Recommendation: Turn Off Motors After Landing

[PhantomFanatic]

So I guess I'm not the only one this has happened to....

Warning: leaving the motors running after landing or trying to take off again landing (without shutting down / restarting) may cause a tip-over or crash.

I guess the real issue here, is that if you stall a motor the drop in back-EMF is likely to damage the ESC. Usually it is instantaneous, but there is no reason to assume that damage has not been done to an ESC which appears OK after an incident, it could then fail at a later time.

It has been suggested that the same is true if a motor stalls in flight (for example by making abrupt changes in direction), i.e. it could be possible to damage or destroy an ESC whilst in the air.

I too suffered a tip over on my first or second flight. Luckily for me the ESC was fried so I did not have the option of attempting another trip with it.

Note that stalling the motor and taking out the ESC is an inherent design issue of the hardware and is not considered to relate to the recent ESC/motor upgrades.

Are you saying that by a rotor turning backward, that would damage the ESC? If so, what makes it turn backward? I don't know how an electric motor "stalls." It is either on or off. Are you speaking of a high speed shut off?

I'm not challenging you, I'm just tying to understand what you meant. . If there is a such a problem, a rectifier, that is rated for the current, could be installed on the positive or negative, or both leads. This would block a reverse voltage coming back to the ESC.

If you mean a single spike, it would still work, but so would a capacitor (non-electrolytic). Also, other components and circuits could do the job. A Zener diode could be employed, etc. But, is there REALLY such a problem? If so, I could design a fix.

 

[Hughie]

I guess the real issue here, is that if you stall a motor the drop in back-EMF is likely to damage the ESC. Usually it is instantaneous, but there is no reason to assume that damage has not been done to an ESC which appears OK after an incident, it could then fail at a later time.

It has been suggested that the same is true if a motor stalls in flight (for example by making abrupt changes in direction), i.e. it could be possible to damage or destroy an ESC whilst in the air.

I too suffered a tip over on my first or second flight. Luckily for me the ESC was fried so I did not have the option of attempting another trip with it.

Note that stalling the motor and taking out the ESC is an inherent design issue of the hardware and is not considered to relate to the recent ESC/motor upgrades.

Are you saying that by a rotor turning backward, that would damage the ESC? If so, what makes it turn backward? I don't know how an electric motor "stalls." It is either on or off. Are you speaking of a high speed shut off?

I'm not challenging you, I'm just tying to understand what you meant. . If there is a such a problem, a rectifier, that is rated for the current, could be installed on the positive or negative, or both leads. This would block a reverse voltage coming back to the ESC.

If you mean a single spike, it would still work, but so would a capacitor (non-electrolytic). Also, other components and circuits could do the job. A Zener diode could be employed, etc. But, is there REALLY such a problem? If so, I could design a fix.

I am basing much of what I say on this report

"Causes of motor/ESC failures. Test results"
http://forum.dji.com/forum.php?mod=view ... es&lang=en

The guy also speculates that it may be possible to destroy an ESC in midair if you stall a prop, and there is a recent case of something reported on these forums which may fit this hypothesis where a Phantom pilot was doing 48mph in ATTI and then used GPS to arrest the speed to zero. This sort of thing could cause a prop to temporarily stop. Inconclusive, but immediately after doing this manoeuvre the quad dropped out of the sky with a suspected ESC failure.

The report author seems to suggest that whilst the ESC design is actually quite elegant, there is nothing which can be done in its current form to "design out" the flaw.

 

[Andrey320]

I've submitted a suggestion to DJI to look into either fixing this issue or adding a point to the manual. Will let y'all know if I get a reply.

 

[Andrey320]

Got a reply from DJI... not a very favorable one....

"Your request (84809) has been solved. Please submit a new request if you have any further questions.
Dear Andrey,

Thank you for your email.

We appreciate the feedback you have provided and will pass it on to the appropriate department to handle.

Thank you for choosing DJI,
"The Future of Possible!”

DJI Customer Service North America"

 

[justin00]

I asked the same question too... i.e "wtf does it try take off again"... Believe its due to the GPS / trying to maintain altitude + / - few feet.
Never tried...and almost 12am now... but if you change to Atti (or land in that mode) it would not try take off again yeah ?

hmm.. I shall try it out next time..

 

[N017RW]

I asked the same question too... i.e "wtf does it try take off again"... Believe its due to the GPS / trying to maintain altitude + / - few feet.
Never tried...and almost 12am now... but if you change to Atti (or land in that mode) it would not try take off again yeah ?

hmm.. I shall try it out next time..

Not GPS or mode related.

The a/c assumes it is still flying.

Thus the a/c is trying to maintain altitude based on a drifting barometer reading due to changing ambient conditions and overall sensor accuracy, repeatability, filtering, sampling algorithm, quality, etc.

 

[justin00]

fair point you raise.. forgot atti still holds altitude..

 

[ZeekArmy]

Wow...*

 

[Prophecies]

I've had this issue myself. I was just flying it around my living room to make sure my ESC's weren't acting up. I land, and let the motors idle a little, since I was planning on taking off again soon. I wasn't touching the sticks, and out of nowhere, the motors rev up, start spinning faster, and it almost takes off. Had I not been paying close attention, it would have tipped itself over and possibly could have damaged my props/ESC's.

