DJI Phantom 4 Malfuntion - Climbed On Its Own, Would Not Descend

[gerickson9301]

Hello everyone. Last Saturday, 7/21/18 I had an incident with my DJI Phantom 4 Drone. I was operating the drone from a lake-front beach, so it was very open and clear. When I took off my battery was at 70% and the homepoint was recorded prior to take-off. I took off and was flying for a few minutes and everything seemed normal. All of a sudden, my drone started climbing in altitude on its own, I tried to use the joystick to direct it to descend and it would not respond, it kept going up. I hit Return to Home, it did not respond, it kept climbing. I tried AutoLand, it would still not respond. It climbed until it reached 500 meters in altitude, which is the maximum allowed in the settings. I tried Return to Home & Auto Land again, nothing would happen. I tried to change the maximum height in the settings, and this would not make it come down either.

I still had full control of the drone EXCEPT to descend, Return to Home, and Auto Land. The battery was dying and time was ticking, meanwhile the drone was still at 1,640 Feet. I tried to call DJI technical support, but, this was after hours. We were quickly realizing that the drone would likely fall out of the sky once the battery completely drained. We were around many people on the beach and did not want to risk hitting anyone, so, we made sure the drone was above the lake that we were next to and set out in our boat as fast as we could to get near it.

When the battery was critically low, the drone began to descent on its own. I kept receiving low voltage warnings and Not Enough Force / ESC Error messages. The last recorded height in the flight log is 817 Feet high at 18 minutes into the flight, the drone dropped from there and splashed hard into the lake sinking like a rock. We tried to get close to retrieve it but we were not successful.

I have called DJI several times to report this incident and to try to get help. I have flight logs, pictures, and videos of this entire incident happening.

DJI refuses to help because my drone is over 1 year old. What can I do? The drone CLEARLY had a malfuntion of some sort. I tried everything to get this drone down safely and unfortunately I was not able to.
Any tips, help, or advice? Thank you!

 

[Fly Dawg]

Share your flight logs and we can look to see what happened. Either upload the .txt file from your device to the link below, or the .txt file directly here if it is not too large. Instructions are in the link.
Affter you upload, share a link back here to that, if you upload to Phantom Help.
DJI Flight Log Viewer - Phantom Help

 

[gerickson9301]

Share your flight logs and we can look to see what happened. Either upload the .txt file from your device to the link below, or the .txt file directly here if it is not too large. Instructions are in the link.
Affter you upload, share a link back here to that, if you upload to Phantom Help.
DJI Flight Log Viewer - Phantom Help

I believe I uploaded correctly, here's the link:
DJI Flight Log Viewer - PhantomHelp.com

Also, here's 1 quick video I made while this was happening:

 

[Fly Dawg]

This will take some time to review fully. Others may chime in as well. Bear with us.

 

[Fly Dawg]

At a quick look, It appears your RC throttle was stuck at MAX for quite a long duration, but recovered near the end. I am just putting this up as a reference for the moment.

 

[gerickson9301]

Wow, would this mean that my controller somehow malfunctioned? As you can see from the video, we had the throttle on full descend and it wasn't responding.

 

[Fly Dawg]

Wow, would this mean that my controller somehow malfunctioned?

A definite possibility, still looking through the other.

 

[captainmilehigh]

I wonder if turning off the r/c (controller) at any point in the flight would have triggered an autoland/RTH? I once accidentally turned off my r/c during flight, and she headed home.

 

[gerickson9301]

I wonder if turning off the r/c (controller) at any point in the flight would have triggered an autoland/RTH? I once accidentally turned off my r/c during flight, and she headed home.

That's one thing I did not try, looking back I think I should have tried it. I guess I had thought it wouldn't make a difference since it was not responding to the RTH/Autoland when I tried to initiate it.

 

[Fly Dawg]

As shown below, all of the RC controls were responding apparently with the exception of the Throttle which is odd because the rudder was. Your RTH commands were also transmitted it appears. ( I will have to look back at the distance from home, due to the aircraft will autoland if the RTH command is given too close to the home point.) Again however, assuming this throttle data is correct, and will ask @sar104 to look at this as well, the aircraft could not RTH because it was receiving full up throttle commands. Again, I need to look back at the timeline, but just from this it appears to be the case. As a note, it also appears that at one point the aircraft was on its way back home as commanded. Either you cancelled it accidentally thinking that it was not recieving or some other anomaly occurred.

 

[gerickson9301]

As shown below, all of the RC controls were responding apparently with the exception of the Throttle which is odd because the rudder was. Your RTH commands were also transmitted it appears. ( I will have to look back at the distance from home, due to the aircraft will autoland if the RTH command is given too close to the home point.) Again however, assuming this throttle data is correct, and will ask @sar104 to look at this as well, the aircraft could not RTH because it was receiving full up throttle commands. Again, I need to look back at the timeline, but just from this it appears to be the case. As a note, it also appears that at one point the aircraft was on its way back home as commanded. Either you cancelled it accidentally thinking that it was not recieving or some other anomaly occurred.

