Uncontrollable altitude, drone goes above 500m

@RYZPHOTO · Jul 20, 2020

My Mavic 2 Pro is having weird altitude issues, I have no warnings displayed, updated firmware multiple times, IMU and compass calibrated multiple times, vision system calibration multiple times, controller calibration multiple times etc.... I just can't figure this one out!
When flying at low speeds I don't seem to have any issues, as soon as I'm at around half or full throttle in any direction the drone climbs altitude, usually extremely fast and without touching the up/down stick, it will even continue climbing fast when I'm holding down with the left stick if I continue flying in any direction, then after letting go of the sticks it will descend lower than the original altitude from where I originally accelerated from and sometimes it takes as many as 10 corrections with the left stick pushing up agin for it to stay level without wanting to descend, this is quite dangerous as I'm flying over water a lot and if I don't keep an eye on it it would easily drop into the ocean.
I fly with the downward positioning sensors off as they cause issues from the waters surface, I have been flying with Inspires and Phantoms for years previously and have never come across this problem, I have had this drone for around 1 year and have not come across this problem until the last couple of months, it used to just randomly happen but now it is very consistent and seems to start happening after about 5 mins of flight time which just makes no sense to me.
And to top it off all 3 of my batteries have just bloated at the same time.... wonderful.
Any help would be much appreciated!
Thanks!!!

WithTheBirds · Jul 20, 2020

Does the reported altitude appear to correspond with the actual height of the Mavic?

@RYZPHOTO · Jul 20, 2020

WithTheBirds said:
Does the reported altitude appear to correspond with the actual height of the Mavic?

Yes it seems correct

MGVision · Jul 20, 2020

@RYZPHOTO said:
My Mavic 2 Pro is having weird altitude issues, I have no warnings displayed, updated firmware multiple times, IMU and compass calibrated multiple times, vision system calibration multiple times, controller calibration multiple times etc.... I just can't figure this one out!
When flying at low speeds I don't seem to have any issues, as soon as I'm at around half or full throttle in any direction the drone climbs altitude, usually extremely fast and without touching the up/down stick, it will even continue climbing fast when I'm holding down with the left stick if I continue flying in any direction, then after letting go of the sticks it will descend lower than the original altitude from where I originally accelerated from and sometimes it takes as many as 10 corrections with the left stick pushing up agin for it to stay level without wanting to descend, this is quite dangerous as I'm flying over water a lot and if I don't keep an eye on it it would easily drop into the ocean.
I fly with the downward positioning sensors off as they cause issues from the waters surface, I have been flying with Inspires and Phantoms for years previously and have never come across this problem, I have had this drone for around 1 year and have not come across this problem until the last couple of months, it used to just randomly happen but now it is very consistent and seems to start happening after about 5 mins of flight time which just makes no sense to me.
And to top it off all 3 of my batteries have just bloated at the same time.... wonderful.
Any help would be much appreciated!
Thanks!!!

As for bloated, swelling batteries, this is their trademark. They just want to take more money from their customers.

monty.roman · Jul 20, 2020

what about RC sticks inputs showed in flight log? do they act as expected or they're making some invalid movements (for exapmle some unwanted input from left stick when flying forward)?

boblui · Jul 20, 2020

Pls post the flight logs . both .TXT and .DAT files. can be found in your phone. guide on how to get them : Mavic Flight Log Retrieval and Analysis Guide

Deleted member 103366 · Jul 20, 2020

Did you buy this new or second hand?
The only DJI drone I have seen that could get over 500m above take off point was a P3 with hacked firmware on the drone.

Open question not directed at the OP, is it possible a malfunctioning barometer could cause this?

boblui · Jul 20, 2020

PhiliusFoggg said:
Open question not directed at the OP, is it possible a malfunctioning barometer could cause this?

It can also be due to IMU problems as the IMU senses vertical acceleration. The fact that the problem occurs only when the craft is tilted suggests that it may be more IMU related. Anyway, the .DAT file will be needed for identifying the reasons.

Meta4 · Jul 20, 2020

@RYZPHOTO said:
My Mavic 2 Pro is having weird altitude issues, I have no warnings displayed, updated firmware multiple times, IMU and compass calibrated multiple times, vision system calibration multiple times, controller calibration multiple times etc.... I just can't figure this one out!

Without flight data, all anyone can do is make guesses.
But the solution to the mystery is probably in the flight data.
Go to DJI Flight Log Viewer | Phantom Help
Follow the instructions there to upload your flight record from your phone or tablet.
That will give you a detailed report on the flight data.
Come back and post a link to the report it provides and someone might be able to analyse it and give you an understanding of the cause of the incident.

@RYZPHOTO · Jul 20, 2020

Sorry, I have not dealt with flight logs before, have I done it right?
The drone was bought new and the time that it went above 500 meters from when I was flying around a volcano, went flying straight up it stopped at 500 meters, then when I fly in any direction at speed it just starts climbing by itself, no input on the left stick at all, pulling down does not do anything until I stop flying in a horizontal direction....

DJI Flight Log Viewer | Phantom Help

Upload and view flight logs from your DJI drones

www.phantomhelp.com

Meta4 · Jul 20, 2020

@RYZPHOTO said:
Sorry, I have not dealt with flight logs before, have I done it right?

That's good but it's for a flight that never exceeded 33 metres.
It was on the 16th July a little after midday.

If you can find the data for the flight where the Mavic went high we can see what it shows.

@RYZPHOTO · Jul 20, 2020

That flight was quite a while ago now but I'll see what I can find...
But that one that I uploaded was my last flight, I never attempted to go to 33 meters, it does it by itself....
when tracking a surfer as soon as I speed up the drone starts to climb rapidly, no input from me....

