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

Compass error

There are a couple of other factors to consider when determining if close proximity to a ship will cause a "compass issue".

The geomagnetic field is distorted by the ship. Some locations more than others. If the AC is between the ship and magnetic north the field direction is probably pretty accurate. Off to either side the field is bent more.

The other factor is due to the delay that magnetometer data has before Yaw is affected. Within a couple of secs after power up magnetometer data is used to initialize the Yaw value. After that it's mostly IMU data that determines Yaw with magnetometer data being used at low gain - i.e., there will be a delay in any effect caused by magnetometer data. Landing on a manhole cover is OK as long as it's quick. Spend some time hovering over the cover and the incorrect magnetometer data will cause Yaw to be compromised leading to erratic flight.

From @Fufar Visual 's description I get the impression that he had been flying around for a while in a distorted geomagnetic field - long enough for Yaw to be compromised.

......
Try it yourself sometime by descending toward a steel roof.
As you get closer, maybe 6 feet or so above the roof, you'll see the drone start to wander a little or slowly spiral.
That's the compass being affected by the magnetic field in the steel interfering with the compass reading of the earth's normal field.
Lift the drone a few feet higher and it hovers properly, lower it and you see the effect.
........
I'm a little surprised by this. I would think it that it would take 30 secs or more before the magenetometer data could affect Yaw enough to see any erratic flight.
 
Hello Mavic specialists ! I have uploaded dat file from the day of the crash. Looking forward to your comments.
 

Attachments

  • 2019-04-05_10-39-22_FLY096.DAT
    3.2 MB · Views: 7
  • 2019-04-05_10-20-57_FLY095.DAT
    1 MB · Views: 7
  • 2019-04-05_10-15-27_FLY094.DAT
    99.7 KB · Views: 3
  • 2019-04-05_10-06-18_FLY093.DAT
    3.2 MB · Views: 7
Hello Mavic specialists ! I have uploaded dat file from the day of the crash. Looking forward to your comments.
This one is over my head. Best I can tell you have a GPS consistency issue where some the information regarding GPS location and height is conflicting. There seems to be a GPS problem of some kind from what I can tell but if it has to do with the magnetic interference or GPS reflections off the steel or something else I am not sure.

This is one for @sar104 and he's the only one I would trust to find an answer.
 
I know I have mentioned this before on another string, but something I have started to do as one of my preflight checks is to point the unit toward a known street. When you fire up the drone and get everything stabilized (recalibrate if necessary), check the map at full size to be sure the drone is pointing to the known street. At least you will know it is within several degrees of correct, and not like 90 degrees out due to compass magnetic interference.
 
I find the MA to be the only dji drone I have ever owned to be so fussy on compass calibrations. I have owned one of each phantom and the MP and spark and now have a M2z yet the only one that requires constant compass calibration is the MA. Plus it tends to veer off course at random times. I like the drone but am considering dumping it as it is unreliable.
 
HI
I just had a crash with Mavic Air in a shipyard,because I suddenly got the message 'compass error' (or similar) and completely lost control as the Magic flew into a newly painted supertanker in the dock ( the subject for my filming). Unfortunately I forgot to mount the propeller guards so the drone fell to the bottom of the concrete dock from about 8 m. I had previously at the same shipyard in the bottom of the dry dock had to re calibrate the compass several times. Anyone with similar experiences and a possible solution for the future ? kr
Torben
The metal hulls of the large ship's are what's causing your compass errors. Try launching from farther away from the ships and be sure to keep at least 40 feet of distance between your drone and the ship.
 
I always fly a mavic pro from a 86 foot fishing boat ( all steel )a few times saw magnetic field interference after taking off disappear never crash or loose gps maybe mavic air behavior is different. Only ones can’t get gps 100 miles from Newbedford on ocean
 
HI
I just had a crash with Mavic Air in a shipyard,because I suddenly got the message 'compass error' (or similar) and completely lost control as the Magic flew into a newly painted supertanker in the dock ( the subject for my filming). Unfortunately I forgot to mount the propeller guards so the drone fell to the bottom of the concrete dock from about 8 m. I had previously at the same shipyard in the bottom of the dry dock had to re calibrate the compass several times. Anyone with similar experiences and a possible solution for the future ? kr
Torben
Did the supertanker have a steel hull? Bet it did!
 
Sorry for my ignorance ! But here is the file attached.Appreciate your interest.
 

Attachments

  • DJIFlightRecord_2019-04-05_[10-10-59].txt
    494.1 KB · Views: 7
  • Like
Reactions: brett8883
I just had a crash with Mavic Air in a shipyard,because I suddenly got the message 'compass error' (or similar) and completely lost control as the Magic flew into a newly painted supertanker in the dock ( the subject for my filming). Unfortunately I forgot to mount the propeller guards so the drone fell to the bottom of the concrete dock from about 8 m. I had previously at the same shipyard in the bottom of the dry dock had to re calibrate the compass several times. Anyone with similar experiences and a possible solution for the future ?
Your description has very little detail and left people confused and guessing what really happened.
Here's what the flight data looks like: DJI Flight Log Viewer - PhantomHelp.com
It shows very poor GPS conditions for much of the flight with the drone down in the dry dock and the drone's sky view blocked by the walls of the dry dock and the tanker when you get in close to it.
GPS health varies between 4 and 1 and for some of the time, the drone would have had no position holding ability.
You only had good GPS conditions for a small part of the flight.

