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

My mavic pro escaped!

RAS666

Well-Known Member
Joined
Sep 15, 2017
Messages
58
Reactions
11
Age
36
So first proper flight after replacing a motor on my mavic and it being layed up for almost a year it all seemed to be working fine so decided to take a panoramic from abut 200 ft and that's where it all went wrong. From looking at the logs it shows an imu error and yaw errors not sure what caused it as it appears to fly fine now (low down) but my main concern is the difference between the dji flight log play back and the air data results and the fact that it apears that i didnt touch the controls when it set off evenue thouh i did. The dji log looks like I lost the GPS signal but air data shows the GPS track where it went walkies to?
This is the dji flight record it starts to go wrong during the pano at about 5:30

And this is the air data log from the same source as the dji log!

Some slight difference?
 
Looks like you likely took off from a magnetically dirty site, which is shown by the compass errors right away.

...others will chime in with more info.
I saw those too on the air data but none appeared while I was actually flying plus I took off where I normally take off for flying round the back garden it's a large bit of concrete with no reinforcement but there is a scrap pile I flew over on the way into the garden but not normally an issue. Plus the imu error seemed to occur after quite some time. I don't understand the difference between the 2 flight logs even though they come from the exact same file.
 
The IMU error happened late because you didn't make any yaw adjustments really. Slight adjustments here and there won't cause enough of a discrepancy. Once it started shooting the Pano and rotated far enough, the IMU went haywires...

Or so it appears to me
 
I've checked through my logs and true there are absolutely no compass errors in any other flight taking off at the same location before or today after very odd maybe I had something in my hand when I turned it on. Also it's not the first time I had used it since the motor replacement I have hovered it 2 ft off the ground and tested various manovers before this flight. Bit worrying that the dji app still had the thing as hovering above me on the map on screen when in reality I was climbing a tree to catch sight of the think nearly 1/4 of a mile away ?
 
So first proper flight after replacing a motor on my mavic and it being layed up for almost a year it all seemed to be working fine so decided to take a panoramic from abut 200 ft and that's where it all went wrong. From looking at the logs it shows an imu error and yaw errors not sure what caused it as it appears to fly fine now (low down) but my main concern is the difference between the dji flight log play back and the air data results and the fact that it apears that i didnt touch the controls when it set off evenue thouh i did. The dji log looks like I lost the GPS signal but air data shows the GPS track where it went walkies to?
This is the dji flight record it starts to go wrong during the pano at about 5:30

And this is the air data log from the same source as the dji log!

Some slight difference?
Can you retrieve and post the .DAT for this flight? It'll have file name ending in FLY113.DAT.
 
The .TXT flight log file provides limited information but two things can be seen :

Compass error
There was indeed a compass error as can be seen in the following part of the flight. There was no wind and only forward elevator was applied. The nose of the craft should be pointing to the direction of flying but the direction sensed by the IMU/compass was 30 degrees off. Such magnitude of yaw error is insufficient to cause fly away and the response of the craft to stick inputs may felt to be normal.

compass error.gif


DJI playback vs AirDataAT
At 480 seconds the flight controller suddenly considered the GPS / compass data to be totally unreliable ( navHealth 5 -> 0 ) despite the fact that there were 15 ~ 16 satellites. May be the flight controller saw some anomalies in the magnetic field but the .TXT file does not contain such information.

Anyway the situation ( navHealth = 0 ) last all the way to the end of the flight. I believe that's why "weak GPS" is indicated in the DJI playback. It probably meant the combination of GPS signal quality and compass accuracy is not good enough for GPS to be used for navigation purpose.

After that the flight path shown in the DJI playback starts to deviate from that shown on AirData. I think the AirData flight path is wrong because the craft was shown to be moving at 15 ~ 19 mph without any tilting and there was only very little wind. If you view the flight log in PhantomHelp, the flight path that is absent from the DJI playback is shown in a different color ( red ).

1602998910441.png

If you can dig out the .DAT flight log files, may be more insight can be obtained.
 
Last edited:
Is it the flight logs off the aircraft you want? I plugged it in the pc and it all looks different than what it used to the lost of logs doesn't include the day when it happened. Altho the date could be wrong in the mavic as I saw flight logs for September when it was packed away. I will look at them in more detail tonight when I get in. What I can say is when it went it went at some fair speed maybe a random strong gust of wind or it flew itself. Also the location where I regained control and dropped it down to about 20ft the position on the airdata map looks pretty accurate. I saw the tree and when I panned round I saw the boats in the marina and decided to fly it home with the display.
 
The .DAT file in the phone is needed. From Mavic Flight Log Retrieval and Analysis Guide :

The DAT file naming convention, based on the date and time of the start of the file, is: YY-MM-DD-hr-min-sec_FLYXXX.DAT., where XXX is the flight recorder file index from the HOME_dataRecorderFileIndex field in the txt log.

These are retrieved by the same method as the TXT logs. Under both iOS and Android they are in a subfolder, MCDatFlightRecords, in the folder that contains the TXT logs. In some cases, for reasons not fully explained but possibly mobile-device hardware related, and most often under Android, DAT files are not created and that folder is empty. Uninstalling and reinstalling the app sometimes fixes that. One user also found that the process required manual deletion of the app folder (apparently not deleted automatically in the uninstall process) before reinstalling the app.

The DJI Fly app deletes the DAT files if and when it syncs flight records with the DJI servers, and so they are often missing from the MCDatFlightRecords folder. The iOS version of the DJI GO 4 app does the same since version 4.3.24. The Android version of DJI GO 4 does not delete the DAT files as of 4.3.32.

