MM: Total loss of control - Help appreciated

[Tobias_MavicMini]

Hey guys,

I recently got a DJI Mavic Mini drone and I really enjoy flying with it, doing videos and pictures. I also try to learn a lot by reading your articles and watch videos / tutorials. So far also by flying very cautiously, everything went really well.

However, a few days ago I had a really serious drone malfunction: I flew on a beach and suddenly my drone went nuts and was completely out of control for about 2-3 minutes (which felt a lot longer). It sped up and flew some kind of curve at a very high speed and didn’t react to any commands I tried. Luckily, after 2-3 minutes I regained control and could land it safely again.

Can you see / guess / analyse / help me explaining what has happened here? I am actually quiet concerned about my next flight as long as I don’t understand what has caused the problem.

Attached you’ll find the data concerning the flight (unfortunately I cannot retrieve the .dat file out of the drone, cause it doesn’t show up in the dji assistant app running on my mac).

Here are some knowledge bits you might need:

• Loss of control / fly away starts at time stamp 5m 23s
• The compass data shows that the drone yawed by about 180° - however, the video shows that this yaw did not happen
• The drone did not react to any commands, although connection is good and there were no errors shown
• Firmware was 1.0*3*.** (not the newest one, I know / already updated)
• I started from a parking lot ca. 5m away from cars on a sidewalk
• There are no bridges / big metal constructions at the beach
• 5 minutes earlier, I did the exact same flight with no problems at all (there was once a short interference warning); however no problems with flying at all. Warning was gone after one second

If you need any more information please let me know. I would love any hints / comments / help regarding what has happened here and what I can do in future to prevent such a mishap.

Link to drone video: temp_video_for_share.mp4


Thank you very much and best regards
Toby

 

[MAvic_South_Oz]

DJI Flight Log Viewer | Phantom Help

Upload and view flight logs from your DJI drones

www.phantomhelp.com

Glad you got it back !
This will be interesting, there were a lot of weird stick and flight contrasts in various places.

That is an amazing part of Australia too, Lucky Bay it was for you.

In your opinion was there any wind that would be of concern ?
I know from travel through there it is almost always blowing pretty hard.
Glad you kept it to under 100', that's sensible.

 

[Meta4]

Can you see / guess / analyse / help me explaining what has happened here? I am actually quiet concerned about my next flight as long as I don’t understand what has caused the problem.

Congratulations .. you had a yaw error and your drone survived.
Most aren't so lucky.

• Loss of control / fly away starts at time stamp 5m 23s
• The compass data shows that the drone yawed by about 180° - however, the video shows that this yaw did not happen
• The drone did not react to any commands, although connection is good and there were no errors shown
• Firmware was 1.0*3*.** (not the newest one, I know / already updated)
• I started from a parking lot ca. 5m away from cars on a sidewalk
• There are no bridges / big metal constructions at the beach
• 5 minutes earlier, I did the exact same flight with no problems at all (there was once a short interference warning); however no problems with flying at all. Warning was gone after one second

Classic symptoms of a yaw error
Launching from the concrete sidewalk was probably the cause
It's best to avoid reinforced concrete surfaces for launching, it's a common cause of these problems.
It doesn't happen every time because of the variations in the positioning of your drone relative to steel reinforcing in the concrete.

You can confirm that the drone is safe to fly by checking the drone icon in the map view is oriented correctly with respect to compass directions.
If the drone is pointing north but the icon points east, you have a problem and should power off and move away from the problem and start again.

 

[Deleted member 94047]

unfortunately I cannot retrieve the .dat file out of the drone, cause it doesn’t show up in the dji assistant app running on my mac

The .DAT file you want is stored in the mobile device and not the Aircraft. See this post by sar104 for details

The DAT file naming convention, based on the date and time of the start of the file, is: YY-MM-DD-hr-min-sec_FLYXXX.DAT., where XXX is the flight recorder file index from the HOME_dataRecorderFileIndex field in the txt log.

These are retrieved by the same method as the TXT logs. Under both iOS and Android they are in a subfolder, MCDatFlightRecords, in the folder that contains the TXT logs.

 

[Tobias_MavicMini]

Hi guys!

Thank you so far for your posts! Just to add some information:

Wind: Probably the wind let to some of deviation of the drone at first but definitely did not lead to complete loss of control problem.

.dat File: Thanks for the info: Obviously the dat file was not saved on my iPhone either...


Are there any other methods to check if your compass is working correctly right after take-off?

Cheers
Toby

 

[Deleted member 94047]

Are there any other methods to check if your compass is working correctly right after take-off?

After take off, no. At least none I know of. But checking the orientation of the drone symbol (the little triangle representing your drone on the onscreen map) against the orientation of the actual drone itself before take off is a good way of catching issues with the compass. I usually place the drone on the ground such that it is pointing to a very recognizable feature of the area (a river, a big building, a road, a junction - whatever is in the area that I can quickly identify on a map) and then check to see if the drone symbol on the map is also pointing to the same.