The data also shows you flying very slowly in Tripod Mode for the whole flight.
From around 3:06 the data shows a compass error that is present until the end of the data about one minute later.
It appears that for this part of the flight you were in close to the tanker (?) and the steady compass error was due to being so close to a very large amount of steel.
The GPS track shows unsteady wandering for this part of the flight.
Had you moved away from the ship, the compass error should have disappeared, but you stayed close.
You made no right stick inputs after 3:45 and only a little left stick input to vary the height.
Speeds for the last 40 seconds were less than 0.5 mph and the drone was just drifting/hovering.
The crash happens at 4:08.9 and the drone tumbles to the ground.

The site looks like a tricky flying site with so much steel and reinforced concrete making it hard to find a safe launch point.
Flying very close to large steel objects caused the compass error and with poor GPS conditions and a failure to keep the drone away from danger, a collision happened.
Here's Google Earth imagery to help anyone interested in understanding of the flying environment.
i-KVLb9GZ-X2.jpg
 
Your description has very little detail and left people confused and guessing what really happened.
Here's what the flight data looks like: DJI Flight Log Viewer - PhantomHelp.com
It shows very poor GPS conditions for much of the flight with the drone down in the dry dock and the drone's sky view blocked by the walls of the dry dock and the tanker when you get in close to it.
GPS health varies between 4 and 1 and for some of the time, the drone would have had no position holding ability.
You only had good GPS conditions for a small part of the flight.

The data also shows you flying very slowly in Tripod Mode for the whole flight.
From around 3:06 the data shows a compass error that is present until the end of the data about one minute later.
It appears that for this part of the flight you were in close to the tanker (?) and the steady compass error was due to being so close to a very large amount of steel.
The GPS track shows unsteady wandering for this part of the flight.
Had you moved away from the ship, the compass error should have disappeared, but you stayed close.
You made no right stick inputs after 3:45 and only a little left stick input to vary the height.
Speeds for the last 40 seconds were less than 0.5 mph and the drone was just drifting/hovering.
The crash happens at 4:08.9 and the drone tumbles to the ground.

The site looks like a tricky flying site with so much steel and reinforced concrete making it hard to find a safe launch point.
Flying very close to large steel objects caused the compass error and with poor GPS conditions and a failure to keep the drone away from danger, a collision happened.
Here's Google Earth imagery to help anyone interested in understanding of the flying environment.
i-KVLb9GZ-X2.jpg
Your description has very little detail and left people confused and guessing what really happened.
Here's what the flight data looks like: DJI Flight Log Viewer - PhantomHelp.com
It shows very poor GPS conditions for much of the flight with the drone down in the dry dock and the drone's sky view blocked by the walls of the dry dock and the tanker when you get in close to it.
GPS health varies between 4 and 1 and for some of the time, the drone would have had no position holding ability.
You only had good GPS conditions for a small part of the flight.

The data also shows you flying very slowly in Tripod Mode for the whole flight.
From around 3:06 the data shows a compass error that is present until the end of the data about one minute later.
It appears that for this part of the flight you were in close to the tanker (?) and the steady compass error was due to being so close to a very large amount of steel.
The GPS track shows unsteady wandering for this part of the flight.
Had you moved away from the ship, the compass error should have disappeared, but you stayed close.
You made no right stick inputs after 3:45 and only a little left stick input to vary the height.
Speeds for the last 40 seconds were less than 0.5 mph and the drone was just drifting/hovering.
The crash happens at 4:08.9 and the drone tumbles to the ground.

The site looks like a tricky flying site with so much steel and reinforced concrete making it hard to find a safe launch point.
Flying very close to large steel objects caused the compass error and with poor GPS conditions and a failure to keep the drone away from danger, a collision happened.
Here's Google Earth imagery to help anyone interested in understanding of the flying environment.
i-KVLb9GZ-X2.jpg
Hi Meta. Thanks for your thorough analysis. You indicate that if had chosen a 'safe launch point' (which I interpret as a point without any compass/GPS issues) I could had flown into the 'unsafe' area closer to the ship, without encountering the problems and the eventually crash ? I am new to drone flying, but fortunately the Mavic Air survived the crash, except for a few scratches.
 
  • Like
Reactions: Trollistika
Lycus Tech Mavic Air 3 Case

DJI Drone Deals

New Threads

Forum statistics

Threads
134,374
Messages
1,594,159
Members
162,941
Latest member
lpcdan