switch popping problem that seems weird to me

[Wounded Paw]

I've got a switch popping problem on an effects loop blender pedal I'm making.  I've followed all the troubleshooting for switch pops including replacing the switch, different pull down resistors etc.  I've been able to make the pop quiet enough that it doesn't intrude on the sound but only by reducing the input pull down resistor to 47k.  While this works it seems odd to me that I had to go that low and I'm wondering if this will start affecting the input signal, impedance or anything else?  Is there a better way to get rid of the pop? 
Here's my schematic with the original 1M input pull down resistor underlined.  The 1uf caps are all non-polar.  And yes, this design takes a lot from the Paralooper by Moosapotamus and the B. Blender by Sean M.

[R.G.]

What are you switching? I don't see any switches in your schemo.

[Wounded Paw]

Sorry, the switching is just your basic true bypass for the whole pedal, before the input and after the output on the schematic.

[R.G.]

That being the case, if it were mine I would check across the input pulldown resistor and output potentiometer to see if there is any DC voltage to ground. That could happen by the input and/or output caps being backwards.

[Wounded Paw]

11 mV across the output potentiometer and 7 mV across the input pulldown resistor (47K).
When I replace the input pulldown resistor with the original 1M resistor the DC voltage across it goes up to 85 mV.  Is that high enough to cause the pop?  The input cap is non-polar so it should be fine whichever way it is, right?

So with the input pulldown resistor at 47k there is a barely audible pop but at 1M it's very loud.  Is there any problem with using the 47k as the pulldown resistor?  It seems really low to me.  Is there some other design change I can make to avoid this?

[R.G.]

11 mV across the output potentiometer and 7 mV across the input pulldown resistor (47K).
When I replace the input pulldown resistor with the original 1M resistor the DC voltage across it goes up to 85 mV.  Is that high enough to cause the pop? 

Mother Nature has just given you a lesson in what size pops are audible and not. 7mv is pretty trivial, and 85mV is not. Since it pops, yes, it can cause the pop.

The input cap is non-polar so it should be fine whichever way it is, right?

Should be, yes.

So with the input pulldown resistor at 47k there is a barely audible pop but at 1M it's very loud.  Is there any problem with using the 47k as the pulldown resistor?  It seems really low to me.  Is there some other design change I can make to avoid this?

Yes, there is. For a guitar plugged directly into that, you will see quite a bit of treble loss.

I think that your input cap is leaky.

[Wounded Paw]

Nope, not the input cap.  I tried a 10 uf and a 0.1 uf in it's place and still got really high voltages across the pulldown resistor.  Higher than with the original 1uf NP cap.  So it would seem there is a problem with the design but I cannot see what it is.  The DC should be stopped at the input cap, no?
I also added a 100uf cap from +9v to ground for some more filtering.

[R.G.]

Nope, not the input cap.  I tried a 10 uf and a 0.1 uf in it's place and still got really high voltages across the pulldown resistor.  Higher than with the original 1uf NP cap.  So it would seem there is a problem with the design but I cannot see what it is.  The DC should be stopped at the input cap, no?
I also added a 100uf cap from +9v to ground for some more filtering.

Let's try a 'speriment. If it's not the input cap or the output cap leaking into those resistors what can be leaking into them?

So let's see - if we pull the input cap loose from the 1M, leaving everything else the same, that's like putting a 1,000,000,000,000 k resistor in series with the cap, right?  So do that, just pull up one leg of the input cap, away from the 1M input pulldown. Now measure the voltage across the 1M resistor, exactly the same circumstances. If it has remained the same, then the leakage is not coming from your circuit, but from whatever is connected to the input. If it drops to nearly zero, then the current had to be coming from the input cap, right?

[Wounded Paw]

Okay, it's definitely coming from the cap.  The voltage disappears if I pull up a leg of the cap.  So now what?  Several different caps leaked in pretty much the same way.  I'm stuck.

[R.G.]

Good caps don't leak so much that they pull a 1M resistor up by 85mV.

Where - and when - did you get your caps?

Seriously, a 1M resistor will pull a good 1uF or 10uF electro cap to within millivolts of ground.

