MA2 goes up and down without interaction.

[Bjorn.Cotteleer]

Today I lost my drone in the Rio De La Plata in Uruguay. After it started descending by itself. I made the mistake of getting distracted and lost sight of the drone.
In the past the drone had been going up and down in the air without any interaction.
Does anyone have a clue why the drone changes altitude without any command being given. At the time of flight it had a gps fix of 13/14 satellites.
I will need to order a new drone now .
The crash is on my Instagram.com/Bjorn.Cotteleer

 

[Meta4]

In the past the drone had been going up and down in the air without any interaction.
Does anyone have a clue why the drone changes altitude without any command being given.

Your flight data might help solve the mystery.
Go to DJI Flight Log Viewer | Phantom Help
Follow the instructions there to upload your flight record from your phone or tablet.
That will give you a detailed report of the flight.
Come back and post a link to the report it gives you.
Or .. just post the txt file here.

 

[GadgetGuy]

Your flight data might help solve the mystery.

...and the potential of a possible free warranty replacement, if no pilot error was involved.

 

[G Randy Brown]

I don't take my MA2 out a lot but the last time I did I noticed it descending slowly while hovering on several flights.
I just did the update so fingers crossed that fixes it.

 

[db4476]

Interesting, I have not experienced this yet...

Cheers!

 

[Bjorn.Cotteleer]

Your flight data might help solve the mystery.
Go to DJI Flight Log Viewer | Phantom Help
Follow the instructions there to upload your flight record from your phone or tablet.
That will give you a detailed report of the flight.
Come back and post a link to the report it gives you.
Or .. just post the txt file here.

When I replay my flight log, it clearly shows that I’m not moving the joysticks. But the strange thing is that the altitude given by the drone remained exactly the same. I took of from a bollard on the ship (+/-10m ASL). The altitude given was 1.6m above Homepoint. And it remained 1.6m in the replay until it sank in the water 10m lower.

 

[Meta4]

Today I lost my drone in the Rio De La Plata in Uruguay. After it started descending by itself. I made the mistake of getting distracted and lost sight of the drone.
In the past the drone had been going up and down in the air without any interaction.
Does anyone have a clue why the drone changes altitude without any command being given.

You launched from out at sea, took the drone up 1.3 metres higher than the launch point, flew it a short distance and left it hovering for 22 seconds.

Here is a graph showing the height measured by the IMU sensors and the VPS, downward looking sensors.
Green - IMU sensors are measuring the airpressure to show height above launch point.
Blue - VPS sensors are using visual and infrared sensors to measure height above the water.

You can see that the IMU height is fairly stable with small variations.
These variations are from 1.1 metres to 1.9 metres, a total range of 0.8 metres, which is within the range of hovering precision shown in the specs for the Air 2 (±0.5 m).

But the VPS height (the true height above the water), is not as steady and contains a lot of variations that are larger that the IMU height variations early in the flight with some very large variations toward the end of the graph.

What this is reflecting is that the drone didn't gradually lose height and sink, but that the ocean's surface is not flat and that it rises and falls with the passing of waves or swells.

The peak marked X shows the drone 4 metres above the water surface.
Or to view that a different way, the water surface was 2.5 metres lower than it was a little earlier.
This is followed by the low point (marked Z) where the water level has risen up to the level where the drone was hovering (note that the IMU height hardly moves, indicating the drone was still in a stable hover).
And then the show is over.

The problem was that you left the drone hovering too low above the sea, not conscious of the swells rising and falling.
Had you put the drone higher, or watched and raised it before the swell came through, it would have stayed safe.

At the time of flight it had a gps fix of 13/14 satellites.

Altitude data comes from the barometric sensor in the IMU.
It is not affected by GPS.

 

[TnDronePilot]

Even though I feel read bad that Bjorn lost his drone, that explanation was amazing to read. I got a technical woody

 

[twickers14]

I have very little technical knowledge but Meta4s (love the name) is so clear and understandable. I shall never go low over swiftly changing water levels (don’t fly over a lot of water anyway) thanks to this explanation.

 

[TnDronePilot]

I have very little technical knowledge but Meta4s (love the name) is so clear and understandable. I shall never go low over swiftly changing water levels (don’t fly over a lot of water anyway) thanks to this explanation.

And through a cloud. Did you see one guy's video where he almost couldn't get the drone to land because the downward sensors thought the thick cloud was the ground? I would have shipped my pants.

 

[Capt KO]

Had same problem over an elevated deck painted all one color, grey. The A2 ground sensor was unable to pick up the ground at all, same with water. Solved my water flights with a Rescue Jacket from Phantomrain. Ends the pucker factor. Tough lesson for sure.

 

[Meta4]

Had same problem over an elevated deck painted all one color, grey.

The elevated deck would have had to rise and fall a few metres for you to have had the same problem.

 

[Capt KO]

The elevated deck would have had to rise and fall a few metres for you to have had the same problem.

True. Only referring to the lack of help from the sensor when low over a mono color surface. Sorry for the comparison. I can blame it on dementia, perks of old age.
Enjoy