Phantom 3 malfunction and crash (flight record video)

vinnnocc · May 23, 2015

that sucks. would love to see a video of the log with your stick bottoms showing.

Adam Kontras · May 23, 2015

vinnnocc said:
that sucks. would love to see a video of the log with your stick bottoms showing.

I'll put a video up. They just stop. I didn't do anything once it was going nuts... but I can understand the curiosity.

Adam Kontras · May 23, 2015

vinnnocc said:
that sucks. would love to see a video of the log with your stick bottoms showing.

Here ya go:

F6Rider · May 23, 2015

You can use iFunBox to get to some of the flight records on the Pilot App, (don't need the bird plugged in) but the real data DJI will want is on the internal SD card i am thinking. Sorry you lost your bird, I know you were having a blast with it. Hopefully DJI or the Dealer will do right by you.

vinnnocc · May 23, 2015

******* thats bad.

isopro · May 23, 2015

DJI might claim that this was caused by a Pilot error collision at 0:10 sec into the Video you input forward collective pitch into the remote and exactly after it act like if it collided with something, then at 0:12 is look like there is a secondary impact, that make it spin even harder until it dies on the ground.

Good luck !

Edit: Also if you look the forward and vertical speed indicator it does give data that would confirm this theory...

My 2 cents..

Also i noticed you often give 100% stick into the controls, might i advise against such aggressive flight in urban area...No need to input more then 60% stick input, unless you need to make a emergency altitude/trajectory change...

Pumpfixer · May 23, 2015

isopro said:
DJI might claim that this was caused by a Pilot error collision at 0:10 sec into the Video you input forward collective pitch into the remote and exactly after it act like if it collided with something, then at 0:12 is look like there is a secondary impact, that make it spin even harder until it dies on the ground.

Good luck !

Edit: Also if you look the forward and vertical speed indicator it does give data that would confirm this theory...

My 2 cents..

Also i noticed you often give 100% stick into the controls, might i advise against such aggressive flight in urban area...No need to input more then 60% stick input, unless you need to make a emergency altitude/trajectory change...

Wow what do you do for work? Are you a detective? Haha nice catch

isopro · May 23, 2015

Pumpfixer said:
Wow what do you do for work? Are you a detective? Haha nice catch

No, i'm normally what the detectives are looking for !

Adam Kontras · May 23, 2015

isopro said:
DJI might claim that this was caused by a Pilot error collision at 0:10 sec into the Video you input forward collective pitch into the remote and exactly after it act like if it collided with something, then at 0:12 is look like there is a secondary impact, that make it spin even harder until it dies on the ground.

Good luck !

Edit: Also if you look the forward and vertical speed indicator it does give data that would confirm this theory...

My 2 cents..

Also i noticed you often give 100% stick into the controls, might i advise against such aggressive flight in urban area...No need to input more then 60% stick input, unless you need to make a emergency altitude/trajectory change...

Thankfully you can look at this flight in a 3D space and see there was literally nothing around me.

Adam Kontras · May 23, 2015

...but by all means, make up something like "secondary impact" to give a corporation an out to avoid responsibility.

Alright, someone please help me get these txt files out of the drone so we can see this flight in a 3D space so it's obvious there is nothing impacting the flight, and that no, pushing the sticks at 90 or 100% doesn't cause complete motor failure.

F6Rider · May 23, 2015

Adam Kontras said:
Thankfully you can look at this flight in a 3D space and see there was literally nothing around me.

Bird?? You were 90' up and looking at the app. Just a possibility.

Skynet1 · May 23, 2015

What concerns me is that the new motors on the P-3 get fairly hot to the touch, and I don't mean just warm. I've got three DJI Phantoms, and none of the motors on the earlier models have ever gotten as hot to the touch.

I say this not as result of a long flight or aggressive flying, but simply as result of hovering, and some simple maneuvers. I usually break in the motors this way for the first several flights, and will end the flight with 50% battery remaining during these initial flights. I had not found any resistance in any of the motors as they all turn freely with little resistance.
Has anyone else noticed the temperature of the motors after flight, more than mildly warm, but hot.

Adam Kontras · May 23, 2015

F6Rider said:
Bird?? You were 90' up and looking at the app. Just a possibility.

Your theory is I hit a bird.

And no, I wasn't looking at my app, I was looking directly at my drone as I was bringing it back down to me. ?!?!?! SO you're saying because DJI can claim I hit a bird - they have no responsibility!??! Seriously?

F6Rider · May 23, 2015

Adam Kontras said:
Your theory is I hit a bird.

And no, I wasn't looking at my app, I was looking directly at my drone as I was bringing it back down to me. ?!?!?! SO you're saying because DJI can claim I hit a bird - they have no responsibility!??! Seriously?

No, I didn't say it was my theory, I asked YOU if it was a possibility, not the same thing. And you did say you were making setting changes, so I didn't know if you were eyeballing the bird when it started its death dance or not.

isopro · May 23, 2015

Adam Kontras said:
...but by all means, make up something like "secondary impact" to give a corporation an out to avoid responsibility.

Alright, someone please help me get these txt files out of the drone so we can see this flight in a 3D space so it's obvious there is nothing impacting the flight, and that no, pushing the sticks at 90 or 100% doesn't cause complete motor failure.

I'm just stating the obvious that DJI might claim, because from first look of this data, it looks like you gave forward pitch until .2 second before the start of the (change of attitude) and you also give a fast -100% throttle .1 sec before the crash that can be interpreted as a panic move to try to go under something like a light pole, electricity wire, or else, but failed and it hit a first time, making the aircraft suddenly loose 2m altitude and starting to spin and then 2 second later there is a second change of attitude that align with the line of a building on the map and then it fall even faster.