[Wounded Paw]

Most of the caps I tried were from the local electronics store but the 10uf was a Xicon from Small Bear.  Same result. 

[rogeryu_ph]

Sorry to interrupt, but this is also the case where i am only different circuit.
Mine, the output has a 0.35mV and the input is just 0.01mV.
@ R.G. Does this mean my output coupling cap is the problem? It's brand new and it's electro non polar co'z i can't get 1uf film cap and it's bulky in size and my box space is quite small. Second question, that being said that the cap is leaky can i still use those cap for other circuit except for coupling purposes co'z my project is working perfectly but just that pop is the issue. I'm using 3pdt blue bypass switch both with 1meg on input/output.

BTW Wounded Paw, give us your feedback when you find your solution there buddy and many thanks.

[Wounded Paw]

Aha!
It was the cap after all.  After trying at least 5 different caps I found one that doesn't leak at all.  It was a 0.1uf metallized film capacitor.  I guess I just have a crappy collection of caps right now.  Even the Xicon cap leaked quite a bit.

[R.G.]

Sorry to interrupt, but this is also the case where i am only different circuit.
Mine, the output has a 0.35mV and the input is just 0.01mV.

Did you misplace a decimal point? 0.35mV is too small to hear unless the circuit is a very high gain distortion pedal. Pops on the order of 10mv or less are non-intrusive and are often missed on the output. But if you have an offset of 350 MICROvolts on the input and a gain of 200, that's 70mV on the output, and that's definitely unacceptable.

If you meant .35 VOLTS and 0.01 VOLTS then you have a problem of leaky caps.

@ R.G. Does this mean my output coupling cap is the problem? It's brand new and it's electro non polar co'z i can't get 1uf film cap and it's bulky in size and my box space is quite small. Second question, that being said that the cap is leaky can i still use those cap for other circuit except for coupling purposes co'z my project is working perfectly but just that pop is the issue. I'm using 3pdt blue bypass switch both with 1meg on input/output.

In general, the output cap is not such a problem, because you can use lower pulldown resistors on the output for many effects  - although not all of them! - than you can on an input. The input is one place where leakage is really critical because you can't in general lower that pulldown resistor to less than about 1M without starting to lose treble.

[rogeryu_ph]

Sorry to interrupt, but this is also the case where i am only different circuit.
Mine, the output has a 0.35mV and the input is just 0.01mV.

Did you misplace a decimal point? 0.35mV is too small to hear unless the circuit is a very high gain distortion pedal. Pops on the order of 10mv or less are non-intrusive and are often missed on the output. But if you have an offset of 350 MICROvolts on the input and a gain of 200, that's 70mV on the output, and that's definitely unacceptable.

If you meant .35 VOLTS and 0.01 VOLTS then you have a problem of leaky caps.

@ R.G. Does this mean my output coupling cap is the problem? It's brand new and it's electro non polar co'z i can't get 1uf film cap and it's bulky in size and my box space is quite small. Second question, that being said that the cap is leaky can i still use those cap for other circuit except for coupling purposes co'z my project is working perfectly but just that pop is the issue. I'm using 3pdt blue bypass switch both with 1meg on input/output.

In general, the output cap is not such a problem, because you can use lower pulldown resistors on the output for many effects  - although not all of them! - than you can on an input. The input is one place where leakage is really critical because you can't in general lower that pulldown resistor to less than about 1M without starting to lose treble.

Sorry that's 35mV or 0.35V and on the output. So R.G. what's the ideal lower pulldown value resistor next from 1meg for the case of mine if i don't have anymore CAP laying around  appreciate your reply for this.
BTW Paw, glad you solved out the problem, did you not hear anymore pop?  Also does those previous crappy cap no longer useful? Thanks also.

[R.G.]

Sorry that's 35mV or 0.35V and on the output. So R.G. what's the ideal lower pulldown value resistor next from 1meg for the case of mine if i don't have anymore CAP laying around  appreciate your reply for this.

The lower the resistor, the more treble loss. Your ears will tell you when to stop.

The *right* answer is to get caps that don't leak. Use my standard advice and use whatever you have until the perfect thing can be obtained.