I am still curious how they were able to determine this chip as the failure. Do you have some watchdog test to see what the timeout is ?

I am not informed what the actual problem is; however we did, with a previous board layout (1" trace to decoupling cap), have horrendous problems with the Microchip super picking up noise which the MAX did not.
The WD test is simply that if the WD kicks in, the board is failed. It could be noise pickup or timeout outside the tolerance.

So, whatever the problem is, the MAX does not fail, the MC fails at a high rate.

Erik