Motor failure does not let the aircraft fly another 30m from 30m altitude, if motors fail it probably drops pretty fast considering it was already at low altitude, but it could be possible also that a ESC failed and the bird was trying to compensate during the crash trajectory.

Can we have the google map coordinate (address/street of where this flight occurred), then it will be easy to see if there's anything that could of caused the first change of attitude and or impact ...?

DrChris · May 23, 2015

Good eyes, isopro. I looked at this video in detail too and found:

The bird was in a hover like Adam described for some time (extending to before this section of the video). Then the throttle was moved down by 91% resulting in the phantom going from 118 to 91 feet. At this point it still seems to be in control. Next 100% left is applied to the left stick resulting in a rotation counterclockwise by about 100 degrees. The bird still seems in control at this point.

Then at the 00:10 mark 81% forward stick was applied until the phantom had 9.1 MPH forward speed (before spinning out of control). Just before the forward stick was released, the left stick was applied at 91% in the downward direction resulting in a 4.9 MPH downward speed. After this both sticks were released and the bird started spinning out of control. After this point, there are no inputs to the sticks except for a 2% applied to the right stick in the reverse direction at about 00:15 very briefly.

Adam, could you have accidentally hit something at that point? It was about 82 feet in the air at the time. I know that you were adjusting settings in the app. Could you have accidentally hit the control sticks causing first mild descent, then rotation by about 100 degrees, then forward acceleration, then descent again?

Please understand that we're not trying to inadvertently place blame on you! You've posted some awesome video on this site that is simply amazing. I'm just hoping to help figure out what happened.

So sorry for the loss of your phantom! (I've come to realize that there's no point in being in this hobby if you can afford to lose everything you have invested. That may be a lot like going to Vegas anyway!)

Best wishes and very interested to hear the end of this,

Chris

Adam Kontras · May 23, 2015

Skynet1 said:
What concerns me is that the new motors on the P-3 get fairly hot to the touch, and I don't mean just warm. I've got three DJI Phantoms, and none of the motors on the earlier models have ever gotten as hot to the touch.

I say this not as result of a long flight or aggressive flying, but simply as result of hovering along with some simple maneuvers. I usually break in the motors this way for the first several flights, and will end the flight with 50% battery remaining during these initial flights. I had not found any resistance in any of the motors as they all turn freely with little resistance.
Has anyone else noticed the temperature of the motors after flight, hot.

Just so you know, this was my 27th flight with my P3 (nearly 3 hours flight time) and it's also not my first drone, I had the P2 before this. As you can see from my Vegas drone footage, I'm by no means a novice. This flight was less than 4 minutes and I spent most of the time hovering and trying to recreate the "jello" effect and then changing settings.

I brought it from 200 feet to 90 feet, then turned the aircraft towards me. Went forward at 2-3 mph and began the descent. It was during the decent that it began to spin out of control. I did not hit a bird, but I guess I can't prove that I didn't hit a bird so DJI will pretend this never happened and I'm just out $1250.

Thank you for giving them that, everyone. Meanwhile, there's a very serious issue with this craft that is either a motor issue or a firmware issue that cause it to spin violently out of control and crash with absolutely no user-error.

Adam Kontras · May 23, 2015

Guys, just look at the map. There is a tree at the left of my yard, but my drone doesn't come within 20 feet of it. It starts spinning 80 feet up and gets faster and faster as it loses control. There are no powerlines in our front yards, those are all at the back edge of our properties.

Serpent · May 23, 2015

Sorry to hear of your loss. Recommend pull the flight logs and see if there's any data for each motor that could indicate a motor/ESC failure. Hopefully DJI have the means. Although near 100% stick inputs are quite agressive it should be designed for it.

Adam Kontras · May 23, 2015

isopro said:
I'm just stating the obvious that DJI might claim, because from first look of this data, it looks like you gave forward pitch until .2 second before the start of the (change of attitude) and you also give a fast -100% throttle .1 sec before the crash that can be interpreted as a panic move to try to go under something like a light pole, electricity wire, or else, but failed and it hit a first time, making the aircraft suddenly loose 2m altitude and starting to spin and then 2 second later there is a second change of attitude that align with the line of a building on the map and then it fall even faster.

Motor failure does not let the aircraft fly another 30m from 30m altitude, if motors fail it probably drops pretty fast considering it was already at low altitude, but it could be possible also that a ESC failed and the bird was trying to compensate during the crash trajectory.

Can we have the google map coordinate (address/street of where this flight occurred), then it will be easy to see if there's anything that could of caused the first change of attitude and or impact ...?

I'll instant message you my address and I'll give you the co-ordinates. Please don't make it public (street names, etc)

1 out of 4 motor failures could ABSOLUTELY let the aircraft spin out of control and at 90 feet high, it could certainly continue to travel. However, all the readings after it freaked out are obviously not accurate. The GPS is going in and out - it says it's final resting spot is -56 feet. Myself and all my neighbors watched it spin, rock and generally freak out. Was it trying to right itself? Surely it was. It didn't just stop spinning and drop, it was freaking out. Something happened and clearly it was to one arm of the copter which caused it to spin.

Listen, I know I'm a little defensive here and some of you guys are indeed trying to figure out what happened because it's scary to think this can happen without any user error. I will try to be as unemotional as possible when talking about this, but if you're trolling to make it look like my fault because you want to feel better about your purchase, you're kidding yourselves. This effects ALL of us.

(And for the love of GOD, stop acting like any movement of the stick where it hits the side of the controller is worthy of a CRASH. As a guy only interested in video production, 99% of my movements are in the 5-10% range for the slowest possible movements, but you can absolutely go up and down --when over a hundred feet in the air-- at 100% movement on the sticks without issue)