DJI Mavic Pro crashed! NEED help to determine causes!

[rogerjiangcf]

Background info:
I have recently bought nolimitdronez mod software trying to remove the maximum height limit of my Mavic. Downgraded from V01.04.0400 to V01.03.0900 in order to get the height limit removed. Planning to test the limits in the countryside of course. However, the firmware of the controller is on V01.04.0400 and has not been touched. I'm was using the MODDED DJI GO APP (For the first time). DJI GO 4 MOD 2.0 | DJI GO 4 MOD LITE [ Additional Features | Android | App Ver. 4.1.15 ]

Before crash:
I was flying my Mavic on a grass field 100 meters away from a tall building to take a look. Everything was working normally and suddenly the motor stopped working or a prop flew off or something. The aircraft started spiraling down, I tried to restart the motors with success but according to the log file my Mavic fell even faster after the attempted motor restart.

My Suspection:
I either lost a prop in flight or it was the MOD that caused it. However, I would like to know the exact cause.
I have checked the log file and there are quite a few errors. Such as MOTOR IDLE, GPS POSITION NO MATCH, there was also missing data for 16s right before the motor went out.


Note: NO ONE WAS INJURIED! My Mavic crashed on a grass field and there was no one around(Except for me). Fell 2 meters away from me. Definitely scared the crap out of me

Zip file attached includes the DAT log file, TXT log file, and xlsx file converted from the TXT log file.

Thanks in advance for the community's help!

 

[SpinItUp]

Yep, it sure lost a prop and getting faster is the distance it was covering when coming down but there seems to be erratic up and down shown in the google earth kml
But I get another path from your records that show this?
... This is just my quick look but no doubt @sar104 may input his diagnosis!
Sorry but it does remind me of this;

Prop came off

Edit;
From what I can gather your front right prop Failed

Edit, Edit;
Maybe something hit the prop
The log shows

impact_in_air

 

[Meta4]

I was flying my Mavic on a grass field 100 meters away from a tall building to take a look. Everything was working normally and suddenly the motor stopped working or a prop flew off or something. The aircraft started spiraling down, I tried to restart the motors with success but according to the log file my Mavic fell even faster after the attempted motor restart.

You were just taking the Mavic straight up with full left stick forward and no other stick input.
The climb was stable until0:43.5 and 531 feet up when the mavic started to rotate clockwise, roll to the right and pitch downwards.
Its upward momentum helped it to continued to gain altitude for another second but it was out of control spinning and tumbling.
That indicates that the front right prop stopped doing its thing and the outcome from then was inevitable.

 

[sar104]

Yes - it lost a prop - the front left prop to be precise, at 43.5 seconds. The resulting attitude progression is characteristic of 3-prop flight.



Looking at the motor speed data:



Left front runs to maximum as the load vanishes. The event stream log notes the resulting loss of control:

43.722 : 6426 [L-FDI][CTRL]: fault on , tilt_ctrl_fail​

43.742 : 6427 [L-RC]craft ctrl failed!!!​

It also includes the amusingly descriptive diagnosis:

46.058 : 6543 [L-FMU/MOTOR]some motor is unload and some are not​

 

[rogerjiangcf]

Yes - it lost a prop - the front left prop to be precise, at 43.5 seconds. The resulting attitude progression is characteristic of 3-prop flight.

View attachment 50603

Looking at the motor speed data:

View attachment 50604

Left front runs to maximum as the load vanishes. The event stream log notes the resulting loss of control:

43.722 : 6426 [L-FDI][CTRL]: fault on , tilt_ctrl_fail​

43.742 : 6427 [L-RC]craft ctrl failed!!!​

It also includes the amusingly descriptive diagnosis:

46.058 : 6543 [L-FMU/MOTOR]some motor is unload and some are not​

Thank you so much for the analysis!
Such a stupid mistake from me. However, the prop that appears to be missing is from the front right prop.
Also, what software were you using to get the motor data and the stream log? I was trying to retrieve the motor pwm info with various software including DJI assistant 2 but couldn't find it.

 

[rogerjiangcf]

You were just taking the Mavic straight up with full left stick forward and no other stick input.
The climb was stable until0:43.5 and 531 feet up when the mavic started to rotate clockwise, roll to the right and pitch downwards.
Its upward momentum helped it to continued to gain altitude for another second but it was out of control spinning and tumbling.
That indicates that the front right prop stopped doing its thing and the outcome from then was inevitable.

Thank you for the analysis! Indeed I couldn't find my front right prop at the crash site.

 

[rogerjiangcf]

Yep, it sure lost a prop and getting faster is the distance it was covering when coming down but there seems to be erratic up and down shown in the google earth kmlView attachment 50579
But I get another path from your records that show thisView attachment 50580?
... This is just my quick look but no doubt @sar104 may input his diagnosis!
Sorry but it does remind me of this;

Prop came off

Edit;
From what I can gather your front right prop Failed

Edit, Edit;
Maybe something hit the prop
The log shows

impact_in_air

Thank you so much!

 

[sar104]

Thank you so much for the analysis!
Such a stupid mistake from me. However, the prop that appears to be missing is from the front right prop.
Also, what software were you using to get the motor data and the stream log? I was trying to retrieve the motor pwm info with various software including DJI assistant 2 but couldn't find it.

That's very strange - as you can see above the data indicate that front left exhibited the no load behavior. I guess it's possible that the data fields are mis-labelled. I'll have to do some tests on that.

 

[rogerjiangcf]

That's very strange - as you can see above the data indicate that front left exhibited the no load behavior. I guess it's possible that the data fields are mis-labelled. I'll have to do some tests on that.

Thanks again! And would you mind providing the name of the tool that you used for the analysis?

 

[SpinItUp]

ditto that...those graphs are great
DatCon was a tool pointed to from sar104 a while ago but for the graphs, he pulls out of the hat I have no idea

 

[sar104]

Thanks again! And would you mind providing the name of the tool that you used for the analysis?

ditto that...those graphs are great
DatCon was a tool pointed to from sar104 a while ago but for the graphs, he pulls out of the hat I have no idea

I use Wavemetrics Igor Pro for the analysis and visualization.