Pyboard suddenly stopped working and becomes extremely hot – has anyone experienced this?

[oreoneuro]

Hi everyone,

I'm trying to track down the cause of a failure we've recently encountered with two pyControl boards, and I'm wondering if anyone has seen something similar.

Both boards worked normally for some time (weeks to months), and then suddenly stopped being recognized over USB. When I checked with a multimeter, I confirmed that there is a short in the part connected to the chip, and I was told that this part had failed.

While searching for similar cases, I found a 2022 Google Groups post from Mostafa Safaie (https://groups.google.com/g/pycontrol/c/YVm5sS9NpAc) describing what appears to be the exact same symptoms: pyboards suddenly failing after weeks or months of normal operation and becoming very hot. Unfortunately, since the discussion was on the old Google Groups, I couldn't find any follow-up or whether the root cause was ever identified.

Has anyone else experienced this type of failure?

If so, I'd be very interested to know:

• What was ultimately found to be the cause?
• Was it related to wiring, a connected peripheral, the breakout board, power supply, or something else?
• Were you able to repair or replace the board, or did you identify a damaged component?

I'm mainly trying to understand whether this is an isolated incident or if others have encountered the same failure mode.

Also, it seems that pyboards are becoming increasingly difficult to source. For those still running pyControl setups, where are you currently obtaining replacement pyboards? Have you found any reliable suppliers or alternative solutions?

Any insights or experiences would be greatly appreciated. Thanks!

[alustig3]

Unfortunately, if the chip on the pyboard itself is getting hot, it is unlikely that the board is recoverable.

@AtMostafa may have some insight on the problems he ran into? If I recall correctly, separate outlets powering his computer and powering his breakout boards had separate grounds and that was causing problems? Hopefully he will chime in.

Also, it seems that pyboards are becoming increasingly difficult to source. For those still running pyControl setups, where are you currently obtaining replacement pyboards? Have you found any reliable suppliers or alternative solutions?

We are in the process of developing an new pyControl breakout (see discussion). The new work has been an update to the entire breakout board, with a focus around designing around a more powerful and more easily sourced microcontroller.

Thomas and I have also talked about developing a board with the H7 microcontroller that could be a drop-in replacement for the pyboards in the existing pyControl breakouts. This seems like it would be very useful to existing users. We will keep you posted on progress!

[AtMostafa]

Yes, we fried several of our pyboards while we were building the rig. As I recall, it was exactly like what @oreoneuro described. Since then however, we've run our experiments using a single pyboard for a few years.
My advice would be to test all the bit and pieces that should be ground, using a multimeter, while there is no power in the circuit. In our case, DC power sources were not internally connected to the ground (even if you see a metal GND prong on the plug, it might not be used internally). And going forward, always ground any parts that are being added to the rig first, before wiring them up.