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

First Crash - Mavic 2 "flipped out"

KennyL

Active Member
Joined
Sep 30, 2019
Messages
38
Reactions
13
Age
54
Location
southern indiana
So after a few months of flying without incident - -
Was flying at dusk - brought it back using return home..
Mavic 2 was overhead at 120' and began descending.
I was planning on landing on my deck and noticed the drone was off course about 10' while descending.
Turned off return home and right after, battery warning alarm sounded (15%) -
was bringing it down manually - got to about 5' off the deck when Mavic 2 slowly started going forward on it's own........
Then.... Mavic flipped upside down and flew full speed right into the brick wall (missing me by a foot or two)
UPDATE - - The speed has been debunked - - guess my brain was on overload
Seemed like full speed but in fact only like 5mph
In my defense - it was only about 4' from the wall so it happened quickly
Flight recorder also showed that it did not flip until after it hit the wall

Mavic hit the ground upside down and continued to accelerate.
Had a rough time manually stopping.
Props destroyed - - Drone seems OK
NO IDEA WTH HAPPENED
UPDATE - - PILOT ERROR due to home position not set correctly and Low Battery RTH took over
 
Last edited:
Sorry to hear of the accident- definite bummer; glad the damage to it and other weren't too bad.

Odd it kept running after inverting- usually the motors shut off after something like a 90 degree roll angle. Definitely submit your log- there are some super smart folks who will analyze it.

Good luck and safe flying going forward-

Neil.
 
The recorded flight data usually helps to understand flight incidents.
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 of the flight.
Come back and post a link to the report it gives you.

Would highly appreciate your time
 

Attachments

  • DJIFlightRecord_2019-10-18_[19-36-46].txt
    1.8 MB · Views: 958
Sorry to hear of the accident- definite bummer; glad the damage to it and other weren't too bad.

Odd it kept running after inverting- usually the motors shut off after something like a 90 degree roll angle. Definitely submit your log- there are some super smart folks who will analyze it.

Good luck and safe flying going forward-

Neil.

Yep - - it was on it's back and props spinning at full speed
To be honest - - Scared the crap outta me
Just missed me - - That would have been ugly
 
Is that the file your talking about ???
Here's the story the data is telling me:

9:36 RTH initiated, drone is at 345 ft altitude and 190 ft from home
Drone flies toward a point above home at up to 20 mph and slows as it approaches its target.
At 9:49.1 drone is 0.9 ft from the RTH target point.
It rotates to face the way it was at launch and drifts to 2.4 ft off the target before descent starts at 9:59.4
During descent the drone slowly drifts to 7 feet off the target.
RTH Landing cancelled at 10:23.3 and 122ft height
Drone is brought down manually to 5ft AGL 18 ft from the launch point when Low Battery RTH is initiated at 10:58.9.
Drone turns toward the original launch point and flies toward home 18 feet away.
At 11:04.1 with the drone in RTH, the pilot applied joystick input with both joysticks to make the drone climb as well as move forward and change direction.
At 11:04.4 there is a warning: Upward Obstacle Detected
Joystick input was changed to descend and rotate anticlockwise.

VPS was indicating height of 6.9 ft AGL and there was a sudden drastic change in pitch, roll and yaw data at 11:04.7 indicating the drone collided with an object.
The drone pitched backward nearly 90 degrees and rolled upside down as it fell 6ft to the ground.

It looks like the issue was due to collision with an obstacle close to the point the drone was trying to RTH to.
An additional factor was the pilot not realising Low Battery RTH was operating and fighting against this with the joysticks.
 
Here's the story the data is telling me:

9:36 RTH initiated, drone is at 345 ft altitude and 190 ft from home
Drone flies toward a point above home at up to 20 mph and slows as it approaches its target.
At 9:49.1 drone is 0.9 ft from the RTH target point.
It rotates to face the way it was at launch and drifts to 2.4 ft off the target before descent starts at 9:59.4
During descent the drone slowly drifts to 7 feet off the target.
RTH Landing cancelled at 10:23.3 and 122ft height
Drone is brought down manually to 5ft AGL 18 ft from the launch point when Low Battery RTH is initiated at 10:58.9.
Drone turns toward the original launch point and flies toward home 18 feet away.
At 11:04.1 with the drone in RTH, the pilot applied joystick input with both joysticks to make the drone climb as well as move forward and change direction.
At 11:04.4 there is a warning: Upward Obstacle Detected
Joystick input was changed to descend and rotate anticlockwise.

VPS was indicating height of 6.9 ft AGL and there was a sudden drastic change in pitch, roll and yaw data at 11:04.7 indicating the drone collided with an object.
The drone pitched backward nearly 90 degrees and rolled upside down as it fell 6ft to the ground.

