I posted in another thread that I was having problems keeping altitude with my P3A, and was trying to determine if it was unique to my unit, or just part of the situation of flying these thing.  I've been doint all the things I've read to do (calibrate contorler, upgrade firmware, calibrate IMU, etc...) and running test flights on my patio to see if it resulted in a change.  I would take off and try to hover at about 6 feet, and watch as within a minute or two it would be about a foot off the ground.  I did a handful of these testing all my callibrations, upgrades, etc.

Anyway, I was reading that temperature could have a big impact on the baramoter, so on this flight, I sat outside for a bit with the Phantom to get it close to ambient temp.  I then powered on the motors, looked things over for a bit (25 seconds according to the logs) to check GPS, homepoint, etc,  and left-sticked up to take-off.  The bird immediately took off about a foot or two, went hard left, and crashed a second or two later about 15 feet away, landing upside down (and then obviously then losing sats).  I did a CSC as soon as I could once it was on the ground upside down, and got it shut down quickly.

Since I was within my Amazon 30 day return window, I was considering exchanging it anyway due to the altitude issues, bit with this latest event, I didn't even investigate the latest status of the unit -- I processed the exchange.  New P3A arriving today, thanks to Amazon's good policies and prime shipping, and this one will go back in the box to them tomorrow.

While I'm not keeping it, I thought it would be good to try to figure out what happened to avoid it in the future if possible.  For reference, on this flight, Controler and Phantom were on the latest firmware (where I had done one test flight and one "real" flight before, both without event except the losing altutude issue), IMU and Controler had been calibrated 2 flights before.  The compass was not calibrated immediately before this flight, although I did not get any compass errors.  GPS was showing 10 sats at takeoff, and then homepoint had been set.  Battery was listed as 98% charged on takeoff.

The healthdrone log is at: P3A Crash on takeoff log
and the Unit's .dat file is at  P3A Crash on Takeoff .DAT

Anyone have any ideas?  

Looks like you may have been trying to fly it in a restricted area... I took a look at the first link you provided with a log and there are 3 warnings - check those out. Looks like it switched to ATTI after takeoff - in which case is not as easily controlled as when it's in GPS. The restricted flying zone thing could have had some effect on the flight mode - I'm really not sure. Are you near an airport?

B        00m 00s        0.0 ft        0 ft        Warning        The aircraft is in a Warning Zone (Unpaved Airports). Fly with caution here.

I'm sorry you had an accident.
What is your email and I will look at the flight to see if i can see what happened.
Sync your flight records and I'll take a look at the last flight.It does look like it took off in ATTI, but DJI does not use any 3rd party analysis sources.
If I don't see anything from the flight, I have have the data analysis team analyze the DAT file.

Are you sure you didn't inadvertently change the switch on your RC to A ? The log says that the flight mode was changed to ATTI at 26 seconds.
Also you said that you took off with the left stick whle the logs show that 'Assisted takeoff' was used.
Furthermore how do you explain the negative heigths after takeoff ? Is your patio higher than the surroundings ?
Finally the AC will fall upside down only when the motors are stopped. Are you sure you didn't perform a CSC in the stressy moment when you saw the AC fly away because it was in ATTI and not stabilised in place anymore ?

Adam Flurk Posted at 2016-8-4 04:30
Looks like you may have been trying to fly it in a restricted area... I took a look at the first lin ...

If you zoom out on the Healthydrones map there's a BIG airport nearby

... could be part of the problem

When the new drone comes drive a long way from home and try again in an open space

Thanks all for the quick replies.  I don't see a multi-quote function, so I'll answer what I can:

1)Yes - I live near a major airport.  I am outside of the restricted space, but according to airmap (which uses the center of the airport) I'm just barely inside the 5 mile zone.  When I fly for real, I do go outside the 5 miles, or call operations and the tower before hand  if I'm using one of the open spaces still within the 5 mile zone  (although I admittedly do not call them if I'm staying below my own roof line like I did to test the altitude hold here). The new firmware pops up a couple of check boxes that I'm in the warning (not no fly) zone, and I have to check those to continue.  Annoying, but it works.  Odd thought that the warning is "unpaved" airport, as that is a major international airport.  I'm actually just outside of the 5 mile buffer from a local airport, so there is only a small strip I can fly without notifying the airports, but I have that routine down and they are good with it.  Since I'm well outside the true restricted space, this shouldn't cause a problem, and the airport has always been ok with me flying from a nearby open location.

2 - I saw the Atti switch in the logs and went back to the controller to see what position it was in, and it was still in P position.  I don't know if someone with more experience than I can look at the logs and know for sure, but based upon what I heard and saw I believe that it went into Atti mode once it was upside down and lost the sats.  The flight was really only about a second (below 25 seconds was me idling on the ground.0

3- @PhanFran, great questions.  I think it turned upside down because it hit the ground at an angle at a decent rate of speed.  The motors were still spinning when it hit the ground  (hence the obstruction warning), so no, I do not believe that I did a CSC in mid flight.  I have no explanation for the negative altitude.  My patio is probably about 6 inches above the rest of the ground, and the entire "flight" was over the patio.  As noted, I did this test flight because of some altitude problems I was having (posted in another thread).  Maybe that is the entire cause of the whole event.

4- @DJI-KEN.  Thanks for offering to have someone look at it.  I did sync my logs.  I sent you a friend request so I can PM my email address.  I can post it here if need be, but have tried to avoid it showing up on searches.  Let me know if I need to post it here instead.


Thanks again for all the quick replies.  Hopefully this will be a good lesson for me if it was pilot error, or some data if it was not.