Hi, We are experiencing "glitches" on the ACcelerometer values once in a while. we are using 52Hz rate. expected values are 1000mg +-100mg on the Z Axis. Once in a while we see several samples going down significantly.

[Ytepp.1]

Hi,

We have been using LSM6DSOX in our Pool Alarm Unit.

The Device is floating horizontaly in the pool and valid Z axis Acceleration is around 1000mg.

We are currently facing a problem that is caused by seldom "glitches" on the Accelerometer values.

Device is configured at 52Hz for Both Acc and Gyro.

Gyro values seem not to be effected by this glitch.

Also, we record video of the device, and there is no external event at the time of the glitch.

After 6 "glitch" samples, the sample are normal again.

Attached is the excel with the 6 axis logging.

You can see the "glitch" samples in yellow.

Please advise what can cause this,

Thanks,

yuval,

Lifebuoy.

[Eleon BORLINI]

Hi yuval @Ytepp.1​ ,

you probably posted a log without the related graph, but ok I can see a "glitch" there.

I'm not sure however this is a "glitch", or at least an electrical glitch, since this is not a point spike but it has somehow a dynamic, lasting at least 3 ODR points.

I'm wondering if it could be related to an acquisition issue, I mean the reconstruction of the signal and the LSB to physical units conversion: are you using you custom board? If so, how did you configured the sensor? For example, did you set the BDU bit to guarantee the (see for example lsm6dsox_read_data_polling.c)? Or did you use the FIFO to acquire data?

Moreover, from an application point of view, is the board waterproof, so that there is no interaction of the water with -for example- the Vdd / VddIO lines, where some interferences could cause a voltage drop on the line (that's strange since the gyro should be ok, as you say)?

I would however prefer explaining the issue referring to a mechanical event: did you monitor the board with an high-speed camera (high frame rate)?

-Eleon