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

Drone crash into side of building - mag interference - how do I get logs

ischonfeld

Active Member
Joined
Feb 21, 2020
Messages
32
Reactions
9
Age
70
Location
Riverton, New Jersey
One of our guys was flying our Mavic Enterprise Dual. He was on the 2nd battery, bringing it back for a landing at approx 8' AGL. The LZ was about 10' from the side of the building. As he approached, the drone stopped responding to controls and the display indicated MAG INTERF. The drone continued in the direction it was going (toward the building) and could not be corrected by operator control. It crashed and luckily did not sustain significant damage.

I would like to pull the flight logs - can you direct me to instructions? Is the log in the controller or drone? We're flying with DJI Pilot using an iPad Air with the Mavic Enterprise Dual. I believe the controller was in P mode but would like to verify that from the logs too.

Thank you
 
Follow the instructions from this site.


Another option is to set up a free AIR DATA account.


There are some experienced pilots here that can break down the flight for you after you post the link here.

Good luck.

.
 
  • Like
Reactions: MARK (LI)
This is the in depth thread at the top of the Mavic Crash & Flyaway section.


Part 3 has the phantomhelp and airdata methods of getting your txt report, and the dat if that is available (and sometimes helpful).
Getting the report is easy, understanding and dissemination of the data is quite complex and involves other specialised programmes and interpretation, so please post up the report when you have it, one of the numerous analysis folk will come along and do their magic.
 
I'm in the process of getting the flight log (it is on an individual's phone instead of the department ipad).

But a question: Since the drone wasn't under control of the RC and possibly lost communications with it, just how much information will be there of the event itself? Does the drone itself keep any information, and if so how do you get it?
 
...But a question: Since the drone wasn't under control of the RC and possibly lost communications with it, just how much information will be there of the event itself? Does the drone itself keep any information, and if so how do you get it?
Not sure you can get the DAT off the drone, being a M2 model it’s usually only DJI that can do that. Or it’s encrypted.
One of the analytical folk here might elaborate / confirm.
A craft .DAT log from a M2 is encrypted ... it's only DJI staff that can make use of that.

You have instead 2 other logs stored in the mobile device the craft was flown with ... the mobile device .TXT log (containing telemetry, battery data, gimbal & camera data, flight status and error flags) and the mobile device .DAT log (containing nearly all of the raw sensor data).

If the craft/RC or the mobile device/RC connection is lost the log will stop ... if the connection returns the logging will proceed. This will be clearly seen in the log if it's the case ...

From your incident description it sounds as the mobile device .TXT will not be enough to diagnose this ... the mobile device .DAT will be needed (& that one will not be as easy as just upload to PhantomHelp or Airdata as both these sites will not show the needed sensor data ... a tool like CsvView will be needed instead).

If you want assistance start up with attaching the mobile device .TXT log in a new post here ... if the mobile device .DAT is needed we guide you to which .DAT belongs to this flight.
 
I have flight logs!

The individual flying at the time used his personal phone instead of the designated iPad, so it took a few days to get the flight logs. I have attached the last few TXT and DAT files from that day. Hopefully they have the information as to the magnetic interference issue and why that caused the M2ED to fly off into the side of the building.

Any help would be appreciated. If this is related to the area where he was flying, well then we learned not to fly there. But if there's something we're missing, we'd certainly want to know in order to avoid that in the future.

Thank you!
 

Attachments

  • 2021-09-25_15-22-43_FLY034.DAT
    9 MB · Views: 10
  • 2021-09-25_15-47-33_FLY035.DAT
    8.2 MB · Views: 5
  • DJIFlightRecord_2021-09-25_[15-55-41].txt
    575.5 KB · Views: 8
  • DJIFlightRecord_2021-09-25_[15-58-50].txt
    69.8 KB · Views: 5
  • DJIFlightRecord_2021-09-25_[15-59-17].txt
    257.1 KB · Views: 6
It's only the DAT file FLY034.DAT that is relevant, the crash occurs in the second flight in that power on period (a DAT records from power on to power off) ... would be good to have the corresponding TXT file, is it any earlier ones beside those you've shared?

