P4P flipped in gusting winds?

[phil99]

Hi there guys, hoping someone has some sensible input on this... flying P4P in 15mph gusting to 25mph winds... windy but not too much... the drone has around 200 plus flights on it and have flown in worse... however the only thing I can think of is that it was hit by some kind of swirling wind... So I was hovering at 50 meters after flying for around 17 minutes... when all of a sudden the camera goes into gimbal over limit and hangs loose so I cannot see clearly what is happening.... I put in no stick commands hoping it is a gust and will come back.... then the drone just falls from the sky at 10m/s (thats a free fall?) I am looking at the screen and did not see what the drone looked like as it fell... it was literally over in seconds.. as you can see.... so I go and find the drone and luckily it had a soft landing in a bush and is fine.....

Looking at the data it gave a motor speed error as it fell... compass errors and that the compass had changed direction too much in a fraction of a second... this is what made me initially think a propellor had come off. But when I found the drone, it is all ok... and the weird thing is, it works perfectly and flies perfectly now... Also it was running at maximum amps as it fell something like 24amps? Its very strange.

So.... is there something wrong with the drone, or did I get hit by a freak wind? I have flown it again now around 15 times without issue.

DJI Flight Log Viewer | Phantom Help

Upload and view flight logs from your DJI drones

www.phantomhelp.com

And log is here

 

[phil99]

To add on to this, here are the warnings I get from airdata...: during the final fall:


N 17m 34s 53.5 m 139 m Warning Gimbal Roll Reached Movement Limit O 17m 35s 53.4 m 140 m Low Risk Compass Error (repeated 9 times) P 17m 36s 51.9 m 143 m Low Risk Low severity, impact on safety is likely not significantCompass Error. Low severity, impact on safety is likely not significantNot Enough Force/Esc Error (repeated 26 times)Q 17m 38s 33.9 m 174 m Low Risk Speed Error. Low severity, impact on safety is likely not significantNot Enough Force/Esc Error (repeated 25 times)

 

[phil99]

As said when looking at Airdata... the amps pulled during the whole flight were almost all green below 15A... but during its free fall it was purple... pulling over 25A... giving 25 Not Enough Force/Esc Error warnings, and 25 compass error warnings (compass could be due to it spinning?)

 

[Meta4]

flying P4P in 15mph gusting to 25mph winds... windy but not too much... the drone has around 200 plus flights on it and have flown in worse... however the only thing I can think of is that it was hit by some kind of swirling wind.

It would have to be a very unusual wind to flip a drone.
Yours was not flipped at all, as confirmed by the pitch and roll data.

then the drone just falls from the sky at 10m/s (thats a free fall?)

In free fall a phantom quickly reaches a terminal velocity of 15-17 metres/sec.
Yours never exceeded 10 m/s.
In free fall, a Phantom tumbles, yours maintained a stable attitude, with pitch and roll around 10 degrees.

Looking at the data it gave a motor speed error as it fell... compass errors and that the compass had changed direction too much in a fraction of a second... this is what made me initially think a propellor had come off. But when I found the drone, it is all ok... and the weird thing is, it works perfectly and flies perfectly now... Also it was running at maximum amps as it fell something like 24amps? Its very strange.

So.... is there something wrong with the drone, or did I get hit by a freak wind? I have flown it again now around 15 times without issue.

Your incident is mysterious, particularly given that there was no damage and the drone flies properly now.
Your drone was hovering at 175 ft with you rotating slowly clockwise.
At 17:34.6 the drone rolled heavily (24 degrees) to the left and started rotating anti-clockwise, quickly picking up speed.
Within half a second it was losing height and spinning.
But the spinning stopped shortly afterwards and the data stops well before the drone crashed.

Something went wrong, but I can't tell what it was.

 

[phil99]

It would have to be a very unusual wind to flip a drone.
Yours was not flipped at all, as confirmed by the pitch and roll data.


In free fall a phantom quickly reaches a terminal velocity of 15-17 metres/sec.
Yours never exceeded 10 m/s.
In free fall, a Phantom tumbles, yours maintained a stable attitude, with pitch and roll around 10 degrees.