Meta4 · Jul 20, 2020

@RYZPHOTO said:
That flight was quite a while ago now but I'll see what I can find...
But that one that I uploaded was my last flight, I never attempted to go to 33 meters, it does it by itself....
when tracking a surfer as soon as I speed up the drone starts to climb rapidly, no input from me....

OK ... I'll look closer at it.

boblui · Jul 20, 2020

It looks like there were some problem with the IMUs.

Abnormality occurred in time period A:

Some elevator / aileron input applied
Steady half throttle-down input was applied BUT
All four motors had the speed increased
Both the GPS and the barometer indicated that the craft went steadily upward as the result
The accelerometers of the two IMUs indicated that the craft was accelerating downward on average. I judge it from the size of the spikes of accelZ in different directions so I can be wrong here. But if I have interpreted the data correctly, the accelerometers were making the flight controller believe that the craft was going down too fast so the FC commanded the motors to spin faster to keep the descending speed steady.

Things returned to normal in time period B:

NO elevator / aileron input applied
Full down-throttle was applied
All four motors had the speed reduced subsequently
Both the GPS and the barometer indicated that the craft was going down
The abnormal spikes have disappeared from the outputs of the accelerometers which indicated to the FC that the craft’s vertical speed was steady.

This is the second time I see the FC ignoring the GPS and barometer data and trusting the IMU only. Sensor fusion seems not working properly here.

@RYZPHOTO · Jul 20, 2020

boblui said:
It looks like there were some problem with the IMUs.

Abnormality occurred in time period A:

Steady half throttle-down input was applied BUT

All four motors had the speed increased

Both the GPS and the barometer indicated that the craft went steadily upward as the result

The accelerometers of the two IMUs indicated that the craft was accelerating downward on average. I judge it from the size of the spikes of accelZ in different directions so I can be wrong here. But if I have interpreted the data correctly, the accelerometers were making the flight controller believe that the craft was going down too fast so the FC commanded the motors to spin faster to keep the descending speed steady.

Things returned to normal in time period B:

Full down-throttle was applied

All four motors had the speed reduced subsequently

Both the GPS and the barometer indicated that the craft was going down

The abnormal spikes have disappeared from the outputs of the accelerometers which indicated to the FC that the craft’s vertical speed was steady.

This is the second time I see the FC ignoring the GPS and barometer data and trusting the IMU only. Sensor fusion seems not working properly here.

View attachment 108345
View attachment 108346

Thanks for the info! would that mean both IMU's are faulty?

boblui · Jul 20, 2020

@RYZPHOTO said:
Thanks for the info! would that mean both IMU's are faulty?

I am not sure but it looks like so. In the other case I have seen in this forum, both IMUs were giving the same wrong info to the FC. My analysis above is kind of rough. More rigorous analysis will require finding the vertical components of accelX, Y, Z and integrating the vertical acceleration over time to get the vertical velocity, etc, etc but I don't have the set up to do so. I would suggest to wait for the replies of other more experienced analysts here before coming to any conclusion.

boblui · Jul 20, 2020

@RYZPHOTO said:
Thanks for the info! would that mean both IMU's are faulty?

BTW, the data confirmed your observation that the irregular behavior occurred when elevator / aileron input were applied.

BudWalker · Jul 20, 2020

boblui said:
.......

The accelerometers of the two IMUs indicated that the craft was accelerating downward on average. I judge it from the size of the spikes of accelZ in different directions so I can be wrong here. But if I have interpreted the data correctly, the accelerometers were making the flight controller believe that the craft was going down too fast so the FC commanded the motors to spin faster to keep the descending speed steady.

......

I haven't had a close look. But, you might find the time integrated accel signals useful. IMUCalcs:totalAccelZ:C. Since the M2 is tilting IMUCalcs:totalAccelX:C will also be relevant.

boblui · Jul 20, 2020

BudWalker said:
I haven't had a close look. But, you might find the time integrated accel signals useful. IMUCalcs:totalAccelZ:C. Since the M2 is tilting IMUCalcs:totalAccelX:C will also be relevant.

IMUCalcs:totalAccelZ:C has been increasing throughout the flight. ie, the craft was sensed to be going downward / backward on average throughout the flight. Looks like the IMU has sever bias but the OP has already calibrated the IMU for multiple times. Can't figure out why.

BudWalker · Jul 20, 2020

boblui said:
IMUCalcs:totalAccelZ:C has been increasing throughout the flight. ie, the craft was sensed to be going downward / backward on average throughout the flight. Looks like the IMU has sever bias but the OP has already calibrated the IMU for multiple times. Can't figure out why.

View attachment 108358

During hover AccelZ will be 0.0. But, it won't be 0.0 when there is horizontal acceleration and the AC is tilted. The Z axis dips down into the horizontal acceleration and AccelZ = sin(tilt) * horizontal acceleration. Since flights are typically flown in the positive X axis direction IMUCalcs:totalAccelZ:C will show a positive bias. If a flight were flown backwards then IMUCalcs:totalAccelZ:C would have a negative bias.

Uncontrollable altitude, drone goes above 500m

Member

Well-Known Member

Member

Well-Known Member

Well-Known Member

Well-Known Member

Deleted member 103366

Guest

Well-Known Member

Well-Known Member

Member

Attachments

Well-Known Member

Member

Well-Known Member

Well-Known Member

Member

Well-Known Member

Well-Known Member

Well-Known Member

Well-Known Member

Well-Known Member

Similar threads

DJI Drone Deals

Forum statistics