DJI Mavic, Air and Mini Drones
Friendly, Helpful & Knowledgeable Community
Join Us Now

Mavic mini 2 flyaway and crash

mattvenn

Member
Joined
Jul 20, 2021
Messages
20
Reactions
27
Age
47
Location
Spain
Hi all, new member and mavic owner here. I've flown 250s for a few years, always 2d and with minimal gyro assistance so this is all amazing to me.
I got the mavic mini 2 drone yesterday and flew about 10 times. Very impressive.

Today I tried another flight and was hovering about 6m up, full battery, about 10m away from me, almost no wind.
The drone started moving on its own and a message on the screen said a motor was blocked. It crashed into a tree about 50m away. I was able to retrieve it undamaged.

I enabled the emergency off double stick motion and tried a take off at a safe distance. It seems totally fine! So now I don't know what to do, I certainly don't feel safe flying this anymore.

Any advice? I've already contacted dji but wanted to see what people with more experience of this drone thought.

Thanks for any guidance,
Matt
 
Today I tried another flight and was hovering about 6m up, full battery, about 10m away from me, almost no wind.
The drone started moving on its own and a message on the screen said a motor was blocked. It crashed into a tree about 50m away. I was able to retrieve it undamaged.
The recorded flight data is always the first place to look to explain flight incidents.
And it would also help if you could describe the surface you launched from.

For flight data ....
Go to DJI Flight Log Viewer | Phantom Help
Follow the instructions there to upload your flight record from your phone or tablet.
That will give you a detailed report on the flight data.
Come back and post a link to the report it provides and someone might be able to analyse it and give you an understanding of the cause of the incident.
 
Thanks Meta4 for the guidance.

Launched from a flat concrete pad. Are you asking because one possibility is a bit of sand or something got stuck in the motor at takeoff? I could understand that if it crashed straight away, but it was about 40 seconds in. I just posted the video in case that gives any clues. Interesting to me was that it's a relatively slow crash. On my 250, if something breaks, like a motor cable, then it's usually much more spectacular!


And here's the log:


That viewer is cool. I'd say the moment before it flew away was at 0:46, and I noticed and tried to correct with the right stick at 0:51. Then I think I hit the break/pause button (doesn't seem to be shown in the log).

Thanks,
Matt
 
Launched from a flat concrete pad. Are you asking because one possibility is a bit of sand or something got stuck in the motor at takeoff?
I was asking because launching from reinforced concrete surfaces is a common cause of incidents where the drone appears to run away without any control input.
The Phantomhelp log viewer shows the drone pointing to the north at launch.
Is that direction correct or was it pointing somewhere else?

 
  • Like
Reactions: MARK (LI)
I’m no expert at this, but it seems like the compass might’ve not been calibrated accurately. Look at the heading of the aircraft and compare that your video. Unless you adjusted the video after the fact, the aircraft seems to think it was pointing a different direction than it actually was. I know taking off on concrete can do odd things because of the rebar within it.

E2613393-9A35-4DE9-B23F-CC0B91919A2C.jpeg
 
@skeixnamelxlaldxniejs is right, it wasn't pointing north, it was pointing towards the tallest leaf of the marijuana leaf (see linked video in my post above). So yes, seems compass was wrong.

I'll see what DJI says, but this is good news because it seems like something I can make sure doesn't happen in the future rather than a random failure. Out of interest, is there any official info about this? I don't remember reading anything in the guides.

I suppose a good pre-flight check is making sure the compass is correct? If I started it on grass, but took off from concrete is that OK?

Thanks for the help,
Matt
 
  • Like
Reactions: maggior
@skeixnamelxlaldxniejs is right, it wasn't pointing north, it was pointing towards the tallest leaf of the marijuana leaf (see linked video in my post above). So yes, seems compass was wrong.

I'll see what DJI says, but this is good news because it seems like something I can make sure doesn't happen in the future rather than a random failure. Out of interest, is there any official info about this? I don't remember reading anything in the guides.

I suppose a good pre-flight check is making sure the compass is correct? If I started it on grass, but took off from concrete is that OK?

Thanks for the help,
Matt
It’s not a compass calibration problem, but it is a compass error. The aircraft heading is tracked primarily by the IMU rate gyros, which are initialized by the magnetic compass on power up. This problem arises when the aircraft is powered up at a location where the earth’s magnetic field is distorted by an external magnetic field - in this case almost certainly magnetised rebar in the concrete. So the IMU thought that the aircraft was facing NW, when in fact it was facing south at takeoff.

As a result, after takeoff, when it was pushed slightly towards the west by the wind, it corrected by applying right thrust, which would have been correct if it had actually been facing NW. But since it was facing south it exacerbated the drift and pushed it faster to the west. That’s the expected mostly linear flight error for a heading error of the order of 180°.

The guaranteed way to avoid this problem is to check the aircraft heading arrow on the map before takeoff, and make sure that it is pointing in the same direction as the aircraft.

The thread linked below explains this in a bit more detail:

 
@skeixnamelxlaldxniejs is right, it wasn't pointing north, it was pointing towards the tallest leaf of the marijuana leaf (see linked video in my post above). So yes, seems compass was wrong.
The problem isn't that the compass was malfunctioning.
It was deflected by the invisible steel under the concrete to give a false reading.
The big problem with this is that on startup the gyro sensors take their initialisation value from the compass.
Straight after launch, the compass and gyro sensor are both showing the same direction, but as the drone gets further from the influence of the steelwork, the compass returns to showing the correct direction.
The drone detects that it is off course and is attempting to correct but each correction puts it further off course.
Within seconds, it is flying at crazy speed.
 
It’s not a compass calibration problem, but it is a compass error. The aircraft heading is tracked primarily by the IMU rate gyros, which are initialized by the magnetic compass on power up. This problem arises when the aircraft is powered up at a location where the earth’s magnetic field is distorted by an external magnetic field - in this case almost certainly magnetised rebar in the concrete. So the IMU thought that the aircraft was facing NW, when in fact it was facing south at takeoff.

As a result, after takeoff, when it was pushed slightly towards the west by the wind, it corrected by applying right thrust, which would have been correct if it had actually been facing NW. But since it was facing south it exacerbated the drift and pushed it faster to the west. That’s the expected mostly linear flight error for a heading error of the order of 180°.

The guaranteed way to avoid this problem is to check the aircraft heading arrow on the map before takeoff, and make sure that it is pointing in the same direction as the aircraft.

The thread linked below explains this in a bit more detail:

just read that post - very useful info, thanks a lot.
 
Lycus Tech Mavic Air 3 Case

DJI Drone Deals

New Threads

Forum statistics

Threads
130,988
Messages
1,558,677
Members
159,981
Latest member
bbj5143