Hello all, I'm hoping someone here can help me. Today for the first time I started using DJI GO 4 because that is what Dji is recommending because they are not going to support DjiGo anymore. I flew at my usual location where I normally fly. Today, I was within Vlos and approx 200' altitude. I was hovering in place for a few seconds when all of the sudden my P4 started decending rapidly. At the time this happened, I was giving no stick input at all. When I realized it was coming down fast, I started pushing up on throttle all the way, but it was acting like it was fighting me. It did stop decending after fighting it back up. I flew back to my home point so it was overhead. I let it hover there for a few seconds again, then it started climbing for no reason at all, then decending again. Seems very erratic to me. So, I pulled the throttle down to land. As it was coming down, it was coming down very slowly, not it's usual speed at all. I did have over 15 satellites during this event, I looked at that several times. I don't know what would cause this to happen for no reason at all. After I landed, I shut everything down, restarted everything except this time I am using litchi. I took off, went back up to fly around where I was before, and it started again, now with litchi, so I'm guessing it's something with the P4 itself? Or could I be wrong about that? I did a calibration on Everything on a perfectly level surface outdoors with nothing near me. Has anyone had this experience or know someone on here that did? I am very nervous to fly now because of this. I don't want to lose my first nice P4. Thank you for any help you guys can give me.

The altitude drop on its own would indicate low battery. But whatever the reason, a persistent message should indicate the autonomous motion.

Sounds somewhat like a compass error. Is it spinning around like it is lost? Have you calibrated everything in a good environment? I'm sure you'll hear from others, but I would start off by doing an IMU calibration along with the compass and maybe even re-link your aircraft to make sure everything is communicating the way it should, and give it another try.

I have done that. It's not spinning around. Battery was well over 70%. But it did act like it was forcing me to land, but showed no indications of that. My battery was around 75%. But I know when the battery is too low, if forces you to land, you can fight it to stay up, but it will keep trying to land when you stop pushing up. That's what it seemed like it was doing. But the other thing was the really slow decent. I had my throttle all the way down and it was coming down painfully slow. I am looking at the .dat files off of the P4 in Assistant 2 and i can see on the graph both the barometer and gps altitude agree and I can see the climb with no throttle input and then the decent with no throttle input until I try to stop it from dropping. When I pushed the throttle up and it started climbing back up again, I let go of the throttle, it stopped climbing and hovered for a few seconds, then it started climbing on it's own up to 15m higher then dropped again 10m lower with no input at all. When trying to bring it back in, the controls seemed very erratic, like it wanted to fight me. I have never had a problem with this P4 before. Nothing has changed except that I installed Go4. But it did the same thing using litchi.

This is a screen shot of the log viewer.  Red is the barometer, purple is the gps altitude and green shows where I started throttle up and when I let go. Starting from the left, you can see the alt. increasing for no reason at all. Then it starts decending rapidly until I push up on the throttle. You can see it kept decending for a couple seconds even after I hit the throttle up. Then the next green line shows where I let go of the throttle. When I let go of the up throttle, it dropped a little again then started climbing on its own for 8 seconds then dropped back down. I didn't touch the sticks again until 450 then started bringing it down.

Please forgive my poor drawing skills here, just trying to show what is going on.

Are there other fields in the dat log that I can compare or look at to help figure out what was going on here?

Added forward pitch end point.

C. Dave Pappas Posted at 2017-2-7 13:07
Are there other fields in the dat log that I can compare or look at to help figure out what was going on here?

WOW! Strange for sure. I am hoping someone else can offer up something for you to try. I agree, that being you tried both apps, after re-calibrating everything and no errors, and had issues, I  would be concerned too. How old is you phantom? I am sure you don't want to go that route, but you might have a hardware problem, and maybe you could return it for another one. I have never had to call DJI for support, but I would try to get some feed back from them if possible. To me, I don't know why it wouldn't respond to your controls when all seems to be in order. It's not like it is trying to RTH or something, and yes it will land when low on battery but that should be in the 10% range...not at 70%. It would also by saying to you that it is landing, which you haven't heard. Good luck and let us know how you proceed. Sorry, I'm no help to you.

