Help with Mavic 2 Pro Crash Location

[NorcalExplorer]

Put the Mavic up for a quick flight a few nights ago and haven't seen it since. Afterward, I noticed in the flight record that I got a "battery not installed properly" warning at 8.9s into the flight. Unfortunately, I failed to see the warning and must have been watching the drone at that time. At 3:55 into the flight I got a weak signal warning and the DJI Go app went dark. I hit RTH on the controller but noticed that the distance wasn't decreasing and thus the drone was not returning to me. I went to the area of it's last position and looked and looked to no avail. It was flying at around 380ft moving at about 30 mph when it seems to have disconnected. Maybe I've underestimated where it crashed and it went much further. If anyone can provide some insight, I would greatly appreciate it.

 

[sar104]

That's fairly easy to calculate using a simple drag model in a finite difference code, and I have that implemented for the Mavic 2 aerodynamics, with initial conditions for aircraft height, velocity, wind speed and direction. That gives the following in this case:



On the map, the descent path is the green trace:

 

[JAW]

That's fairly easy to calculate using a simple drag model in a finite difference code..........

please don’t take this the wrong way..........it’s a backhanded compliment, but that statement hit my funny bone.
,

 

[itsneedtokno]

That's fairly easy to calculate using a simple drag model in a finite difference code

Your knowledge continues to amaze me. I'm not sure if your are an electrical engineer, or a physicist, or maybe just a lamination technician like myself, but man... Your worth here, is priceless.

Keep it up!

 

[sar104]

please don’t take this the wrong way..........it’s a backhanded compliment, but that statement hit my funny bone.

I guess I wrote that rather poorly - what I meant was that it only takes a few lines of code to solve these kinds of differential equations numerically.

Your knowledge continues to amaze me. I'm not sure if your are an electrical engineer, or a physicist, or maybe just a lamination technician like myself, but man... Your worth here, is priceless.

Keep it up!

Thanks. Physicist. Just trying to share the power of physics.

 

[NorcalExplorer]

sar104,

Thanks for those calculations. I spent 3 hours yesterday looking and still haven't found it. That's in addition to the two other times I've already scoured that general area.

I doubt someone found it and took it. There just aren't that many people around there.

I had been assuming that the drone experienced a catastrophic power loss and crashed, but now I'm starting to wonder if it just kept on flying.

Thoughts?

 

[JAW]

....the power of physics.

Well, to me the power of physics is what happens the next day after eating a box of prunes.
??

 

[peterc4]

Same happend to me recently lost my m2p dji didnt want to know but did give me 30% of a new one, i think since i lost mine you are the 3rd on here to loose under very smilar circumstances, lost mine in april

 

[Paulf10]

Perhaps if you use SAR104 crash site coordinates and your mobile phone to track your movements and the begin a slow, methodical, overlapping grid search starting from SAR’s estimated crash site location you may have better luck. Just a thought. Best wishes for your success.

 

[sar104]

sar104,

Thanks for those calculations. I spent 3 hours yesterday looking and still haven't found it. That's in addition to the two other times I've already scoured that general area.

I doubt someone found it and took it. There just aren't that many people around there.

I had been assuming that the drone experienced a catastrophic power loss and crashed, but now I'm starting to wonder if it just kept on flying.

Thoughts?

That's strange - I be confident in those coordinates to within 20 ft or so if it was a mid-air shutdown. It won't have just kept on flying - with loss of connection it will have switched to failsafe, which was probably RTH. You might want to see if you can retrieve the mobile device DAT file ending FLY092.DAT - it's possible that it will contain some other clues.

 

[NorcalExplorer]

That's strange - I be confident in those coordinates to within 20 ft or so if it was a mid-air shutdown. It won't have just kept on flying - with loss of connection it will have switched to failsafe, which was probably RTH. You might want to see if you can retrieve the mobile device DAT file ending FLY092.DAT - it's possible that it will contain some other clues.

