I'm working on an Effects Layout Sasquatch II - a modded Big Muff with a green ringer on a second foot switch, which is the issue. With the muff section bypassed, the octave foot switch makes a loud buzz (not pop) when activated, with the muff section engaged there is a half-second of silence before the signal kicks in. Everything works great otherwise.

Going through the octave section I checked all values and reflowed the pads, changed the buffer jfet and replaced the foot switch but the problem remains. I have built 5 of these, so I know the board is good. Any ideas gratefully received, thanks!

It's not possible to be sure from the photos, but many of the solder joints look a bit inadequate. Maybe try reflowing all the joints.

Clean build. However, you were a bit conservative with the solder there.

Typically we see the opposite, where too much solder is applied causing shorts or hiding cold joints. Here it is the opposite which is uncommon, but demonstrates discipline. You're on your way!

Do an image search for "good solder joints".

Thanks for the solder advice, I went back over each joint and reflowed with extra solder but the switch buzz/delay is still the same, very consistent every time not like a loose connection. Could it be a bad cap or diode? The time element (length of buzz and silent delay) seems like a clue.

Without the schemo of the footswitch wiring, it's hard to guess anything other than soldering. Maybe an electrolytic cap inserted backwards would do it.

I did check all values and orientation of parts, plus I changed the foot switch and daughterboard. I made five of these without issue, it's driving me crazy.

There is a jfet buffer at the very beginning of the octave circuit. The schematic calls for a PF5102 but I used a 2n5457 (like the other ones I made) but changed this to a J113 to see if it helped. No change in the issue.

Quote from: FootSolder on April 29, 2024, 08:12:19 PMI did check all values and orientation of parts, plus I changed the foot switch and daughterboard. I made five of these without issue, it's driving me crazy.

There is a jfet buffer at the very beginning of the octave circuit. The schematic calls for a PF5102 but I used a 2n5457 (like the other ones I made) but changed this to a J113 to see if it helped. No change in the issue.

Do a DC check with your multimeter on both sides of C1 and pins 3 and 1 of the Octave pot. Might help determine if you have a leaky cap. The signal cutting out/buzzing is pretty strange, though.

A J113 is more typically used for analog switching so I think a 2n5457 (or compatible N channel amplifier) is the right choice here. Maybe someone else can tell me if I'm wrong about that.

Quote from: FootSolder on April 29, 2024, 01:41:06 PMwhen activated, with the muff section engaged there is a half-second of silence before the signal kicks in.

That is usually an indication that a coupling capacitor has no pulldown (or pull up) resistor after it, or the resistor is open circuit.

I'd start from disconnecting BOTH indicator LEDs..

P.S.
Regardless of the result, I'd place both LEDs (and CLRs, of course) before D7..

jfet gate to ground resistor?

Okay, it's the LED. When I took it out the buzz/signal-drop completely goes, put a new LED in and it comes back. I replaced everything in the buffer section (see pic) and switched to a 2n5457 jfet, as folks seemed to be pointing to those parts as being a potential culprit. Still the same issue.

I then tried the AMZ LED pop fix (see pic), still the same issue. Why does the LED quack or duck (pun intended) when the switch is activated??

Did you relocate the LED before D17..??

D7? I'm not able to on the existing PCB. The second LED for the fuzz section is in the same place on the schematic and works fine, as did both LEDs in the 5 units I previously built. Are you thinking it is something specific to D7?

Sorry, just rigged it up using some patch cables - with LED before D7 the problem is the same

If so, I'd put the blame on C22..

I replaced C22, it still buzzes/cuts out when switch is activated.

FWIW, the noise is like a burst of static, not a pop or grounding hum. The LED is a standard diffuse 5mm and the CLR is 4K7, tried a water-clear but same deal.

I really appreciate all the suggestions, hoping someone might be able to suggest another fix, I am completely baffled!

is your circuit dia drawn and layouted with Eagle?

I ask because the jfet is the only part on the schem showing junction dots where they would not otherwise be. this suggests to me that a part on a 0.05" grid was placed on a 0.1" gridded circuit. then the pins don't auto connect, and need jumpering links from/to the grid differences.

that's surely 2 cents worth.

It's actually a premade PCB from Effects Layouts, one that I've successfully used a number of times. Your suggestions makes sense but I don't think it applies in this case, thank you tho!

What are you using for a power supply? Try without C22 as a test. Sounds like C22 is discharging quickly when the LED is activated, and then immediately needs recharging which current-starves the rest of the circuit and/or pulls down the V+ voltage until C22 is back up to full 9V charge. This should not happen if C22 is not present (I think) but that's not a permanent solution.

If this is the case, the remedy is to connect the LEDs before D7, as Antonis has recommended, which prevents current from flowing out of C22 to the LEDs. It might also help to add caps in parallel with LED1 and LED2 (experiment to find the right value - maybe start with 1µF).

EDIT: a cap alone in parallel with the LED will probably blow the LED- include a resistor in series with the cap (the combination in parallel with the LED) with the resistor being a bit smaller than the CLR.

Larger LED CLR values might also help.

It's not clear why you haven't had this problem with previous builds of the same circuit, but changing the FETs could be related, as could the actual power supply itself.

There is another way to quiet LED switching.  Instead of switching the LED on and off with a series switch, use a transistor to ground in parallel with the LED.  The circuit pulls nearly the same current in the on and off case since you have a series resistor that takes a voltage of 9 volts minus the LED voltage of 1.88 (measured on one of my projects) for red or 2.05 for green (from another project) and switches to 9 volts minus the 0.2 volts of a saturated transistor (if you are using a logic switch).

This can be made even closer by adding one or two diodes in series with the transistor so you get 9 volts minus 0.9 or 1.6 volts, almost no difference between the voltage drop across the resistor and therefore the current in the on or off state.  Even if you use a mechanical switch, two diode drops at 1.4 volts can be almost equivalent to the 1.88 volts of the LED.  This would still limit the difference in current between the on and off state, thus limiting the noise. The only penalty is constant current consumption in the LED circuit even when the the effect is off.

Quote from: FootSolder on April 30, 2024, 05:57:31 PMFWIW, the noise is like a burst of static, not a pop or grounding hum. The LED is a standard diffuse 5mm and the CLR is 4K7, tried a water-clear but same deal.

As a longer shot, RF self oscillation noise can sometimes sound like an angry hiss. And follower circuits are prone to self oscillation at times. If this was a MOSFET follower, this would be even more likely, and the cure would be to put 10R to 100R as close to the gate pin on the FET package as you can get it. The mechanism of the LED causing this would be a transient on the power supply causing the follower to ring, and the parasitic oscillation dying out from not being able to quite self-sustain.

If I were debugging this, I'd try measuring the DC voltage at the source pin to see if it rose or dipped when the static was heard. Scoping it would be definitive, but a JFET could easily be self oscillating at frequencies above those a DIYer scope can see. And changing the JFET type could well account for the occurrence. I really should compare the datasheets for the P5102, 2N5457, and J113, but I've been too lazy to do that yet.

A possible cure could be something like a drain resistor to the power supply, perhaps split in half, with a local decoupling cap to the bottom of the source resistor ground. A 0.1uF >ceramic< cap from drain to bottom of source resistor might be another trick, but this would be better with a resistor from the drain to power supply as well. The resistance doesn't have to be big; 100R to 1K ought to do it. If either of these tricks even changes the static, it's very likely that you're chasing a self-oscillation.