New P4 Pro. After a number of flights. 1200ft away drone just drops out the sky. It was within los and no obstacles. I just saw it drop with a aircraft disconnected message on the DJI App.

Flight logs attached DJIFlightRecord_2018-05-15_[13-19-34]csv.zip (27.02 KB, Down times: 7)

2018-5-15 Upload

Click on the file name to download attach

Any insight would be appreciated

Are you sure the battery was fully inserted?

Grumpy D Posted at 2018-5-15 05:16
Are you sure the battery was fully inserted?

That I always make sure of. It was.

http://www.phantomhelp.com/logviewer/upload/        Go here, follow the instructions, come back and post the link to the flight log.  Maybe it will provide some insight to what happened.
   

The recorded data is no help as it just stops while the drone was in stable flight with a strong battery.
Whatever the cause was, it's not in the data.
It looks like a case for DJI to investigate.

They may be able to tell more from the internal flight data recorder.

Does the battery still work, or might it have been a faulty battery?

Contact DJI Support, they will look at the logs and since it sounds like a faulty aircraft or battery and it was new are quite likely to replace it, depending on what they find.

battery still works fine and was undamaged. unlike the rest of the drone. gimbal and camera a write off. legs broken. motors still spin fine

flight log from AC https://drive.google.com/open?id ... -_rciVv5YLsH8tEmL7a

Same happened to my P4P last Friday. Still good battery, no issues, new machine on 5th flight, just disconnected and lost forever

Keen to hear your outcome.

Odysseus Posted at 2018-5-15 06:40
flight log from AC https://drive.google.com/open?id=1LrWVGAZSGZXUQ-_rciVv5YLsH8tEmL7a

Well for what it's worth I had a look...

Your RC log tells me nothing. As previously stated it looks like an uneventful flight until a sudden and abrupt end in the recording:

http://www.phantomhelp.com/logviewer/VYEJSRM4R988RH7VC6VN/

The FC log is almost as unevetful except for the very last few data frames (~100ms) before it too ends abruptly. Right before the recording stops:

VoltageVsSpeed

Voltage on all 4 motors suddenly drops (~16V to ~6V) confirmed by the decrease in motor speed before recording ends.

You were flying steady in a relatively straight line at ~20mph, ~120ft above a road which looks (?) clear from abstructions * based on sat photos.

It's probably safe to assume the recording stopped because the Flight Controller lost power.

I wonder if it tried and/or had enough time to reset before it hit the pavement, do you have the next FC log after this? Want to share it?

Really sorry about your loss, better open up a ticket with DJI support and send it in for an assessment. How old is the Phantom?

PS I checked all the other obvious telemetry data like battery / ESC stats, etc... I couldn't find anything abnormal in my experience. Maybe some of the other more experienced members can have a look and find something I missed, also post your logs on phantompilots.com / rcgroups.

The software used to decipher the FC logs can be found here:

https://datfile.net/

The RC log parser previously linked can be found here:

http://www.phantomhelp.com/logviewer/Upload/

We're so sorry to hear the accident. It is recommended to conduct data analysis to check the exact status if the unit is still under warranty. You can contact support@dji.com to start a case.

That sucks.... Hope you get it figured out.

I could be way out of the ball park here but i think the battery fell out, the accelerometer shows hard braking (unless im reading it the wrong way round) prior to the moment the logs stop, this means the vehicle was tipped on its ass end thus the battery falling out or disconnecting with the vehicle, as there was no more in the logs then im inclined to go with the battery falling out or catastrophic failure mid flight

Bashy Posted at 2018-5-15 21:20
I could be way out of the ball park here but i think the battery fell out, the accelerometer shows hard braking (unless im reading it the wrong way round) prior to the moment the logs stop, this means the vehicle was tipped on its ass end thus the battery falling out or disconnecting with the vehicle, as there was no more in the logs then im inclined to go with the battery falling out or catastrophic failure mid flight

"the vehicle was tipped on its ass end thus the battery falling out "
If you are in an airliner and it makes a sharp turn, do you fall to the side?   No, the only thing you feel is an increase in gravity.