It looks like the issue was due to collision with an obstacle close to the point the drone was trying to RTH to.
An additional factor was the pilot not realising Low Battery RTH was operating and fighting against this with the joysticks.

That's what has me confused
The joystick for decent was slightly down only when the thing went bat crap crazy on me
wonder if there was some sort of interference ???

Can you elaborate on the low battery rth fighting - - I thought the rth was turned off and i was manually landing
did I have a "brain fart" ?? LOL
What do you mean by fighting with it ?
Was it trying to return on it's own since the battery was low while I was thinking it was still in manual landing mode ?

Thanks so much for this info

How did you convert all that code stuff into readable information ?
 
That's what has me confused
The joystick for decent was slightly down only when the thing went bat crap crazy on me
wonder if there was some sort of interference ???
Interference doesn't make the drone do crazy things.
Can you elaborate on the low battery rth fighting - - I thought the rth was turned off and i was manually landing
did I have a "brain fart" ?? LOL
What do you mean by fighting with it ?
Was it trying to return on it's own since the battery was low while I was thinking it was still in manual landing mode ?
Low Battery RTH was initiated at 10:58.9 and you probably didn't notice it.
The drone was trying to RTH and you were trying to fight against it to do something else.
What did the drone collide with at 11:04.7?

How did you convert all that code stuff into readable information ?
The Phantomhelp log viewer did it. Thanks @msinger for creating such a useful tool.
You could do it yourself with the instructions in post #2
To see the real data that I analysed, click on the Download CSV link:
 
Interference doesn't make the drone do crazy things.

Low Battery RTH was initiated at 10:58.9 and you probably didn't notice it.
The drone was trying to RTH and you were trying to fight against it to do something else.
What did the drone collide with at 11:04.7?


The Phantomhelp log viewer did it. Thanks @msinger for creating such a useful tool.
You could do it yourself with the instructions in post #2
To see the real data that I analysed, click on the Download CSV link:

Slammed a brick wall at what seemed to me to be full speed
Scared the _____ out of me
Just missed me
After hitting the wall - it fell on it's back (upside down) and the props sounded to be at full stroke / full speed
Props did not stop - - had to do an e-stop on them
 
Last edited:
Slammed a brick wall at what seemed to me to be full speed
I'm not home to check now but the speed would be shown in the data.
I think it would be fairly slow.
It's always a good idea to launch in an area clear of obstacles to prevent issues with a potential RTH.
 
Yes ... collision speed was about 4 mph
 
Last edited:
Getting worried.
I see more M2P crashes and fly aways
every day. Might be a planned adsolance?
Most I see in this forum are a result of operator errors rather than some type of conspiratorial DJI planned obsolescence.
 
  • Like
Reactions: Johnmcl7 and Gindra
Would highly appreciate your time

To add some additional detail to @Meta4's explanation, this was one of those cases where a couple of subtle factors interacted to produce an unwanted behavior.

The first issue was that you launched without waiting for a good enough position lock - only 9 satellites and a NavHealth of 4/5. That resulted in a home point, recorded with inaccurate position data, that was around 15 ft NW of the actual launch point.

gps.png

That's why, on RTH, it was descending to the wrong location. You cancelled RTH and brought it back manually to the correct location, but in doing so moved it away from the recorded home point, which caused the smart battery GoHome level to increase. That triggered smart battery RTH because the battery was low:

Battery.png

As a result, at 658 seconds it turned to the NW to face the incorrectly recorded home point and flew into the wall of your house:

Attitude.png

FYI - impact velocity was 4.5 mph, heading NW.
 
  • Like
Reactions: Meta4
Getting worried.
I see more M2P crashes and fly aways every day. Might be a planned adsolance?
I'm not worried.
Looking at the flight data from the same incidents, gives me confidence in the reliability of these machines.
Here's an interesting perspective on the issue to give you something to think about:
 
Another reason not to take off before getting a GPS lock.
There's a reason the lady says "homepoint has been updated, please check it on the map".
 
- - it was on it's back and props spinning at full speed
To be honest - - Scared the crap outta me
Just missed me - - That would have been ugly
OMG, it's begun...The Drone.

Sorry KennyL, I couldn't resist. Hope you have a sense of humor.
 
It probably was going full speed trying to right itself. My P3 would be stubborn to respond to CSC in such situations.
 
Lycus Tech Mavic Air 3 Case

DJI Drone Deals

New Threads

Forum statistics

Threads
130,585
Messages
1,554,095
Members
159,585
Latest member
maniac2000