Strange behavior on IMU.GyrX and IMU.GyrY

Archive
1

Hi all,

Following my issues described here, I made some test with the quad standing still on the bench and pre-arm logging enabled.

I’m seeing strange behaviors for IMU.GyrX and IMU.GyrY that manifest into huge HUD drifts. IMU.GyrZ, IMU2.GyrX, IMU2.GyrY and IMU2.GyrZ look good.

Here is what I see at random times. This are pre-arm data with the quad standing still on the table:

bad-04.png800×438 49.3 KB

bad-08.png800×438 50.6 KB

When this happens, the HUD drift a lot (up to more than 45°) and I think that flighting in these situations will not be a good idea.

Sometimes, there is no suddenly changes in the values but a (fast?) regular drift of the values:

not-too-bad.png800×438 56.4 KB

In this case, the HUD drift only a bit (few degrees) and I think in this case the software is able to adapt to the bias.

Do you also see this kind of drift in IMU.Gyr? Is this normal? Do I have bad hardware (it’s an ebay clone)? Thank you

My logs are available here.

Regards,
Cyril

2

Anyone seeing similar behavior or is this just me and a faulty hardware?

3

So replying to my own post. I manage to replace the MPU-6000 on my Pixhawk (actually my brother did it, he has all the required tooling at work). Since then, everything is fine :sunglasses: No more gyro drift and leans during flight. Both IMU report consistent data.