Turn props off in mid-air for fast descent (and on again :-)

[LazyBird]

Theoretically you could restart the drone in free fall provided it is in take-off posture I.e props up at the moment you engage. I know for sure, running props turn off if the aircraft is tilted to either side. But you can always give it a try and let us know what happened.

 

[FoxhallGH]

The firmware limits the vertical speed independently of the horizontal speed.

Ahhh - makes sense ... thanks. But I guess in the case of a 'bird-escape' it might make the Mavic less of a target!

 

[sar104]

Theoretically you could restart the drone in free fall provided it is in take-off posture I.e props up at the moment you engage. I know for sure, running props turn off if the aircraft is tilted to either side. But you can always give it a try and let us know what happened.

But we see Mavics tumble after losing a prop without the motors shutting down.

 

[JAW]

Would have been nice if there was video from the falling Mavic.

 

[Rayray]

Hello,
a rather theoretical question:
The Mavic 2 props can be turned off by pulling both sticks down/inside. With the same function they can be turned on again.
Can that be used in mid-air for a fast descent?

When flying up to 500m, the descent takes very long because the descent rate is rather slow (~2-3 m/s). So what would happen if I just turn off the props at 500m above ground, and turn them on ~200m above ground? Would the Mavic 2 stabilize again, or crash?
Anyoney ever tried?

best regards
mariachi76

If you are 107 certified, the only circumstance where this would be legal is in an emergency situation. And 500 meters is also generally not legal.

 

[ff22]

But we see Mavics tumble after losing a prop without the motors shutting down.

Unless there was a change in firmware, I watched my Zoom mow the lawn after it hit a small branch and tumbled and inverted and hit the ground. In my panic I don't know if it turned itself off or I pushed the sticks to stop it.

 

[jc44224]

Dont try this!!! If motors are shot down in mid-air, the drone will just fall like a stone, and it will rotate on all axis due to air forces against the drone frame...
So if you turn motors on again, and the drone is not in flight position, it will just flip and flip again until it hits the ground...(already tried with a small toy-drone, and it crashed on each trial...)

Even if the drone fall flat (keeping its flight position), I think it will get lot of vertical speed very quickly, and the motors and props will not get enough power to stop the fall...

Bad idea!!

BTW it's true that descent speed is very low with mavics, but I guess it's normal, because when you pull throttle stick fully down, you can ear motors and esc working hard to keep the drone as flat as possible...
Motors speed is very low at this time, and it can't stabilize the drone as well as it does when you're climbing or in normal flight...

To view this content we will need your consent to set third party cookies.
For more detailed information, see our cookies page.

Accept third party cookies

 

[Inspirephil]

Had similar situation with my I2 and it crashed doing about 50-60mph and ended up a pile at my feet.
IMU went crazy and shut the motors off

 

[hiflyer201]

Flying at 500 m??? that would show a definite lack of ...uh...sense...in the first place

 

[ff22]

To view this content we will need your consent to set third party cookies.
For more detailed information, see our cookies page.

Accept third party cookies

Well, I realize he noted it was a safe location without many people. What kind of ghost town was it. Too many houses to try this maneuver if it failed. Sorry - INSANE in that location unless it was Chernoble

 

[Daviska]

(I'm not sure if its possible with a mavic 2 as i tried at low level and nothing happened) I could imagine that if the motors restarted, even if it was in an upright position it could easily induce vortex ring state as it would be descending too fast to begin with, and so would crash anyway, the slow descent is designed to prevent this in the first place. In most places it is illegal to fly above 400ft and a slow rate of descent should be taken into account when planning emergency drills. In the event of an air incursion it is sometimes quicker to send the drone up instead of down in order to avoid a collision.

 

[Phill B12]

Out of an abundance of caution I would not risk my drone for the sake of saving a few seconds of descent time. Others with big wallets that like to live on the wild side will likely attempt this if only for the bragging rights no matter the outcome. Which raises the question, "How high will it bounce"?

