My Mavic was only about 100 feet away from me hovering when I noticed it 'toilet bowling' off to the right just before a 'compass error, exiting GPS mode' message appeared. I landed, rebooted and checked my compass values which were all completely normal. I tried another brief flight which went ok with no further errors.

Replaying the flight record, I noticed a couple of other error messages, one of which makes no sense. Although I was dozens of miles from any no fly zone, the error states that the RTH path would cross one. Really? With the craft 131 feet from me?

From chit chat on other forums I've noticed this situation seems to be affecting the Spark too. Anyone know what's going on as this is knocking my confidence severely?

If you upload the DAT file from the flight I'll take a look at it and may be able to figure out what happened. The instructions for retrieving it are here and the correct file will be tens or hundreds of megabytes in size. Once you've obtained the correct file upload it to DropBox and provide a link to it in this thread.

As far as the NFZ message goes, I've seen people report that before and I think it's just because DJI changed the firmware a while back to assume that the Mavic is in an NFZ when there's no GPS (e.g., indoors).

I'll follow this thread since I live within NFZ and I can still fly with no problems.

GOING KOOKOO for NFZ KOOKOO POPS!

DroneFlying Posted at 2017-7-30 12:44
If you upload the DAT file from the flight I'll take a look at it and may be able to figure out what happened. The instructions for retrieving it are here and the correct file will be tens or hundreds of megabytes in size. Once you've obtained the correct file upload it to DropBox and provide a link to it in this thread.

As far as the NFZ message goes, I've seen people report that before and I think it's just because DJI changed the firmware a while back to assume that the Mavic is in an NFZ when there's no GPS (e.g., indoors).

Hi Droneflying,

Thanks in advance for this. Here's a link to the google drive location, hope it works ok.

https://doc-0o-b0-docs.googleuse ... lNFBISnc?e=download

Watty2000 Posted at 2017-7-30 15:32
Hi Droneflying,

Thanks in advance for this. Here's a link to the google drive location, hope it works ok.

No problem. However, I don't seem to be able to download the file; it looks like I don't have access to it.

DroneFlying Posted at 2017-7-30 17:02
No problem. However, I don't seem to be able to download the file; it looks like I don't have access to it.

Arghh, sorry. Try this:

https://www.dropbox.com/s/h4x634 ... 1_00-25-20.DAT?dl=0

Watty2000 Posted at 2017-7-31 00:35
Arghh, sorry. Try this:

https://www.dropbox.com/s/h4x634au3ac83gk/DJI_ASSISTANT_EXPORT_FILE_2017-07-31_00-25-20.DAT?dl=0

Yep, that worked. This looks very much like a case of taking off from a geomagnetically distorted site, which is usually because there's some metal nearby that affects the compass.

Notice how in the graph below that yaw (what the Mavic think its heading is / should be in yellow) and magYaw (the heading detected by the compass in red) were closely aligned until you took off (around 0 on the horizontal axis), at which point magYaw began bouncing around and deviated a bit more from yaw, which in turn triggered the errors you saw. That happens because the compass readings fluctuate as it begins to get clear of whatever source of interference had influenced it before.

The variation was large enough at around 12.5 seconds into the flight to trigger a YAW_ERROR_LARGE (shown by the blue shading), and then about a second later it went into ATTI mode (the purple shading). And since GPS is disabled / ignored in ATTI mode, the Mavic reverted to the behavior I mentioned earlier where it assumes that it's in an NFZ when it has no GPS (e.g., indoors), which is why you got the NFZ warning message.

The bottom line is that this is all normal behavior, so it should fine to fly in the future as long as you take off from a location where there's no magnetic interference.

DroneFlying Posted at 2017-7-31 02:49
Yep, that worked. This looks very much like a case of taking off from a geomagnetically distorted site, which is usually because there's some metal nearby that affects the compass.

Notice how in the graph below that yaw (what the Mavic think its heading is / should be in yellow) and magYaw (the heading detected by the compass in red) were closely aligned until you took off (around 0 on the horizontal axis), at which point magYaw began bouncing around and deviated a bit more from yaw, which in turn triggered the errors you saw. That happens because the compass readings fluctuate as it begins to get clear of whatever source of interference had influenced it before.

Thanks for looking into this although I'm a bit dubious about the findings. Immediately after this problem flight I power cycled the Mavic and controller and flew from exactly the same spot with no adverse effects at all. Surely if there was a magnetic anomaly where I flew the first time, wouldn't it be reproduced taking off from the same spot?

Again, thank you for taking the time to check this out.

Watty2000 Posted at 2017-7-31 04:32
Thanks for looking into this although I'm a bit dubious about the findings. Immediately after this problem flight I power cycled the Mavic and controller and flew from exactly the same spot with no adverse effects at all. Surely if there was a magnetic anomaly where I flew the first time, wouldn't it be reproduced taking off from the same spot?

Again, thank you for taking the time to check this out.

BudWalker also has his doubts about this one, so I've asked him to weigh in on it with his own theory about what might have caused it.

Watty2000 Posted at 2017-7-31 04:32
Thanks for looking into this although I'm a bit dubious about the findings. Immediately after this problem flight I power cycled the Mavic and controller and flew from exactly the same spot with no adverse effects at all. Surely if there was a magnetic anomaly where I flew the first time, wouldn't it be reproduced taking off from the same spot?

Again, thank you for taking the time to check this out.

I'm also a bit dubious. Typically, the effects of a geomagnetic distortion are within 1 meter or so of the launch site. And, once clear of those effects the magYaw stops changing. There are exceptions though. I once looked at an incident where the launch was next to a 5 meter high metal water tank.


Could you post a link to the .DAT for the flight you did after the incident flight?

Thanks Bud. I'll pull the .DAT off the aircraft when I get home later tonight. The flight was started on a remote country road with no pylons or telegraph poles around and there were definitely no metallic objects anywhere near the Mavic in the air so it's all a bit confusing.