The area where it went down has a lot of sagebrush and manzanita and it's very possible that it's still there and I missed it. I do trust your coordinates though and will be going back for a 4th time to look again. Maybe the .dat file can help. I was able to look at it on airdata and can see that it shows a 38 second longer flight time than the .txt file. There is also a rapid decrease in battery from 60% down to 10% that then bounces back to 83% at the end.

 

[NorcalExplorer]

Same happend to me recently lost my m2p dji didnt want to know but did give me 30% of a new one, i think since i lost mine you are the 3rd on here to loose under very smilar circumstances, lost mine in april

Sorry to hear that. I know I'm bummed out to lose mine. It's been such a reliable drone that this really surprised me. Learning lots of lessons though. And, not giving up the search yet. I know it's gotta be out there somewhere!

 

[NorcalExplorer]

Perhaps if you use SAR104 crash site coordinates and your mobile phone to track your movements and the begin a slow, methodical, overlapping grid search starting from SAR’s estimated crash site location you may have better luck. Just a thought. Best wishes for your success.

That is essentially how I was looking for it. Very methodically. I'll be going back to conduct another search. The thick brush in that area could have easily concealed it from me. Thanks for the suggestion. I'll take all the help I can get!

 

[sar104]

The area where it went down has a lot of sagebrush and manzanita and it's very possible that it's still there and I missed it. I do trust your coordinates though and will be going back for a 4th time to look again. Maybe the .dat file can help. I was able to look at it on airdata and can see that it shows a 38 second longer flight time than the .txt file. There is also a rapid decrease in battery from 60% down to 10% that then bounces back to 83% at the end.

The DAT file also shows multiple battery error entries starting at 2 seconds:

2.080 : 1980 [L-BATTERY][ERROR] BATTERY_STATE_ABSENT​

but no other obvious clues that I can see. It looks like a power disconnect shut down. It has to be there.

 

[NorcalExplorer]

The DAT file also shows multiple battery error entries starting at 2 seconds:

2.080 : 1980 [L-BATTERY][ERROR] BATTERY_STATE_ABSENT​

but no other obvious clues that I can see. It looks like a power disconnect shut down. It has to be there.

Thanks for your expertise. I will continue to search the area.

 

[BudWalker]

That's fairly easy to calculate using a simple drag model in a finite difference code, and I have that implemented for the Mavic 2 aerodynamics, with initial conditions for aircraft height, velocity, wind speed and direction. That gives the following in this case:

View attachment 103948

On the map, the descent path is the green trace:

View attachment 103949

I'm a little surprised that the north and east curves are so straight - i.e the velocities were constant. The drag would have decreased both velocities. Also, the M2 was travelling into the wind which would have decreased those velocities even more.

 

[BudWalker]

The area where it went down has a lot of sagebrush and manzanita and it's very possible that it's still there and I missed it. I do trust your coordinates though and will be going back for a 4th time to look again. Maybe the .dat file can help. I was able to look at it on airdata and can see that it shows a 38 second longer flight time than the .txt file. There is also a rapid decrease in battery from 60% down to 10% that then bounces back to 83% at the end.

Those extra 38 secs occurred before the flight. The .txt will only contain data for the actual flight.

 

[NorcalExplorer]

Those extra 38 secs occurred before the flight. The .txt will only contain data for the actual flight.

Good to know. I was wondering if that was why there was a difference. Thanks for that clarification.

 

[sar104]

I'm a little surprised that the north and east curves are so straight - i.e the velocities were constant. The drag would have decreased both velocities. Also, the M2 was travelling into the wind which would have decreased those velocities even more.

No - it was traveling with the wind - around 6 m/s out of the NW.

 

[BudWalker]

No - it was traveling with the wind - around 6 m/s out of the NW.

You're right. I was looking at the battery_info:wind_dir and assuming the usual DJI convention was being used.

I printed your graph and laid a straight edge on the North and East plots and there is a slight bend to those curves.