it behave with seeing source impedance from my pot and 50K load from the CE-1 input. Not optimal?

The return section impedance is 10K so I believe MET-09 will work nicely here. The CE-1 output Z is 1K.

Is there anything more in form of resistors or capacitors that I should add? Or values that I picked that can be better? Will the Kill Dry switch "pop"? Let me know what I can do to improve the overall design of this schematic.

Thanks

[antonis]

Will the Kill Dry switch "pop"?

In theory, no..
C6 is permanently grounded (via 5k pot & R12) and C9 has no DC voltage difference between its legs..

[Buffalo Tom]

Thanks Antonis! So the rest look good?

[fryingpan]

Why a transformer? It seems overkill. An attenuator is fine, as long as the resistances in play are kept small. The OPA2134 has good drive capabilities, so you can use smallish resistors.

[R.G.]

Why a transformer? It seems overkill.

I suspect that he's trying to prevent hum pre-emptively. Using an AC powered device in the pedal chain to drive an AC powered amplifier could lead to hum because of AC mains leakage in the amp or the device. Same issue as needing isolation when driving two amplifiers at the same time. It's probably more about hum than circuit performance.

[Buffalo Tom]

Why a transformer? It seems overkill.

I suspect that he's trying to prevent hum pre-emptively. Using an AC powered device in the pedal chain to drive an AC powered amplifier could lead to hum because of AC mains leakage in the amp or the device. Same issue as needing isolation when driving two amplifiers at the same time. It's probably more about hum than circuit performance.

That's correct—without the transformers, there was a ground loop. I tested this setup using an RJM Mixer, a boost pedal, and an EHX Signal Pad to achieve the functions I needed. However, without isolation transformers, the setup produced a significant amount of hum. My plan now is to build a single unit to replace those pedals.

[antonis]

without the transformers, there was a ground loop. I tested this setup using an RJM Mixer, a boost pedal, and an EHX Signal Pad to achieve the functions I needed. However, without isolation transformers, the setup produced a significant amount of hum.

That could be remedied via a single series resistor (10R or so) between DC(audio) and AC(chassis) grounds..
(and 100nF cap in parallel for RF bypass..)

[fryingpan]

without the transformers, there was a ground loop. I tested this setup using an RJM Mixer, a boost pedal, and an EHX Signal Pad to achieve the functions I needed. However, without isolation transformers, the setup produced a significant amount of hum.

That could be remedied via a single series resistor (10R or so) between DC(audio) and AC(chassis) grounds..
(and 100nF cap in parallel for RF bypass..)

The so called "ground lift".

[WangoFett]

Will the Kill Dry switch "pop"?

In theory, no..
C6 is permanently grounded (via 5k pot & R12) and C9 has no DC voltage difference between its legs..

What happens when the Kill Dry switch feeds that ground potential to the input of the following op-amp?

[R.G.]

A few comments about ground lifts.
> Many AC powered things don't have a third-wire safety ground. They rely on the "neutral" AC wire being connected to ground-rod-ground in the breaker box, so they cannot have a true earth ground reference because of the V = I * R in the neutral wire.
> All AC power supplies leak some of the high voltage AC mains currents. Some more, some less, all of them leak. This is why medical electronics has dramatically stringent requirements on wall-socket power wiring and on tested leakage limits and extra isolation/shielding.
> Two-wire guitar amps had the "death cap" to connect signal ground to the incoming neutral. This helped with RF buzz, but if the "line reverse" switch is in the wrong position, the cap itself, even if it's not broken, leaks AC directly into signal ground.
> Two guitar amps used in tandem, with signal ground connected between the two inputs by cords, show off their difference in leakage by running leakage currents between them. How much hum you get depends on the leakages and the circuit details inside the amps. An AC powered effect can do the same. So can an AC powered PA mixer in a run-down bar. And we all know by now that grabbing a mike stand can shock you. That's leakage currents.
> Ground lifts were invented when people noticed that mixing two-wire and three-wire equipment could still hum. Detaching the safety ground wire in the properly grounded equipment could usually float the signal grounds and reduce the hum. The evolution of the 10 ohm lift was fairly long, but the safer ones parallel the 10 ohm resistor (or similar) with a 25A or more rated diode bridge to allow fault currents to be shunted to safety ground while not holding up enough voltage to kill you. Usually. Whether they work well depends on how much leakage is coming out of each piece of equipment. 10 ohms is a compromise between zero ohms for safety and a high resistance for not letting leakage currents flow.

If this sounds complicated, it is. The right answer is known - make each piece of power equipment be three-wire grounded and plug them into the same breaker circuit. This connects all of the safety grounds together  in the wall wiring, and does not raise them all above ground-rod voltage with I * R currents.

