mavic mini crashed without a reason

[sar104]

The cause of that was pretty obvious, and not one I've seen previously. The IMU gyro package appears to have failed at 400 seconds, leading to complete loss of control:



The motors don't appear to have stopped, although the motor data are missing from the file conversion (@BudWalker?), but craft control failed. That's a warranty issue.

 

[WooZie]

Thank you very much
do they have acces to these logs online, or are saved in the craft? or they will ask me for it?

 

[sar104]

Thank you very much
do they have acces to these logs online, or are saved in the craft? or they will ask me for it?

There is a detailed DAT file on the aircraft, but they will likely ask for the txt log.

 

[WooZie]

Motor failure perhaps? Or perhaps it collided with a bird (outside of camera view)?

no, it was in front of me, i saw it, no bird, otherwise if it was a bird, the drone would have been recover itself in the air, not rotating until down

 

[WooZie]

There is a detailed DAT file on the aircraft, but they will likely ask for the txt log.

thank you sir, now i have a little hope due to you that my drone will be changed with a new one without any cost

 

[Greg6032]

there is nothing to repair, an arm is broken, the case is broken, the motors are blocked, the gimbal is broken, the drone fall of 60 meters height, nothing to repair, but only change. The drone was takeoff and landed only on the clean concrete, not grass, not soil. And especially after takeoff i stayed with it a few seconds in air, to warmup the battery( i read that is good for drones), all the motors works very well, no strange noise or something else, very stable...so nothing strange in his behavior

Hi
If I remember correctly if your drone failed because there is some problem with electronics and not your fault and you have drone for less then 2 weeks they will replace it with new one with no any cost for you
Open case with DJI ASAP

 

[WooZie]

I have done it imediatelly after the incident, the ticket is open but today i have to take the package to go to the store in the original box, i have everything prepared, but here is 6 AM so i can't sleep i am nervous, i smoked the whole night, what bad luck i had.....500 E lost in a second, but it has only 5 days since i bought it. Bad luck for me

 

[mavicn00b]

Single IMU evidently - not great when it fails.

 

[DanMan32]

Anyone able to sync their logs? I get a message in Chinese when I tried, but link to sync goes away as if successful. But if I come back, says they didn't sync.
Plus sync doesn't carry over to other devices like it does with Go.

 

[MAvic_South_Oz]

I have done it imediatelly after the incident, the ticket is open but today i have to take the package to go to the store in the original box, i have everything prepared, but here is 6 AM so i can't sleep i am nervous, i smoked the whole night, what bad luck i had.....500 E lost in a second, but it has only 5 days since i bought it. Bad luck for me

Don't be nervous, you'll be fine.
Maybe copy and print relative parts of this thread (the data analysis bits) and include in the package.
They can see someone knows it's not pilot error, but then I think I'd trust them reasonably enough to do the right thing, based on posts I've read.
Just be prepared to go higher if needed, but hopefully they'll just admit a build problem.

 

[WooZie]

Anyone able to sync their logs? I get a message in Chinese when I tried, but link to sync goes away as if successful. But if I come back, says they didn't sync.
Plus sync doesn't carry over to other devices like it does with Go.

yes, i did succes but i had to try many times, or let the phone on auto sync and will be uploaded all the flights, i think the servers are busy, but it will be uploaded finally

 

[WooZie]

Don't be nervous, you'll be fine.
Maybe copy and print relative parts of this thread (the data analysis bits) and include in the package.
They can see someone knows it's not pilot error, but then I think I'd trust them reasonably enough to do the right thing, based on posts I've read.
Just be prepared to go higher if needed, but hopefully they'll just admit a build problem.

i already saved all from here and from the app logs, and again i want to thank you all for your cooperation and help.

 

[BudWalker]

It's also possible this was a propulsion issue. Looking at the integrated gyro data it can be seen that the MM starts tumbling forward back to front, rolling CW and rotating CCW. These are all consistent with a right front propulsion issue.

I've come to rely on these signals to determine if there was a propulsion issue. Looking at roll, pitch and yaw is problematical because of the relatively low sample rate that causes aliasing that's hard to see past. Also, roll, pitch and yaw are computations done by the FC and it's not always clear they are correct.

There is also the Motor Obstructed messages in the .txt log.

DatCon doesn't decode the motor data because the MM is a new platform. One MM .DAT isn't enough to do the reverse engineering necessary to include the signals that are different from the other DJI platforms. This particular .DAT has some corruption problems.

I did notice the MM has a much different clockrate than the preceding DJI platforms.

I also noticed the message about the stuck gyro. Looking at the gyro data

it can be seen that it was the X axis gyro that appeared to go through a jump discontinuity from +1800 to -2000 when the stuck gyro was reported in the eventLog stream. It's possible this was the result of a number representation problem where adding one to +2048 results in -2048. Just a theory...

@WooZie would it be possible to get the .txt for this flight? That will help some in the reverse engineering effort.

@sar104

 

[WooZie]

yes of course, here it is, thanks

 

[sar104]

It's also possible this was a propulsion issue. Looking at the integrated gyro data it can be seen that the MM starts tumbling forward back to front, rolling CW and rotating CCW. These are all consistent with a right front propulsion issue.
View attachment 85999
I've come to rely on these signals to determine if there was a propulsion issue. Looking at roll, pitch and yaw is problematical because of the relatively low sample rate that causes aliasing that's hard to see past. Also, roll, pitch and yaw are computations done by the FC and it's not always clear they are correct.

There is also the Motor Obstructed messages in the .txt log.

DatCon doesn't decode the motor data because the MM is a new platform. One MM .DAT isn't enough to do the reverse engineering necessary to include the signals that are different from the other DJI platforms. This particular .DAT has some corruption problems.

I did notice the MM has a much different clockrate than the preceding DJI platforms.

I also noticed the message about the stuck gyro. Looking at the gyro data
View attachment 86003
it can be seen that it was the X axis gyro that appeared to go through a jump discontinuity from +1800 to -2000 when the stuck gyro was reported in the eventLog stream. It's possible this was the result of a number representation problem where adding one to +2048 results in -2048. Just a theory...

@WooZie would it be possible to get the .txt for this flight? That will help some in the reverse engineering effort.

@sar104

Looking at that I'm inclined to agree. I prematurely dismissed the gyro data as unphysical and combined with the error message I assumed gyro failure. Of course that makes no sense, because the attitude data are primarily derived from the gyros. It does suggest motor failure.

 

[WooZie]

but, i saw on the flight text, the drone has changed very fast itself the mode p in sport, in just a second, for many times, i couldn't have do that even if i would have wanted

 

[Greg6032]

I have done it imediatelly after the incident, the ticket is open but today i have to take the package to go to the store in the original box, i have everything prepared, but here is 6 AM so i can't sleep i am nervous, i smoked the whole night, what bad luck i had.....500 E lost in a second, but it has only 5 days since i bought it. Bad luck for me

Do not be nervous
Everything with be ok and you will get new drone
It is only question of the time

 

[WooZie]

do you believe that? Is this company so seriously? Or they will try to accuse me from doing something wrong

 

[GerdS]

I guess they will take a look onto the wreckage and log files and in this case just take a new one out of their stock and replace it for free.

 

[WooZie]

i hope so, because has only 2 flights and 4 days old too young for die so soon