İ wasn’t flying over water…just near it 
. That’s why I just put it above my head as I was standing on solidish ground
It’s one thing it not being happy over water…it’s another going full throttle nuts and the Pause not even working .
But who the heck knows. I’ll be avoiding puddles in future just in case 
Your guess is as good as mine. About 30 minutes down the coast is Göcek where all the seriously rich guys have their super yachts and Russian ones with their tracking turned off…allegedly. Abromovich had one of his there for a while. Wouldn’t surprise me there!
Yeah " the drone’s downward vision positioning system may malfunction over reflective surfaces like water, causing erratic behavior or loss of control" - " avoid flying over water and at night.]** DJI Neo does not support Obstacle Avoidance, and Palm Control is not available in environments with poor lighting . It is recommended to fly in open areas that have good lighting without obstacles.
All i am doing is duplicating what dji have said, and yes people have flown over water etc, but if you do something that the manual or advises not to, a company could argue you are missing using the product. This is a general statement and just highlighting what i have found online via dji and other social media sites,
Is it possible that some sort of gps spoofing or jamming is in play here? Not that it would be easy to find out because if there is, it will be covert and possibly illegal…
Yes, DJI typically (if not always) sends a replacement drone rather than repairing the one sent in. Most people report positive experiences with those replacements.
Your video shows the Neo was facing a different direction than what’s recorded in the flight log. On the right side in the image below, you can see about a 180-degree discrepancy. This is often a sign that the compass was disrupted by an external magnetic source.
Mmmm interesting.
So the red arrow is based on the compass ?
That explains why from the data it looked like it turned to fly in the direction it was travelling but the video shows it was pointing out to sea and flying backwards.
The Flyaway from 27th December doesn’t show a red arrow at all ?
So this is the reply from DJI.
It’s good they will replace it FOC and I have accepted their kind offer but I have gone back to them to look at it again as I’m just not happy it should have behaved like that if it was simply being beside water.
Why they picked out 22 seconds is a mystery as the drone kept climbing until 31 seconds when it reached 27 meters high and didn’t go out of control untill 1:19 immediately after I centred Yaw
I am fairly persistent so won’t let them away quitely
" We have now received results of the data analysis for the flyaway case and please inform us, on how you wish to proceed in regards to your remote controller.
The result is as follows:
1. The aircraft worked under GPS mode;
2. Flight Time T=00:22, Relative Height H=8.4 m, Distance to Home Point D=5.2 m, The aircraft stayed close to the water. Drift of aircraft and accidents due to ineffective positioning of aircraft on the water surface.
3. The incident coordinate: 36.6227207, 29.0998615;
4. The DJI Neo user manual has instructions that the aircraft cannot be flown over water; Please keep the water at least 10m above the surface to fly.
According to the analysis, the incident was caused by improper operation, please pay attention to the flight environment.
We understand that it takes time to familiarize yourself with your Neo. Considering your experience with DJI’s products, we would like to make a special gesture and offer you a free replacement for this case.
For safe flights in the future, we kindly recommend paying close attention to your flight environment and maintaining a minimum distance of 3 meters when flying over bodies of water."
Good to hear that DJI have replaced it FOC 
DJI will never admit to an issue but will replace FOC if there is one
The red arrow represents the yaw direction, indicating which way the drone is pointed. Yes, data from the compass is used to orient the drone correctly.
The yaw for that flight was -161.4 at takeoff. The arrow doesn’t appear on the map until 0m 44.0s into the flight since the latitude and longitude data was not available until that point.
Did you send the video to DJI? If not, the flight log suggests you were flying over water. Even if that were the case, the drone shouldn’t have behaved the way it did.
Mike, they had the Video Cache so they will have had the video. I ddi send the YT link again also
I’ve challenged them politely on their (lazy) analysis 
and pointed out the video and the glaring inconsistencies with their conclusion and expected behaviour even if they were right.
We will see. They have normally been very quick to reply but they haven’t replied to my email from this morning (its now evening here) so maybe they are having another look at it.
