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

Mavic acted like a runaway horse and I had no control

RickJW

Active Member
Joined
Jan 11, 2019
Messages
28
Reactions
19
Age
84
GPS mode changed to Atti and message came up (app error). Ac kept going *til it hit a tree. Bounced around in the branches until it came free and was hovering about 4’ off the ground. I was able to fly it into the clear and land I. 3 props destroyed . Now I have to find out why I lost control.
Any ideas?
Rick
 
hope these go thru.
 

Attachments

Need a little help here. You posted the logs for 11 flights. Can you be a bit more specific as to which one had the problem? It doesn't appear to be the last one.
Because I have no idea what I'm doing with this drone stuff, I wouldn't know how to recognize which flight it was. Maybe from the 3rd to the 7th, but that is just a guess.
I was inadvertently connected to wifi. It connected automatically to daughters router, and an update might have been trying to upload to the ipad while the AC was in the air. Could that cause the problem?
 
Because I have no idea what I'm doing with this drone stuff, I wouldn't know how to recognize which flight it was. Maybe from the 3rd to the 7th, but that is just a guess.
I was inadvertently connected to wifi. It connected automatically to daughters router, and an update might have been trying to upload to the ipad while the AC was in the air. Could that cause the problem?

Okay - the flight in question was DJIFlightRecord_2019-02-02_[13-19-16].txt, and the problem was caused by yaw errors immediately after takeoff, leading to incorrect attempts by the FC to hold position and then a switch to ATTI when control failed. It was likely due to magnetic interference on startup, and it looks like the yaw was initialized at 56° when the aircraft was actually pointing a little west of north. The mobile device DAT file (FLY107.DAT) would be needed to confirm that.

This shows the discrepancy between the track that the aircraft followed and the track that the IMU was estimating:

imu_gps.png
 
Nope, don't re
Okay - the flight in question was DJIFlightRecord_2019-02-02_[13-19-16].txt, and the problem was caused by yaw errors immediately after takeoff, leading to incorrect attempts by the FC to hold position and then a switch to ATTI when control failed. It was likely due to magnetic interference on startup, and it looks like the yaw was initialized at 56° when the aircraft was actually pointing a little west of north. The mobile device DAT file (FLY107.DAT) would be needed to confirm that.

This shows the discrepancy between the track that the aircraft followed and the track that the IMU was estimating:

View attachment 61683
OK, that makes sense to me now. There was a chain link fence near the take off point. But, shouldn't I have gotten a warning?
 
  • Like
Reactions: Wangodetango
Nope, don't re

OK, that makes sense to me now. There was a chain link fence near the take off point. But, shouldn't I have gotten a warning?

Unfortunately - not necessarily. The problem is that when the aircraft is powered up its only information on its heading (i.e. the direction it's facing) is from the compass. If the earth's magnetic field is locally distorted by ferromagnetic material then, unless the magnetic field is clearly wrong, the FC will accept the indicated heading and initialize the IMU yaw value, and it has no way to know that's not the correct value.

The important test, which is not suggested anywhere by DJI, is to check that the aircraft orientation arrow on the map is pointing in the same direction, relative to north, that the aircraft is actually facing. I try to start the aircraft exactly facing one of the cardinal directions (use your phone compass to align) so that it's easy to check on the map.
 
Unfortunately - not necessarily. The problem is that when the aircraft is powered up its only information on its heading (i.e. the direction it's facing) is from the compass. If the earth's magnetic field is locally distorted by ferromagnetic material then, unless the magnetic field is clearly wrong, the FC will accept the indicated heading and initialize the IMU yaw value, and it has no way to know that's not the correct value.

The important test, which is not suggested anywhere by DJI, is to check that the aircraft orientation arrow on the map is pointing in the same direction, relative to north, that the aircraft is actually facing. I try to start the aircraft exactly facing one of the cardinal directions (use your phone compass to align) so that it's easy to check on the map.
What a great forum to get expert help and advice.
I thank you sir for taking your time to assess and answer my questions so directly and to the point. Thanks,
Rick
 
"The important test, which is not suggested anywhere by DJI, is to check that the aircraft orientation arrow on the map is pointing in the same direction, relative to north, that the aircraft is actually facing. I try to start the aircraft exactly facing one of the cardinal directions (use your phone compass to align) so that it's easy to check on the map. "

I'm adding that to my pre-flight checklist right now. Thanks.
 
Never even thought about s pre-flight check list. Just been so excited to fire up and take off. After 2 crashes and 2 sets of props destroyed, I believe it's time to fly like a real pilot. My AC is more than a toy.
 
I've only had the Air for a short while, but I have a lot of experience flying freestyle quads. Flying the Air is easier, but learning a lot of sophisticated technology that comes with it, and may jump in to "help" you is a challenge. I slightly modified this from recommendations on this forum - the original I think came from someone named "Alan (or Adam) the Beast".
 

Attachments

  • Like
Reactions: IgorD2

DJI Drone Deals

New Threads

Forum statistics

Threads
134,669
Messages
1,597,369
Members
163,149
Latest member
strudladi
Want to Remove this Ad? Simply login or create a free account