Your incident is mysterious, particularly given that there was no damage and the drone flies properly now.
Your drone was hovering at 175 ft with you rotating slowly clockwise.
At 17:34.6 the drone rolled heavily (24 degrees) to the left and started rotating anti-clockwise, quickly picking up speed.
Within half a second it was losing height and spinning.
But the spinning stopped shortly afterwards and the data stops well before the drone crashed.

Something went wrong, but I can't tell what it was.

Thanks for your input... it is for sure strange to me.

Is it safe to assume that the compass is correct seeing as it also reported 25 compass errors on the way down.... ie it may show it was steady, but could be spinning? Although the rudder input does seem to correspond with a change in the heading slightly before the descent. So yes...

Also cannot see why when it is descending at 10m/s the battery current draw goes up to over 25amps... the only time is has done that all flight.

It hit a tree I assume when the log stops, as I found the battery 20 meters behind the tree, and the drone in a bush below the tree. It is all in perfect condition.. not even a scratch on it... I assume that is just luck as the last data it is travelling at over 30mph.

A very very strange down draft wind would result in the motor esc errors,, the props had not enough load on them? And may result in high current as the drone would push as hard as it could? Although without enough load on the props to fly, it wouldn't draw max current?

Or could it have had a bird on top of it?! (I laughed at first at this... but it would correspond with it dropping at the same time as the motor current going out of the roof?!)

My main concern is that it is now dangerous to fly this drone. Or is there nothing wrong with it.... It has since done around 10 test flights in stronger winds, all as normal totally fine...

 

[phil99]

All the way through 17.35 I am giving a right rudder input and the nose is still turning to the left..... its only at 17.36 the nose starts turning to the right...

 

[phil99]

Also it is not then in free fall but dropping at 10m/s is going down super fast, way above its allowed max descent speed?

 

[Meta4]

Also it is not then in free fall but dropping at 10m/s is going down super fast, way above its allowed max descent speed?

It didn't free fall because three motors were working hard to try to compensate.
That accounts for the extra current drain too.
But three motors don't have enough thrust to prevent the drone from losing height and they can't work against the unbalanced torque of two motors rotating one way and one, the opposite.
And that accounts for the induced spin, which probably caused the false indication of compass errors.

The cause of the speed errors is unclear.
Usually they indicate a GPS issue, with speed calculated from GPS data differing from speed dsta from the gyro sensors.

The wind had nothing to do with the incident and there's no indication of a collision.
It's unlikely that a collision would have caused propulsion fault that disappears after landing/crashing.

And the amazing recovery of the drone is a big mystery.
Whatever caused the problem was only temporary.
I don't know what it was or if it will recur.
Sorry .. but I haven't run into something like this before.

 

[WV. Rootman]

Could a bird have caught the drone and flew it to the tree?

 

[Andy9]

Yes. The bird would be the most logical cause. Did you see the drone at the time of that happening?
With one leg it could grab the gimbal and with the other leg or wing or whatever it partially blocked one of the motors.
This is only my imagination but if you didn't see it at that moment, this can be true as well.

 

[GFields]

Glad the story ended on a good note. Thanks for sharing.

 

[phil99]

On a serious note.... if a bird had stopped one of the propellors the drone would give a motor obstructed warning message? However yes the theory that 3 of the motors were working overtime does seem to tie in with the current draw and descent rate. If I had not found the drone it would look like it had lost a propellor mid flight?

The fact that the drone works now, makes me think it was an outside force..... a bird would have also let go of it before it hit the tree and the battery came out?

I know it seems ridiculous but maybe like a mini whirlwind... or a bird are the most likely?

 

[Meta4]

I know it seems ridiculous but maybe like a mini whirlwind... or a bird are the most likely?

Having looked closely at the actual data, I would rate both of those suggestions as extremely unlikely.
A hardware fault is a lot more believable.

 

[phil99]

Having looked closely at the actual data, I would rate both of those suggestions as extremely unlikely.
A hardware fault is a lot more believable.

We can rule out the battery and connection as the voltages and current all all reading and working? It would be a fault that has not shown itself since.... so maybe a connection to one of the motors or escs? Dry joint or a bad connection? Any problems with these things that often happen? Connections between boards?

 

[Andy9]

The drone works correctly after the incident as you say, so I think you'll never find out what was really the reason.
I hope it won't happen again.

 

[phil99]

I am now thinking of opening up the drone and checking all the connections between the boards are sound... the locking ribbon cables still have their retaining screws etc...