Was it any extra equipment attached to the drone in the flight that ended with the crash that wasn't on the flight before?
 
Here are the rest of the TXT files - I'm not sure which one corresponds to the flight.

The broadcast speaker (comes with the M2ED) was attached. I will check with the pilot to see if it was attached for both flights or just the last one. Thanks.
 

Attachments

  • DJIFlightRecord_2021-09-25_[14-55-01].txt
    1.6 MB · Views: 3
  • DJIFlightRecord_2021-09-25_[15-02-31].txt
    957.6 KB · Views: 3
  • DJIFlightRecord_2021-09-25_[15-09-11].txt
    2 MB · Views: 3
  • DJIFlightRecord_2021-09-25_[15-19-32].txt
    181.2 KB · Views: 4
  • DJIFlightRecord_2021-09-25_[15-20-34].txt
    198.9 KB · Views: 3
  • DJIFlightRecord_2021-09-25_[15-22-55].txt
    444.9 KB · Views: 4
  • DJIFlightRecord_2021-09-25_[15-26-24].txt
    1.5 MB · Views: 4
  • DJIFlightRecord_2021-09-25_[15-50-23].txt
    19.5 KB · Views: 4
  • DJIFlightRecord_2021-09-25_[15-53-55].txt
    272.7 KB · Views: 4
  • DJIFlightRecord_2021-09-25_[15-59-17].txt
    257.1 KB · Views: 3
  • DJIFlightRecord_2021-09-25_[15-58-50].txt
    69.8 KB · Views: 3
  • DJIFlightRecord_2021-09-25_[15-55-41].txt
    575.5 KB · Views: 3
1633296681290.png

Yeah ... this is what I faced when I should post my idea to what happened ... Took screen dumps before I deleted all, so unfortunately no charts can be made larger by clicking on them, but think I explain in the text clear enough & the pics can be seen somewhat OK.

1633296843026.png
1633296869174.png
1633296883570.png
1633296897673.png
1633296912573.png
1633296939626.png
1633296953430.png
1633296968329.png
1633296981094.png
1633296993069.png
1633297005502.png
1633297019190.png
 
Last edited:
View attachment 135775

Yeah ... this is what I faced when I should post my idea to what happened ... Took screen dumps before I deleted all, so unfortunately no charts can be made larger by clicking on them, but think I explain in the text clear enough & the pics can be seen somewhat OK.

View attachment 135776
View attachment 135777
View attachment 135778
View attachment 135779
View attachment 135780
View attachment 135781
View attachment 135782
View attachment 135783
View attachment 135784
View attachment 135785
View attachment 135786
View attachment 135787
208.640 : 11485 [L-CTRL]accessory:3 detected
Good find. I took a look at some other *.log.txt files that came from Mavic 2. It's just the Mavic 2 Enterprise that looks for accessories - the Mavic 2 doesn't.

@ischonfeld have you been able to determine if the speaker was installed on the ME2 after the first flight?
 
Thank you, Slup, for all of this work - it's clear you spend a lot of time analyzing the telemetry (plus writing it all up).

I know what the accessory they added was - it was the broadcast speaker that comes with the M2ED. We've used it before several times, including me personally. All without incident up until this time.

It's a speaker, so there's most certainly a magnet in it.

Now I'm not so sure how to proceed. Do we abandon the use of the speaker? Recalibrate when attaching it and then again after removing it? I guess the pilot was lucky that this was a practice flight and only minor damage was done. Could have been a lot worse.

Thanks again.
 
Thank you, Slup, for all of this work - it's clear you spend a lot of time analyzing the telemetry (plus writing it all up).

