Mavic bricked mid flight need answers !

[BudWalker]

@BudWalker Do you have any tricks to view flight009.dat? I threw it into csvview and it seems corrupt (not much of a surprise given the circumstances of its creation...) It's obvious to me that this crash is not due to pilot error, however, I have no clue what to look for to prove it! Is there some sort of error log or command line output hidden somewhere that could shed some light?

FLY009.DAT seems OK to me. Why do you think it's corrupted?

 

[BudWalker]

So I know you guys are very busy looking at the data and I have to say thank you (i did not get this kind of help on dji site)

But from what you have read on the flight 008, is there anything you can tell was pilot error? I mean thats when I lost all control I had good gps lock and everything seemed good across the board and boom my rc juts said connecting and the dji4 app said disconnected from aircraft. I just dont see what I did wrong to say pilot error, Im mean if i was being dumb and was 3 miles away behind a mountain and crashed or if I rammed it into a tree yes its my fault, but from all this im just kinda upset at this point not getting a clear answer...

We're still looking at the incident. But, it does appear that there was an abrupt mid flight power off.

 

[Drew77]

I guess he tried to open the file and could not and due to the weird nauture of this problem of what happened maybe he thinks that's why it can't open?
So bud I would love your opinion on what you think has happened, and do you think this was piloting error or a mechanical issue?

 

[Drew77]

Thank you bud and jack maybe DJI should take note of how helpful you guys are, please take your time I don't want to rush you as long as we get to the bottom of this I will be happy, not DJI rush response of pilot error pay right now, with absolutely no explanation at all.

 

[Logger]

FLY009.DAT seems OK to me. Why do you think it's corrupted?

Perhaps because he "threw it into csvview" without actually converting it with dat con first.

 

[BudWalker]

Looks to me like there was an abrupt power interruption at 584 secs while the Mavic was at 97.1 meters height. As was mentioned above this comes from the motorVoltage data where there was a drop from 10.4 volts to 6.4 volts. The data recording for FLY 008.DAT ended here


FLY009.DAT then starts on the ground a short distance away

It would appear that the descent was not a free fall; consistent with the pilot's report of minimal damage. From the timeStamps on the two .DAT files the descent took 79 seconds. That's 1.22 meters/sec. The accelerometer data does not indicate the Mavic hit the ground after FLY009 started recording. So it must've hit the ground before. I'll speculate that the power interruption was not a total shutdown and that some propulsion was available to slow the descent. When the Mavic hit the ground full power was restored which allowed the recording of FLY009.Just a theory.

The remainder of FLY009 was uneventful as it was sitting there waiting to be retrieved.

 

[BudWalker]

Perhaps because he "threw it into csvview" without actually converting it with dat con first.

CsvView accepts a .DAT which it then converts internally with DatCon.

 

[BudWalker]

@BudWalker Do you have any tricks to view flight009.dat? I threw it into csvview and it seems corrupt (not much of a surprise given the circumstances of its creation...) It's obvious to me that this crash is not due to pilot error, however, I have no clue what to look for to prove it! Is there some sort of error log or command line output hidden somewhere that could shed some light?

You're right, FLY009 looks kinda weird Some of that comes from the fact that it was a new flight, not the continuation of an ongoing flight. FLY009 starts all over; the relHeight starts from 0. It would have been -12 meters if the home point altitude in the first flight was used. Also, gpsAlt and baroAlt are 0 for the first couple of seconds after batteryOn.

 

[Logger]

CsvView accepts a .DAT which it then converts internally with DatCon.

Ah thanks - that's good to know. I Didn't know that.

 

[Logger]

....It would appear that the descent was not a free fall; consistent with the pilot's report of minimal damage. From the timeStamps on the two .DAT files the descent took 79 seconds. That's 1.22 meters/sec. ...

That descent rate is even less than what you see when it does a critical battery auto-land descent which is ~3.5 meters/sec or 7mph. Just checked one of mine from 270'. Might have been non linear with a hover at the bottom prior to LDG. Good sleuthing BTW.

 

[Drew77]

Bud walker you are absolutely amazing I have to give you major credit for that report,that is a true 100% professional response .
So it looks like you solved what happened even tho I don't know what caused it? But flight 009 might be when I found the drone (that is exactly where I found it in the persons driveway) and started to drive home.
So anyway DJI looked at the flight logs and determined it was a malfunction with the drone and would be covered under warranty.So thank you for everyone's help once again this forum is amazing!!

