Feeling lost after loss of drone

@Casper2015: That's precisely the reason why I didn't calibrate before this flight. I've tried to keep it to the minimum as my flights happen in a small area.

@TAZ: You mean the whole .txt log file from that flight? Unedited, no conversion?

Right now studying the data in Google Earth Pro.
 
Google Earth Pro shows the same pattern as Google Maps inside the Pilot app.

Not however too convinced about the accuracy of this location data, because it puts the drone too close to the tower roof. The shots were certainly much further away taken. But if the last known position is true, then there's now another possible location for the wreck to be... On the roof of the tower. :S The other one to be investigated on Wednesday is the roof at the base of the tower. Will get a call if anything is found.

Not sure how accurate this 3D rendering of the path is, but the last position is at the edge of the roof in this renderation. However I never myself flew it that low. It would anyway be stupid to fly close to a building that high up with unexpected gales. And if RTH really shouldn't decrease the altitude, then something has definetly gone haywire inside the drone.
 
The elevation curve is misleading. The pictures were taken before that wierd zigzag, which must have taken place when I lost the live feed. As the image doesn't show the depth, you don't see how far away the drone was in that direction.
 

Attachments

  • flight_simulation.jpg
    flight_simulation.jpg
    148.4 KB · Views: 312
Google Earth Pro shows the same pattern as Google Maps inside the Pilot app.

Not however too convinced about the accuracy of this location data, because it puts the drone too close to the tower roof. The shots were certainly much further away taken. But if the last known position is true, then there's now another possible location for the wreck to be... On the roof of the tower. :S The other one to be investigated on Wednesday is the roof at the base of the tower. Will get a call if anything is found.

Not sure how accurate this 3D rendering of the path is, but the last position is at the edge of the roof in this renderation. However I never myself flew it that low. It would anyway be stupid to fly close to a building that high up with unexpected gales. And if RTH really shouldn't decrease the altitude, then something has definetly gone haywire inside the drone.
I don't know if you are going to get another phantom soon.
but if you are, and you can buy it now, if I were you I would buy it right now, and use it to search for the last drone.
 
If it's on the roof, then it might be the only viable option. Not exactly daily stuff for a handyman to walk around there and stumble upon it. :D
 
There is video somewhere on youtube of a guy rescuing his friends drone from a high building, with his drone. I'm not saying you can fish it out with another drone..maybe you can, but if you can have visual confirmation of its location, I bet it would be a big step forward :)
the guy on the video just managed to drag the other drone to the edge, and a couple of friends catch it with a blanket.
 
Okay dokey, here we go: DJIFlightRecord.txt.

Now it's time to sleep as a new day dawns. :) Hopefully going to get some answers and maybe a better end to this.

This .txt file seems to be corrupt somehow. Doesn't upload to djilogs.com. Gets "unknown file type". Perhaps the web server you put it on is attaching header/footer information...or something happened to the file when you renamed/saved it. I can tell it's the correct file, as I can see reference to Tampere...which is where the tower is located...but something wrong with the upload. Can you test it out and maybe provide another upload?
 
To those who are saying they calibrate the compass every time, please read this:

http://www.phantompilots.com/threads/compass-calibration-a-complete-primer.32829/

A faulty calibration can be a source of a flyaway or unexplained behaviour, and calibration is not necessary unless you have travelled at least a hundred miles from the site of your last calibration.

Interesting and very important topic.
I read that thread and it seems to make sense but a few things leave me confused.

The DJI manual for MY Phantom says this....

7. GPS & Compass Calibration
If the Compass Module is not used, you can skip this step.
The GPS module has a built-in magnetic field sensor for measuring the geomagnetic field, which is not the same indifferent areas. The GPS module will not work unless the Compass Module has been connected. Make sure theCompass Module connections are correct.
Please always keep the compass module away from the magnet. If this situation occurs please change the compass module before flying. Otherwise it may damage the compass module and lead the aircraft to work abnormally or evenbe out of control.
Calibrate the compass before the first flight or when flying in a different area.