 

[sar104]

Congratulations .. you had a yaw error and your drone survived.
Most aren't so lucky.

Classic symptoms of a yaw error
Launching from the concrete sidewalk was probably the cause
It's best to avoid reinforced concrete surfaces for launching, it's a common cause of these problems.
It doesn't happen every time because of the variations in the positioning of your drone relative to steel reinforcing in the concrete.

You can confirm that the drone is safe to fly by checking the drone icon in the map view is oriented correctly with respect to compass directions.
If the drone is pointing north but the icon points east, you have a problem and should power off and move away from the problem and start again.

This was definitely caused by a compass problem, but it was not caused by magnetic interference at the takeoff point. The initial IMU yaw at takeoff was 73°, which corresponds exactly with the view at takeoff and the initial flight track. The problem arose at 80 seconds when the IMU mysteriously flipped the yaw value by 190° in 3 seconds, leading to immediate large disagreements between the IMU velocity and position.





Unfortunately there is no way to diagnose this further without the mobile device DAT file ending in FLY018.DAT..

@Tobias_MavicMini - are you quite sure that it's not on the phone?

 

[Prismatic]

This was definitely caused by a compass problem, but it was not caused by magnetic interference at the takeoff point. The initial IMU yaw at takeoff was 73°, which corresponds exactly with the view at takeoff and the initial flight track. The problem arose at 80 seconds when the IMU mysteriously flipped the yaw value by 190° in 3 seconds, leading to immediate large disagreements between the IMU velocity and position.

View attachment 92675

View attachment 92676

Unfortunately there is no way to diagnose this further without the mobile device DAT file ending in FLY018.DAT..

@Tobias_MavicMini - are you quite sure that it's not on the phone?

With what you've been able to show with even limited data, this looks like a major system failure. The drone should be replaced under warranty.

 

[sar104]

With what you've been able to show with even limited data, this looks like a major system failure. The drone should be replaced under warranty.

It's not an issue that I've seen before, at least not quite like that. My suspicion would be a problem with the z-axis rate gyro but, as I mentioned, without the DAT file there is no way to confirm that.

 

[PhantomFandom]

Once again @sar104 I must thank you and commend you for not only your expertise, but also for your time and dedication here. You do us all a great service!! BRAVO !!

This was definitely caused by a compass problem, but it was not caused by magnetic interference at the takeoff point. The initial IMU yaw at takeoff was 73°, which corresponds exactly with the view at takeoff and the initial flight track. The problem arose at 80 seconds when the IMU mysteriously flipped the yaw value by 190° in 3 seconds, leading to immediate large disagreements between the IMU velocity and position.

View attachment 92675

View attachment 92676

Unfortunately there is no way to diagnose this further without the mobile device DAT file ending in FLY018.DAT..

@Tobias_MavicMini - are you quite sure that it's not on the phone?

 

[Tobias_MavicMini]

Unfortunately there is no way to diagnose this further without the mobile device DAT file ending in FLY018.DAT..

@Tobias_MavicMini - are you quite sure that it's not on the phone?

Hi sar104,
hi community,

thank you very much for your analysis - it is highly appreciated! Unfortunately there is not a single .dat file on my phone... (checked it twice). Unless there is another way to get that data, it cannot be provided.

What would you recommend doing now?

Thanks again and best regards!
Toby

 

[sar104]

Hi sar104,
hi community,

thank you very much for your analysis - it is highly appreciated! Unfortunately there is not a single .dat file on my phone... (checked it twice). Unless there is another way to get that data, it cannot be provided.

What would you recommend doing now?

Thanks again and best regards!
Toby

So just to clarify - DJI Fly » FlightRecords » MCDatFlightRecords exists, but it is empty?

 

[Tobias_MavicMini]

So just to clarify - DJI Fly » FlightRecords » MCDatFlightRecords exists, but it is empty?

Yes, the folder exists, but it is empty...

 

[m80116]

Have you considered you might see it empty because a .dat file isn't exactly user friendly ? It might be there but hidden from view... (just speculating... I don't come from the Apple side).

 

[Tobias_MavicMini]

Have you considered you might see it empty because a .dat file isn't exactly user friendly ? It might be there but hidden from view... (just speculating... I don't come from the Apple side).

It's not hidden or anything. It is just empty (0 bytes folder size)

 

[Bbarr]

It's not hidden or anything. It is just empty (0 bytes folder size)

I have the same problem . Experimenting,and it seems the dat files are created,but disappear when you sync the .txt files.
I don’t believe the dat files are synced. Thus you can create dat files if you don’t start the motors.
But have a txt file and a dat file. The dat file will disappear when you sync the txt file.

 

[Bbarr]

The dat file id is contained in the txt file . It is my hypothesis that when you sync ,it reads that id and deletes it. ..