Here's my flightrecord and where it flew. (Airdata) https://app.airdata.com/flight/fb9ebcede334dd60ef8e96a2d8c21ab3
cant find where i suppose to report this but if you have done it can you comment below how to go on taking contact with dji?

Normally AIrdata isn't much use for investigating incidents but in this case it looks pretty clear.
At 4 and 5 seconds the app was screaming at you about a compass error and at 6 seconds it is showing a Yaw Error.
The compass error/yaw error combination is usually due to launching from a magnetically dirty site.
The most common cause is launching from a steel or reinforced concrete surface.

Another one for my list...

My understanding is your warranty is no good without the recovered aircraft.

Sorry for your loss.

Labroides Posted at 2018-8-22 02:22
Normally AIrdata isn't much use for investigating incidents but in this case it looks pretty clear.
At 4 and 5 seconds the app was screaming at you about a compass error and at 6 seconds it is showing a Yaw Error.
The compass error/yaw error combination is usually due to launching from a magnetically dirty site.

Okay thanks! should I just email them and see what they can do about it?

fans36c1a17e Posted at 2018-8-22 02:32
Okay thanks! should I just email them and see what they can do about it?

Can you comment on where you launched from?

DJI are only going to be interested if there's evidence to show it was a defective Phantom and if it's less than one year old.

Labroides Posted at 2018-8-22 02:49
Can you comment on where you launched from?

DJI are only going to be interested if there's evidence to show it was a defective Phantom and if it's less than one year old.

It was Cobblestone, since i was on a tourist Island (SveaBorg) in Finland, and i launched the phantom from a more ''unpopular'' place. My drone would have been exactly 1 year today.

Sorry to hear the lost of your drone. I would recommend you to contact our support team to start up a case: (https://www.dji.com/support?site=brandsite&from=nav). We have the professional data analysis team who would do their best to find out the reason of the flyaway incident and then the corresponding resolution would be provided. Thanks for your support.

Mark The Droner Posted at 2018-8-22 02:29
Another one for my list...

My understanding is your warranty is no good without the recovered aircraft.

Would this be the “don’t take off with the battery  at 66%” list ? It’s getting longer sadly.

tokenbrit Posted at 2018-8-22 03:58
Would this be the “don’t take off with the battery  at 66%” list ? It’s getting longer sadly.

It's not clear if the battery was at 66% because of a short flight immediately before or if it had been sitting around for days.

It's never a good idea to launch with a battery that's been sitting and discharging battery but the battery had nothing to do with this incident.

Save your time and effort about trying to make a case with DJI.  Clearly pilot error.  Best to just invest in another and chalk this up to experience.

It looks to me like the drone should never have left the ground.  It also looks like the wind got it at altitude.  What actions did you take while this was happening?  Did you try to lower altitude and bring it back?  I'm not trying to belittle you.  I just would like to know what action people take when things like this happen.

djordan2 Posted at 2018-8-22 07:08
It looks to me like the drone should never have left the ground.  It also looks like the wind got it at altitude.  What actions did you take while this was happening?  Did you try to lower altitude and bring it back?  I'm not trying to belittle you.  I just would like to know what action people take when things like this happen.

What I do is check a wind app before I fly and then plan the flight accordingly and carefully.  Experienced pilots know to fly away high with a tail wind, turn around at an appropriate point, and fly back home low into the head wind.  Otherwise fly away low into a head wind and fly back home high with the tail wind.  I usually don't fly when there's a strong cross wind.

I don't believe in "Wind Apps".  I rely on real-time wind testing at altitude on location.  I think that is much more accurate than a wind app.  I also believe is watching those numbers on the screen to get an accurate idea of what is going on up there.

djordan2 Posted at 2018-8-22 07:08
It looks to me like the drone should never have left the ground.  It also looks like the wind got it at altitude.  What actions did you take while this was happening?  Did you try to lower altitude and bring it back?  I'm not trying to belittle you.  I just would like to know what action people take when things like this happen.

It also looks like the wind got it at altitude.
When a Phantom is launched from close to steel and gets the compass error/yaw error combination, it is common for it to be uncontrollable.
This was not a case of wind blowing the Phantom away.
It was uncontrollable just seconds after launching.

Labroides Posted at 2018-8-22 13:43
It also looks like the wind got it at altitude.
When a Phantom is launched from close to steel and gets the compass error/yaw error combination, it is common for it to be uncontrollable.
This was not a case of wind blowing the Phantom away.

Where does it say that he launched from an area with a lot of metal?  He said it was Cobblestone.

Labroides Posted at 2018-8-22 02:22
Normally AIrdata isn't much use for investigating incidents but in this case it looks pretty clear.
At 4 and 5 seconds the app was screaming at you about a compass error and at 6 seconds it is showing a Yaw Error.
The compass error/yaw error combination is usually due to launching from a magnetically dirty site.

