Mavic air out of control

[Murmansken]

Yesterday I tried to fly my Air and suddenly it flew away on it´s own, couldnt do anything and it didnt respond on my commands. Unfortunately it crashed into a tree and broke my propellers but otherwise it seems fine. Does anyone have an idea why this happened?

DJI Flight Log Viewer - PhantomHelp.com log file

 

[slup]

Hi there fellow countryman, welcome to the forum ... even though it's under these unfortunate circumstances.

This looks like a take off from a magnetic interfered area ... To determine that we need your DAT log located in your mobile device you flew with (it's the one with FLY076.DAT in the end). See below how to retrieve it ...

Mobile device DAT files

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.

iOS: Apps » DJI GO 4 » FlightRecords » MCDatFlightRecords

Android: DJI » dji.go.v4 » FlightRecord » MCDatFlightRecords


Do you remember if the drone was pointing in the direction shown in the pic. below ... or was it in fact directed in another direction?

 

[Deleted member 94047]

Welcome to the forum @Murmansken. Sorry about what happened to you but you have come to the right place to figure out what happened. I took a look at the flight log in CsvView and yeah, there is definitely something wrong with this flight (you will need @slup and @sar104 for the analysis - OK I just saw @slup posting while writing this). Locate and upload the DAT file and hopefully this will be figured out.

@slup, this is not making any sense to me at all. Why is VPS height seemingly stuck at 10m? The aircraft never went above 12m, so it should have been active throughout the flight and definitely once the aircraft descended (assuming the data is valid)

 

[Murmansken]

Hi there fellow countryman, welcome to the forum ... even though it's under these unfortunate circumstances.

This looks like a take off from a magnetic interfered area ... To determine that we need your DAT log located in your mobile device you flew with (it's the one with FLY076.DAT in the end). See below how to retrieve it ...

Mobile device DAT files

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.

iOS: Apps » DJI GO 4 » FlightRecords » MCDatFlightRecords

Android: DJI » dji.go.v4 » FlightRecord » MCDatFlightRecords


Do you remember if the drone was pointing in the direction shown in the pic. below ... or was it in fact directed in another direction?

View attachment 96063

The drone was positioned in the direction as shown when it was in the air, at the takeoff Im unsure. Tried to flew it away from the trees as you can see but it wouldn´t respond at all. I thought it was a pilot error because im new to this with drones and still learning.

 

[old man mavic]

welcome to the forum sorry its not in better circumstances

 

[Murmansken]

welcome to the forum sorry its not in better circumstances

Thank you, well I want to know what went wrong. I hope it is an pilot error so I can learn something from it.

 

[Deleted member 94047]

I hope it is an pilot error so I can learn something from it.

This really doesn't look like one to me but let's wait for sar104 to have a look at the logs.

 

[slup]

Welcome to the forum @Murmansken. Sorry about what happened to you but you have come to the right place to figure out what happened. I took a look at the flight log in CsvView and yeah, there is definitely something wrong with this flight (you will need @slup and @sar104 for the analysis - OK I just saw @slup posting while writing this). Locate and upload the DAT file and hopefully this will be figured out.

@slup, this is not making any sense to me at all. Why is VPS height seemingly stuck at 10m? The aircraft never went above 12m, so it should have been active throughout the flight and definitely once the aircraft descended (assuming the data is valid)

View attachment 96066

The crash occurred at 26sec ... so all odd thing's there after can possible be a result of that. The only strange thing I see now without access to the DAT, beside the flyaway itself ... is that all RC related signals doesn't show up until 8sec.

@Murmansken ... did you fly with the RC or only your mobile?

 

[slup]

Thank you, well I want to know what went wrong. I hope it is an pilot error so I can learn something from it.

@Murmansken ... retrieve & attach your DAT log here, then we hopefully can tell, without that all will only be speculation.

 

[Murmansken]

The crash occurred at 26sec ... so all odd thing's there after can possible be a result of that. The only strange thing I see now without access to the DAT, beside the flyaway itself ... is that all RC related signals doesn't show up until 8sec.

@Murmansken ... did you fly with the RC or only your mobile?

View attachment 96071

With rc and mobile attached for video. I did attach the .dat file in previous message but it awaits moderator approval.

 

[Deleted member 94047]

@Murmansken ... retrieve & attach your DAT log here, then we hopefully can tell, without that all will only be speculation.

DAT was already posted but I see that that post has disappeared @Murmansken, did you delete the post? Also, I have already downloaded it, I can upload it from my end if you didn't mean to delete it.

EDIT: Just saw Murmansken's reply above. Attaching the DAT I downloaded earlier for expedience. Here you go @slup

 

[Murmansken]

DAT was already posted but I see that that post has disappeared @Murmansken, did you delete the post? Also, I have already downloaded it, I can upload it from my end if you didn't mean to delete it.