If you can't do that (or the manufacturer would not pay for it when they made the equipment) get "double insulated" stuff. It's guaranteed not to let enough leak to kill you. It _probably_ has had enough attention paid to leakage to not hum badly.

If you can't do that use isolating transformers so the guitar=>FX signal ground doesn't have to be tied to the possibly leaky amplifier chassis ground.

If there is just nothing else to be done, use a ground lift and pray. Hum isn't the worst thing that can happen to you. Even better, if you play gigs in random places, put in some prep time. Get one of those outlet-tester things at a home improvement store and test the AC outlet you're going to plug your amp and other stuff into. The outlet itself can sometimes be wired wrong. If so and you then use a ground lift... oops! Test and vet your setup at home, getting rid of both hum and shock hazards by design and testing, not by throwing in a ground lift at the last minute and just leaving it there forever.

Kudos to the original poster. Good thinking!

[antonis]

What happens when the Kill Dry switch feeds that ground potential to the input of the following op-amp?

It happens to off-set following op-amp's DC output by VB x [(47k+5k+620)/(47k+47k+5k+620)]..
(practically permanently topped..)

@Buffalo Tom: Did you catch what WangoFett's eagle eye did..??

[Buffalo Tom]

What happens when the Kill Dry switch feeds that ground potential to the input of the following op-amp?

It happens to off-set following op-amp's DC output by VB x [47k / (47k+5k+620)]..
(practically permanently topped..)

@Buffalo Tom: Did you catch what WagoFett's eagle eye did..??  

No, I don't quite understand. In Kill Dry mode, R6 changes its role and becomes a 47k pull-down resistor to ground (not VB). Would that be an issue?

[antonis]

No, I don't quite understand. In Kill Dry mode, R6 changes its role and becomes a 47k pull-down resistor to ground (not VB). Would that be an issue?

That wouldn't be the issue..

The "other" (R6 connected to inverting input) should be a major issue..
You see, there is no DC blocking capacitor between R6 and op-amp's inverting input (which capacitor should serve for rolling off DC gain to unity) so op-amp's output is hitting the positive rail struggling to bring inverting input to VB..

More simply, R12 and R6+pot+R12(the one going to GND) form a DC voltage divider with its middle forced at VB..
What should the divider upper DC voltage (op-amp's output) be..??
(see below..)



P.S.
I let you judge the right place for putting that DC gain isolation cap..

[amptramp]

There are linear optical and RF frequency isolation circuits that are often cheaper than an audio transformer.  It can be worthwhile to investigate those.  Even with a transformer, you have capacitance between the input and output windings that can pass a considerable amount of hum.

[Buffalo Tom]

P.S.
I let you judge the right place for putting that DC gain isolation cap..

Thanks, Antonis! I've added a 1µF DC gain isolation capacitor (C12). This should be the correct placement, right? Im not sure... feels strange to have one leg of the cap floating... Also, a big thanks to @WangoFett for catching this design oversight!

VERSION 2

[Buffalo Tom]

Should I do the Kill Dry switch like this instead?

[antonis]

Version 3 is good enough..

Do you mind if I ask the reason for choosing 1μF for C12 (and also making C9 of equal value)..??
(other than you might been confused about C12 right polarity so making it non-polar should get you out of puzzlement - just joking..  )

Anyway, even with new caps values, HPFs corner frequencies still remain inaudible so no big deal..

[Buffalo Tom]

Version 3 is good enough..

Do you mind if I ask the reason for choosing 1μF for C12 (and also making C9 of equal value)..??
(other than you might been confused about C12 right polarity so making it non-polar should get you out of puzzlement - just joking..  )

Anyway, even with new caps values, HPFs corner frequencies still remain inaudible so no big deal..

Regarding the series capacitors on op-amp outputs, I'm confident about their orientation  As you mentioned, 1µF should be sufficient here since it's feeding another op-amp further down in the circuit. There's no need for a larger value in this case, right? That said, I usually go with 22µF after op-amps when they're closer to the output. The reason I like to match the capacitor values is mostly due to my OCD—it just feels better when they're the same.

Now, about the Kill Dry switch... For simplicity, I've decided to go with version 3, but I have one more question.

The original idea was inspired by a Suhr mixer, as you can see here. It uses an SPDT switch with a 49.8kΩ resistor to ground, which is very similar to what I was trying to implement. However, I suspect the reason it works in the Suhr mixer but not in my version is that the Suhr mixer uses a dual supply, while I'm working with a single supply. Is that correct?