Bufferz 'a poppin'

[SamJenkinsRides]

I built a Quaverato tremolo pedal from a kit made by Zepplin Design Labs. The build went great, and I love the pedal. But when I use it with my 8 channel loop controller (Joyo PXL Live), it reacts differently depending on which pedals are in loops before and after it.

If the Quaverato is placed anywhere in the guitar signal, there is a very noticeable and very loud jump in volume when other active (on) pedals are introduced (via loop controller) into the signal path along with the Quaverato. Take the Quaverato back out of the circuit, the volume returns to normal. Take all other active pedals out of the circuit besides the Quaverato, and the volume returns to normal.

I purchased a Walrus Audio Monument, and it causes/reveals the exact same issues as the Quaverato.
No ther pedals in my 50+ pedal arsenal cause this issue.

The only way I found that eliminates the volume jump (more specifically, permanently engages the jump) is to connect an (off, but powered) Boss pedal directly to the Quaverato's output. I thought this was a buffer/impedance issue, so I placed an inexpensive IC opamp buffer to the Quaveratos output. That permanently engaged the volume jump, but caused some of the switches to start loudly popping on the PXL. SO then I figured I needed a discreet buffer circuit like the ones used in the Boss pedal that worked so well at the Quaveratos output. So with some diagnostic help from the fine folks on this forum, I was able to complete a working discrete buffer. But alas, the popping is still there (albeit the volume jump is permanently engaged).

I went back and forth with the Zepplin Designs guy who designed that pedal for weeks, and he doesn't have any clue as to why it's happening.

I don't have room on my board to add another pedal, otherwise I would simply hook a random Boss to the output.

I understand that the PXL is probably causing some weirdness in the circuit. But that's what I've got, and it is what it is.

I'll figure out a way to use my rig, but I will never not be curious as to WHY this is happening. To me, the solution is secondary to knowledge and understanding.

Any ideas?

I have multimeters and the ability to use them. But I don't have specialty electronics equipment like spectrum analyzers or anything like that.

Could this be a matter of resistance, capacitance, polarity? If I active the switches on the PXL enough times in a row, the popping gets a little less loud.....as if it's discharging some stored energy?
I really don't know much about how all this stuff works, but I am very eager to learn.

Thank you for your time.

[ElectricDruid]

https://www.zeppelindesignlabs.com/wp-content/uploads/docs/Quaverato-Assembly-Instructions-ZD3672-AND-UP.pdf

The Quaverato output is from an op-amp, via 1K, a 1uF bipolar, and then a 50K volume pot. That seems pretty standard stuff to me, so I don't see why that would cause all the problems you've described, but perhaps someone else can step in.

Edit: Ooops, output pot is 500K, not 50K. Wasn't reading carefully, sorry.

[Rob Strand]

More details required for exactly when the level rises/drops.

The Quaverato looks like true-bypass in bypass mode so it's not possible to affect the level when bypassed with it's own footswitch, unless there's a build error.   If it did then that would be some crazy fault.

Of course through the PXL and with the Quaverato's footswitch to continuously set to effect the Quaverato's input stage would be always be connected to the PXL.

I was thinking the wrong values around R1, R2, R3, C1 could cause a volume *drop* in effects mode and/or bypass mode.   We don't have a schematic of the PXL.   The pedal has a output level control so there's not unity gain as such, you set it with the volume knob.   If there was a gain drop in effects mode then that would be perceived as non-unity.   Excessive loading of the PXL output due to a low  Quaverato input impedance might do it.

The same issue on the Walrus Audio Monument would mean it's more than a build error.

Something doesn't add up.

[SamJenkinsRides]

Every pedal I mention is always on, with the exception of the off Boss pedal placed directly at the Quaverato's output.

Any switching involved is only done with the PXL, taking active (on) pedals in and out of the circuit.

I have tried swapping any and all types (that i have at my immediate disposal) of pedals and cables and power supplies in any and all combinations and orders and arrays that I can possibly think of. And the only way I can use the Quaverato in one of the loops of the PXL and not have the volume jumping problem is to permanently activate the volume jump by placing an active buffer directly after the Quaverato in the same loop. An off Boss pedal does this without causing the PXL switches to pop loudly. The two stand alone buffers I have used (IC opamp based, and discrete transistor based) cause popping.