The HD playback is interesting.  It looks like the ac was in atti mode drifting rapidly downwind straight after takeoff. It was also flying backwards, so when the operator pulled back it flew away even faster. The stick movements seem to be consistent with the track, so he had controls until presumably he lost signal.  It switched to WiFi for some reason.

Amongst all of the warnings, low battery finally came up with RTH - my guess is it couldn’t get back upwind.

djordan2 Posted at 2018-8-22 15:03
Where does it say that he launched from an area with a lot of metal?  He said it was Cobblestone.

Read post #2.
He may have taken off from cobblestone but getting compass error/yaw error combination like that indicates the compass was seriously messed up.
You don't get that without a strong magnetic influence at the launch point.

Perhaps there was a steel pipe or similar just under the cobblestones?

tokenbrit Posted at 2018-8-22 15:51
The HD playback is interesting.  It looks like the ac was in atti mode drifting rapidly downwind straight after takeoff. It was also flying backwards, so when the operator pulled back it flew away even faster. The stick movements seem to be consistent with the track, so he had controls until presumably he lost signal.  It switched to WiFi for some reason.

Amongst all of the warnings, low battery finally came up with RTH - my guess is it couldn’t get back upwind.

It looks like the ac was in atti mode drifting rapidly downwind straight after takeoff. It was also flying backwards, so when the operator pulled back it flew away even faster. The stick movements seem to be consistent with the track, so he had controls until presumably he lost signal.  It switched to WiFi for some reason.
You aren't reading the data correctly.
There's no hint that wind had any effect but there's plenty of evidence to show that the Phantom was flying uncontrollably as is very commonly seen in compass error/yaw error conditions.
If you think the joystick input matches the Phantom's flight, just look at 0:22.5 - 0:41.3 when he had his right stick centred but the Phantom accelerated from 18-59 mph.
Switched to wifi?
The P3 Standard is a wifi drone.  It was never using anything but wifi.


Amongst all of the warnings, low battery finally came up with RTH - my guess is it couldn’t get back upwind.
Low battery??
There's no hint of any low battery warning and the battery is still healthy when the data stops.
Your guess about wind is incorrect - his Phantom was uncontrollable.

use the find my drone feature to  find it

BeastMaster_101 Posted at 2018-8-22 16:59
use the find my drone feature to  find it

That won't be any help.
Find my drone can only tell you where the drone was last recorded - not necessarily where it is now.
And in this case the drone was last recorded out at sea and headed further away.

Anyone else want to offer some half-baked "advice" that ignores what we can clearly see in the data?

Labroides Posted at 2018-8-22 16:54
It looks like the ac was in atti mode drifting rapidly downwind straight after takeoff. It was also flying backwards, so when the operator pulled back it flew away even faster. The stick movements seem to be consistent with the track, so he had controls until presumably he lost signal.  It switched to WiFi for some reason.
You aren't reading the data correctly.
There's no hint that wind had any effect but there's plenty of evidence to show that the Phantom was flying uncontrollably as is very commonly seen in compass error/yaw error conditions.

Wind speed close to 20mph and the last warning “Battery at maximum distance:”. All the stick movements were taking the drone higher and further downwind.  

tokenbrit Posted at 2018-8-22 19:19
Wind speed close to 20mph and the last warning “Battery at maximum distance:”. All the stick movements were taking the drone higher and further downwind.

You'd be better sticking to things you understand.
A look at the battery message you think you see is simply stating the battery level at the max distance in the data.
It's not saying anything about the battery being low.
In fact the battery level is 58% and looking closer it's showing 3.6 volts per cell.
That's still quite healthy and not a warning of low battery level.

As for the wind, it did not help but it was not the cause of the issues observed.  

You obviously haven't ever looked at a runaway yaw error situation have you.
Look at the error messages, it's non-stop yaw error + speed error with the Phantom getting to 59 mph !!

The Phantom is actively running away - not being blown away.
Have a look at the time from 0:22.5 till 0:39.1.
With his right stick centred the Phantom starts at 18 mph and accelerates to 59 mph !!
No-one would go out to fly in a wind that could do that.
It would be much more than 20mph.

Looks like it took off from a quay, looking at the size of those things that ya tie a large boat to i bet its all seriously reinforced under those flag and cobblestones

wow, this is new
https://app.airdata.com/flight/f ... ations&val=detailed

Ive not noticed it before, so new to me :p

Labroides Posted at 2018-8-22 02:22
Normally AIrdata isn't much use for investigating incidents but in this case it looks pretty clear.
At 4 and 5 seconds the app was screaming at you about a compass error and at 6 seconds it is showing a Yaw Error.
The compass error/yaw error combination is usually due to launching from a magnetically dirty site.