 

[3man]

Well I have done it many times to speed up m descent with Mavic Pro and NLD. Even at max settings the drone would not descent faster than 10m per second. I had an acccident once when I didn't make it home. Battery went dead at around 270m. I usually fly in the fields outside of any towns but I was close to a lake so when I realised I am not gonna make it, I just checked the wind (direction) and maneuvered my drone to such a position that when free falling the wind would not push it into the lake. At around 270m the gimball went limp and connection was lost. I was just looking at it falling down and what hit me it was not tumbling, it was falling like a flat frog with limbs spread. I guess centre of gravity must be low. I did it once with my phantom 2 and it was tumbling down like crazy. Anyway since I had no control, all I could do was to watch the bird go down. It hit the ground out of my sight behind some bushes. I was sure there would be not much left to recover. When I finally found it, it was 100% intact. Not a slightest damage. Freshly ploughed field saved my drone. I just had to dedust it. I was amazed and happy and I just couldn't believe my luck. Since that time I started to use mid flight shut down to descent very often. I just set my restart of motors to 0.1 second or something like that (the lowest possible in NLD). It happened maybe only twice in 10-15 descents that the drone flipped. Remember that propellers create drag so the drone behaves sort of like an awkward dart. The only think I have notieced is that original propellers are best for the job. These are much more stiff than the quiet ones. When restarting the drone with quiet props on they usually bend and hit the chassis and tips get destroyed. Original props don't have that problem. However the case with MP2 may be different since this drone is not as flat as MP so the center of gravity is higher. For now I am using my drone for photos and filming and MP is left for "crazy" stuff. Anyway back to free falling. Props and the body creates drag so the highest speed reached falling down was about 19-22 m/s which is around twice as fast when going down with NLD mod set to max descent which is 10m/s. And also there is always a little adrenaline when doing the stunt. However I have never had any problems with MP and free falling. It always recovered instantly when motors were turned on. The only draw back is that the height meter is reset to zero on restart.

 

[G Lee]

This may have already been said , but to the OP i find that a cork screw or long zig zag descent with nose down works well. That way there is no prop wash in the way to slow you down. Just 2cents

 

[Victorian Aerial Vision]

Hmm... just because it CAN be done, doesn't mean it SHOULD be done.

If you keep VLOS you tend to see a potential hazard before something happens. I recently flew in an area where some eagles took an interest in my Mavic, I did the sensible thing & landed to save my Mavic & possibly the eagles from harm. To try and turn off the motors for a quick descent would have resulted in disaster (according to the data) as I was only 50 metres or so high.

Yes, I do realise the original theoretical question was talking about 500 metres altitude, personally I try not to fly that high is it is against our local regulations.

Victorian Aerial Vision Videos

 

[Dingo]

watch it being done here

To view this content we will need your consent to set third party cookies.
For more detailed information, see our cookies page.

Accept third party cookies

 

[sar104]

This thread has a bizarre alternative reality flavor to it. Even though I posted the data from a flight log of exactly this maneuver early on in the thread, several posters described having successfully tried it, and there are links to videos of the maneuver, we still have multiple posters continuing to speculate that it's not possible.

 

[Point Zero]

This thread has a bizarre alternative reality flavor to it. Even though I posted the data from a flight log of exactly this maneuver early on in the thread, several posters described having successfully tried it, and there are links to videos of the maneuver, we still have multiple posters continuing to speculate that it's not possible.

There are lots of us who do read and appreciate your factually accurate posts and explanations.

It’s great to know what the data supports. I wonder if DJI did all of these kinds of tests and know that these drones can easily be restarted mid free fall and landed safely.

 

[sar104]

There are lots of us who do read and appreciate your factually accurate posts and explanations.

It’s great to know what the data supports. I wonder if DJI did all of these kinds of tests and know that these drones can easily be restarted mid free fall and landed safely.

That's a good question. I suspect that the guys at DJI have tested all kinds of maneuvers and scenarios that never get reported, but I don't know that to be the case. It always seems to be the case that there is an adventurous base of users who explore all kinds of things that a manufacturer never intended, and possibly never considered.

 

[netfolks2000]

On one of the other forums about this matter I did an experiment of holding the drone sideways and upside down and it would not start. The speculation was that the drone is not designed to upright itself perfectly in a freefall but if it did get back to right side up then the motors will stop, perhaps while even in a tumble at the moment it is upright long enough.