One potmeter of my Pepper stopped working well.
All other functions are fine, but the potmeter (adc~5 in Pd) doesn't react to changes.
Only sometimes, when the Pepper front plate is pushed, the knob values come through.
Same result with (lop~ 100) filtering.
When I insert a jack on that input (adc~ 5), all comes in fine, but the knob stays not functioning well.

Solder joints look good. Where should I check on the PCB?
Because of the function coming back when the front plate is slightly pushed, maybe this could the shield connection?
After careful pushing on the headers, the connection came back, but it only woks under a certain tension. When installed back into the rack, the connection gets lost again.

    Floris the potmeter (adc~5 in Pd) doesn't react to changes.

    Is the issue affecting only that one channel or all of them?

    The relevant parts for that channel are R1, C2 and the potentiometer itself. It's worth checking that they are soldered properly and there are no shorts, dirt or blobs of solders. Reflow any solder joint that looks opaque and not lucid.

    Can you tell whether the signal is stuck at the maximum or minimum or otherwise? Does it get affected from signals or pots from neighbouring channels?

      Thanks, I'll check R1 and C2 too.
      The signal is stuck somewhere halfway between the minimum and maximum.
      No other channels are affected, apart from that pot, all works fine.

      a year later

      Reopening this one after a year.
      R1, C2 and the potmeter were re-soldered and still adc~5 in Pd stayed stuck.
      2 months ago, halfway during a tour, the input/pot connected to adc~10 in Pd also stopped working.
      When pressure is applied to the front plate, sometimes the signal comes back, but after a while this jumps back to the minimum value.
      Maybe it's a bad connection between the pin headers?
      While always housed in a wooden box with lid and traveling in a Peli case, Pepper still gets countless bumps while touring.

      If a new version of Pepper arrives in the next half year or so, I'd be fine with waiting and getting a new one.
      Otherwise, I'd like to continue troubleshooting of wordt case, try a new Pepper shield. Any suggestions welcome.

      13 days later

      hi sorry I forgot about this until a new Pepper-related post reminded me of it.

      I would suspect that the issue you have here is that one or more of the pins on the analog input header are slightly too short and barely make contact with the mating socket header on the Bela cape. There are a couple of things that you could try:

      • ugly, but you could try soldering wires to them. These wires would go around the back of the Bela cape to the top layer of the Pepper PCB ( the one closest to the faceplate)
      • alternatively, if you have a bit of experience reworking PCBs and you know when hot is too hot, you could try one of these (and if any fails, revert to the above):
        • remove the pin header from pepper, clean the holes, source and install a header with longer pins
        • heat up each pin in turn from the top layer of the PCB, add flux and/or solder and push it slightly deeper into the hole till the back of the pin is flush with the PCB. This should give you more than 1mm extra depth which should be enough to improve the reliability of the contact

      thanks, soldering wires seems like a good idea!