As long as they don’t reconsider their FOC offer
Yeah, maybe don’t push them on it. As we’ve been discussing in your other thread, it seems likely there were magnetic metal objects, such as nails, bolts, or screws, in the dock and handrail. This is a common issue when taking off from concrete sidewalks in the US, as many contain metal rebar.
For others reading along, magnetic interference at takeoff isn’t always obvious. If you’re unsure, check the map in the DJI Fly app and ensure the drone’s direction on the map matches its actual orientation. If the direction is incorrect (like in the image I shared above), something is interfering with the drone’s compass. In such cases, power the drone down, move it to a different location, and try again. Don’t take off until the compass shows the correct direction.
DJI are going to replace it free of charge so that’s is good.
I think a prop failure could case the same results. Motor spins faster causing the ESC fault, the lack of one prop causes the spinning and yawning.
Heck but free of charge, smile and take the ~second~ third replacement.
The replacement Neo is in its way
DJI say I can’t speak to their Data Analysis team (but that may be a language barrier !), but they said their Technical Department can phone me. I said yes.
Mikes theory about the compass being almost 180 degrees out , probably because I took off from a steel pier covered in wood sounds like a contributory factor
DJI haven’t identified that (to me). The drone was stable in ascent and yaw then shot off so I’m curious what they say about why it shot off. If the compass was 180 degree out and if surrounding water confused the sensors that’s all well and good . Hopefully they will have a view on what then happened or can feed it back to their development team. There should be some failsafe in there
If the compass isn’t functioning properly, I don’t think the drone could do much other than land at its current location. Even if such a feature could be implemented, the flight path might still be risky until the drone reaches the ground.
As mentioned earlier, the safest approach is to ensure the drone is correctly oriented in DJI Fly before takeoff. This quick preflight check is especially crucial when flying in unfamiliar locations.
Sorry to read about your experience, I noted how smooth your flight was at the begining and how good the conditions were.
I had a Phantom 4. I took off from an industrial unit to about 8’ tested fwd, rev, lt, rt, up, down as usual. No errors, everything normal. I then started my flight but the drone just drifted away and crashed into an industrial unit. When I investigated it turned out to be a compass error caused by steel in the concrete apron of the industrial unit! Here’s the thing, there were no warnings on the DJI app, at the time of take off. Had there been I’d have brought the drone straight back down. Tough lesson, I now have an Argos portable picnic table in the back of the car for take off, if I suspect anything, not ideal but assembles in less than a minute.
Hi Zaphod. Another HHG fan
Yes that was probably the issue.
A new check for me will be checking the arrow is pointing the way the drone is facing, but yes I will be much more cautious about metalwork near take-off!
DJI have delivered the free replacement drone.
I am waiting for a call with their Tech Dept today to dicuss what they determined just for my own curiosity
Happy flying 
To draw a line under this the replacement Neo arrived today in Scotland. It will take me a while to get it out to Turkey
I had a chat with the DJI Tech Department . Pleasant to talk to and didn’t seem rushed or defensive.
I talked through the incident and my main safety concerns about not being able to stop the drone with Pause and it flying 4 x faster than Normal mode should allow.
He said they were incorrect when they said it was caused by flying over water. It never did.
He said it wasn’t Operator Error but in hindsight and with more knowledge , in my mind, flying it from where I did was an error and a new check has been added to my pre-flight.
I owned up about the compass pointing in the opposite direction to the direction the drone was facing. He didn’t leap on that. Just said lots of things can interfere with the GPS signal and / or compass. There are assorted failsafes built in. I think he said they were missing some metadata from when the GPS complained briefly about signal. He said they just don’t see this type of behaviour.
He tried the “wind may have caused it to drift” but there was no wind. The bay was like a mirror.
Bottom line is they can’t explain it , or why it flew away from Home at 52.6 mph and couldn’t be Paused
But I shouldn’t worry … it will probably never happen again
Thanks everyone for your input 
Good Work, Happy Flying too
Best Wishes