More likely that the extra power required by the motors burned out a poor connection or poor component.

Looks like it will be replaced under warrantee.

Bashy Posted at 2018-5-15 21:20
I could be way out of the ball park here but i think the battery fell out, the accelerometer shows hard braking (unless im reading it the wrong way round) prior to the moment the logs stop, this means the vehicle was tipped on its ass end thus the battery falling out or disconnecting with the vehicle, as there was no more in the logs then im inclined to go with the battery falling out or catastrophic failure mid flight

I believe negative acceleration in the Z plane means it began accelerating towards the ground (I.E. fall).

When the phantom (quad-copter) travels forward; aft motors apply more power than the forward motors -- this translates into a pitch forward configuration which translates into forward travel. You can see this in the data. Breaking does the opposite; increased power on the forward motors, decreased power on the aft motors -- I don't see this in his data.

I don't have the graph on me right now but I remember seeing the IMU sending an increase power command to all ESCs/motors. Seeing your finding this now makes sense; It detected negative acceleration in the z plane so it attempted to compensate by increasing power in the z plane. Unfortunately the motors didn't respond due to the evident power starvation.

Nigel_ Posted at 2018-5-15 23:43
"the vehicle was tipped on its ass end thus the battery falling out "
If you are in an airliner and it makes a sharp turn, do you fall to the side?   No, the only thing you feel is an increase in gravity.

There is a parallel conversation taking place about this incident here:

https://phantompilots.com/thread ... ck-midflight.134756

Spirytus Posted at 2018-5-16 04:00
There is a parallel conversation taking place about this incident here:

https://phantompilots.com/threads/p4-dropped-like-a-brick-midflight.134756

• There appears to have been some sort of power failure.
• You say the battery is still working OK.
• You also say you inserted it correctly.
• You also say that the motors still spin up.
  

So it seems there was some sort of bad connection, but there is no way to find what without disassembling aircraft and battery to investigate.  Possibly the battery did come loose, but that should not be possible, so if it did then it is due to a manufacturing defect in the connector or latch and will be covered by warrantee.

You need to send it in to DJI and see what they say, probably they will replace it without telling us what happened.

It appears that they do have a few quality issues with the battery connection, resulting in the V2 having a sensor to detect battery latching problems, not a common problem and I get the impression that it is almost always with new aircraft which does suggest a manufacturing defect in some aircraft or batteries.

Wait and see what they say... and then let us all know...

Bashy Posted at 2018-5-15 21:20
I could be way out of the ball park here but i think the battery fell out, the accelerometer shows hard braking (unless im reading it the wrong way round) prior to the moment the logs stop, this means the vehicle was tipped on its ass end thus the battery falling out or disconnecting with the vehicle, as there was no more in the logs then im inclined to go with the battery falling out or catastrophic failure mid flight

At rest the accelZ will have a value of -1.0. I.e. sitting on the ground or hovering the accelZ will be -1.0. In free fall accleZ will be 0.0. In the plot you presented the accelZ shows the halfway point in the transition from normal flight to free fall.

The accel:Composite value is computed from sqrt (accelX^2 + accelY^2 + accelZ^2). So it's value will always be greater than or equal to 0.0.

BudWalker Posted at 2018-5-16 05:38
At rest the accelZ will have a value of -1.0. I.e. sitting on the ground or hovering the accelZ will be -1.0. In free fall accleZ will be 0.0. In the plot you presented the accelZ shows the halfway point in the transition from normal flight to free fall.

The accel:Composite value is computed from sqrt (accelX^2 + accelY^2 + accelZ^2). So it's value will always be greater than or equal to 0.0.

The Legend speaks... Hey BudWalker!

So a change from -1.0 > 0 on accelZ translates to acceleration in the up/down plane? In this case it's clearly going down but in the future how can I tell if it's accelerating up instead?

Spirytus Posted at 2018-5-16 06:37
The Legend speaks... Hey BudWalker!

