Mavic 2 Pro became an acrobat while on ground

[vinsentto]

Hello everyone. I recently bought M2P and fly more kit (received the delivery package last May 24, but didn't avail Care Refresh). I have been reading threads here to minimize the chances of crashing my first drone. Most notably, I learned from posts by sar104 to check the correctness of the direction of arrow before flying, so as not to make the drone behave out of control due to magnetic interference. I also purchased the propeller guard to be on the safe side.

A few hours ago, I flew my drone to record a footage of my dormitory less than 10 kilometers from an airbase (yellow NFZ in the app). The flight lasted a good 14 minutes before I landed the drone with 30% battery. I turned off the drone and replaced the battery. I turned the drone back on as I wanted to edit some settings (decrease the sensitivity when rotating and disable the light in front of the drone). To my surprise, the right motors (I think) suddenly started causing the drone to flip sideways and slam the ground maybe about a meter or half away. I didn't even start the motors manually. The drone survived in one piece but a small part of the body near the back leg got chipped off.

What just happened there? Here are all the files including those from the successful flight, which may provide a hint or something. Kindly help me determine whether I caused the crash or not, before I consider making a request for replacement/repair (which I also don't know how). Please excuse my lack of knowledge for that. Thank you so much in advance!

 

[sar104]

Unfortunately there is no record of that event in those logs. The 14 minute flight that you described is DJIFlightRecord_2019-06-05_[19-24-13].txt and FLY017.DAT, starting at 19:22:28 and ending at 19:39:15. None of the other logs show a motor start. FLY018.DAT is a 7 minute power up starting 2 minutes later at 19:41:19, with no motor start and the aircraft was stationary and horizontal on the ground for the entire period. It ended at 19:48:18.

FLY020 (19:48:47 - 19:48:52), FLY021 (19:50:43 - 19:50:46) and FLY022 (19:51:21 - 19:52:21) are very short power-up events with no motor start or significant movement.

FLY019 is missing but I'm not really even seeing a time period when that event might have occurred - how do these times relate to what you remember?

 

[vinsentto]

Starting from powering the drone up after changing the battery, I didn't turned it off the whole time I'm changing settings in the app. I turned it off only after the incident, in fear that something worse may happen. I promptly turned it on again just to see if the drone completely died. I was relieved it still powered up. I turned it off for the last time and went inside my room to inspect the drone thoroughly. I don't know if that sequence matches up the number of .dat files.

I am scared to learn that something like this was not recorded. I wish I pushed the record button but no one would do that while idly changing the settings. It appears I cannot just request for repair/replacement, right?

Thank you so much for the assistance. I hope I don't encounter this anymore. It's already dark here, so I will have to wait until tomorrow morning to see if the drone can still function satisfactorily. In the meanwhile, what actions are usually done to check for damage? If the app displays nothing unusual tomorrow, will that be enough to assume that the drone is still intact? Again, thank you so much.

 

[sar104]

Starting from powering the drone up after changing the battery, I didn't turned it off the whole time I'm changing settings in the app. I turned it off only after the incident, in fear that something worse may happen. I promptly turned it on again just to see if the drone completely died. I was relieved it still powered up. I turned it off for the last time and went inside my room to inspect the drone thoroughly. I don't know if that sequence matches up the number of .dat files.

I am scared to learn that something like this was not recorded. I wish I pushed the record button but no one would do that while idly changing the settings. It appears I cannot just request for repair/replacement, right?

Thank you so much for the assistance. I hope I don't encounter this anymore. It's already dark here, so I will have to wait until tomorrow morning to see if the drone can still function satisfactorily. In the meanwhile, what actions are usually done to check for damage? If the app displays nothing unusual tomorrow, will that be enough to assume that the drone is still intact? Again, thank you so much.

Well the sequence shown by the logs is:

14 minute flight

2-minute gap​

7-minute power up

20-second gap​

5-second power up

2-minute gap​

3-second power up

35 second gap​

60-second power up

 

[cfd2474]

I had the same thing happen to mine. Best I was able to tell, I was using my original props with about 14 hrs of flight time. (Don't use props that long!!) The locking tab broke off one of the props and caused some friction in the motor, and the system tried to compensate- sending it into a full dive/flip.

