Has anybody ever used CSC intentionally (power-OFF, then power-ON) for rapid descent from extremely high altitudes (1000ft or more)?


I'm sure it's crossed a lot of minds of pilots here in the DJI forum - I can't be the only one!
There are plenty of videos of our copters in/above the clouds. Several with failed landings!
Usually, these birds end up too far down range (fighting extreme winds at altitude), or they run out of battery during a controlled descent.
A "Get Down Quick" procedure may have safely saved these drones.

Don't get me wrong - I'm not looking for something new/crazy to try with my brand new $1500 bird.
It would just be nice to know that if we get into a situation like that, that there's an option for us.

(BTW, this procedure works on the DJI Go simulator - if you restart above 400ft! ... just sayin' !!)

Well, Germans are known for doing destructive things on things.. perhaps there's a German engineer here, with a spare phantom, to test it...  Too bad mythbusters are done.. I am sure they could test it ;)
Perhaps DJI in a vertical wind tunnel could test it?
Actually. seriously, some of those vertical wind tunnels people use to practice falling... if anyone has access to it.. that could be a somewhat safe test.. Set the airflow to a speed that will hover a phantom in off state, then power on the motors.. oh and crash against the ceiling... opps.... damn!

Maybe DJI could create an 'Autorotate' feature for safe, no-power (or reduced-power) descents (like real choppers do).

It has been done, but I don't know how reliable it is, I suspect it may be possible for it to get into a position where it can't recover.

When you do the CSC, left stick down and right first, then wait a few seconds to get the aircraft spinning, then cut the motors.   The gyroscopic forces from the spinning will keep the craft stable and upright, then the props will work as a parachute and keep the vertical speed reasonable as well as keeping it upright.

On the simulator you can cut the motors at 400ft and if your quick still start them back up in time, but I wouldn't try that in real flight as it may rather longer than in the simulation to stabilise!

And make sure you don't exceed any height restrictions that may exist in your location.

Mike_97 Posted at 2017-1-30 09:47
Maybe DJI could create an 'Autorotate' feature for safe, no-power (or reduced-power) descents (like real choppers do).

that is impossible on a quad for a variety of reasons.

AlecW Posted at 2017-1-30 09:16
Give it a go and let us know what happens ;)

Yeah, give it a shot and let us know how that works for you!

If your drone is falling out of the sky at 1000 ft, it is more than likely that motors have stopped, or at least 1 has stopped, so the likelihood of your props doing more damage than the actual aircraft dropping on someone's is minute , so csc is more likely to be used at very low altitude where there is more of a chance of an out of control aircraft doing damage to people or animals.

And that is the way we should think about it.  Good luck..

Nigel_ Posted at 2017-1-30 10:05
It has been done, but I don't know how reliable it is, I suspect it may be possible for it to get into a position where it can't recover.

When you do the CSC, left stick down and right first, then wait a few seconds to get the aircraft spinning, then cut the motors.   The gyroscopic forces from the spinning will keep the craft stable and upright, then the props will work as a parachute and keep the vertical speed reasonable as well as keeping it upright.

Not a bad idea, 'get it spinning so the gyroscopic forces keep it upright'. Thanks.

But just like a jet fighter ejector seat ... I'm not gonna try it 'just for fun' !

Mike_97 Posted at 2017-1-30 10:42
Not a bad idea, 'get it spinning so the gyroscopic forces keep it upright'. Thanks.

But just like a jet fighter ejector seat ... I'm not gonna try it 'just for fun' !

The phantom tends to tumble when free falling , you have to be at a very high altitude and have a lot of luck ,  I heard of someone doing this and most attempts were unsuccessful !

taking this a bit further, as I've been thinking not so much about rapidly losing altitude but rather recovering a malfunctioning UAV, why wouldn't a small parachute be a preferred option?

It could be spring or pyrotechnic loaded to (hopefully) clear a spinning UAV.  I know I've lost a prop or two (P2 and FW days-- nothing since the P3) in flight and a parachute would have kept my shell intact.  

Just something I was thinking on today.

NO way!!!! Is that a clone of it?!!!

http://www.gearbest.com/rc-quadc ... FBFvoGQgaAufY8P8HAQ

Autorotation would be possible given enough speed for the props to spin up in the opposite direction - but the rpm would have to be really high in order to slow down the descent, and certainly would not be enough to make a safe landing. Now, starting the motors will have to overcome that force, and spin the props in the proper direction again, which would make the aircraft tumble, most likely....
Deploying parachute would depend on the position and actions of the aircraft.. in a spin may not work well,, even if deployed from the landing gear......
Now, if someone finally wend mainstream with a multirotor that is a collective pitch, that would solve many such issues, but would make it mechanically more complex....
There could be redundant motors, perhaps only 2 of them, autorotation, 3D aerobatics.. it's all doable with today's technology.. so, who's gonna mod their phantom to be a CP device?


