Mystery Crash P4 flewn with Litschi

[Lylatwars]

Hello,
I flew my P4 pro in 300m distance and 70m height, I had some seconds before 3 signal aborts which never happend earlier. I used the Litschi flight app for the first time in free flight mode together with a Samsung Note 4 (which I am using since 2 years for flights). After the signal was lost I tried the Homo Button but nothing, the battery was filled by 70% and the last video is showing the drone hovering over its crash position in 70m height.
I found it as attached so it felt of its back or roof, so the propeller side which is visible in the track. After I analysed the peaces (of my life...) I found the battery contacts melted. It is still possible to activate the drone with an other battery, there seems to be only melted plastic on the contacts. The crash battery is showing the seconds light on and the first blinking. Its difficult to discribe but the battery button feels different, seems it is possible to press it much more into the battery as usual.
Now I would like to now why the crash happen, should I stay away from Litschi or was is a technical problem with the battery? The Battery was original DJI 5870mah.

I would really appreciate your help because I will buy the same model again and maybe there is something up with the warranty

Merci in advance

 

[GrammatonxXXxCleric]

Melted plastic is highly suspect of a problem.

 

[GrammatonxXXxCleric]

Was the motor that lost its prop hot or show signs of overheating?

 

[Lylatwars]

There was no issue regarding the rotors. The pictures showing the connection between drone and battery. The rotors are still working

 

[GrammatonxXXxCleric]

I realize that. I asked in case one motor was hung up while spinning. That'd explain (possibly) the melted plastic.

 

[Neon Euc]

Def a DJI warranty. I remember someone had the same problem a few weeks back on the same model as yours (I think it was the sane model) def go and contact DJI and you should get a free repair or replacement as this was a fault with the product

 

[WV. Rootman]

It is possible, you didn't have a good connection, there will be high resistance. High resistance produces Heat. That is why the plastic melted.

 

[Lylatwars]

My flight was in the desert, far from everything, the drone flew on the same position 2 hours earlier a 2.5 km x 300 m mission without problems or signal abort but with another app. Could that transmission problem with Litschi be the problem? Signal problem with Litschi --> overheat --> burning contacts-->crash?

 

[sar104]

I suggest that you recover and post the aircraft DAT file - that will show what happened.

How to retrieve a V3 .DAT File from the AC

 

[Lylatwars]

Fly182 is the last number but the other 3 files are carrying the latest time by 1:31 pm. Is that what you need for an analysis? I dont know what to do with that data

 

[Lylatwars]

Fly182 is the last number but the other 3 files are carrying the latest time by 1:31 pm. Is that what you need for an analysis? I dont know what to do with that data

I it possible to get any information out of this data? Please response even when no, so I can tell it to DJI regarding the warranty, Thanks

 

[sar104]

I it possible to get any information out of this data? Please response even when no, so I can tell it to DJI regarding the warranty, Thanks

Unfortunately none of those files are flight files - they are just short aircraft power up events that look like they occurred on 2/25, probably after the crash since many of the onboard systems are reporting failure. The actual flight DAT should be much larger. Look for the last recorded larger file that you can find.

 

[Lylatwars]

Unfortunately none of those files are flight files - they are just short aircraft power up events that look like they occurred on 2/25, probably after the crash since many of the onboard systems are reporting failure. The actual flight DAT should be much larger. Look for the last recorded larger file that you can find.

This is the biggest and 1 min before the crash, is that working out? (unfortunately to big for phantompilots with 52Mb so I placed in in dropbox)
Dropbox - FLY171.DAT

 

[sar104]

This is the biggest and 1 min before the crash, is that working out? (unfortunately to big for phantompilots with 52Mb so I placed in in dropbox)
Dropbox - FLY171.DAT

That's the one. A quick glance indicates a significant power failure at around 210 s, leading to very low motor voltage and interrupted data logging. Strangely the logging resumes at 246 s, but only for around 1/10th of a second. This is going to need a bit more analysis. Do you have the Litchi flight log?

 

[BudWalker]

... Strangely the logging resumes at 246 s, but only for around 1/10th of a second. .....

I think I know what happened. But first, you have to tell me how you got the last little but at 246 secs.

 

[Lylatwars]

That's the one. A quick glance indicates a significant power failure at around 210 s, leading to very low motor voltage and interrupted data logging. Strangely the logging resumes at 246 s, but only for around 1/10th of a second. This is going to need a bit more analysis. Do you have the Litchi flight log?

First of all thank you so much for taking care of my problem!
Here are the Litschi Logs, strange that there is an 14:44 pm log. I have not used Litschi after the crash
Dropbox - LitschiLogs

 

[Lylatwars]

I think I know what happened. But first, you have to tell me how you got the last little but at 246 secs.

What happend at 246secs??? I am sooo interested like before Xmas

 

[sar104]

What happend at 246secs??? I am sooo interested like before Xmas

As @BudWalker mentioned - that appears to have been an artifact of the DatCon version that I used (2.6.9) and is not real, but he is checking it out. The log almost certainly ends at 210 s and does look like a battery disconnect.

 

[BudWalker]

.. Strangely the logging resumes at 246 s, but only for around 1/10th of a second. ....

That log data at time 246 secs are from a different previous flight.

The .DAT file is written in blocks - just like in a normal file system. Records are written into a block until the block is full. At that point the block is "closed" and future records start being written into a new block. When a .DAT file is deleted it's blocks are then made available to be used in a new .DAT file. When the last record of the log is written to the .DAT file the location of the last byte is recorded within the block. If that didn't happen then it looks like there are extra records - past the last record that was actually written. The bytes for those extra bogus records are whatever was in that block from the last time the block was used in a file.

That's what happened here. There was an abrupt power loss or crash and the current block didn't get closed. The extra bogus records where from a previous flight. The incident flight had a last recorded position of 46.64057343 25.19290684 while the next bogus position was 46.48202829 24.65086564. Shown here as the ends of the red line

Also, looking at the GPS info in the bogus records it's possible to see the previous flight occurred 2017-11-20 11:25:41 GMT

Note, also, that it "appears" that the motorVolts recovered from having dropped to the 7 volt range. But, those values came from the previous flight.

 

[sar104]

That log data at time 246 secs are from a different previous flight.

The .DAT file is written in blocks - just like in a normal file system. Records are written into a block until the block is full. At that point the block is "closed" and future records start being written into a new block. When a .DAT file is deleted it's blocks are then made available to be used in a new .DAT file. When the last record of the log is written to the .DAT file the location of the last byte is recorded within the block. If that didn't happen then it looks like there are extra records - past the last record that was actually written. The bytes for those extra bogus records are whatever was in that block from the last time the block was used in a file.

That's what happened here. There was an abrupt power loss or crash and the current block didn't get closed. The extra bogus records where from a previous flight. The incident flight had a last recorded position of 46.64057343 25.19290684 while the next bogus position was 46.48202829 24.65086564. Shown here as the ends of the red line
View attachment 95638

Also, looking at the GPS info in the bogus records it's possible to see the previous flight occurred 2017-11-20 11:25:41 GMT

Note, also, that it "appears" that the motorVolts recovered from having dropped to the 7 volt range. But, those values came from the previous flight.
View attachment 95639

It may be worth reiterating that although the graph above shows voltage traces between 210 and 246, there are no actual data points in that interval. The data logging stops at 210 s when the voltage drops below around 7 V or so - pretty much identical to what we see on other battery disconnect or sudden power failure events.