High gain build popping

[PBE6]

@bluebunny this is the foam my painted boxes came in, not the kind components come in. I just double-checked it using voltage, current and continuity tests and it doesn't appear to conduct at all.

(Whew!)

[PBE6]

@GibsonGM no not yet, I'll give that a try shortly and measure again.

[PBE6]

@duck_arse will do. That should be easy enough, it's just guitar-->pedal-->practice amp.

[PBE6]

@duck_arse ok so with the normal signal chain, the input it at 0.000 V DC, and the output can get as high as -0.008 V. One thing I did notice this time is that the "bang" turns into a quieter "whomp" if I leave it off for 5-10 seconds before re-engaging the circuit. Maybe the time constant of the RC filter on the output is just too high? 3.3uF and 500k makes a time constant of 1.65 seconds. That sounds high, doesn't it?

I used a 500k pot for the output because that's what's in the schematic, but I usually try to keep volume pots to 10k, maybe 100k, to keep the output impedance low. Why does the OCD have such a big volume pot anyway? Is it a nod to the high output impedance of a standard guitar?

[GibsonGM]

It's actually 1.65 seconds for ONE time constant...it takes 5 TC for a cap to fully charge or discharge, believe it or not. 1 TC = 63% charge.  It's 8.25 seconds for "fully" charged/discharged.  Could be related, let's see what the Duck has to say!

I don't see that you posted what schematic you're working from...forgive me, but I don't see a 3.3u cap at the output on the one I have; just the 1u output cap going to switchable resistors as a filter.  My schematic for the OCD is 'singed' "Stratotasto".  Don't know where I got it...

[bluebunny]

(Whew!)

Just checking.   

[PBE6]

@GibsonGM here's the as-built circuit (minus the LED fix):

[PBE6]

Ok just did a few things:

- changed the 500kA pot to a 50kA pot, no effect
- added a wire from input jack ground to output jack ground, no effect
- cut the wire from LED negative to ground, no effect

The last thing to try is replacing the switch, but I'll get to that later. Is there a simple way to test a switch to see which side makes its connection first? It could be that my switch is connecting the output before the input, hence the switching noise. If I could be sure which side latches first, that might solve it.

[Kipper4]

Ok have you tried just putting a cap in series on the output after the tone stack?
I dunno say a 100nf.
Forgive me if you have tried this or its been suggested before. I'm just thinking out loud.
I hope you get it sorted.
Rich

[PBE6]

All suggestions welcome! I will give it a try

[GibsonGM]

That's a decent idea, Rich. 

There's no source for DC after the 3.3u, tho.   What about a pulldown resistor to ground, 470k to 1M, right after the 3.3u cap?  I know that's kind of reaching, but it would be easy to try. If there is ANY DC building up/leaking there, wouldn't that bleed it off pretty fast? 

The cap Rich suggests would also block anything there, so would probably do the same. 

[induction]

Are you sure that 3u3 isn't installed backwards? I can't tell from the pictures without a layout, but it might explain the popping.

[GibsonGM]

Are you sure that 3u3 isn't installed backwards? I can't tell from the pictures without a layout, but it might explain the popping.

+1    I mean, a properly functioning output cap like that should essentially block all DC....there should be none there....the stuff I / we're suggesting are fixes for something that shouldn't be there

[duck_arse]

me? I got nuthin. I was hoping for some "boss" silliness somewhere in the chain. it looks like we're down to the switch.

looking at those doubled vrefs makes me wonder if you could run them in parallel instead of series, then collect all the audio earths that sometimes go to V/2, and send to there their own vref, separate from the opamp inputs. that would only be something for trying on a breadboard, though, so, like I said, "nuthin".

[edit :] spellink

[GibsonGM]

Yeah - short of a bad cap (3.3u), or something connected to Vref instead of ground - or a short somewhere - I'm just not 'getting it'.   There should be no DC going to the switch at all.   

All I could really come up with beyond this is that its at the switch, and shorting somehow internally (??) that's a far reach.    As a last resort, I'd change that cap, being sure of orientation, and maybe add a pulldown after it - the cutoff freq. for that is so low it should be transparent, but still allow the DC to equalize.

[PBE6]

I replaced the cap with another 3u3 cap and the noise is still there but it's better. It's actually fairly quiet compared to the guitar signal, but the switch noise would still be audible if it were turned on during a quiet section (although it's not often that a performance calls for delicate plucking followed immediately by loud, wailing distortion).

If anyone thinks of something brilliant in the meantime, let's know!

[PBE6]

I did a quick recording with the volume at 3/4, and the gain full up, and the pop is about 17 dB lower than the signal. Not perfect, but workable.

[GibsonGM]

A lot of my builds DO have *some* pop when switching...I write that off as a mechanical thing...I mean, you're actively engaging a "live" circuit, so SOMETHING has to make a sound.   There are other schemes that ground in/out while in bypass - maybe you could try one of them out?

I just "get used to" whatever a pedal does, and act accordingly on stage.  I don't stomp until it's REALLY time to stomp, that kind of thing...like, on the "1" where my hand is already moving to hit the "big chord"...switch and chord happen very nearly at the same time...

[induction]

I replaced the cap with another 3u3 cap and the noise is still there but it's better. It's actually fairly quiet compared to the guitar signal, but the switch noise would still be audible if it were turned on during a quiet section (although it's not often that a performance calls for delicate plucking followed immediately by loud, wailing distortion).

If anyone thinks of something brilliant in the meantime, let's know!

I'll ask again: Are you sure that 3u3 is oriented correctly?  Electrolytics pass DC when they are reverse-biased.

[PBE6]

It's connected as per the diagram. I think that's correct, the negative leg faces the output where the voltage eventually runs down to ground, and the positive side faces the opamp where voltage is highest.