 

[Strafe1]

This guy is a genius when it comes to DAT analysis and drone mechanics. I am impressed every time I read his stuff. I think its safe to say that we all appreciate your time, effort and expertise.

 

[pgitug]

I read that you can configure "Both Sticks Down" to power off during normal flight OR program two sticks down to power off only when drone is in distress. Better the latter. Just a video tutorial I saw.


Sent from my iPhone using MavicPilots

 

[Logger]

I read that you can configure "Both Sticks Down" to power off during normal flight OR program two sticks down to power off only when drone is in distress. Better the latter. Just a video tutorial I saw.

It is not clear, of the relevance of your post to this thread? If it is an inference that OP might have issued a CSC command to inadvertently cause the crash, then I suggest this would be incorrect. The stick positions to issue the CSC would surely be logged and visible well before the abrupt power interruption. The command itself may well get logged too? Furthermore CSC only kills the motors. The rest of the Device including the camera & video feed keeps operating if on. Until the mavic is psychically turned off or something catastrophic stops is. Seemingly not the case here.
Personally I prefer the CSC available at all times option set & have used it in the past. You need to hold the sticks in the down & in or down and out position for 3 secs, so pretty hard to do inadvertently.

 

[Drew77]

I'm well aware of the CSC command and it was clearly NOT used just look at the flight log, Budwalker prettt much nailed it with his explanation, and I'm seriously impressed with his work, this guy needs a high position job at DJI, but his response is the response I would expect from a company like DJI, hats off to Budwalker the true professional!!

 

[BudWalker]

That descent rate is even less than what you see when it does a critical battery auto-land descent which is ~3.5 meters/sec or 7mph. Just checked one of mine from 270'. Might have been non linear with a hover at the bottom prior to LDG. Good sleuthing BTW.

The descent rate that I calculated isn't correct. The problem is in determining elapsed time between the end of recording in FLY008 and the beginning of recording in FLY009. I had originally calculated that elapsed time to be 79 seconds. I failed to take into account the offset when the timeStamps are recorded in the .DATs. Using those offsets yields an elapsed time of 1.4 secs. That's way too small since it yields a descent rate of 68 meters/sec. But, I suspect it's a lot closer to the actual elapsed time.

 

[Drew77]

Thanks again bud for the info,Like I said it did have some damage gimble/all propellers/ minor dings and dents but I assume if it came down that hard and fast it would be pretty destroyed. Anyway im not sure what caused the problem but I know it was NOT pilot error and WAS ERROR WITH THE DRONE,and after DJI did there job and went over the flight data, came to the same conclusion, thus they covered under warranty.

 

[FlyingDnutts]

Thanks Anthony for the response I know I thought the same thing even the flight log the drone just stops recording and everything weird?? I tried to get in contact with dji no response yet it's been one full week?
That's why I was hoping to get a more technical answer from here I don't have a lot of faith in dji response right now , I'm going to try to get the whole DAT file up so people can read but as you can see above from the few I got something deff went wrong

Use the online chat. I had 2 great experience with them.

Dnuttz

 

[insainflow]

I'm well aware of the CSC command and it was clearly NOT used just look at the flight log, Budwalker prettt much nailed it with his explanation, and I'm seriously impressed with his work, this guy needs a high position job at DJI, but his response is the response I would expect from a company like DJI, hats off to Budwalker the true professional!!

Honestly could have just been a malfunction, glad DJI and the guys around here are taking care of you. I think it could have been severe electromagnetic interference. In an extreme improbability and attempted hijacking or disabling of your drone with a rf blaster. I have heard of these being designed even though they are very illegal. Whatever it was you are getting it fixed so cheers to that!!!


Sent from my iPhone using MavicPilots

 

[Drew77]

Use the online chat. I had 2 great experience with them.

Dnuttz

To be honest the DJI support is a absolute disgrace, calling on the phone ahaha ok an absolute joke they are more interested in rushing you off the phone then actually helping, online chat is also a disgrace, the resonance from the chat was very scripted and took so long it actually shut down a few times due to inactivity for 5 min, the forum help on DJI site was only someone useful after one of the mods actually pm me back that got the case actually to make progress, mavicpiolts along with its members is a different story, any problem with DJI products should just be instructed to go here, they treat you like an actual paying costumer and they actually get RESULTS!! I got more info and answers on my case here in 2 days than over 1 week on DJI, and to this point gave no explanation at all. But in the long run they are fixing my drone and I will be happy for that