I bought it in  November of 2016. Never had a problem until now.

You could post your file to 'Phantom Help' (Instructions on how to find files on webpage), and post the link here, some may be able to pinpoint what might have happened.

Do you have a Gimbal guard fitted, one of the plastic clip-on devices designed to protect the gimbal from below ?

Aardvark Posted at 2017-2-7 15:16
You could post your file to 'Phantom Help' (Instructions on how to find files on webpage), and post the link here, some may be able to pinpoint what might have happened.

http://www.phantomhelp.com/LogViewer/FC09C68VAOUZBVSLWLHD/

I get more information out of the dat file than I get at PhantomHelp.

C. Dave Pappas Posted at 2017-2-7 16:57
http://www.phantomhelp.com/LogViewer/FC09C68VAOUZBVSLWLHD/

Start looking around 3 min 10 sec and later. Thats where the altitude starts going erratic.

Man0Man that had to make for some heart thumping. I had one flight about a week ago where things went haywire with the home point and twice now I've had the altitude climb way above the set point during a RTH. The only common threads to all these recent problems that people have been having is Go 4. They do still support 3 and in fact just had an update a few days ago. Even at that I still have 2.9.2 on my ipad and I'm not giving it up for anything. There are quite a few posts on here about how to go back in time with iOS if you don't want the latest Go 3.

C. Dave Pappas Posted at 2017-2-7 17:01
Start looking around 3 min 10 sec and later. Thats where the altitude starts going erratic.

Is your aircraft on latest firmware ?

Aardvark Posted at 2017-2-7 18:01
Is your aircraft on latest firmware ?

Yes, latest firmware. Another question I have is about the compass calibration. There seems to be a conflict in the procedure. The "typed" instructions tell you to point the nose of the aircraft down when doing the second part of the calibration, but the picture shows you turning it on it's side. Also some videos I watched on how other people do it say to turn it on it's side. So what or who is correct? Dji says to do it differently than the dji picture shows. Does it actually matter or make a difference?

C. Dave Pappas Posted at 2017-2-7 18:18
Yes, latest firmware. Another question I have is about the compass calibration. There seems to be a conflict in the procedure. The "typed" instructions tell you to point the nose of the aircraft down when doing the second part of the calibration, but the picture shows you turning it on it's side. Also some videos I watched on how other people do it say to turn it on it's side. So what or who is correct? Dji says to do it differently than the dji picture shows. Does it actually matter or make a difference?

I don't think it makes much difference, last time I did mine was camera down. The current guidance is that compass calibration does not need to be performed each flight. I last did mine in an open field in the middle of nowhere. The logic is that if the compass is calibrated often then there is a higher risk of getting a 'bad' calibration. i.e. the calibration itself is erroneous due to effects of surrounding materials. For example on a concrete structure with lots of steel reinforcement, beside a large metal clad building etc. A compass error would not cause this change in altitude that you've seen.
My reason for asking about firmware was that there was a fix in V1.2.503 to overcome an altitude drop problem after the aircraft had come to a sudden stop, with drops in altitude of 20 feet not being uncommon.
V1.2.503 fixed that for me although I believe it did not affect all aircraft.

Aardvark Posted at 2017-2-7 18:37
I don't think it makes much difference, last time I did mine was camera down. The current guidance is that compass calibration does not need to be performed each flight. I last did mine in an open field in the middle of nowhere. The logic is that if the compass is calibrated often then there is a higher risk of getting a 'bad' calibration. i.e. the calibration itself is erroneous due to effects of surrounding materials. For example on a concrete structure with lots of steel reinforcement, beside a large metal clad building etc. A compass error would not cause this change in altitude that you've seen.
My reason for asking about firmware was that there was a fix in V1.2.503 to overcome an altitude drop problem after the aircraft had come to a sudden stop, with drops in altitude of 20 feet not being uncommon.
V1.2.503 fixed that for me although I believe it did not affect all aircraft.