Curious. Wouldn't the magnetic interference resolve after take off and distancing from the launch site and site of assumed Magnetic interference?

stuka75 Posted at 2018-8-22 21:45
Curious. Wouldn't the magnetic interference resolve after take off and distancing from the launch site and site of assumed Magnetic interference?

In dozens of similar case it doesn't.
When you power up the Phantom it determines where north is from the earth's normal magnetic field.
But when powering up where the distorted magnetic field associated with a significant amount of steel close to the compass, the Phantom gets a false north.
After launching and leaving the Magnetic distortion it is comparing the movement recorded by sensors against the false north and runs into problems.
It tries to correct by moving in the direction it calculates is necessary and runs into more problems.
This continues and the result is the Phantom runs away, often at considerable speed.
The effect is variable depending on how big the difference between real north and false north is.


Never launch from steel or reinforced concrete surfaces or close to significant steel objects.

Labroides Posted at 2018-8-22 22:10
In dozens of similar case it doesn't.
When you power up the Phantom it determines where north is from the earth's normal magnetic field.
But when powering up where the distorted magnetic field associated with a significant amount of steel close to the compass, the Phantom gets a false north.

Great explanation. Thanks

Labroides Posted at 2018-8-22 19:47
You'd be better sticking to things you understand.
A look at the battery message you think you see is simply stating the battery level at the max distance in the data.
It's not saying anything about the battery being low.

“When it comes to atti, as well as wind strength and the factors mentioned above, that is up to the ability of the pilot. Some flyers could come unstuck in a gentle breeze while another with experience may deal with 20+ knots with ease.”

tokenbrit Posted at 2018-8-23 03:39
“When it comes to atti, as well as wind strength and the factors mentioned above, that is up to the ability of the pilot. Some flyers could come unstuck in a gentle breeze while another with experience may deal with 20+ knots with ease.”

When your Phantom runs away with constant yaw error, it's not just a Phantom in atti mode.
It's a Phantom running away and you CAN'T control it.
How hard is that to understand?
The thing was racing at 59 mph with the right stick centred!!
The wind wasn't a factor.  Atti wasn't a factor.
You wouldn't have been able to control it, I wouldn't have been able to control it.


Read my posts again (particularly #26) because I'm sick of explaining the same thing over and over.

Labroides Posted at 2018-8-23 04:28
When your Phantom runs away with constant yaw error, it's not just a Phantom in atti mode.
It's a Phantom running away and you CAN'T control it.
How hard is that to understand?

Calm down Labroides - I'm not disputing any of your points - there were clearly many obvious technical issues going on that should have been cause to end the flight before it started.  But rather than focus on the obvious, I'm pointing out that the aircraft was responding to user input and these inputs also contributed to the demise of the flight.

He was flying backwards, in ATTI, downwind in full reverse. How hard is that to understand?

tokenbrit Posted at 2018-8-23 05:57
Calm down Labroides - I'm not disputing any of your points - there were clearly many obvious technical issues going on that should have been cause to end the flight before it started.  But rather than focus on the obvious, I'm pointing out that the aircraft was responding to user input and these inputs also contributed to the demise of the flight.

He was flying backwards, in ATTI, downwind in full reverse. How hard is that to understand?

It was flying itself.
It didn't matter much what he did with the controls.
It was in atti because of the compass error but the yaw error made it uncontrollable.
Any additional factors were trivial in effect.

Labroides Posted at 2018-8-23 07:07
It was flying itself.
It didn't matter much what he did with the controls.
It was in atti because of the compass error but the yaw error made it uncontrollable.

Thanks for seeing the error here, looks like I'm getting a replacement after all. For those who didn't believe. Dear Customer,


Thanks for your patience.

Got an email that said this from dji ''The unfortunate incident that occurred to your aircraft has been confirmed as a warranty case according to our data analysis.

If the aircraft couldn’t be retrieved, we would like to offer you a replacement, Phantom 3 Standard without charger and remote. Please kindly inform us below''

Looks like I will be getting a replacement guys

Labroides Posted at 2018-8-22 02:22
Normally AIrdata isn't much use for investigating incidents but in this case it looks pretty clear.
At 4 and 5 seconds the app was screaming at you about a compass error and at 6 seconds it is showing a Yaw Error.
The compass error/yaw error combination is usually due to launching from a magnetically dirty site.

I can not control the camera on the drone. I have a picture but that’s all

He is 100 percent correct the compass/yawerror is from launching from a magnetic surface or area ! You will loose your drone if you do this

fans36c1a17e Posted at 8-22 02:54
It was Cobblestone, since i was on a tourist Island (SveaBorg) in Finland, and i launched the phantom from a more ''unpopular'' place. My drone would have been exactly 1 year today.

If you are EU resident ... bought in EU ... then Warranty is greater than 1 year ... as even DJI has to abide by EU rules.

Good to see DJI coughing up the replacement

Nigel