I know what the accessory they added was - it was the broadcast speaker that comes with the M2ED. We've used it before several times, including me personally. All without incident up until this time.

It's a speaker, so there's most certainly a magnet in it.

Now I'm not so sure how to proceed. Do we abandon the use of the speaker? Recalibrate when attaching it and then again after removing it? I guess the pilot was lucky that this was a practice flight and only minor damage was done. Could have been a lot worse.

Thanks again.
The speaker has a detectable magnetic field, strongest at the back of the unit, but it has no measurable effect that I can see on the aircraft compass, whether the aircraft is powered up with it in place or it is connected afterwards. It would be a strangely poor design if it did cause those kinds of problems.

So something else was going on in your case most likely unrelated to the accessory.
 
Anything else taken-off or put on the AC besides that speaker ... was anything done out of the ordinary in between the 2 flights? Something starts to disturb the mag.field from the moment the AC is handled between the flights ... & after that compass errors start to occur in the log in a steady stream.
 
The speaker has a detectable magnetic field, strongest at the back of the unit, but it has no measurable effect that I can see on the aircraft compass, whether the aircraft is powered up with it in place or it is connected afterwards. It would be a strangely poor design if it did cause those kinds of problems.

So something else was going on in your case most likely unrelated to the accessory.
I agree. But, it's just too coincidental this occurred after the speaker was attached. I tried to find out if it's a piezo speaker that wouldn't have a permanent magnet but didn't get anywhere.

The magnetic field generated by the current flowing to the speaker might be strong enough to cause a problem. But, audio frequencies are probably above what can be seen by the magnetometers. An anomaly where the signal is stuck on, or biased, could cause a steady magnetic distortion. This would be the same as a "hard iron" distortion. Just a theory - I don't have any experience with a speaker.
@slup
 
Last edited:
I agree. But, it's just too coincidental this occurred after the speaker was attached. I tried to find out if it's a piezo speaker that wouldn't have a permanent magnet but didn't get anywhere.

The magnetic field generated by the current flowing to the speaker might be strong enough to cause a problem. But, audio frequencies are probably above what can be seen by the magnetometers. An anomaly where the signal is stuck on, or biased, could cause a steady magnetic distortion. This would be the same as a "hard iron" distortion. Just a theory - I don't have any experience with a speaker.
@slup
I have the same speaker on my M2EA - it has a permanent magnetic driver, so that’s not the problem.
 
  • Like
Reactions: MARK (LI)
I have the same speaker on my M2EA - it has a permanent magnetic driver, so that’s not the problem.

I don't have any experience with a speaker on a drone, but I wonder if the difference is that you put your speaker on the drone before firing up the drone, whereas it appears that the OP installed his after the drone was on. And maybe the problem had some relevance to the craft being rotated several times like slup says it was.

As you say, the speaker itself might not be the problem, but the sequence of events involving it might be.
 
  • Like
Reactions: MARK (LI)
I don't have any experience with a speaker on a drone, but I wonder if the difference is that you put your speaker on the drone before firing up the drone, whereas it appears that the OP installed his after the drone was on. And maybe the problem had some relevance to the craft being rotated several times like slup says it was.

As you say, the speaker itself might not be the problem, but the sequence of events involving it might be.
I explained in post #14 that the sequence makes no difference. And as I said before, this is a DJI accessory that comes with the M2ED and M2EA - DJI didn't make such a simple and obvious mistake.
 
  • Like
Reactions: MARK (LI)
I explained in post #14 that the sequence makes no difference. And as I said before, this is a DJI accessory that comes with the M2ED and M2EA - DJI didn't make such a simple and obvious mistake.

I didn't see anything in your post #14 about you spinning the drone around several times like slup says happened with the OP.
 
Lycus Tech Mavic Air 3 Case

DJI Drone Deals

New Threads

Forum statistics

Threads
130,585
Messages
1,554,093
Members
159,585
Latest member
maniac2000