MP2 Crash_Inconclusive Data-analysis result from Dji_PLEASE HELP

[Justin.G]

Hello fellow Mavic Pilots.


This is my first post, I’m not 100% sure how the posting process works but I hope my research is correct. Apologies in advance for the long post.


Sadly, my Mavic 2 Pro had an ‘unexplained’ incident recently and I am having some trouble communicating my concerns with Dji. Understandably, Dji cannot simply take every crash at face value and provide replacements for every case but I am convinced that the fall was not pilot error. My understanding of their data-analysis process is to investigate and rule-out pilot or manufacturing fault. I assume crashes are binary, either pilot or manufacturing.

In a nut-shell, I was nearing the end of a normal condition flight, no strong wind and the weather was great. After completing a clear horizontal movement past a nearby building, I complete a sensor, battery and GPS check and proceed to line up the aircraft safely between 2 structures. I pitch forward to begin a slow movement.

After clearing past the structures and approximately 40 seconds into the movement, with no inference or low signal warning, the video feed cuts out and all control is lost. I release the sticks and wait approximately 20-30 seconds for the RC to re-connect before hitting the RTH. After a some time passed and still no connection or response, I check Find my Drone and proceed to find the aircraft. After searching, I find it approximately 25-30 meters from the last position recorded. It had fallen through a tree and was heavily damaged, gimbal hanging by a thread, bottom plate broken off etc. The battery was still attached with 2 bars remaining. I later attempted putting in a fully charged battery but the craft did not power on.


I have attached both the log.txt file file from the flight as well as details of the last known position and the physical location of where I found the aircraft.

Link to last cached video:

If the veterans of the forum could PLEASE help me make some sense of what happened to my much loved bird, it would be highly appreciated.

Thank you!

 

[slup]

Nothing in the flight telemetries indicates anything odd ... only thing that change when you proceed in between those 2 buildings is that the locked satellites drops quickly. But that hadn't anything to do with this, you still had navhealth 4/5 & 10 sats locked.

My thinking is that you quickly lost your connection due to going out of line of sight behind the building ... as the RTH height was set at 30m the AC started it's failsafe RTH at the present height which was 127m. In between your HP & the AC position was the building ... either the AC crashed into it on the way back or the OD sensors stopped it & the AC low battery force landed (threshold 16%) there close to the building wall & straight down into the tree, it was roughly 6 minutes from 37% battery you had when the log ended until the landing should start.

Further one reason can be possible ... the sats dropped further when you lost connection & the AC were left in ATTI mode which had blocked failsafe RTH & wind had blown it into the building. But I think that theory is highly unlikely depending on where the AC was found ... it doesn't correlates with the wind direction which was coming from south - southwest...

 

[Phantomrain.org]

Hello fellow Mavic Pilots.


This is my first post, I’m not 100% sure how the posting process works but I hope my research is correct. Apologies in advance for the long post.


Sadly, my Mavic 2 Pro had an ‘unexplained’ incident recently and I am having some trouble communicating my concerns with Dji. Understandably, Dji cannot simply take every crash at face value and provide replacements for every case but I am convinced that the fall was not pilot error. My understanding of their data-analysis process is to investigate and rule-out pilot or manufacturing fault. I assume crashes are binary, either pilot or manufacturing.

In a nut-shell, I was nearing the end of a normal condition flight, no strong wind and the weather was great. After completing a clear horizontal movement past a nearby building, I complete a sensor, battery and GPS check and proceed to line up the aircraft safely between 2 structures. I pitch forward to begin a slow movement.

After clearing past the structures and approximately 40 seconds into the movement, with no inference or low signal warning, the video feed cuts out and all control is lost. I release the sticks and wait approximately 20-30 seconds for the RC to re-connect before hitting the RTH. After a some time passed and still no connection or response, I check Find my Drone and proceed to find the aircraft. After searching, I find it approximately 25-30 meters from the last position recorded. It had fallen through a tree and was heavily damaged, gimbal hanging by a thread, bottom plate broken off etc. The battery was still attached with 2 bars remaining. I later attempted putting in a fully charged battery but the craft did not power on.


I have attached both the log.txt file file from the flight as well as details of the last known position and the physical location of where I found the aircraft.

Link to last cached video:

If the veterans of the forum could PLEASE help me make some sense of what happened to my much loved bird, it would be highly appreciated.

Thank you!

So without seeing the data ,

In this case a few errors were made:

Flying between building is a risk .
Flying between building a little to low for comfort a risk
Letting go of the Sticks when the drone possibly went into Atti Mode not good.

So a little risk with some Pilot error and that is a Recipe for disaster for any pilot.

My advice : When taking risk you need to have an exit plan ready to go which in this case was to raise the drone straight up ready for Atti mode.
Sorry about your drone.

Phantomrain.org
Gear to fly your Mavic in the Rain
Coal

 

[Keule]

In between your HP & the AC position was the building ...