As mentioned by Bud, the file with the name ending in FLY113.DAT is the one you are looking for.
 
The .TXT flight log file provides limited information but two things can be seen :

Compass error
There was indeed a compass error as can be seen in the following part of the flight. There was no wind and only forward elevator was applied. The nose of the craft should be pointing to the direction of flying but the direction sensed by the IMU/compass was 30 degrees off. Such magnitude of yaw error is insufficient to cause fly away and the response of the craft to stick inputs may felt to be normal.

View attachment 115256


DJI playback vs AirData
The IMU made a yaw correction of 40 degrees during the time period from 423 to 480 seconds. The amout of correction is close to the yaw error magnitude of 30 degrees mentioned above so this is normal. However, immediately after that the flight controller considered the GPS / compass data to be totally unreliable ( navHealth 5 -> 0 ) despite the fact that there were 15 ~ 16 satellites. May be the flight controller saw some anomalies in the magnetic field but the .TXT file does not contain such information.

Anyway the situation ( navHealth = 0 ) last all the way to the end of the flight. I believe that's why "weak GPS" is indicated in the DJI playback. It probably meant the combination of GPS signal quality and compass accuracy is not good enough for GPS to be used for navigation purpose.

View attachment 115257

After that the flight path shown in the DJI playback starts to deviate from that shown on AirData. I think the AirData flight path is wrong because the craft was shown to be moving at 15 ~ 19 mph without any tilting and there was only very little wind. If you view the flight log in PhantomHelp, the part of flight path that is absent from the DJI playback is shown in a different color ( red ).

View attachment 115258

If you can dig out the .DAT flight log files, may be more insight can be obtained.
Just a couple of observations.

The compass error seen in the AirData report is probably when the nonGPS Cause shows the value COMPASS_ERROR_LARGE. This is usually an indication that the FC is seeing a difference between what it's computing to be Yaw and Yaw derived from magnetometer data.

The yawing seen in the [423 -480] secs interval is because the Mavic Pro was in Pano mode
1603028064071.png

When navHealth drops to 0 it's because the FC has lost confidence in the location solution. This can occur for several reasons - low satellite count being one reason. But, usually it's because the FC has determined it can't be sure of the Yaw value. That appears to be the reason here.

@RAS666 If the .DAT on the mobile device isn't available the .DAT on the MP itself will be available. The MP is particularly easy since you don't have to use the DJI Assistant 2 to retrieve the .DAT from the MP. Take a look at these instructions
How to retrieve a V3 .DAT File from the AC
and look in the
Mounted Disk Drive
section
 
  • Like
Reactions: new2mavic
OK only fly117.dat is avaliable on my phone so trying the aircraft itself the mounted disc method doesn't work I can see 2 disc drives but neither do anything and this is what I can see on my list of available logs
 

Attachments

  • 20201018_193528.jpg
    20201018_193528.jpg
    3 MB · Views: 36
Ah well we may never know I have flown it a few times since from the same place even and thrown it around in sports mode and not had any errors again. I have downloaded all of the flight logs individually via dji assistant and checked them in csv view but cannot find 113. Weirdly there are some flight logs in there from a few years ago logs in the 20s and they check out too matching some early flight logs. Maybe I could remove the internal sd card and put it in a reader or is it in some odd format?
 
Ah well we may never know I have flown it a few times since from the same place even and thrown it around in sports mode and not had any errors again. I have downloaded all of the flight logs individually via dji assistant and checked them in csv view but cannot find 113. Weirdly there are some flight logs in there from a few years ago logs in the 20s and they check out too matching some early flight logs. Maybe I could remove the internal sd card and put it in a reader or is it in some odd format?
There were some firmware versions that would cause the on board .DATs to not be recorded. You may have got caught by that. Check out
Keeping .DATs on the Mavic Pro accessible
 
Wow. Sure seems there should be 'a' or 'one' single place a person could go, either from the AC or Phone... grab it (file folder), and inside would be an easy to understand root structure to follow to what file a person needs. My PC is a total mess, thanks to my being lazy when it comes to being organized. But, when I need a file, I can find it. This file system for drones seems absolutely crazy! My admiration to anyone that knows how to decipher these "hieroglyphics" !!! Hope to goodness I never need to go thru this treasure hunt, but sure am glad there are those who are gracious enough to help out non computer code folks like me. Thanks to all who contribute!
 
  • Like
Reactions: WWMav
Think your right about it not deleting some old data files I bet the one we were looking for had been over written because I had flown it after the incident plus messed around with turning it on and off. I've deleted all but the most recent logs and will see how it behaves now. Is there a rtc battery inside the mavic that needs replacing or are the incorrect dates normal? 20201024_141217.jpg

Also checkout the picture I found on airdata that's where the drone came to a stop! That photo isn't on the dji app either nor the memory card and I don't remember taking it either ?Oct_6th__2020_05_54pm_pic-1.jpg
 
I think a lesson learnt here would be to extract the logs immediately after it all goes wrong rather than mess with it because they soon get over written and also in the effort of syncing the flight logs to view them on air data also removes them from the phone itself :rolleyes: it's almost like dji don't want you to know what happened.
 
Lycus Tech Mavic Air 3 Case

DJI Drone Deals

New Threads

Forum statistics

Threads
130,583
Messages
1,554,086
Members
159,587
Latest member
Bigbrute