We are currently implementing a LIS2DE12 accelerometer into a project, but we have issues where they sometimes get "stuck" and continue reporting back the exact same values from when it was in motion.
Current implementation uses 1hz refresh-rate, using int1 for triggering readout. Interrupts keep triggering as usual @1hz when getting "stuck".
We have indications that this may be more frequent when used in situations where it is subjected to higher g-forces (attached to a ground-compactor) but i don't see us even getting close to the 3000g for 0.5s that the device is rated for.
We have not ruled out issues with the PCB, but it is strange that the reads returns the same values over and over when it gets "stuck" in this mode. A power-cycle does seem to get it "unstuck" sometimes but not always.
Are there any erratas on the LIS2DE12 line perhaps?
