P3 Pro Crash. Drone falls from Sky no warning.

Joined
Aug 21, 2015
Messages
26
Reaction score
24
Location
Worldwide
Well I really didn't want to have to be posting one of these but wondered if some more knowledgeable folk could shed some more light on the situation.

The 2 flights the night before the incident seemed fine. 2 days earlier I had some compass issues which was a first since ownership (60+ flights). This happened in Northern BC, Canada some 1500km away. The app put the craft in Atti mode so I could fly it home, turn off drone, re-calibrate compass and everything appeared back to normal.

Bring it forward to the 17th and I was out at Kluane Lake here in the Yukon (Canada) and the compass starts acting crazy, the drone is all over the place. I bring it back in Atti mode. Power down/up, re-calibrate compass again, take off GPS-20 SATs, everything seemed fine. Was flying around to test all ok (just out of eye-sight) up and down the lake, then all of a sudden it dies on me, no word of warning. I saw on the screen it just cut out. Also what I find very weird is the video cuts out exactly where it falls.

I located the drone via the Maps on GPS, fortunately not in the lake! Although it make well just have been, seems the motors are clogged with grit same will the gimbal, blades and shell ruined. The P3 was still powered on when I located it.

Attached flight log.
 

Attachments

  • DJIFlightRecord_2015-09-17_[13-58-31].txt
    86 KB · Views: 588
Well I really didn't want to have to be posting one of these but wondered if some more knowledgeable folk could shed some more light on the situation.

The 2 flights the night before the incident seemed fine. 2 days earlier I had some compass issues which was a first since ownership (60+ flights). This happened in Northern BC, Canada some 1500km away. The app put the craft in Atti mode so I could fly it home, turn off drone, re-calibrate compass and everything appeared back to normal.

Bring it forward to the 17th and I was out at Kluane Lake here in the Yukon (Canada) and the compass starts acting crazy, the drone is all over the place. I bring it back in Atti mode. Power down/up, re-calibrate compass again, take off GPS-20 SATs, everything seemed fine. Was flying around to test all ok (just out of eye-sight) up and down the lake, then all of a sudden it dies on me, no word of warning. I saw on the screen it just cut out. Also what I find very weird is the video cuts out exactly where it falls.

I located the drone via the Maps on GPS, fortunately not in the lake! Although it make well just have been, seems the motors are clogged with grit same will the gimbal, blades and shell ruined. The P3 was still powered on when I located it.

Attached flight log.

Could you possibly also pull the internal flight log off the bird itself?

The one accessed via first putting the bird into "enter data flight log" mode or whatever it's called, and then via the bird's USB port.

I'd be interested to see how that compares with you controller side logs.

And sorry for your crash/loss.
 
Running your controller side logs though HealthyDrones...

HealthyDrones.com - Innovative Flight Data Analysis That Matters

Looks like your compass numbers went a bit crazy.

Hard to tell if the numbers are crazy cause the drone was spinning OR the drone was spinning because the numbers were crazy.

Overall, all the data looks okay, and a quick glance through your logs shows no major issues prior to the incident.

Only thing I will say is that you should never take-off with a less-than-full battery. Logs show you started with 68% battery. For some unknown reasons, the P3 doesn't like to start with less than 95-100% of battery.

If the drone still flies, I'd do this:

1) Do a full IMU re-cal. Followed by a Controller cal and a compass cal in a known good place.

2) Do a few flights at 5-10 ft AGL, very close range, and then run those controller-side logs through HealthyDrones. See if you see any more compass abnormalities.

3) If you do see compass issues, either replace yourself, or send to DJI. Might be a hardware failure.

Just my $0.02.
 
Last edited:
Running your controller side logs though HealthyDrones...

HealthyDrones.com - Innovative Flight Data Analysis That Matters

Looks like your compass numbers went a bit crazy.

Hard to tell if the numbers are crazy cause the drone was spinning OR the drone was spinning because the numbers were crazy.

Overall, all the data looks okay, and a quick glance through your logs shows to major issues prior to the incident.

Only thing I will say is that you should never take-off with a less-than-full battery. Logs show you started with 68% battery. For some unknown reasons, the P3 doesn't like to start with less than 95-100% of battery.

If the drone still flies, I'd do this:

1) Do a full IMU re-cal. Followed by a Controller cal and a compass cal in a known good place.

2) Do a few flights at 5-10 ft AGL, very close range, and then run those controller-side logs through HealthyDrones. See if you see any more compass abnormalities.

3) If you do see compass issues, either replace yourself, or send to DJI. Might be a hardware failure.

Just my $0.02.

So if you take off at 100%, land and change filters you should not use that battery until it's fully charged again?
 
i wouldn't think landing and taking off again on the same battery would cause any problems, i do it all the time.
 
i wouldn't think landing and taking off again on the same battery would cause any problems, i do it all the time.

I agree, I don't recall reading that in the manual but I could have missed it.. Although I have found out quickly everyone's situation is different.
 
Could you possibly also pull the internal flight log off the bird itself?

The one accessed via first putting the bird into "enter data flight log" mode or whatever it's called, and then via the bird's USB port.

I'd be interested to see how that compares with you controller side logs.

And sorry for your crash/loss.

Hey thanks. How do you get the flight logs or .DAT files off the Phantom? I tried connecting the bird with cable to my Mac OSX but nothing comes up. Using the DJI Go App. I can see the Blackbox MicroSD Card in the battery compartment but it looks secured.
 
