We have a LIS3DH accelerometer that have ended up in a latch-up state. The output data is -205/-1149/-4759 (X/Y/Z in mg). When performing the self test, the absolute difference between the 2 samples are 7/6/3 (X/Y/Z in mg) which is well below the min limit (> 60mg). Is there any way to recover from this? Do I have to make the accelerometer powerless?