[Rob Strand]

I have tried swapping any and all types (that i have at my immediate disposal) of pedals and cables and power supplies in any and all combinations and orders and arrays that I can possibly think of. And the only way I can use the Quaverato in one of the loops of the PXL and not have the volume jumping problem is to permanently activate the volume jump by placing an active buffer directly after the Quaverato in the same loop. An off Boss pedal does this without causing the PXL switches to pop loudly. The two stand alone buffers I have used (IC opamp based, and discrete transistor based) cause popping.

The Beavis Audio buffer does have the potential to pop in a stand alone configuration.   The Barker circuit I posted has components R1 and R4 which specifically prevent popping under all conditions.   With R4 at 100k it would be best to use at least 1uF for C3, the 100nF on the Beavis Audio buffer is too small for R3=100k.  Slightly less desirable would be to use C4 100nF and R4 1M.  The Boss pedals have a resistor in the R4 position on their *output* buffer.

As for the problem it seems to be related to the high output impedance on the Quaverato, as set by the 500k output pot.    Most modern pedals don't use such a high pot value, unless they are cloning some old fuzz.   

Without a schematic of the PXL we can't see *why* the problem is occurring.   We can only see cause and effect from experiment.

I'm still not sure about the what conditions you are saying the level changes.  The Quaverato has the output level so why can't the level be set higher or lower to match the dry signal?

[SamJenkinsRides]

I have tried swapping any and all types (that i have at my immediate disposal) of pedals and cables and power supplies in any and all combinations and orders and arrays that I can possibly think of. And the only way I can use the Quaverato in one of the loops of the PXL and not have the volume jumping problem is to permanently activate the volume jump by placing an active buffer directly after the Quaverato in the same loop. An off Boss pedal does this without causing the PXL switches to pop loudly. The two stand alone buffers I have used (IC opamp based, and discrete transistor based) cause popping.

The Beavis Audio buffer does have the potential to pop in a stand alone configuration.   The Barker circuit I posted has components R1 and R4 which specifically prevent popping under all conditions.   With R4 at 100k it would be best to use at least 1uF for C3, the 100nF on the Beavis Audio buffer is too small for R3=100k.  Slightly less desirable would be to use C4 100nF and R4 1M.  The Boss pedals have a resistor in the R4 position on their *output* buffer.

As for the problem it seems to be related to the high output impedance on the Quaverato, as set by the 500k output pot.    Most modern pedals don't use such a high pot value, unless they are cloning some old fuzz.   

Without a schematic of the PXL we can't see *why* the problem is occurring.   We can only see cause and effect from experiment.

I'm still not sure about the what conditions you are saying the level changes.  The Quaverato has the output level so why can't the level be set higher or lower to match the dry signal?

What exactly is the purpose of the Boss's R4, and is that what prevents Boss pedals from causing popping?

What causes popping in general?

What would happen if I changed the 500k output pot on the Quaverato to a different value?

[Rob Strand]

What exactly is the purpose of the Boss's R4, and is that what prevents Boss pedals from causing popping?

What causes popping in general?

It ensure the output cap is charged and that the DC voltage at the output terminal is 0V at all times.   The output cap is charged at power up via R4.

Popping is caused when a discharged cap charges through a DC path (ie via a resistor) when two circuits are connected.  There is a change in the DC condition for a short period while the cap charges and that causes a pulse.

The main points are summarized in these panels.   The case where you connect a de-popped output to a non de-popped input is similar except the roles of the output cap and input cap are swapped.

[Note the resistor numbers have changed.  The output cap charges through R2 or R3.]

What would happen if I changed the 500k output pot on the Quaverato to a different value?

The way the circuit is designed you should be able to change the pot value.   50k would be something to try.    The output cap is only just large enough for a 50k pot.

However, I suspect there's something odd about the PXL inputs which is promoting the problem.  It's quite possible the problem still exists with a 50k pot but you might not notice it.

You could use smaller pot like 10k to reduce the issue more but the output cap should be increased to say 10uF.

Assuming we have got the cause worked out.  The PXL is likely to show this problem on pedals with high output impedance, pedals with output volume controls and in general pedals without buffered outputs.

We could also be on the wrong track.  Tricky problems often require a few experiments or measurements.

[SamJenkinsRides]

What exactly is the purpose of the Boss's R4, and is that what prevents Boss pedals from causing popping?

What causes popping in general?

It ensure the output cap is charged and that the DC voltage at the output terminal is 0V at all times.   The output cap is charged at power up via R4.