What was the weather conditions???

Cold, but not below 0c, maybe around 5-6c. Slight wind gusts but nothing the Phantom can't handle.

I also thought that it could be a large bird of prey like an eagle or similar. But I didn't see any circling around also doesn't explain the instant loss of video when the drone drops.
 
  • Like
Reactions: chamanea
i wouldn't think landing and taking off again on the same battery would cause any problems, i do it all the time.
It could create a cell imbalance that could result in a faster than normal discharge of the total voltage if one cell discharged faster than the others--

however, the Healthy Drones report should indicate a battery failure if that were the case.
 
Cold, but not below 0c, maybe around 5-6c. Slight wind gusts but nothing the Phantom can't handle.

I also thought that it could be a large bird of prey like an eagle or similar. But I didn't see any circling around also doesn't explain the instant loss of video when the drone drops.
That is a distinct possibility and you wont always see the bird-- and eagle could create instant catastrophic failure and you would never know what hit it.
 
Did you check for solar storms? I seem to recall some last week.

No Solar Storms the last week, but I do wonder if that has something to do with messing with the Compass, there were heavy storms in Alberta a few weeks back but the drone behaved fine.

Looking at Shammyh's link : HealthyDrones The last recorded Altitude of 30ft, 9metres. There are no trees that tall there, so that rules out an air collison.
 
That is a distinct possibility and you wont always see the bird-- and eagle could create instant catastrophic failure and you would never know what hit it.

True, but I'm positive I would've seen it flying around afterwards, the area is way too open to miss something like of bird of Prey.
 
1) So you can definitely fly without fully charged batteries... just in my own personal experience, the Phantom can act a bit odd. Don't know exactly why that is... but the only few times either my P2V+ or P3P have acted up, it was with 60-70% charged batteries at take-off. Completely speculative, so take it however you will.

2) Solar storms have a negligible effect here on Earth. You can thank our atmosphere and molten iron core (and resultant magnetic field) for that. Otherwise, we'd have fried millions of years ago from intense radiation... ;-) Some people do take it into consideration, but personally (I'm not a scientist and amateur astrophysicist at best) I think it's a complete red herring.

3) It's possible there was a mid-air collision. Did you recover video? Or were not recording at the time? (I make a habit of automatically switching back to video recording when not taking photos for any extended period.) Small birds can come and go pretty quick, I wouldn't trust your eyesight/awareness on that unless you literally never took your eyes off the Phantom.

4) It doesn't look like a power system failure, as the bird was still "powered on" on the ground when you recovered it and flight logs show no sign of low voltage prior to uncontrolled descent (aka, crash).

~~~

As to the bird-side logs, follow this procedure: Link

They are recorded independently of the controller-side logs and while they contain less granularity of data, they are "higher resolution", aka, more data points per time, and as far as I know, are the "definitive" record of what the Phantom was doing at the time.
 
1) So you can definitely fly without fully charged batteries... just in my own personal experience, the Phantom can act a bit odd. Don't know exactly why that is... but the only few times either my P2V+ or P3P have acted up, it was with 60-70% charged batteries at take-off. Completely speculative, so take it however you will.

2) Solar storms have a negligible effect here on Earth. You can thank our atmosphere and molten iron core (and resultant magnetic field) for that. Otherwise, we'd have fried millions of years ago from intense radiation... ;-) Some people do take it into consideration, but personally (I'm not a scientist and amateur astrophysicist at best) I think it's a complete red herring.

3) It's possible there was a mid-air collision. Did you recover video? Or were not recording at the time? (I make a habit of automatically switching back to video recording when not taking photos for any extended period.) Small birds can come and go pretty quick, I wouldn't trust your eyesight/awareness on that unless you literally never took your eyes off the Phantom.

4) It doesn't look like a power system failure, as the bird was still "powered on" on the ground when you recovered it and flight logs show no sign of low voltage prior to uncontrolled descent (aka, crash).

~~~

As to the bird-side logs, follow this procedure: Link

They are recorded independently of the controller-side logs and while they contain less granularity of data, they are "higher resolution", aka, more data points per time, and as far as I know, are the "definitive" record of what the Phantom was doing at the time.

Thanks a-lot for the link, I will give it a go! The video stops exactly at 1:18 where it crashes..You don't see it falling and I didn't switch-off recording.
 
Thanks a-lot for the link, I will give it a go! The video stops exactly at 1:18 where it crashes..You don't see it falling and I didn't switch-off recording.

Glad to help.

Would you mind posting the video as well? Dropbox/Google Drive/Youtube/etc to share it.

Since you have a lot of good data here (controller side logs & bird side logs & video), I think it'd be interesting to really get to the bottom of this problem if possible.
 
Were all 4 props still strewed to the motors??

Solar activity?? What a joke!

We conclude with several key implications of our results. First, our initial estimate of the energy, EK ≈ 3 × 1050 erg at ∆t ≈ 22 d (see SI Table 2), corresponds to the Eddington luminosity of a 106 M⊙ black hole, lending support to the tidal disruption scenario, and suggesting that the X-ray/γ-ray emission is collimated by a factor of ∼ 103 . Long-term radio monitoring will test this result by providing precise beaming-independent calorimetry15,16 of the true energy release.Thus the so called Phantom effectus dropus.
smily.jpg

Sorry Steve needed an update. :)
 

Members online

No members online now.

Forum statistics

Threads
143,066
Messages
1,467,358
Members
104,936
Latest member
hirehackers