 

[FormerMember07122019_2]

I had the same thing happen to mine. Best I was able to tell, I was using my original props with about 14 hrs of flight time. (Don't use props that long!!) The locking tab broke off one of the props and caused some friction in the motor, and the system tried to compensate- sending it into a full dive/flip.

Sorry for your problem but I am fairly certain worn props wasn’t the cause after 14 hours. There are a whole lot of folks with hundreds and even a thousand hours on the same props with no issues. Your props could have had a defect or got damaged when taking them on or off.

 

[Kilrah]

That's completely unrelated, your aircraft didn't just start motors on its own like OP. And yeah people use props for hundreds of hours with no problem.

 

[navegabem]

Sorry for your problem but I am fairly certain worn props wasn’t the cause after 14 hours. There are a whole lot of folks with hundreds and even a thousand hours on the same props with no issues. Your props could have had a defect or got damaged when taking them on or off.

What would you suggest is the max amount of time to use a set of props? I have a sneaking suspicion that some long time pilots have never changed their props.

 

[FormerMember07122019_2]

That's completely unrelated, your aircraft didn't just start motors on its own like OP. And yeah people use props for hundreds of hours with no problem.

I was addressing cfd2474’s statement regarding worn props after 14 hours not the OP’s issue.

 

[FormerMember07122019_2]

What would you suggest is the max amount of time to use a set of props? I have a sneaking suspicion that some long time pilots have never changed their props.

I have no idea what the max prop usage should be. Most people recommend checking them very thoroughly for cracks or roughness with extra focus on the hubs and mounting tabs before each flight.

 

[runeOfRune]

@sar104 ummm

811.973 : {task_e}INFO:i2c3 core recovery||
811.973 : 46231 [L-BATTERY][ERROR] smbus_read CMD_CELLV1 error||
811.983 : {task_e}INFO:i2c3 core recovery||
811.983 : 46231 [L-BATTERY][ERROR] cell1_vol error||
811.995 : {task_e}INFO:i2c3 core recovery||
811.995 : 46232 [L-BATTERY][ERROR] cell2_vol error||
812.007 : {task_e}INFO:i2c3 core recovery||
812.007 : 46232 [L-BATTERY][ERROR] cell3_vol error||
812.018 : {task_e}INFO:i2c3 core recovery||
812.018 : 46233 [L-BATTERY][ERROR] cell4_vol error||
812.030 : {task_e}INFO:i2c3 core recovery||
812.030 : 46233 [L-BATTERY][ERROR] CMD_BLACK_BOX read error||
812.042 : {task_e}INFO:i2c3 core recovery||
812.042 : 46234 [L-BATTERY][ERROR] smbus_read CMD_PRESENT_STATE error||
813.212 : 46293 [L-VISION][VIO]receive vio data befor current ts||
813.217 : 46293 [L-NS][AHRS] Bad sample tick measure or recapture signal: 0.010000
831.704 : {task_e}INFO:i2c3 core recovery||
831.704 : 47219 [L-BATTERY][ERROR] smbus_read CMD_POWER_STATE error||
849.894 : 48130 [L-NS][AHRS],qg obv buffer size[218/270]
849.894 : 48130 [L-NS][AHRS],pva obv buffer size[238/270]
854.498 : {task_e}INFO:i2c3 core recovery||
854.498 : 48361 [L-BATTERY][ERROR] smbus_read CMD_POWER_STATE error||

Also, is CsvView broken with M2 logs? It seems like it's broken, because battery data makes no sense.

 

[BudWalker]

@sar104 ummm



Also, is CsvView broken with M2 logs? It seems like it's broken, because battery data makes no sense.

For the M2 .DATs the Battery(0) signals don't work. They actually don't exist in the M2 .DAT - DatCon is confused.