Popping is caused when a discharged cap charges through a DC path (ie via a resistor) when two circuits are connected.  There is a change in the DC condition for a short period while the cap charges and that causes a pulse.

The main points are summarized in these panels.   The case where you connect a de-popped output to a non de-popped input is similar except the roles of the output cap and input cap are swapped.

[Note the resistor numbers have changed.  The output has cap charges through R2 or R3.]

What would happen if I changed the 500k output pot on the Quaverato to a different value?

The way the circuit is designed you should be able to change the pot value.   50k would be something to try.    The output cap is only just large enough for a 50k pot.

However, I suspect there's something odd about the PXL inputs which is promoting the problem.  It's quite possible the problem still exists with a 50k pot but you might not notice it.

You could use smaller pot like 10k to reduce the issue more but the output cap should be increased to say 10uF.

Assuming we have got the cause worked out.  The PXL is likely to show this problem on pedals with high output impedance, pedals with output volume controls and in general pedals without buffered outputs.

We could also be on the wrong track.  Tricky problems often require a few experiments or measurements.

Thank you for the thorough reply! This was extremely helpful.
I assumed the popping was a result of discharging caps, not the other way around.

I may very well decide to change the output pot and cap on the Quaverato in an effort to pinpoint the issue.

Armed with some newly aquired understanding (thank you), I'll do some more experimenting with pedal and buffer order and placement and I'll report back if I can paint a clearer picture of what is happening.

[SamJenkinsRides]

Ok, I have been swapping components like a madman. I have come to the realization that the Quaverato is interacting with everything downstream of it. Nothing I do at or before it's input seems to make a difference.

My signal comes out of the PXL, to a DI box, to a mixer XLR input. If I change just one of these components, the Quaverato reacts differently. The main issues seem to lay within the Quaveratos reaction to/relationship with the DI and also the mixer.

Here's one of the routing scenarios I tried:
Guitar-> PXL (Quaverato in loop 8, loop engaged and Quav on)-> DI box-> Acoustic amp (XLR input)

On certain DI boxes (I have several different ones), when I plug a cable into the parallel 1/4" Input of the DI box ( the PXL output is in the other input) I get treble roll-off and a decrease in volume of the Quaverato. No other pedal I own does this (except for the Walrus Monument), and my old ProCo DB-1 DI box is the only one I have (old Rolls, a couple cheap Pyle and Behringer) that doesn't cause the frequency and volume changes in the Quaverato.
But the old ProCo will turn the Quaverato to quiet mud if I plug the XLR out into the input of a cheap little mixer....but won't if I plug the XLR into a Blackstar acoustic amp.

I have no idea what's happening here. I am inclined to agree with the high impedance idea, but I don't know why a DI box wouldn't decrease the impedance and therefore rectify the situation. I also don't know why plugging into a parallel input changes the audio signal.

If I just run everything into the Princeton, and forgo any DI boxes or mixers or anything else, the Quaverato behaves properly and predictably. I didn't realize before that the DI box/mixer combination was the catalyst or sole cause of the issue, and I'll probably be kicking myself in the hind end for a while over that. Regardless, running my board into a mixer is what I need it to do, and the only reason I built this board.

The guy who designed the Quaverato said the output is out of phase with the input. I have no idea if this is relevant.

Side note: I disassembled the Walrus Monument (the only other pedal I own that reacts in the same manner as the Quaverato, only not quite to the same degree), and it's output pot is 10k. Not sure if this is helpful.

Any ideas?

[SamJenkinsRides]

 Upon a bit more reflection:

Does a DI box increase the impedance of the input source, while decreasing impedance of its output?

Could the difference in impedance between the inputs of the acoustic amp and the mixer explain the difference in the way the two react differently with the Quaverato?
Both acoustic amp and mixer have switchable phantom power (off, obviously), not sure if that circutry is contributing to the issue.
I don't have enough knowledge about this stuff to understand how the inputs of the amp and mixer can be different enough for one to cause an issue and one to not.

Should I crack open my DI boxes and see what the difference is between how the parallel inputs are configured on the ProCo vs the others? I can't imagine why simply plugging a cable into the input parallel to the Quaverato-PXL connection would cause such a drastic drop in treble and volume.

It would be easy to simply purchase a different mixer. But how do I know what to look for in a mixer that won't possibly cause the same issue?

The deeper I dig into this issue, the more confused I become.

