Magnetometer outputs -1 often

[ntlhui]

Seen prior to v2.0.0.3 on SN0001

[ntlhui]

v2.0.0.4 on SN0005 shows reasonable magnetometer values

[ntlhui]

Test plan: test v2.0.0.4 on SN0001 and SN0003

[ntlhui]

Using the monitor Sensors command on SN0001 and SN0003, I see mostly ones. This suggests hardware failure.

[ntlhui]

scoping the bus does not show evidence of HW fail. Failure appears to be the magnetometer sampling is timing out. Need to review magnetometer configuration. Punting to future release.

[ntlhui]

Reference https://www.y-ic.es/datasheet/78/SMDSW.020-2OZ.pdf

[ntlhui]

I'm very confused. The firmware specifies using the AK09916 magnetometer (verified from smartfin-fw). However, we don't seem to actually have an AK09916 on the PCB!

Even more odd, smartfin-fw2 at 59ea863814c4e942ca6c62b23f18b761c4d74a9c seems to not indicate I2C addressing issues - that is, there is a device on the I2C bus that acknowledges to the 0x0C address, has the right WHOAMI value at the correct address, and properly passes the self-test checks!

@SUPScientist do you have any insight into the history of including the AK09916?

[ntlhui]

@SUPScientist disregard, ICM-20948 includes an AK09916

[ntlhui]

Reference https://invensense.tdk.com/wp-content/uploads/2016/06/DS-000189-ICM-20948-v1.3.pdf

[ntlhui]

Still results in occasional outputs of -1

Build: de901b06cdee19d4a2690515ba77064299a87c0f

Sfin: 5d0036001750483553353920

In Monitor Sensors

FLOG indicates timeout

[agard111]

The 4G fin is now also affected by the monitor sensors magnetometer -1 issue. Only difference from before is that the fin battery is dead (however, fin is fully operational when charging) Using fw v2.0.0.11

[ntlhui]

Determined to be hardware issue