Don't forget the M2P retraces its flight path for 60seconds in FailSafe RTH. (Provided sufficient GPS health)
If that's the case the AC would be back in LOS in reference to the HP.

 

[sar104]

I think it's quite clear what happened. First - you lost uplink and downlink by flying behind a tall building:





While a really bad plan, it would probably have been okay if you had not set the failsafe behavior to hover, as seen in the brief period of RTH earlier in the flight:



Finally, that close to the building the sky view was restricted and the GPS reception fell off rapidly:



The result was that the aircraft was hovering and probably struggling to hold position. It either continued doing that until smart battery autoland kicked in and it landed in the trees, or possibly it drifted into the building (you had OA set to off) and fell into the trees.

 

[sar104]

Nothing in the flight telemetries indicates anything odd ... only thing that change when you proceed in between those 2 buildings is that the locked satellites drops quickly. But that hadn't anything to do with this, you still had navhealth 4/5 & 10 sats locked.

My thinking is that you quickly lost your connection due to going out of line of sight behind the building ... as the RTH height was set at 30m the AC started it's failsafe RTH at the present height which was 127m. In between your HP & the AC position was the building ... either the AC crashed into it on the way back or the OD sensors stopped it & the AC low battery force landed (threshold 16%) there close to the building wall & straight down into the tree, it was roughly 6 minutes from 37% battery you had when the log ended until the landing should start.

View attachment 99302

View attachment 99303

Don't forget the M2P retraces its flight path for 60seconds in FailSafe RTH. (Provided sufficient GPS health)
If that's the case the AC would be back in LOS in reference to the HP.

RTH was set to "hover" so that's all moot.

 

[RayOZ]

Conclusion. Pilot error.
1. Flew behind building = lost signal to RC
2. RTH not set correctly.
3. Flew around tall buildings = loss of GPS signal.
Sorry about your loss. Live and learn.

 

[timinator]

That whole flight was a bad idea from start to finish. So many things wrong: highly urban environment with tall buildings, violating vlos in that environment, and no clear mission plan really put passersby and property at risk of injury and damage. Knowing when and where to fly is almost half the battle of a good flight outcome. Please choose a better location for your next flight (after your aircraft is repaired or replaced).

 

[slup]

RTH was set to "hover" so that's all moot.

Jeez ... having it set to hover were so risky in the surroundings that I didn't even think of that

 

[Justin.G]

Nothing in the flight telemetries indicates anything odd ... only thing that change when you proceed in between those 2 buildings is that the locked satellites drops quickly. But that hadn't anything to do with this, you still had navhealth 4/5 & 10 sats locked.

My thinking is that you quickly lost your connection due to going out of line of sight behind the building ... as the RTH height was set at 30m the AC started it's failsafe RTH at the present height which was 127m. In between your HP & the AC position was the building ... either the AC crashed into it on the way back or the OD sensors stopped it & the AC low battery force landed (threshold 16%) there close to the building wall & straight down into the tree, it was roughly 6 minutes from 37% battery you had when the log ended until the landing should start.

Further one reason can be possible ... the sats dropped further when you lost connection & the AC were left in ATTI mode which had blocked failsafe RTH & wind had blown it into the building. But I think that theory is highly unlikely depending on where the AC was found ... it doesn't correlates with the wind direction which was coming from south - southwest...

View attachment 99302

View attachment 99303

Thank you so much for your feedback and for taking the time to illustrate the flight path. I haven’t been flying for quite as long as most here, so I guess for some reason I thought I’d seen the worst and maybe pushed the craft a little too confidently. I’ve learnt a lot just from your response, this 1 and the later message about the hover and risk. I will definitely keep your analysis in mind for the next flight.
Thanks again!

 

[Justin.G]

So without seeing the data ,

In this case a few errors were made:

Flying between building is a risk .
Flying between building a little to low for comfort a risk
Letting go of the Sticks when the drone possibly went into Atti Mode not good.

So a little risk with some Pilot error and that is a Recipe for disaster for any pilot.

My advice : When taking risk you need to have an exit plan ready to go which in this case was to raise the drone straight up ready for Atti mode.
Sorry about your drone.

Phantomrain.org
Gear to fly your Mavic in the Rain
Coal

Thanks a lot for the advice! More than an explanation, you’ve provided great strategy for the next flight. Much appreciated!

 

[Justin.G]

Don't forget the M2P retraces its flight path for 60seconds in FailSafe RTH. (Provided sufficient GPS health)
If that's the case the AC would be back in LOS in reference to the HP.

Although this proved not be the case later (through my own fault) I did not know that the M2P did this. Awesome!