[ElectricDruid]

I would change that 500K output pot on the Quaverato, because that's the only thing about that's not absolutely normal. With the pot at half-value, for example, there's effectively 250K in series with the output. That's a *massive* output resistance, and I can believe that might make many bits of gear act weird or eat treble.

Why the output of one of your DI boxes sounds different into a mixer compared to into an amp can only be down to the DI box, and can't ("shouldn't", certainly) have anything to do with what's plugged into the DI box.

[antonis]

That's a *massive* output resistance

Good for a current source but too bad for a voltage source, which is the case here..

@SamJenkinsRides: There always are some capacitances hanging around to form LPF with output impedance.. 

[Rob Strand]

But the old ProCo will turn the Quaverato to quiet mud if I plug the XLR out into the input of a cheap little mixer....but won't if I plug the XLR into a Blackstar acoustic amp.

I have no idea what's happening here. I am inclined to agree with the high impedance idea, but I don't know why a DI box wouldn't decrease the impedance and therefore rectify the situation. I also don't know why plugging into a parallel input changes the audio signal.

The thing about *passive* DI boxes is they lower the output impedance at the expense of lowering the signal level.   There's no way around it.    Also the input impedance of a passive DI box depends on the load at the output (that's more or less regardless of the impedances written on the DI box).  These boxes don't act like a buffer.  A buffer presents a constant input impedance and a low output (source) impedance.   The impedance on the output does not change the impedance on the input side (at least to first approximation, which applies to this case).

If I just run everything into the Princeton, and forgo any DI boxes or mixers or anything else, the Quaverato behaves properly and predictably. I didn't realize before that the DI box/mixer combination was the catalyst or sole cause of the issue, and I'll probably be kicking myself in the hind end for a while over that. Regardless, running my board into a mixer is what I need it to do, and the only reason I built this board.

The amp input impedance is high, like 1M, so the output impedance of the Quaverato  is not affected by the amp loading.

Side note: I disassembled the Walrus Monument (the only other pedal I own that reacts in the same manner as the Quaverato, only not quite to the same degree), and it's output pot is 10k. Not sure if this is helpful.

It might be 10k inside but unless the 10k pot is wired to the output jack directly (actually via the switching as well) we don't know what extra circuit is between the 10k pot and the output.   I'll admit it does seem like weird behaviour, at least unexpected behaviour.

Any ideas?

I think these issues are more about what you expected from the DI box.   

We already know the problem is solved with an output buffer.  So that supports the argument that the issue is output impedance related.

Does a DI box increase the impedance of the input source, while decreasing impedance of its output?

Could the difference in impedance between the inputs of the acoustic amp and the mixer explain the difference in the way the two react differently with the Quaverato?
Both acoustic amp and mixer have switchable phantom power (off, obviously), not sure if that circutry is contributing to the issue.
I don't have enough knowledge about this stuff to understand how the inputs of the amp and mixer can be different enough for one to cause an issue and one to not.

Should I crack open my DI boxes and see what the difference is between how the parallel inputs are configured on the ProCo vs the others? I can't imagine why simply plugging a cable into the input parallel to the Quaverato-PXL connection would cause such a drastic drop in treble and volume.

It would be easy to simply purchase a different mixer. But how do I know what to look for in a mixer that won't possibly cause the same issue?

The DI box isn't the solution.  The short answer is in general a DI box won't solve the problem but a buffer will.
One caveat: a powered DI (also called Active DI) might help because it has electronics inside instead of , or in addition to, a transformer.  The electronics will provide a buffering behaviour.

[Matthew Sanford]

I think, if you want to bend it, you could make the passive DI box active. Might want to review schematics first, and I don't truly know, but seems adding power, a 10k voltage divider with a couple 1Ms for Vbias, and a TL072 with both set up as voltage followers between input jack & circuit, other circuit & output jack. Probably need coupling caps too.

If you've got a mind to do it, maybe open it and run wires to the rest on a breadboard to verify it first.

Now I'm not truly knowledgeable, Rob covers that, but it's an idea.

[SamJenkinsRides]

The thing about *passive* DI boxes is they lower the output impedance at the expense of lowering the signal level.   There's no way around it.    Also the input impedance of a passive DI box depends on the load at the output (that's more or less regardless of the impedances written on the DI box).  These boxes don't act like a buffer.  A buffer presents a constant input impedance and a low output (source) impedance.   The impedance on the output does not change the impedance on the input side (at least to first approximation, which applies to this case).