So a change from -1.0 > 0 on accelZ translates to acceleration in the up/down plane? In this case it's clearly going down but in the future how can I tell if it's accelerating up instead?

An accelZ value of 0.0 means that the AC is is in free fall with a velocity of 32 ft/sec^2.  It's gaining speed as it falls - every second it falls 32 ft/sec faster than the previous second.  If the accelZ = -1.0 the AC isn't rising or falling. It's -1.0 because of the earth's gravitational field. If accelZ < -1.0 then the AC is rising.

At least that's how I remember it from high school pysics.

And that right there shows that i didnt learn diddly at school   thanks for the wee lesson, just dont quiz me on it tomorrow ;)

Having looked at the "other" thread, it looks like the consensus is a power failure of perhaps the battery coming loose or out.

I have tried with my p4p, i cannot have just one lug clipped in, both do it at the same time, i have tried semi clipped in where they have not latched 100%, there is power, it wouldnt take much for it to work loose and exit the vehicle  though.

Keeping it simple and based on the experts opinion it seems you have as clear a warranty claim as any...I'd open your trouble ticket now and get the process moving for your shipping and replacement.
Good Luck and let us know how it goes; we are always interested in folks experience with DJI in these matters.

Bashy Posted at 2018-5-16 19:55
And that right there shows that i didnt learn diddly at school   thanks for the wee lesson, just dont quiz me on it tomorrow  ;)

Having looked at the "other" thread, it looks like the consensus is a power failure of perhaps the battery coming loose or out.

"I have tried with my p4p, i cannot have just one lug clipped in, both do it at the same time, i have tried semi clipped in where they have not latched 100%, there is power, it wouldnt take much for it to work loose and exit the vehicle  though."

Did it let you start the motors / take off like that?

I would have expected that even though the power pins on the connector make contact before the battery is fully inserted,  there will be one of the contacts that doesn't connect to let it know that the battery is not fully in yet.

Nigel_ Posted at 2018-5-17 05:02
"I have tried with my p4p, i cannot have just one lug clipped in, both do it at the same time, i have tried semi clipped in where they have not latched 100%, there is power, it wouldnt take much for it to work loose and exit the vehicle  though."

Did it let you start the motors / take off like that?

I started the motors, no chance would i risk taking off, here, have a butchers....
excuse the video, itshard doing the rc and the note 3 lol

Bashy Posted at 2018-5-19 04:43
I started the motors, no chance would i risk taking off, here, have a butchers....
excuse the video, itshard doing the rc and the note 3 lol

That's not good, but it is good to hear that the V2 has a sensor to detect when that situation occurs.   

I'm surprised that is possible given the obvious safety issue.  Should have a latch design more like an SD memory card - you can't leave them partially in and they have a short contact to detect fully in, yet it is not a complicated or heavy system.

I had this very thing happen to my P4P last year. DJI did a warranty replacement but never did tell me what the problem was that caused my drone to loose power mid-flight.

Nigel_ Posted at 2018-5-19 08:25
That's not good, but it is good to hear that the V2 has a sensor to detect when that situation occurs.   

I'm surprised that is possible given the obvious safety issue.  Should have a latch design more like an SD memory card - you can't leave them partially in and they have a short contact to detect fully in, yet it is not a complicated or heavy system.

Be interesting to know how the V2  sensor works......

The same thing happen to me, my drone was about 50 feet away and about 40 feet high and the props just shut off and plunged into the lake in my backyard, GOOD LUCK with DJI you are in for a ride the will just shuffle you around in the call center there customer service SUCKS !!! I have bought every Phantom they have made and also the Mavic and Mavic Pro. I will truly look for alternative drones to buy these people are NOT very helpful. I still can't understand how that drone just shutdown in mid air with 97% battery.  GOOD LUCK !!

I have shown  you how its possible with my video a few replies back, i have shown that the drone can power up the motors with the battery incorrectly inserted, in my video, neither of the catches were latched into place yet it still powered up, in this scenario, it wouldnt take much for the battery to work its way out a little and disconnect the power to the drone resulting in a total power failure and fall from the sky