How to protest a DJI crash analysis?

[QuadKid]

If you look at the first graph in post #22 you will see that you used auto takeoff.

Hey Sar, what software are you using for diagnostic, I've been using FRAP developed by a user over in the DJI forum, kinda like the display parameters of the one you are using.

 

[sar104]

Hey Sar, what software are you using for diagnostic, I've been using FRAP developed by a user over in the DJI forum, kinda like the display parameters of the one you are using.

I'm converting txt logs with TXTlogToCSVtool and DAT logs with DatCon. To analyze and display the resulting data I use Igor Pro.

 

[QuadKid]

I'm converting txt logs with TXTlogToCSVtool and DAT logs with DatCon. To analyze and display the resulting data I use Igor Pro.

Thank you Sir, just d/loaded the trial.

 

[7thstring]

I don't see any evidence of movement prior to your elevator input at 6.7 s.

View attachment 42502

Was it your impression that it moved significantly before that? Neither the IMU data nor the GPS data support that.

Isn't that the error in the first place? I mean, if the AC knew it was moving without user input, it should have stopped by logic. I don't think we can't exclude that the OP saw it moving just by reading the graphs. I mean, if he had everything filmed we might have a different story, film vs. logs. Obviously the images don't lie, the readings are... just an interpretation of the reality. Perhaps biased.

 

[sar104]

Isn't that the error in the first place? I mean, if the AC knew it was moving without user input, it should have stopped by logic. I don't think we can't exclude that the OP saw it moving just by reading the graphs. I mean, if he had everything filmed we might have a different story, film vs. logs. Obviously the images don't lie, the readings are... just an interpretation of the reality. Perhaps biased.

I'd have to disagree there. We have two quite separate diagnostics indicating no significant prior movement (GPS and accelerometers) together with pitch and roll measurements consistent with those data. I'm generally going to take redundant, correlated data over an eyewitness report in an event of this kind. We see the unreliability of pilot recollection of these kinds of events repeatedly confirmed on these forums, and, in this case, the data also correctly describe the overall outcome.

 

[7thstring]

Ok so you're suggesting that he didn't attempt the CSC move because he saw the AC going crazy? That DJI are right in saying it was totally pilot error?

 

[sar104]

Ok so you're suggesting that he didn't attempt the CSC move because he saw the AC going crazy? That DJI are right in saying it was totally pilot error?

No - there is clear evidence of the attempted CSC after the aircraft moved, but it was not the first stick input after takeoff. See graph 3 in post #22 - full negative elevator at 6.5 s started the aircraft moving.

 

[RayOZ]

Erm, apart from the yaw and slight drift ~0.5m with no stick input, why the sudden panic and attempt to CSC? It didn't seem to be moving very fast either. Maybe too close to trees? To test full control from RC, I would've move the stick a little bit to see how it reacts and then try to control. Or just wait a few more seconds to see if it stabilises itself. It only took off for about 6s. I guess I wasn't there to see what as going on, but I don't think I would've assumed it was going to fly away and try to CSC.
At ~26m altitude, CSC would crash the drone anyway. It would look like pilot's deliberate attempt to crash the drone, not sure how you can contest that with DJI.

 

[Sportking]

Did you have Refresh as well as a warranty?

As far as analysis, you can post your flight logs here.

This is good to know. Thanks for posting.

 

[SteveH]

Update:
I contacted DJI and asked if they would look over the info from this thread hoping that the compass or yaw anomalies would be enough to warrant repair under warranty. After about a week I received an email that the damage would not be covered under warranty. I paid for the repair and am waiting for shipment. So I guess the main thing I learned was that if a similar episode happens again is to just leave the controller alone. Do not try and save the drone, either let it stabilize on it's own or fly off and or crash. At least then it can't be attributed to operator error.

Thanks again to all that helped with the flight logs.

 

[Cbrokken]

My seven month old Mavic Pro crashed and broke off a rear motor. I sent it to DJI for repair and their analysis was pilot error. (no warranty) Here are the details. I launched the ac and once airborne I noticed the gimbal was not horizontal so I landed and did a shutdown on the ac. Restarted so that it would do a gimbal calibration again. I then did a compass calibration and set the bird down on concrete and waited for gps lock and home point record. When I launched the drone instead of rising up to the normal position it rose up about three or four feet and immediately started to fly off sideways with no stick inputs. So my first reaction was to stop the motors with full down and inward stick control which seemed to have no effect. The drone continued a sideways skid across the yard and smacked into the curb breaking off the rear motor.

What should I do?

I had on my fourth flight mavic air launched got up about 17 feet one of the props fell off it fliped upside down and smashed into the ground skittering acrost the ground wouldnt shut down the log stated it went -50 feet broke rear arm off they charged me 162$ to fix it 30 for parts

 

[Pixar]

Agreed :: concrete contained rebar

 

[7thstring]

Update:
So I guess the main thing I learned was that if a similar episode happens again is to just leave the controller alone. Do not try and save the drone, either let it stabilize on it's own or fly off and or crash. At least then it can't be attributed to operator error.

Or at least if you have concerns about it crashing into any obstacle while it's out of control, try sending it up rather than down.

 

[7thstring]

I had on my fourth flight mavic air launched got up about 17 feet one of the props fell off it fliped upside down and smashed into the ground skittering acrost the ground wouldnt shut down the log stated it went -50 feet broke rear arm off they charged me 162$ to fix it 30 for parts

Can you exclude you didn't properly mount the props?

 

[Bosque]

Replacing the rear motor isn't too difficult, I had to do that with my pro and have no particular electronics skills and did it following this you tube tutorial

 

[sar104]

Can you exclude you didn't properly mount the props?

Having examined the logs, I can exclude that.

 

[7thstring]

Having examined the logs, I can exclude that.

Thanks, but that was for Cbrokken

I had on my fourth flight mavic air launched got up about 17 feet one of the props fell off it fliped upside down and smashed into the ground skittering acrost the ground wouldnt shut down the log stated it went -50 feet broke rear arm off they charged me 162$ to fix it 30 for parts

 

[sar104]

Thanks, but that was for Cbrokken

I know that the question was directed at the OP, but the OP did not analyze the log files.

 

[7thstring]

I know that the question was directed at the OP, but the OP did not analyze the log files.

You should reread more carefully then. My question wasn't directed to the original poster (SteveH), but to the quoted poster (Cbrokken).

 

[sar104]

You should reread more carefully then. My question wasn't directed to the original poster (SteveH), but to the quoted poster (Cbrokken).

So it was - disregard my reply. In that case I completely agree with your comment - that was a likely explanation.