Come on DJI! Hint! Hint!

kndll Posted at 2017-1-30 11:27
The phantom tends to tumble when free falling , you have to be at a very high altitude and have a lot of luck ,  I heard of someone doing this and most attempts were unsuccessful !

"most attempts were unsuccessful"
How many attempts did they have, given that all the unsuccessful attempts would have destroyed the aircraft?

RC drones generaly don't have enough rotor blade area, to autorotate the aircraft down safely.


RedHotPoker

Autorotation here at about 7:30

The Curtis Youngblood stingray uses a modified heli tail rotor assembly. Unlike most common RC drones.
One motor and belt drives, controls all four of its props. Plus it's a Collective pitch system, unlike our fixed blade phantom.
I belong to Our local RC heli association, www.ercha.ca where it has been demonstrated. ;-)

Honestly, much prefer my Heli-Max Voltage 500 3D RX-R
As seen flown here by Matt Botos... Ha

RedHotPoker

AlecW Posted at 2017-1-30 14:01
1 unsuccessful attempt per drone is my guess ;)

I remember reading he made about a dozen attempts start at higher altitudes and coming down ,  I don't remember how much damage or how many drones he had .

"Usually, these birds end up too far down range (fighting extreme winds at altitude), or they run out of battery during a controlled descent.
A "Get Down Quick" procedure may have safely saved these drones."

If flyers put their drones up into extreme winds they can't fight or climb further than their battery can bring them back, that's just natural selection at work.
Flyers that would do that are going to come unstuck anyway.
Protecting them from their stupidity wouldn't achieve much.

kndll Posted at 2017-1-30 14:30
I remember reading he made about a dozen attempts start at higher altitudes and coming down ,  I don't remember how much damage or how many drones he had .

Maybe he was doing it wrong?

Just starting the motors is no good since it then behaves as though it is idling on the ground, and since the props stop spinning backwards the aircraft is no longer being stabilised so you need to go full throttle for a take off ASAP and then back off to let it sort itself out.

Tyrohne Posted at 2017-1-30 13:24
taking this a bit further, as I've been thinking not so much about rapidly losing altitude but rather recovering a malfunctioning UAV, why wouldn't a small parachute be a preferred option?

It could be spring or pyrotechnic loaded to (hopefully) clear a spinning UAV.  I know I've lost a prop or two (P2 and FW days-- nothing since the P3) in flight and a parachute would have kept my shell intact.  

I also think parachutes maybe the answer for falling aircraft,  while dji have created redundancy for IMU and compass, I don't think it's possible for any redundancy in a quadcoptor if one motor fails or one prop comes off, your quad is going to the ground fast.

This is the same for inspire 2, and although it has extra redundancy now with 2x batteries a simple thing like loosing a prop or motor and it will fall straight to the ground.

Unlike a hexocopter octocopter which can loose motors or props and can still be landed safely.

I guess this answers ALL of the questions & conjectures...


Thanks @HeyDrones on YouTube

From what I can gather from the video:
They hover and initiate intentional CSC at 300m
after descending ~10m, the P4P starts freefalling/tumbling at ~290m.
At ~250m, the P4P doesn't auto-rotate, but instead rights itself, probably from drag induced by the propellers.

The pilot allows the drone to continue freefalling (now in an upright position, no longer tumbling).
At ~200m, he initiates an outer-stick CSC to re-start the motors.
After falling another 50m, at ~150m the motors start working and by ~135m the drone is hovering.

So, YES, it's possible to 'get down quick' by stopping & re-starting the motors.
Who knew!

BTW, the video was posted a month before I started this topic - it just didn't pop up on my radar until today.

Interesting, the CSC starts it spinning clockwise, but without power it wants to spin anti-clockwise so it slows it's spin, then tumbles for a bit, then spins up anti-clockwise and is then stable.

Should have done the CSC with left stick down and left so that it was spinning the correct way to start with and it wouldn't have needed to go through the tumble, but does left stick down and left actually do a CSC?  I think not...

Can anyone read the vertical speed while tumbling and after it stabilises in the anti-clockwise spin?  Seems to be falling faster in the spin than I though they did, maybe it didn't have time to slow down after the tumble?


What was broken about the P4?  Looked like it worked pretty well!

Nigel_ Posted at 2017-3-16 06:39
Interesting, the CSC starts it spinning clockwise, but without power it wants to spin anti-clockwise so it slows it's spin, then tumbles for a bit, then spins up anti-clockwise and is then stable.

Should have done the CSC with left stick down and left so that it was spinning the correct way to start with and it wouldn't have needed to go through the tumble, but does left stick down and left actually do a CSC?  I think not...

[What was broken about the P4?  Looked like it worked pretty well!]

I read down the comments section of the video...
(VPS?) "sensors under the drone"

Labroides Posted at 2017-1-30 14:54
"Usually, these birds end up too far down range (fighting extreme winds at altitude), or they run out of battery during a controlled descent.
A "Get Down Quick" procedure may have safely saved these drones."

I love this post. Laughed at the last sentence.  

Not that I would every fly my drone so far away or high enough to try a CSC.