This is interesting. Again, and as you can see from the video I posted, we definitely did not have the throttle in the up 100% position, so I'm assuming it's safe to say that something was malfunctioning? We tried everything we could think of at the time to get this drone to land for about 15 or so minutes.

 

[Fly Dawg]

and as you can see from the video I posted, we definitely did not have the throttle in the up 100% position,

That I am fairly certain even without the video. No one in their right mind would do such a thing on purpose. Just trying to look at some more of the data before offering any sort of conclusion.

 

[gerickson9301]

That I am fairly certain even without the video. No one in their right mind would do such a thing on purpose. Just trying to look at some more of the data before offering any sort of conclusion.

Thank you very much for your insights on this, I appreciate it.

 

[Fly Dawg]

I am "almost" convinced that this was indeed a "stuck" throttle. Judging from all the other RC inputs reacting as they should. The chart below simply shows the states of the AC and RC vs the Altitude. The altitude remains unchanged until the close to the end of the flight. What is very odd however is that right at the end, the throttle seems to have recovered but by then it was only around 70 seconds until the end. I will assume that you were moving the mode switch on the RC in an attempt to get control. That is shown in the bottom chart.

 

[gerickson9301]

I am "almost" convinced that this was indeed a "stuck" throttle. Judging from all the other RC inputs reacting as they should. The chart below simply shows the states of the AC and RC vs the Altitude. The altitude remains unchanged until the close to the end of the flight. What is very odd however is that right at the end, the throttle seems to have recovered but by then it was only around 70 seconds until the end. I will assume that you were moving the mode switch on the RC in an attempt to get control. That is shown in the bottom chart.

We did switch the modes on the control when attempting to gain control back, nothing worked for us though and we never had control over ascend/descend. Does a "stuck throttle" happen often on these? Am I out of luck as far as DJI helping me because the drone is over 1 year old? I'm so upset because I don't feel I did anything wrong and now I've lost my drone.

 

[Fly Dawg]

I don't feel I did anything wrong and now I've lost my drone.

Apparently that is the sad part. I don't think you did anything wrong. But if this issue is a correct diagnosis, and nothing is a sure thing, you could possibly still deal with DJI and provide all the data that you have. It may or may not be successful, but you have nothing to lose by doing that. Granted you do not have the AC but you do have these data files to utilize as a possible case. Let me ask one other @BudWalker for a third opinion on this. If we all agree then you have a good case, I would think. However, this could also be a case of some sort of "debris" interfering with the stick motion, but I think you would have noticed that.

 

[gerickson9301]

We may go diving and recover the drone, it's only 35 deep where it sank. Although, I don't really think this will help me accomplish anything.

 

[Fly Dawg]

Although, I don't really think this will help me accomplish anything.

It could, if ( and a big if ) you can recover the bird. Since this is fresh water you have a chance, assuming there is no significant damage to the aircraft. What you can do is, if you do recover it. Allow it to dry for a LONG period of time. ( A vacuum dryer would speed the process but that is not generally something anyone would have). On quite a few occasions they survive and are still functional. Barring that, if you can get the aircraft to simply power up, you can pull the data file from the AC itself, which provides much more detail than the device data logs. That said, an addition to a possible case, if you were to pursue that.

 

[BudWalker]

Apparently that is the sad part. I don't think you did anything wrong. But if this issue is a correct diagnosis, and nothing is a sure thing, you could possibly still deal with DJI and provide all the data that you have. It may or may not be successful, but you have nothing to lose by doing that. Granted you do not have the AC but you do have these data files to utilize as a possible case. Let me ask one other @BudWalker for a third opinion on this. If we all agree then you have a good case, I would think. However, this could also be a case of some sort of "debris" interfering with the stick motion, but I think you would have noticed that.

I took a quick look. It certainly does seem that the RC thought the throttle was full forward in the interval [171, 1033] secs.

 

[gerickson9301]

It could, if ( and a big if ) you can recover the bird. Since this is fresh water you have a chance, assuming there is no significant damage to the aircraft. What you can do is, if you do recover it. Allow it to dry for a LONG period of time. ( A vacuum dryer would speed the process but that is not generally something anyone would have). On quite a few occasions they survive and are still functional. Barring that, if you can get the aircraft to simply power up, you can pull the data file from the AC itself, which provides much more detail than the device data logs. That said, an addition to a possible case, if you were to pursue that.

I would imagine it's pretty damaged just from the impact, let alone the water.