Mavic 2 Pro Flyaway - Please Help Explain

[WaleyZhang]

Hello everyone, new member here, and I'm very glad I found this forum!

I want to start by giving some context on myself as a flier. I've been flying my Mavic 2 Pro for about a year, and I've never really had any safety issues. I have a recreational drone license and I read the manuals carefully. I'm a fairly frequent flier and I've logged many hours of flying since I've gotten the drone.

Fast forward to today, I was out by a mountainside, somewhat high up (~7000ft elevation) but definitely within the working bounds of the drone, and trying to film a video of the afternoon sun above the clouds. I launched my drone by a roadside, well clear of any vehicles and pedestrians, and started filming my video. 2 minutes later, I lost downlink with the drone, possibly because there was some trees blocking the radio signals. At that point, my controller indicated that it had switched to "Return to Home" mode, which I expected it to do so since in the few times that I've had connections issues with my drone, my drone automatically switched to Return-to-Home and came back successfully. However, I waited for 5 minutes and still the connection had not been restored, so I drove a little further, to an area I believed would no longer be blocked by said trees, and reestablished connection with my drone. However, to my horror, the drone, during the 5 minutes of disconnection, had flown almost a kilometer away and was definitely not heading back towards the original launch site. What was even more troubling was the controller and the DJI GO 4 app were both showing the drone was in Return-to-Home mode. So, I tried to switch to manual control in order to manually fly it back, but quickly realized the drone was not responding to my inputs---despite being on full throttle, it would not move in the direction of my controls. At this moment, the low battery warning began to flash so I decided to just let the drone handle Return-to-Home like it always does when it's on low battery, but again, it failed to fly towards its starting point. Instead, it drifted away from me, decreasing in altitude, and eventually, when the battery ran out, it appears to have crashed. During its last minutes, the low battery warning automatically trigged it to land-in-place, which made it descend along the side of the mountain, The last transmission I received showed it being ~750 feet below my starting altitude. Unfortunately, the crash site is nowhere close to the road and the trek down the mountainside seemed too dangerous, so I wasn't able to retrieve the drone.

I've uploaded my flight record to AirData and the shareable link is here: (you can download the original txt file from it). I've also uploaded my DAT files to a Google Drive folder here. If any of you are able to help me understand why my drone flew away, that'd be greatly appreciated!

A final question: when I bought the drone, I also purchased DJI Care Refresh alongside it (for precisely incidents like today), and the 1-year term on it hasn't yet expired. Since I can't retrieve the drone, I'm unsure if my DJI Care Refresh would cover this incident. Does anyone have any advice on what I can do? Is there still a way to somehow use DJI Care Refresh to get a replacement? It'd really suck if this becomes $2k down the drain...

 

[Meta4]

If any of you are able to help me understand why my drone flew away, that'd be greatly appreciated!

The reason for the incident is very simple and should have been obvious to you.
In just one word it was ..... WIND
You don't seem to have had any idea of the wind and what that was doing to the drone.

However, to my horror, the drone, during the 5 minutes of disconnection, had flown almost a kilometer away and was definitely not heading back towards the original launch site .... So, I tried to switch to manual control in order to manually fly it back, but quickly realized the drone was not responding to my inputs---despite being on full throttle, it would not move in the direction of my controls.

Your drone was responding to your control inputs, but the headwind was more than it could fly against.
Here's a slightly longer version of what happened.

As you climbed above the launch point, you failed to notice that there was a significant wind and the drone was having difficulty holding position.
At almost 400 ft the drone was being blown at 3-8 mph by gusty winds.
There was also a strong wind warning at 50 seconds.

You also didn't pay attention to the wind direction and flew the drone off downwind.
You lost signal at 123.5 seconds with the drone 1470 ft from home.
Signal was re-established at 472 sec with the drone now 3755 feet away in RTH but and being blown backwards at up to 5 mph.
You let the drone be blown further away and started descending at 506.4 sec.
You also tried pushing the right stick forward and the drone started to make slow headway against the headwind at times and going backwards at others.

At 541.8 sec you initiated RTH again and left the drone to be blown further backwards as it tried to RTH against a headwind.
At 695.9 sec you cancelled RTH again and resumed control.
At full stick the drone was still blowing backwards slowly.
At 714.3 sec ith the battery at 12% you tried RTH again.
Shortly after at 726.3 sec, the drone commenced autolanding because the battery had reached critical low voltage.

The battery ran out at 972.7 sec.

The place to look for it is: 20.77686 -156.25629
It's possible that it landed or was very close to the ground when the power ran out so you may be in luck?

 

[WaleyZhang]

Thank you @Meta4 I had suspected after the fact that the wind might have had something to do with it, even though from ground level where I was controlling it I didn't feel a significant gust around me personally. Also, based on the weather data from AirData it seems like the wind had a speed of 15.5mph, though if I recall correctly, the Mavic 2 Pro is rated for up to 22mph. Why was the headwind then more than it could fly against?

I understand that pilot error on my part in the form of not properly assessing the wind conditions is the cause of the crash. How does that factor into the DJI Care Refresh condition? Is there anything that I can do?

Again, I really appreciate your fast response and analysis of my flight data!

 

[Meta4]

You might be in luck if you go searching.
It looks like it came down close to a track near where it crosses a gully.
The yellow line shows the descent path.

 

[Pappy]

Thank you @Meta4 I had suspected after the fact that the wind might have had something to do with it, even though from ground level where I was controlling it I didn't feel a significant gust around me personally. Also, based on the weather data from AirData it seems like the wind had a speed of 15.5mph, though if I recall correctly,

You need to take into account that wind speed INCREASES as the aircraft gains altitude. 15 on the ground could be 30 or more at 400'.

See - WIND SPEED INCREASING WITH HEIGHT.

I forgot to add - When you do fly on windy days always start your flight going up wind and continue the up wind side until you are ready to come back. This helps to ensure you will always have enough fuel in the fuel tank to make it home.

 

[DougMcC]

The reason for the incident is very simple and should have been obvious to you.
In just one word it was ..... WIND
You don't seem to have had any idea of the wind and what that was doing to the drone.

Your drone was responding to your control inputs, but the headwind was more than it could fly against.
Here's a slightly longer version of what happened.

As you climbed above the launch point, you failed to notice that there was a significant wind and the drone was having difficulty holding position.
At almost 400 ft the drone was being blown at 3-8 mph by gusty winds.
There was also a strong wind warning at 50 seconds.

You also didn't pay attention to the wind direction and flew the drone off downwind.
You lost signal at 123.5 seconds with the drone 1470 ft from home.
Signal was re-established at 472 sec with the drone now 3755 feet away in RTH but and being blown backwards at up to 5 mph.
You let the drone be blown further away and started descending at 506.4 sec.
You also tried pushing the right stick forward and the drone started to make slow headway against the headwind at times and going backwards at others.

At 541.8 sec you initiated RTH again and left the drone to be blown further backwards as it tried to RTH against a headwind.
At 695.9 sec you cancelled RTH again and resumed control.
At full stick the drone was still blowing backwards slowly.
At 714.3 sec ith the battery at 12% you tried RTH again.
Shortly after at 726.3 sec, the drone commenced autolanding because the battery had reached critical low voltage.

The battery ran out at 972.7 sec.

The place to look for it is: 20.77686 -156.25629
It's possible that it landed or was very close to the ground when the power ran out so you may be in luck?

Very educational post for ALL of us

 

[HoozierDroneDaddy]

He might also look at the turbulence affect near ridge lines.

 

[tlswift58]

Thank you @Meta4 I had suspected after the fact that the wind might have had something to do with it, even though from ground level where I was controlling it I didn't feel a significant gust around me personally. Also, based on the weather data from AirData it seems like the wind had a speed of 15.5mph, though if I recall correctly, the Mavic 2 Pro is rated for up to 22mph. Why was the headwind then more than it could fly against?

I understand that pilot error on my part in the form of not properly assessing the wind conditions is the cause of the crash. How does that factor into the DJI Care Refresh condition? Is there anything that I can do?

Again, I really appreciate your fast response and analysis of my flight data!

At what altitude was the wind in AirData at the speed you mention - was it 100, 250, or 400 ft. That is what is critical and esp when dealing with wind gusts that you know what the wind can be.

As for Refresh - only DJI will tell if covered under "flyaway" or not. Send them a request - can't hurt - and they'll want the data files to see whether they will or not. "Flyaway" has changed the way DJI looks at things as compared to when it had to be a DJI malfunction and not something by the pilot.

I use UAVForecast and have set my wind setting at 250 to give me an approx wind idea. If it's gusting in the 20-30 mph at that altitude, going higher will only increase.

 

[BossBob]

7000ft on the side of a mountain. I’d have thought that would be a pointer to higher than normal wind speed at ground level without even looking at AirData. I’ve no idea if AirData’s algorithm allows for that high a launch point.

 

[WaleyZhang]

Thank you everyone for all of the input. I also noticed in AirData that the battery life seems much shorter than it should’ve been. Is this because it’s working too hard to try to go against the headwind when it’s in RTH mode?

 

[Arnold LeVine]

As to battery life, it was probably working like crazy just to hold position against the headwind, much less flying against the headwind.
I certainly would have been in panic mode in the moment, but in the calm of reading this, would switching to sport mode and flying manually have been a way to overcome the wind? Or even better, would RTH use the higher speed threshold of Sport mode? Of course, in the heat of the moment, realizing the headwind situation would have to have come to mind. But if that would work, then this discussion would have been, as DougMc stated above, a VERY educational learning moment for all of us. Kinda like pilots training disaster recovery in flight simulators.
Thoughts/Opinions?

 

[HoozierDroneDaddy]

As to battery life, it was probably working like crazy just to hold position against the headwind, much less flying against the headwind.
I certainly would have been in panic mode in the moment, but in the calm of reading this, would switching to sport mode and flying manually have been a way to overcome the wind? Or even better, would RTH use the higher speed threshold of Sport mode? Of course, in the heat of the moment, realizing the headwind situation would have to have come to mind. But if that would work, then this discussion would have been, as DougMc stated above, a VERY educational learning moment for all of us. Kinda like pilots training disaster recovery in flight simulators.
Thoughts/Opinions?

Sport mode would give him more power and speed, so it would depend on what the actual wind speed was. However, going to sport mode may have resulted in the same outcome as it would have reduced his battery faster, so he may still not have been able to make it back. The actual speed at his altitude is the variable that gives the final answer. The M2P can go up to 45mph in sport mode. To gain any speed to make it back before the battery dies, he would need wind speed well below that to battle the headwind. IMHO.

 

[slup]

...based on the weather data from AirData it seems like the wind had a speed of 15.5mph, though if I recall correctly, the Mavic 2 Pro is rated for up to 22mph. Why was the headwind then more than it could fly against?

That 15,5mph wind was on ground level ... the wind your M2P encountered was well over the specified 22mph.

 

[WaleyZhang]

Thank you for the wind data! That would explain why it couldn’t fight against the headwind.

Given these conditions, would there have been any way for me to recover the drone, once it was in the air? What maneuvers would I have needed to perform? Or was drone doomed as soon as I decided to fly it?

 

[Meta4]

Given these conditions, would there have been any way for me to recover the drone, once it was in the air? What maneuvers would I have needed to perform? Or was drone doomed as soon as I decided to fly it?

The best way to have flown safely would have been with an awareness of the wind conditions.
There must have been some indications of the wind's speed and direction from where you flew?
Things like the feel of the wind on your face, the sound of it in your ears, the sight of it blowing tree tops?
Checking your screen would have shown that half-way up, the drone was having trouble holding position.
It was slowly being blown to the west and the distance from home was steadily increasing even though you'd only flown straight up.

At that point you could have tried to fly back against the wind and found out how bad it was, then lowered the drone to 100 ft or lower, where the wind was less and flown home.
At that short distance, you could have tried Sport Mode, but that wouldn't have helped later when you finally realised the problem.
The speed gain would have been marginal, but the battery burn rate would have been faster.

If you had proper wind awareness, you could have climbed straight up, watching the drone's speed and distance, and then tried flying upwind to see how fast your drone could go against the wind.
When you saw how the wind affected the drone, you could have descended, and either abandoned the flight or continued your flight upwind and had an easy flight home.

In general, higher up = higher wind speed.
If you are caught fighting a headwind, bring the drone down where wind speed is less.
Be aware of the wind and what it means for your return flight before it becomes a problem and fly accordingly.

 

[umanbean]

The best way to have flown safely would have been with an awareness of the wind conditions.
There must have been some indications of the wind's speed and direction from where you flew?
Things like the feel of the wind on your face, the sound of it in your ears, the sight of it blowing tree tops?
Checking your screen would have shown that half-way up, the drone was having trouble holding position.
It was slowly being blown to the west and the distance from home was steadily increasing even though you'd only flown straight up.

At that point you could have tried to fly back against the wind and found out how bad it was, then lowered the drone to 100 ft or lower, where the wind was less and flown home.
At that short distance, you could have tried Sport Mode, but that wouldn't have helped later when you finally realised the problem.
The speed gain would have been marginal, but the battery burn rate would have been faster.

If you had proper wind awareness, you could have climbed straight up, watching the drone's speed and distance, and then tried flying upwind to see how fast your drone could go against the wind.
When you saw how the wind affected the drone, you could have descended, and either abandoned the flight or continued your flight upwind and had an easy flight home.

In general, higher up = higher wind speed.
If you are caught fighting a headwind, bring the drone down where wind speed is less.
Be aware of the wind and what it means for your return flight before it becomes a problem and fly accordingly.

That's a lot better than what I had.

 

[Meta4]

based on the weather data from AirData it seems like the wind had a speed of 15.5mph,

The wind speed you found was just a forecast and had nothing to do with the actual wind your Mavic was fighting.
There are lots of reasons that forecast wind strength is different from actual winds at your flying location.

though if I recall correctly, the Mavic 2 Pro is rated for up to 22mph. Why was the headwind then more than it could fly against?

" Rated for up to 22 mph" is misleading.
It doesn't mean that a 22 mph wind won't affect it.
It means that 22 mph is the max wind speed that the drone is able to hold position against.

The speed shown on your screen is the speed the drone makes over the ground, calculated by GPS.
Any headwind will eat into your drone's speed.
A 22 mph headwind wind will take 22 mph off your groundspeed.
If you have obstacle avoidance enabled, your drone would barely make any headway against a 22 mph headwind.
Without OA, you could make approx 5 mph

 

[Dave Maine]

All wind speeds from the web are general forecasts. The actual wind may vary considerably from that forecast, and as mentioned can be very different at height.

 

[Kentucky Ranger]

The reason for the incident is very simple and should have been obvious to you.
In just one word it was ..... WIND
You don't seem to have had any idea of the wind and what that was doing to the drone.

Your drone was responding to your control inputs, but the headwind was more than it could fly against.
Here's a slightly longer version of what happened.

As you climbed above the launch point, you failed to notice that there was a significant wind and the drone was having difficulty holding position.
At almost 400 ft the drone was being blown at 3-8 mph by gusty winds.
There was also a strong wind warning at 50 seconds.

You also didn't pay attention to the wind direction and flew the drone off downwind.
You lost signal at 123.5 seconds with the drone 1470 ft from home.
Signal was re-established at 472 sec with the drone now 3755 feet away in RTH but and being blown backwards at up to 5 mph.
You let the drone be blown further away and started descending at 506.4 sec.
You also tried pushing the right stick forward and the drone started to make slow headway against the headwind at times and going backwards at others.

At 541.8 sec you initiated RTH again and left the drone to be blown further backwards as it tried to RTH against a headwind.
At 695.9 sec you cancelled RTH again and resumed control.
At full stick the drone was still blowing backwards slowly.
At 714.3 sec ith the battery at 12% you tried RTH again.
Shortly after at 726.3 sec, the drone commenced autolanding because the battery had reached critical low voltage.

The battery ran out at 972.7 sec.

The place to look for it is: 20.77686 -156.25629
It's possible that it landed or was very close to the ground when the power ran out so you may be in luck?

I haven’t had this happen to me, yet.
But I’m always watching out for weather conditions, and I’m pretty sure if this would’ve happened to me, I would’ve just landed where I could the moment I got the signal back, and tracked it down.
wind can overwhelm a drone easily, if the weather is looking bad, I won’t fly.
I’ve also read, and watched many videos about drone piloting, and have learned to fly upwind of home, so that if the drone hits some bad gusts, the wind will actually help blow it back home.

 

[Yvan]

Wind seems to be a problem but the battery is damaged too. There is a major deviation on cell 2 throughout the flight.