This seems to be a common problem, and while it's not a big issue (just turn the motors off after landing), it's definitely something DJI needs to fix.

 

[morph000]

So... just to confirm the theory here:

Unless the bird has intentionally landed as a result of auto-land etc it doesn't know it has in fact landed, it thinks it is fact hovering above ground level and so perceives the changes in flight data like elevation, compass and barometer data as unwanted deviations from where it should be and so any attempts, either by itself in the case of excessive positional error data or when lift-off is user initiated, to move to where it thinks it should be which can result in a tip-over, due to the fact it is on the ground and therefore has landing strut/ground friction as well as low prop speed.

If so, this raises 2 questions to me:

1. Why wouldn't full ascend stick cause lift-off and avoid a tip over ? ie Does telemetry data take priority over stick input ?
Effectively, the bird when landed and idling is in the same state as pre first take off now isn't it !

2. Imagine a scenario where the bird is hovering for a while above ground in ideal conditions. ie very minimal flight data change.
Couldn't the same tip happen for the same reason ?

 

[lake_flyer]

I tried it several times intentionally, after reading this thread. In every case it started to rev up again and take off.
All it took to get it from tipping over was applying a little throttle and a little back stick (just as with take off), and it hovered again.


People should just try this behaviour a couple of times so they know how to react to it should it ever happen unintentionally.

 

[lake_flyer]

This should be in a sticky:

"Things you can't find in the manual but generally shouldn't do with any Phantom, unless you really want/need to for some reason".
or maybe:
"Things anyone should try and observe in a controlled way to avoid disaster in the future".

Many learned the hard way, simply because they never flew anything before and DJI told them even their late grand mother could fly the thing without any training.
This SHOULD be in the manual. It is an important behaviour that could cause serious injury.

If a newbie is just leaning over the quad after it's landed for the first time and it still idles, studying the rotors or whatever, even switching off the remote for some reason, and it suddenly takes off in their face.... (of course they had attached the carbon props that came in the package for free because they looked way cooler then the white ones).

 

[gingerbloke]

The real solution is to hand-catch, but there's no way DJI can sanction this (and commit it to print) because someone will get it wrong and stick their hand in the prop. This product appeals to (and sells to) people with no previous R/C experience, DJI have a tough job on their hands IMO.

Why is this important precaution not in the manual?
Why is this not in any of the "beginner guides"?
Why hasn't DJI fixed this?

You are right, it should mentioned clearly in the manual.
A technical/design fix is more complicated and therefore more expensive though.

 

[Casper2015]

What DJI do say is as follows:

7. After landing, leave the throttle stick down for 3 to 5 seconds to stop the motors. Return throttle stick to
middle position after the motors have stopped.

It is always difficult to think of everything that someone might do in such a situation, and leaving the engines idling wasn't envisaged as being something specifically they needed to warn against. I agree, DJI have a tough job in writing these manuals.

 

[lake_flyer]

I just don't believe DJI engineers didn't observe this themselves while testing. I'm sure they did a lot of tests including idling after landing.
Very simple to add "please don't let the phantom idle after landing because it will attempt to take off again."

Another thing is that after a battery low auto landing, the motors DON't stop. They keep spinning until you do a CSC or 2sec throttle down. I found that out when my FC40 went into auto landing some 100m away. I assumed that the engines would have stopped after landing but they kept running. By the time I got to the phantom I even could take off again and hover for a minute before auto landing kicked in again. I have not yet totally drained my P2 into auto landing yet so I'm not sure if this is phantom wide or just Naza-M V2 behaviour. If there was no fresh gale right now I would try it out immediately.

But to be sure, after a battery low auto landing, especially when it happens beyond direct LOS, always pull down the throttle stick until the motors stop. Never assume it's shut off. Never do a CSC to shut off when the quad is out of sight. It might start up again in the case it HAD shut down.

 

[IflyinWY]

So I guess I'm not the only one this has happened to....

Warning: leaving the motors running after landing or trying to take off again landing (without shutting down / restarting) may cause a tip-over or crash.

This is a great bit of information which I feel should become an "Important Sticky".

It's a shame the subject got "Off Topic" and then sat stagnant for almost a month.


Thanks for your post Andrey320.

 

[ukglyn]

I am totally new to this having just got my first phantom 2 days ago. Am I right in thinking then that if your phantom tips & the props come into contact with the ground you will/can damage the motor? I would of thought they would of had some sort of failsafe design built in to stop this happening.

Again sorry if I am being thick here, as this is all new to me. :shock:

 

[IflyinWY]

Hello ukglyn, welcome to the forum.

That wasn't the point of the thread, but apparently some do believe an esc and/or motor can be damaged in a tip over.

I've tipped over and outright crashed my FC40 between 15 and 20 times. I've tipped over in tall grass and flown directly into the top of a 30 foot utility pole and fallen to the gravel road below, and just about everything in between. I know of no issues with my esc's. I do believe if I continue to abruptly stop my motors that they will fail.

 

[Hughie]

That wasn't the point of the thread, but apparently some do believe an esc and/or motor can be damaged in a tip over.

I can assure you this is the case