No I didnt delete it awating moderator approval... I posted it again

 

[Deleted member 94047]

No I didnt delete it awating moderator approval... I posted it again

I attached the DAT to post #10

 

[slup]

Hmm ... this one looked really fishy to say the least.

At power on the compass usually shall initiate the yaw angle ... but here they doesn't at all.

The 2 compasses match each other really good, so does the gyros which both start up with 0 degrees. The 2 IMU yaw's on the other hand firstly doesn't align with the compasses & also have a disagreement of approx 30 degrees ... the VIO yaw seems to hook on to the IMU's as a start up value.

My bet right now goes to a IMU failure ... but will continue to scratch my head & scrutinize a bit more ?

 

[BudWalker]

Hmm ... this one looked really fishy to say the least.

At power on the compass usually shall initiate the yaw angle ... but here they doesn't at all.

The 2 compasses match each other really good, so does the gyros which both start up with 0 degrees. The 2 IMU yaw's on the other hand firstly doesn't align with the compasses & also have a disagreement of approx 30 degrees ... the VIO yaw seems to hook on to the IMU's as a start up value.

My bet right now goes to a IMU failure ... but will continue to scratch my head & scrutinize a bit more ?

View attachment 96083

Power on occurred at about -48 secs but the .DAT didn't start recording until -11 secs. The Yaw/magYaw separation probably happened during the unrecoded interval.

 

[Murmansken]

Hmm ... this one looked really fishy to say the least.

At power on the compass usually shall initiate the yaw angle ... but here they doesn't at all.

The 2 compasses match each other really good, so does the gyros which both start up with 0 degrees. The 2 IMU yaw's on the other hand firstly doesn't align with the compasses & also have a disagreement of approx 30 degrees ... the VIO yaw seems to hook on to the IMU's as a start up value.

My bet right now goes to a IMU failure ... but will continue to scratch my head & scrutinize a bit more ?

View attachment 96083

Ok so no pilot error then, I didnt remember exactly what happened when I took off and which direction the drone pointed but I flew it up to about 10 - 15 meters and then it got crazy and ran away from me and I couldnt do anything. Do I have to replace the IMU? Did a IMU calibration after the crash with no errors.

 

[sar104]

Ok so no pilot error then, I didnt remember exactly what happened when I took off and which direction the drone pointed but I flew it up to about 10 - 15 meters and then it got crazy and ran away from me and I couldnt do anything. Do I have to replace the IMU? Did a IMU calibration after the crash with no errors.

That's not a valid conclusion, at least not without more information. It does appear likely that the aircraft was powered up somewhere with magnetic distortion, and then moved to the takeoff point. Can you confirm whether you did move it after powering it up?

 

[slup]

The drone was positioned in the direction as shown when it was in the air, at the takeoff Im unsure. Tried to flew it away from the trees as you can see but it wouldn´t respond at all. I thought it was a pilot error because im new to this with drones and still learning.

Oh ... didn't see that you edited your earlier post.

Well this makes it even more puzzling ... that can mean that the heading shown in the Airdata picture can be the correct one (approx -135 degrees) which doesn't align with the compasses showing -30.

But anyhow ...

Looking at a period shortly after the auto take off have finished ... you apply first a short throttle to ascend then a 2sec long rudder input. During that input the log shows this differences in approx change of angle, looking at only IMU_ATTI(0)

magYaw = +40 degrees
Gyro = +135 degrees
Yaw = + 133 degrees
VIOYaw = + 131 degrees

This makes the compass reaction odd in relation...

 

[Murmansken]

That's not a valid conclusion, at least not without more information. It does appear likely that the aircraft was powered up somewhere with magnetic distortion, and then moved to the takeoff point. Can you confirm whether you did move it after powering it up?

Yes I powered it up indoor walked outside, about 3 - 4 meters and put it on the ground, then used the automatic take off. But shouldnt it give me a warning for magnetic interferance or at least compass calibration required?

 

[Murmansken]

Oh ... didn't see that you edited your earlier post.

Well this makes it even more puzzling ... that can mean that the heading shown in the Airdata picture can be the correct one (approx -135 degrees) which doesn't align with the compasses showing -30.

But anyhow ...

Looking at a period shortly after the auto take off have finished ... you apply first a short throttle to ascend then a 2sec long rudder input. During that input the log shows this differences in approx change of angle, looking at only IMU_ATTI(0)

magYaw = +40 degrees
Gyro = +135 degrees
Yaw = + 133 degrees
VIOYaw = + 131 degrees

This makes the compass reaction odd in relation...

View attachment 96095

That 2 second rodder input is probably when I ascend to 10-15 meters just before it went crazy.