I just want to step into this discussion as I did have my share on the Click board usage.
I use them with the pocketbeagle, which is posted as being designed so that it is easy to integrate Click boards.
I found this to be not compeltely true:
Mechanical issues: The idea is to integrate at the processor side of the PCB, but the processor is so close to the holes that it is not possible to simply add a header to add the click board.
I decided to add the header on the other side and to adapt the click panel so that the pins stick to the other side.
If you are a bit handy it is feasible to do this, but you have to remain awake as you should never connect a "normal" Click board as it would result in feeding 5V in the 3.3V power of the click board.

Pin organisation issues: it might be a coincidence but the pins that are used to integrate the CAN RX/TX into the PB are only available for CAN after the boot proces, the processor uses them for some messaging while booting.
This is exactly something you don't want: booting of the PB would result in corrupting the live CAN bus by sending rubbish messages in wrong bit rates, I had this issue and was puzzled why my canbus network went silent each time the PB booted.
So I moved the pins to another CANbus on the PB header (white wires)
I decided not to take the route of the overlay, just set up a script that defines the port configuration and other related actions (start my CANbus service) in the right order and enable this script as a service to be run at boot.
This was very convenient ans now I just have to maintain this script.
Gwen
Op vrijdag 29 maart 2019 00:52:44 UTC+1 schreef Mala Dies: