DJI Mavic, Air and Mini Drones
Friendly, Helpful & Knowledgeable Community
Join Us Now

Mavic Air 2 self-ascended to 1.5 km due to sensor error

That looked scary. The altitude readout was stuck at 87 m and the vertical speed was jumping all over the place. It looked like he had RTH activated but it was doing nothing to stop the ascent.

The pilot's heart must have been in his throat during that ordeal. Do you know if he recovered the quad intact?
 
  • Like
Reactions: New England Droning
That looked scary. The altitude readout was stuck at 87 m and the vertical speed was jumping all over the place. It looked like he had RTH activated but it was doing nothing to stop the ascent.

The pilot's heart must have been in his throat during that ordeal. Do you know if he recovered the quad intact?
Take a look at the battery when last available (9% @2:50, it was still climbing, it should have been auto descending), I am thinking given all shown...it fell to earth and who knows where.
 
Last edited:
Is there a reason given that they didn't perform a CSC?

Ehhh... I think that's a tough call.

With it continue to going up, you do have some worry about manned aircraft aloft. But if you CSC, you've got to worry about exactly where that comes crashing down and property damage or personal injury. Obviously, in this event it seems likely it lost power and came crashing down anyway -- so hindsight tells us CSC would have been a better plan, but I think the natural reaction is to think you can get control over it, or it will figure itself out and auto land when battery is low. I think hardly anyone would see their drone going up with plenty of battery left and immediately think"looks like I'll never get control back and it won't auto-land so lets CSC". That's a tough call to make in a panic situation no one has ever really seen before. In my area, I'm near a major airport, so I'm probably more likely to CSC knowing there is probably going to be aircraft nearby. If I'm in an area where air traffic is unlikely, maybe its a different choice.

It's easy for us to dissect the video now and see the barometer is gone, and the drone thinks its descending when it isn't, and that might not fix itself. I think its less likely we would figure that all out in-the-moment.
 
Ehhh... I think that's a tough call.

<SNIP>

It's easy for us to dissect the video now and see the barometer is gone, and the drone thinks its descending when it isn't, and that might not fix itself. I think its less likely we would figure that all out in-the-moment.
I think in the situation given, at least for my area, I would have thrown it all in on the "Hail Mary Pass". I would have CSC'd counted to 5 then CSC'd again...That way I would have depleted almost all altitude, restarted and maybe reset the drone and then had it auto land at a much lower (With hopefully enough power) altitude all things given. I would have much less wanted to hit a possible aircraft (I am super close to airport) than something on the ground had it fail, as it looked like she was coming screaming down anyway. :p
 
  • Like
Reactions: z28lt1
I would have CSC'd counted to 5 then CSC'd again...That way I would have depleted almost all altitude, restarted and maybe reset the drone
That's possibly worth a try except that in 5 seconds it's not going to lose much altitude.
Falling Mavics have a lot of drag and a terminal velocity of 12-14 metres/sec.
 
at Z28lt1, all true, especially hindsight being a wonderful thing and I agree the pilot was between the Devil and the deep blue sea concerning height and someone's head.
I suppose it is something that comes with experience or age but I think that, in the event that something has gone wrong and it is flying into a dangerous area, we must be prepared to ditch whilst we have some control and sight of the drone.
I have had this happen twice, both with a Phantom 3, both probably my fault.
In the first case I was a newbie and suspect I did something wrong, the drone flew away at speed but low and was heading for a busy road, I crashed it into a wall between me and the road. Fortunately for me there was little or no damge, I might have cracked a prop.
When compared to the pilot of this thread I had the advantage in that I knew where it was but this pilot presumably let it climb out of sight, prior to that they might have been able to discern where it would crash.
In the second case the winds above the trees around my house took it and flat-out it couldn't make headway, it was heading towards other peoples' houses. This was sustained wind and not merely a quick gust. I was literally just about to fly it, perpendicular to the wind, out over the sea and ditch when the wind dropped and it was able to make headway and I was able to bring it back. I couldn't descend, to get out of the wind, where it had been blown to as it was over other peoples' gardens.




I am curious though..........how is the final height of this drone known?
 
Last edited by a moderator:
  • Like
Reactions: A.O. and z28lt1
That's possibly worth a try except that in 5 seconds it's not going to lose much altitude.
Falling Mavics have a lot of drag and a terminal velocity of 12-14 metres/sec.
I like to error on the side of caution and take baby steps? :p
 
at Z28lt1, all true, especially hindsight being a wonderful thing and I agree the pilot was between the Devil and the deep blue sea concerning height and someone's head.
I suppose it is something that comes with experience or age but I think that, in the event that something has gone wrong and it is flying into a dangerous area, we must be prepared to ditch whilst we have some control and sight of the drone.
I have had this happen twice, both with a Phantom 3, both probably my fault.
In the first case I was a newbie and suspect I did something wrong, the drone flew away at speed but low and was heading for a busy road, I crashed it into a wall between me and the road. Fortunately for me there was little or no damge, I might have cracked a prop.
When compared to the pilot of this thread I had the advantage in that I knew where it was but this pilot presumably let it climb out of sight, prior to that they might have been able to discern where it would crash.
In the second case the winds above the trees around my house took it and flat-out it couldn't make headway, it was heading towards other peoples' houses. This was sustained wind and not merely a quick gust. I was literally just about to fly it, perpendicular to the wind, out over the sea and ditch when the wind dropped and it was able to make headway and I was able to bring it back. I couldn't descend, to get out of the wind, where it had been blown to as it was over other peoples' gardens.




I am curious though..........how is the final height of this drone known?
Live streamed GPS XYZ Cartesians maybe? Don’t know if they are sent.
 
Wow...
I would have brought it down as soon as the it got stuck on 87.
 
It looks like the RTH altitude was set at 2850M.

(Edited)
No, that's the Max distance. That information is on two screens. The three numbers you see in preflight check or status in order are RTH, Max Altitude, Max Distance.

In the safety tab it is Max Altitude, Max Distance, RTH Altitude.

If we can't read the language we can still tell by the grey heading bar. Max Distance and Altitude are always together, so if there is one thing after the Grey bar, it is RTH. In this case, there are two things after the grey title bar, so those are Max Distance and Altitude.

In this video, RTH is set to 194 Meters. Should be noted that you can't set it past 500 meters (same with Max Altitude).
 
I have had something similar happening with my MA2, in fog. Fog is bad. Difficult to say if that is the issue here though.
 
IF the drone was not "adjusted", this is a serious design issue. Supposing at least one between the gps sensor and the barometric sensor is valid the drone should be able to fly somehow.
 
Is there a reason given that they didn't perform a CSC?
I have read several postings here ask the reason the pilot did not perform a CSC. I just read the manual on the subject. I have a question... If the pilot, in this situation, could stop the motors (from this height) and then restart them before the drone hit the ground?
 
I have read several postings here ask the reason the pilot did not perform a CSC. I just read the manual on the subject. I have a question... If the pilot, in this situation, could stop the motors (from this height) and then restart them before the drone hit the ground?

It's been done before on other Mavics. I'm not sure anyone has tried with a Mavic Air 2, but no reason to think it wouldn't work. It's what @Repaid1 was referring to by "counting to 5".

If you are feeling brave, give it a shot and let us know how it goes ?.

EDIT: Should add it is one thing to restart the motors. Another thing for the drone to right itself from a tumble, but there have been cases where the restart has led to full recovery to normal flight.
 
I have read several postings here ask the reason the pilot did not perform a CSC. I just read the manual on the subject. I have a question... If the pilot, in this situation, could stop the motors (from this height) and then restart them before the drone hit the ground?
As long as the aircraft stays upright and reasonably flat then yes, but it's a bit of a lottery...
 
Lycus Tech Mavic Air 3 Case

DJI Drone Deals

New Threads

Forum statistics

Threads
131,145
Messages
1,560,363
Members
160,116
Latest member
henryairsoft1