Well, it's dark here now. Here is what I did so far. I refreshed the firmware. Calibrated the remote and re-linked it to the P4. I did the IMU calibration and compass calibration. IMU was done on a perfect level surface, and compass calibration was done in my back yard, nothing within 50 feet in any direction. So far everything looks good but nothing looked bad before, so I just don't know. I will have to wait til tomorrow now to do any test flights.

C. Dave Pappas Posted at 2017-2-7 18:49
Well, it's dark here now. Here is what I did so far. I refreshed the firmware. Calibrated the remote and re-linked it to the P4. I did the IMU calibration and compass calibration. IMU was done on a perfect level surface, and compass calibration was done in my back yard, nothing within 50 feet in any direction. So far everything looks good but nothing looked bad before, so I just don't know. I will have to wait til tomorrow now to do any test flights.

Ahh, I've just remembered, try calibrating the 'sensors' using DJI Assistant 2 from HERE.. That has been known to cure some unexplained changes in altitude (sensors incorrectly interpreting data ?).

Video guide here:- https://www.djivideos.com/video_ ... d-afa8-3d4082064a81

That's for an older version of Assistant 2, but same principles, screen layouts and hoops to jump through are slightly different in latest version.

Aardvark Posted at 2017-2-7 18:55
Ahh, I've just remembered, try calibrating the 'sensors' using DJI Assistant 2 from HERE.. That has been known to cure some unexplained changes in altitude (sensors incorrectly interpreting data ?).

Video guide here:- https://www.djivideos.com/video_ ... d-afa8-3d4082064a81

Well, that was no fun. Sensors calibrated.

If it's convenient, please connect your drone to computer and get the .dat file out from your drone and upload it to Dropbox or Google Drive, it's a bit big, and might take a while. Then our engineer can take a look to determine which part went wrong.

DJI-Thor Posted at 2017-2-8 01:19
If it's convenient, please connect your drone to computer and get the .dat file out from your drone and upload it to Dropbox or Google Drive, it's a bit big, and might take a while. Then our engineer can take a look to determine which part went wrong.

I uploaded the dat file using Assistant 2 and included my claim number.

Ok, first test flight today was a sucess. I had no altitude issues at all. It was a short 8 minute flight, staying within 500' of my location. Altitude set at 200. It changed only by 1 foot up or down, but no more that that. So I would say that is acceptable. I will do some more test flights today to make sure there are no more issues or altitude problems. Hopefully this was just a glitch that was corrected with calibrations. Like I said, this thing has been perfect in flying performance since the day I got it. I hope it stays that way now. It will bug me though if it is fixed because I will never know what caused the problem in the first place. Hopefully when Dji is done examining the dat file, they can tell me what was causing it.

C. Dave Pappas Posted at 2017-2-8 07:11
Ok, first test flight today was a sucess. I had no altitude issues at all. It was a short 8 minute flight, staying within 500' of my location. Altitude set at 200. It changed only by 1 foot up or down, but no more that that. So I would say that is acceptable. I will do some more test flights today to make sure there are no more issues or altitude problems. Hopefully this was just a glitch that was corrected with calibrations. Like I said, this thing has been perfect in flying performance since the day I got it. I hope it stays that way now. It will bug me though if it is fixed because I will never know what caused the problem in the first place. Hopefully when Dji is done examining the dat file, they can tell me what was causing it.

Good to hear that it appears to be behaving now, exactly what caused it might remain a mystery, but my money would be on the sensor calibration.

Second flight today went well. Flew at 200' for 9 minutes. Went a little farther away and half way into the flight I switched to sport mode. With a tail wind, I was reaching 75 mph. I don't know if that speed reading in Go4 is correct or not, but it was moving fast. No altitude problems at all. It stayed at 200 most of the flight.