Damn, I was really hoping that would be the silver bullet.
As I said before, the ID Pogo Pin is meant to line up with a data pad on the Zero and Zero W, but this pin is not actually essential. The entire USB Hub HAT is designed so that it can it can make use of the ID Pogo Pin on a Pi that has it, but not to be affected if the Pi does not. This means its compatible with both the Pi Zero and Pi Zero 2 Series.
Iām going to have to test the Ground Loop hypothesis on a couple of Pi Zeros W and Zero 2 Ws with the USB HAT because that would be a flawed design if it were true.
Yeah. Me too. As a hobbyist I buy boards thinking they will be debugged, but this USB Hat is proving difficult.
There is no corresponding ID Pad on the Zero 2W so the fact the hat would not even power up initially was a puzzle, a small pad of thin insulating material seemed to cure that issue so maybe the green solder mask is not quite right or there is RF interference at that spot? I have been a bit busy writing code and investigating the other instability issues (USB resetting and re-enumerating devices when the Zero/UPS/USB combination is connected one CPS UPS abut not on a different model of CPS UPS is a worry ā I am not in control over which UPS devices are used out in the field ā but in the interim I have recommended that the customer does NOT use simulated sine wave type UPS for now. There is definite interference measurable on the USB Data and Power lines that look suspicious. Over the holiday break I will look further into that ā might need some special USB cables whereby I can add ferrites and other noise filtering and break the shielding from UPS USB port to eliminate that as a cause. Never had the problem with the old design using the PiJuice board for UPS and RTC. I need an easily repeatable solution for long term use.