 

[Justin.G]

I think it's quite clear what happened. First - you lost uplink and downlink by flying behind a tall building:

View attachment 99305

View attachment 99306

While a really bad plan, it would probably have been okay if you had not set the failsafe behavior to hover, as seen in the brief period of RTH earlier in the flight:

View attachment 99308

Finally, that close to the building the sky view was restricted and the GPS reception fell off rapidly:

View attachment 99309

The result was that the aircraft was hovering and probably struggling to hold position. It either continued doing that until smart battery autoland kicked in and it landed in the trees, or possibly it drifted into the building (you had OA set to off) and fell into the trees.

This just might be my most important lesson from this whole incident. Thank you so much for your time and analysis. Solid knowledge for the future.

 

[Justin.G]

Conclusion. Pilot error.
1. Flew behind building = lost signal to RC
2. RTH not set correctly.
3. Flew around tall buildings = loss of GPS signal.
Sorry about your loss. Live and learn.

Definitely a big learning curve, to be honest, from this incident, I’ve learnt more than I believe I otherwise would of. So your honesty is much appreciated. Thank you.

 

[Justin.G]

That whole flight was a bad idea from start to finish. So many things wrong: highly urban environment with tall buildings, violating vlos in that environment, and no clear mission plan really put passersby and property at risk of injury and damage. Knowing when and where to fly is almost half the battle of a good flight outcome. Please choose a better location for your next flight (after your aircraft is repaired or replaced).

You’re very much correct. I suppose sometimes flying a little too confidently leads to some very stupid outcomes. I will definitely select locations more responsibly next time. I instantly felt the dread of someone possibly being injured after the fact, very scary feeling.

This might possibly be a silly question but please try to be patient, Dji has not ruled this as my error nor hardware fault. What you believe to your experienced knowledge will be my chances of warranty repair?

 

[Justin.G]

That whole flight was a bad idea from start to finish. So many things wrong: highly urban environment with tall buildings, violating vlos in that environment, and no clear mission plan really put passersby and property at risk of injury and damage. Knowing when and where to fly is almost half the battle of a good flight outcome. Please choose a better location for your next flight (after your aircraft is repaired or replaced).

You’re very much correct. I suppose sometimes flying a little too confidently leads to some very stupid outcomes. I will definitely select locations more responsibly next time. I instantly felt the dread of someone possibly being injured after the fact, very scary feeling.

This might possibly be a silly question but please try to be patient, Dji has not ruled this as my error nor hardware fault. What you believe to your experienced knowledge will be my chances of warranty repair?

 

[sar104]

If they figure out correctly what happened, which is not a given with DJI, then they will deny the warranty claim. So it depends to some extent how carefully they examine the txt log and mobile device DAT file. If you send them the aircraft they will likely pull the aircraft DAT, and that will almost certainly show them that it was not equipment failure, and not a warranty issue.

 

[timinator]

You’re very much correct. I suppose sometimes flying a little too confidently leads to some very stupid outcomes. I will definitely select locations more responsibly next time. I instantly felt the dread of someone possibly being injured after the fact, very scary feeling.

This might possibly be a silly question but please try to be patient, Dji has not ruled this as my error nor hardware fault. What you believe to your experienced knowledge will be my chances of warranty repair?

Slim to none on the warranty repair.
Your over-confidence statement reminds me of an old flying quote: "There are old pilots, and there are bold pilots, but there are no OLD BOLD pilots"

 

[Justin.G]

If they figure out correctly what happened, which is not a given with DJI, then they will deny the warranty claim. So it depends to some extent how carefully they examine the txt log and mobile device DAT file. If you send them the aircraft they will likely pull the aircraft DAT, and that will almost certainly show them that it was not equipment failure, and not a warranty issue.

Thanks @sar104 Understood. From the official DJI forum I’ve seen some pretty strange replacement/repair policies. Cases where DJI ruled obvious pilot error and then later over turned and replaced. Well, I’ve sent them as much information as I possibly could at this point which is basically what I’ve listed in my original post here. They requested me to retrieve said DAT files from the aircraft but everything on the belly of the aircraft is smashed in and the craft is unresponsive. I have been wrongly aggressive with DJI and in hindsight, I should have posted here first. But as was mentioned earlier, my fault, you live and learn. Thanks again.

 

[sar104]

Thanks @sar104 Understood. From the official DJI forum I’ve seen some pretty strange replacement/repair policies. Cases where DJI ruled obvious pilot error and then later over turned and replaced. Well, I’ve sent them as much information as I possibly could at this point which is basically what I’ve listed in my original post here. They requested me to retrieve said DAT files from the aircraft but everything on the belly of the aircraft is smashed in and the craft is unresponsive. I have been wrongly aggressive with DJI and in hindsight, I should have posted here first. But as was mentioned earlier, my fault, you live and learn. Thanks again.

Their first response, obviously based on a very cursory look at the files, is often pilot error. They will subsequently change that if the data show otherwise and it is pointed out to them. In your case it simply was pilot error, and so it depends on whether they get it wrong in the other direction, which happens, but much more rarely.