Px4-autopilot: QGC _altitude issues_CUAV Nano

[Tony3dr]

Describe the bug
We have noticed the altitude issues on the CUAV Nano this issue was noted on QGC altitude reading. (and relative altitude) not being displayed correctly.

To Reproduce
Steps to reproduce the behavior:

1. Drone switched on
2. Uploaded mission
3. Takeoff, Position Mode
4. Switch from Position mode to Mission mode
5. See altitude reading on QGC

Log Files and Screenshots
Log Nano V5:
https://review.px4.io/plot_app?log=e3d7bc0a-891d-4ecf-8ebd-01afb7630ab7

Drone (please complete the following information):
FC- CUAV Nano
Frame: F450 frame
GPS- NEO V2

Additional context
The vehicle flight performance is well but the altitude is off by approximately 10 to 13 meters.@dagar what could be the issue?

[MaEtUgR]

I have multiple CUAV V5 nano board here so I might be able to reproduce quickly.

[Tony3dr]

I have multiple CUAV V5 nano board here so I might be able to reproduce quickly.
@MaEtUgR were you able to reproduce the issue? Thanks

[cuhome]

I tested it, but I didn't find the height of the error.
Logs:
Https://logs.px4.io/plot_app?log=19572b33-284d-457d-8028-a792349e8641

[cuhome]

[MaEtUgR]

Iwasn't able to reproduce on the CUAV V5 nano but I have a similar problem on a Pixhawk 4 mini setup. It was flying perfectly before and without any change at some point it began reading strange diverging altitude even when stationary on the desk and with freshly calibrated accelerometer.

I made a log from boot today with latest master when rebooting just after calibration: https://logs.px4.io/plot_app?log=d3d0d46b-b850-4225-97a6-812aee45ff56

[MaEtUgR]

I quickly checked flashing 1.9.0 and I have the same problem. Let's analyze the log.

[MaEtUgR]

Logs for comparison from the testing team's setup (thanks @Tony3dr !):
https://review.px4.io/plot_app?log=d8b837a4-1881-4895-b9b1-1df9008b5237
https://review.px4.io/plot_app?log=a5f8141a-6991-4217-8520-426b533bfff2

[MaEtUgR]

Thanks to @kamilritz for looking into the logs. He found that the malicious drifting altitude in the tests on the bench come directly from the barometer... I wonder why certain boards have this a lot more than others. We should test if it's temperature dependent or see if the data sheet mentions some compensation.

[stale[bot]]

This issue has been automatically marked as stale because it has not had recent activity. Thank you for your contributions.

[MaEtUgR]

Is this problem still present?

I'd assume this was fixed in the meantime because there were various related issues addressed e.g. accelerometer bias drifting https://github.com/PX4/ecl/pull/749, fusing height multiple times https://github.com/PX4/ecl/pull/767 and a lot more that I don't know of. The platform Holybro QAV250 with pixhawk 4 mini that I had problems with doesn't show them anymore.

Please reopen if you see it again.

[Tony3dr]

Is this problem still present?

I'd assume this was fixed in the meantime because there were various related issues addressed e.g. accelerometer bias drifting PX4/ecl#749, fusing height multiple times PX4/ecl#767 and a lot more that I don't know of. The platform Holybro QAV250 with pixhawk 4 mini that I had problems with doesn't show them anymore.

Please reopen if you see it again.

@MaEtUgR the issue can be closed, we haven't seen this issue anymore. Thanks for all the help!