Bless your heart. This explains alot. So, if I understand it correctly, the input impedance of the mixer must be quite low, and must be loading the Quaverato/PXL combo...both through the DI box or straight in (issues like volume jumping and switch popping are still present with the DI box before the mixer, just not as drastic). The input impedance of the acoustic amp's XLR must be significantly higher so as not to load the Quaverato/PXL combo via DI box.
Am I getting this right?

It might be 10k inside but unless the 10k pot is wired to the output jack directly (actually via the switching as well) we don't know what extra circuit is between the 10k pot and the output.   I'll admit it does seem like weird behaviour, at least unexpected behaviour.

I took pictures. I didn't want to post them on here and muck up the thread unless they'd be useful. I'd be glad to post them if you like.

The DI box isn't the solution.  The short answer is in general a DI box won't solve the problem but a buffer will.
One caveat: a powered DI (also called Active DI) might help because it has electronics inside instead of , or in addition to, a transformer.  The electronics will provide a buffering behaviour.

The purpose of the DI box was to convert to Low Z XLR for my mixer input, and to send to FOH. The little mixer on my board is there so I can adjust eq and level of my monitor (ears or wedge) on the fly and not have to rely on a sound tech. I also don't want my monitor eq and level adjustments to affect FOH....which is why I must split the signal before the mixer.
I tried 2 different active DI boxes a while back. One of them wasn't helpful at all. The other one was helpful, but a monstrosity that wouldn't come anywhere close to fitting on my board.

Did some more work on the rig today, and have found a useful and practical solution.
Routing is as follows:
Guitar-> PXL (a random pedal in every loop, Quaverato in loop 8 )-> off Boss pedal-> "y" splitter-> 1/4" Neutrik channel input of mixer.
The other end of the splitter plugged into a DI for FOH, at minimal (microscopic) change in tone or volume.

So it turns out I was on to something with the buffer idea from the very beginning. I just had it in the wrong spot (original placement was directly at the Quaverato's output, within the same loop. Still not sure why this didn't work). I wasn't using the Neutrik 1/4" input on the mixer because it was heavily modifying the audio signal (treble loss, drastic volume drop, added compression)....but that was with the buffer in suboptimal placement.
No more need for DI between PXL and little mixer (that was screwing me anyway, apparently). DI box for FOH is no longer affecting signal integrity for the rest of the rig. I'm happy.
I don't think I would've been able to come to this exact routing scheme without the education I've received here.

Words can't begin to convey how thankful I am for the help and education I've received on this forum. To every commenter here, especially you, Mr. Strand. I thank you from the bottom of my heart.

[Rob Strand]

Bless your heart. This explains alot. So, if I understand it correctly, the input impedance of the mixer must be quite low, and must be loading the Quaverato/PXL combo...both through the DI box or straight in (issues like volume jumping and switch popping are still present with the DI box before the mixer, just not as drastic). The input impedance of the acoustic amp's XLR must be significantly higher so as not to load the Quaverato/PXL combo via DI box.
Am I getting this right?

It's highly likely that's what is going on. 

I took pictures. I didn't want to post them on here and muck up the thread unless they'd be useful. I'd be glad to post them if you like.

If you get desperate, I can trace it from pics no problem.   (Maybe not in the next couple of days.)

Did some more work on the rig today, and have found a useful and practical solution.
Routing is as follows:
Guitar-> PXL (a random pedal in every loop, Quaverato in loop 8 )-> off Boss pedal-> "y" splitter-> 1/4" Neutrik channel input of mixer.
The other end of the splitter plugged into a DI for FOH, at minimal (microscopic) change in tone or volume.

At least you have a solution.

So it turns out I was on to something with the buffer idea from the very beginning. I just had it in the wrong spot (original placement was directly at the Quaverato's output, within the same loop. Still not sure why this didn't work). I wasn't using the Neutrik 1/4" input on the mixer because it was heavily modifying the audio signal (treble loss, drastic volume drop, added compression)....but that was with the buffer in suboptimal placement.

No more need for DI between PXL and little mixer (that was screwing me anyway, apparently). DI box for FOH is no longer affecting signal integrity for the rest of the rig. I'm happy.
I don't think I would've been able to come to this exact routing scheme without the education I've received here.

Words can't begin to convey how thankful I am for the help and education I've received on this forum. To every commenter here, especially you, Mr. Strand. I thank you from the bottom of my heart.

No problem at all.  If you look at it you actually did all the hard work and you found the solution yourself!   All I've done is explained a few things and maybe helped pick the best solution.