Maybe it's in the Chenglish translation?

I took that to mean "First Flight of the Day"....because surely they don't mean ONLY before the first flight ever, then no more except if you move 100 miles or more away? Or do they?
Maybe I interpreted it wrong.

In the Phantom 2 manual it says....
Please carry out the calibrating procedures in the flight field before flight.


Again, they don't specifically say if they mean only the first flight after purchase or the first flight of the day at the field.

I really can't see how the 100 mile thing could be accurate given the DJI manual says the calibration can require changing due to things as simple as metal in the ground.
I am going to contact a well known compass manufacturer and pose some of these questions.

On the other hand, I can see how you could potentially botch a compass calibration in the field and without connecting to the assistant and checking the mod values, you won't know if it's within specs.

I guess if you were to go through enough threads where flyaways occurred and determine how many occurred without calibration and how many after calibration you might get a more accurate picture?

Let's note that in this thread....
The OP of THIS thread DID NOT do a calibration before the flight in question...and had a flyaway. Perhaps it would not have if he had done one? Hard to say.
 
If atti was activate during this event, wouldnt it saved it. Happened twice with me with my inspire and both times I brought it back. Maybe p3ps are different and switching to atti doesnt work. Also adding the hdmi module so you can run both tablet and monitor may help when you loose video signal to your tablet you would have your hdmi monitor to view with the osd home point marker to follow home in atti mode.
 
The OP of THIS thread DID NOT do a calibration before the flight in question...and had a flyaway. Perhaps it would not have if he had done one?

You omit then the fact the drone accelerated to full speed when things went awry, which is something I did not give a command to do. Also RTH is done in a very controlled manner and it never speeds up. IMO the events don't point to a compass error as the drone should be able to handle it with GPS and GLONASS. Sure, it could be a factor in this ordeal, but not the only issue.

And again: I doubt it's useful to calibrate every time. It would however be useful to have a proper pre-flight routine for all pilots, that would be based on sound facts. Right now there seems to be two camps regarding the compass issue.
 
  • Like
Reactions: JKDSensei
So you did not do a compass recalibration at the start of your flying session?

There's no reason to do so. Every time you calibrate, you introduce a significant possibility of error. Unless you changed equipment, or traveled a significant distance, you should NOT calibrate. Get one good calibration and leave it alone.
 
The service crew will have a look on the roofs, but no guarantee for any success. Other than that, I've pretty much given up now. I do have a great deal of anticipation on the other hand still, but finding something that can still fly is pretty unlikely. Not sure yet if I will get a replacement drone soon or just wait until next spring and see how the stuff develops.
 
The service crew will have a look on the roofs, but no guarantee for any success. Other than that, I've pretty much given up now. I do have a great deal of anticipation on the other hand still, but finding something that can still fly is pretty unlikely. Not sure yet if I will get a replacement drone soon or just wait until next spring and see how the stuff develops.

If you had an inexpensive drone with a camera like a Syma X5C-1 you could at least take a look on the roof tops (depending on how close you can get). Might want to get one for recon next time.
I've seen the tops of large shopping malls with one and would be able to see a drone on top for a good distance. In this image I'm about 500ft from the structure.

rooftop_zpsr1iyl6fu.jpg
 
Last edited:
Found! However the first information I received is that something is broken, just as expected. The drone CSI will be onsite in a few hours to recover the victim. More breaking news as it emerges.
 
  • Like
Reactions: Freebird1
Good news! Was going to mention that if you were considering a move to a Phantom 2 with your GoPro, you could sell your P3 controller and spare battery here to make up at least a significant chunk of, if not all of the cost of a P2. The spare P3 battery will NOT work with a P2 anyway.

Hopefully your P3 is salvageable and it's a moot point. Good luck.
 

Recent Posts

Members online

No members online now.

Forum statistics

Threads
143,086
Messages
1,467,528
Members
104,965
Latest member
Fimaj