3PDT switch wired to live rail?

[CurlsOnKeys]

Hi all,

I'm fixing a EHX Fuzz Graphic Fuzz which has a loud pop when engaging the pedal. Looking for possible culprits, I came across a bizarre wiring on the 3PDT switch. Googled a long time for it, but didn't find anything about this configuration.
I'm fairly new to really trying to "fix" these things, although I have a decent background in electronics.

Here's the link to the schematic: https://smokingtip.files.wordpress.com/2014/03/aagraphicfuzz.pdf

So to make the pop go away, these are the things I tried first (all to no avail):
- hook up pulldown resistors to input/output caps
- unsoldered the LED
- measured with my oscilloscope to see if caps were "floating" or to find out where the DC would come from

In doing this last thing, I noticed that the input - when bypassed - had the whole positive supply rail (12V) on it. It seemed to me that this could possibly be the cause of the loud pop (engaging the pedal means a 12V cap discharge). Does that seem possible?

The 12V comes from a wire that is directly connected from output of the 78L12 to the 3PDT switch (soldered in between U8 and C42 on the schematic). It's not mentioned on the schematic (it has no dedicated pcb wiring terminal) but it's there and it looks like it has always been. The switch is wired as follows, with a jumper wire between 2 and 7 (from input jack).

From PCB out --- From input jack --- Led+
    |                             |                    |
To output jack --- To PCB in --------- Led-
    |                             |                    |
From input jack --- To +12V ---------- not connected

As you can see, in bypass mode, input-jack goes true bypass to output jack, but pcb in hangs to +12V.

I can't figure out why and didn't find other examples of this kind of wiring... Some 3PDT configurations have +V tied to the LED, but here this is not the case. The only thing I've found so far is a guy on the EHX forum asking the same question (without answer): http://www.ehx.com/forums/viewthread/6061/

Any ideas here? I would simply try to desolder it and see what it does, but I was stupid enough (hey, everybody's gotta learn sometime) to try to ground the 12V lug first, thereby frying the zener  So as I'm waiting for a new one, I thought I'd give it a shot here!

Thanks!

[PRR]

Welcome!

> other examples of this kind of wiring...

The footswitch does not touch audio. It sends signals to other parts which turn on/off as needed.

> the input - when bypassed - had the whole positive supply rail (12V) on it.

That schematic is large, complex, and not drawn too obviously.

But my 2 cents says replace C38. Observe polarity. Exact value not critical. 0.1uFd would be ample in most cases, and be a film cap, which is non-polar and also lower-leak longer-life; however larger.

Do C37 while you are in there. Deciding which is failed would take longer than just replacing both.

[CurlsOnKeys]

Hi PRR,

thanks for your 2 cents!

Replaced the zener and the stombox is back up (pfew).

Measured output at C38, sits perfectly at 0V, C37 also. I could replace them, but if they don't bleed DC, that doesn't seem the cause?

Still wondering about the 3PDT switch attached to +12V. Doesn't seem to "send signals" to me. To make it clear: there goes a wire from in between C42 and 78L12 (grabbing the +12V that's there), towards the bottom row of the footswitch, which is then - when the switch is in bypass - coupled to PCB in, so to the input of C1 to be exactly. C1 then of course blocks this +12V DC, so what's the point in sending it there? On the other side of C1 I meassure some 35 mV and I measure the same 35 mV when the switch is not in bypass.

So why on earth would you want +12V on the input capacitor, which just sits there, waiting to be pulled to ground the moment you hit the switch?

[CurlsOnKeys]

Update: desoldered the +12V connection from board to 3PDT switch.
All functions on the stombox still work like they're supposed to and pop is gone.

No hum, ground or pop noises for now so I left that pole of the switch unconnected, since the fx input is already grounded by means of R1, so seemed pointless to ground it a second time...

Still puzzled why it was there...

[Rocket Roll]

Did the same thing on a friend's Graphic Fuzz, with the same results.

This schematic shows something else happening with the footswitch, but the actual wiring in the actual unit was the same as described above.