Most of the BattInfo and battery_info signals work.

 

[BudWalker]

Well the sequence shown by the logs is:

14 minute flight

2-minute gap​

7-minute power up

20-second gap​

5-second power up

2-minute gap​

3-second power up

35 second gap​

60-second power up

I suspect that FLY018 is the .DAT corresponding to the incident. There are several things wrong with it. E.g. there are no MotorCntrl records. The underlying cause may be related to the spontaneous motor starts.

 

[sar104]

I suspect that FLY018 is the .DAT corresponding to the incident. There are several things wrong with it. E.g. there are no MotorCntrl records. The underlying cause may be related to the spontaneous motor starts.

Maybe - but even if the motor data are incorrect, the attitude data show it steady for that cycle. It looks like a real power up with data, unless I'm missing something.

 

[BudWalker]

Maybe - but even if the motor data are incorrect, the attitude data show it steady for that cycle. It looks like a real power up with data, unless I'm missing something.

I think it's a real power up, just that parts of the system didn't start. In addition to the missing MotorCntrl records the ctrl_horiz_module_debug and ctrl_horiz_motor_debug records are missing. Don't want to read too much into it but this kinda suggests (to me at least) that the modules that control the motors weren't functioning when it powered up.

 

[paulatkin73]

Unfortunately there is no record of that event in those logs. The 14 minute flight that you described is DJIFlightRecord_2019-06-05_[19-24-13].txt and FLY017.DAT, starting at 19:22:28 and ending at 19:39:15. None of the other logs show a motor start. FLY018.DAT is a 7 minute power up starting 2 minutes later at 19:41:19, with no motor start and the aircraft was stationary and horizontal on the ground for the entire period. It ended at 19:48:18.

FLY020 (19:48:47 - 19:48:52), FLY021 (19:50:43 - 19:50:46) and FLY022 (19:51:21 - 19:52:21) are very short power-up events with no motor start or significant movement.

FLY019 is missing but I'm not really even seeing a time period when that event might have occurred - how do these times relate to what you remember?

based on battery voltage and current there is no sign of any motors spinning in the provided logs. it is impossible to hide as each motor draws about of 2A to spin enough to flip it. in all logs other than 17 current is steady at the idle level - 1.24A.

so it has to be one of missing logs - the only theory i can offer is the possibility to consider flight controller went into a sudden reboot and ESCs went mad during that reboot spinning motors. very odd. but it is indeed not in the provided logs, and, to be more odd - voltage curve does not really show any loss of power, log 18 ends at the 16.98V. there is an odd jerk in there at the end of the log 18, on voltage, where it drops from 16.98 to 16.96 and then goes back to 16.98 - but it motors would turn, it would sag much more then just a 0.02v.

 

[Kilrah]

I was addressing cfd2474’s statement regarding worn props after 14 hours not the OP’s issue.

And I was also replying to him, not you. Your post just appeared inbetween seconds before I posted mine.

 

[BudWalker]

based on battery voltage and current there is no sign of any motors spinning in the provided logs. it is impossible to hide as each motor draws about of 2A to spin enough to flip it. in all logs other than 17 current is steady at the idle level - 1.24A.

so it has to be one of missing logs - the only theory i can offer is the possibility to consider flight controller went into a sudden reboot and ESCs went mad during that reboot spinning motors. very odd. but it is indeed not in the provided logs, and, to be more odd - voltage curve does not really show any loss of power, log 18 ends at the 16.98V. there is an odd jerk in there at the end of the log 18, on voltage, where it drops from 16.98 to 16.96 and then goes back to 16.98 - but it motors would turn, it would sag much more then just a 0.02v.

I think you're right. The only thing to consider is that recording didn't start until about 9.64 secs and the spontaneous motor start could've happened before that. But, as @sar104 mentioned there were no significant accelerations or gyro data for the next 418 secs. I suspect that @vinsentto didn't just leave it on the ground for roughly 7 minutes

@vinsentto how long after power up did the spontaneous motor start happen? Was it after 424 secs?