MPU6050: Self test acc Y [FAIL] - motor problem?

[tobias]

I've notices it is often the Y axis on the accelerometer that breaks (left and right movement of the Crazyflie). Since there are actually moving parts in the silicon I read that when it breaks it is often because it gets stuck in an end position. An interesting test would be to bang the the Crazyflie on the M2, M4 motor side and see if it possible to get it working again. Preferable without the motors and mounts attached, but doing it with should also work, with the risk of damaging the motors.

[leland]

Hello, interesting update/idea.

Before packing up the CF for its return to China for repair, I did in fact tap (rather than bang) the assembled CF on all 4 arms out of curiosity, but it did not help.

Too late now for me to try to 'bang' instead, will be curious to hear if this works for others.

The Seeedstudio Customer Service folks were very prompt and helpful in their replies, and the CF began its journey 'home' yesterday.

Cheers, Leland

[stefan]

An interesting test would be to bang the the Crazyflie on the M2, M4 motor side and see if it possible to get it working again.

Thanks Tobias, this solution worked at least for me

I have detached the motor and mount from M2 and and banged the Crazyflie a couple of times on the table and after restarting, the self tests passed again!
So for everyone having this problem, I recommend trying the solution described by Tobias.

[foosel]

So... sometimes violence is the solution Too bad I already sent mine away, I'd really love to try that right now

[tobias]

Thanks Tobias, this solution worked at least for me

Wow, didn't really expect it to work, cool!
Can you also report back if it is flying as before and especially when hovering?

[owolf]

Tapping/banging worked for me, too. Flies as before, with hovering still being somewhat difficult.

Oliver

[stefan]

I would also say, that the flight behavior is like before. Hovering wasn't really easy from the beginning and there is still some drifting I cant't get under control. But I am still playing with the trim values for role and pitch...

Stefan

[seba]

Possibly have the same issue here....After power off-on sequence it won't start, flashes about 5 times in red all the time. Don't have CF here to check console now, but after reading this thread it can be broken gyro problem.

I managed to "repair" mine with big dose of luck and re-soldering/re-heating MPU6050 pads. Now it works as new - hope it will last

[idhorst]

I've returned the crzayflie to Seeed and they repaired the IMU. It arrived today. The selft test now runs succesfully. Hopefully I can get to play with it real soon.

[farcop]

+1

It seems there is no any chance to bring my Crazyflie to life

SYS: Crazyflie is up and running!
SYS: Build 31:ba1d111112ad (2013.4) CLEAN
SYS: I am 0x8712482054568549674FF50 and I have 128KB of flash!
IMU: MPU6050 I2C connection [OK].
IMU: HMC5883L I2C connection [FAIL].
IMU: MS5611 I2C connection [FAIL].
MPU6050: Self test acc Y [FAIL]. low: 0.300000, high: 0.949999, measured: 0.0

CF 6-DOF first batch