Possible hover problem (throttle related)

[rockhoundrob]

I just got the Phantom 2 and did the calibrations. The process went as expected...

Just to be on the "safe" side, I wanted to test the controls and reactions of the Phantom inside the house.
We did the compass calibration and satellite links. I assume the satellite links did not work because the green LEDs on the phantom were slowly blinking (but that's OK and was expected).

I had my son lay down on the floor and hold the phantom above him. I started up the propellers and moved the controls. The phantom moved as expected. I even tested RTH by turning off the controller and as expected, the props speeded up because it needed to get to 60 feet (also after a few seconds it tried to change course and apparently it thought "home" was SE). I turned on the controller and got the control of the phantom back again. I am not too worried about the RTH just yet, I will test that part later.

Now the throttle was not doing what I expected. When it was in GPS mode, I would increase throttle and then let go (stick pops back to middle), but it would keep going at high speed. And I had to manually slow it down.I thought what was supposed to happen is I can increase the the throttle until I get the right height, then let go and it should stay there and hover.

On the other hand, when I switch to Atti mode, I would increase the throttle and when I let go, the props would slow down to idle speed (same as when it is sitting on the ground waiting for take off).
I know if I did this outside, the phantom would fall out of the sky. In Atti mode am I supposed to keep the throttle up so stay at that altitude?

What about GPS mode? what should I expect when moving the throttle and letting go?

 

[landmannnn]

You are joking?

If not...

Have you seen photos of damaged humans where a prop has hit their face?
Where does it say on any manual or forum that this is a way to test the phantom?

 

[jumanoc]

OOpppsss, that was not the right way for testing... :shock:

At least you could test everything WITHOUT PROPELLERS ON....

Looks like a RC bad calibration, show your Naza Values after calibrating.

 

[Marlin009]

Wow! I'd say you got lucky. Take it outside. No way to tell anything by holding it.

BTW - Slow flashing greens is normal.

 

[N017RW]

There is nothing mentioned indicating there is a compass or GPS problem.

While it may not be intuitive at first, the quad appears to act normally from your description.
It does not know that it is being held (down) and that you're just doing a "test".

You 'requested' the a/c to take-off or climb by advancing the throttle but it was not able to as detected by it's barometric altimeter. So it added or maintained RPM (thrust) to attempt to carry-out your 'request'.

This issue also manifests itself in another common mistake...
...after landing, turn Off the motors immediately by holding the throttle stick 100% down for 3-5 seconds once on the ground or landing surface. If you land and let it sit or 'idle' the a/c still thinks it's flying (it has no sensors to indicate it is sitting on the ground) so it will slowly increase the motor RPM's attempting to maintain altitude based on the barometric pressure. The barometer in the Phantom (Naza/FC) is of low quality and likely lacks sophisticated filtering or sampling algorithms. As a result the baro sensor's output drifts and when it detects a rising pressure that will be interpreted by the FC as falling and try to compensate with increased motor RPM.
This commonly results in a "tip over" damaging props and in some cases ESC's.

 

[rockhoundrob]

Guys,

Thanks for the concerns about safety. That is why I said "safe" when it really is not. This will not be a habit!

To N107RW, thank you for your answer, this is this kind of answer I was looking for.
Based on your information, I will check the RC calibration again just to be sure.
I also think I will set the height max to 40 ft and distance to 100ft (then tie a strong string). Once I feel like everything is performing as I expect (or I understand how things work), then I will
change the settings to 200 ft height and maybe 600ft distance limits.

BTW, how can I find someone on this forum that lives in Oklahoma City? I think once I get the basics, I would like someone to walk through a few things that I might have overlooked.

Thanks again!

 

[N017RW]

There's a "Regional Meet-Ups" sub-forum you can start with.

 

[Marlin009]

I will set the height max to 40 ft and distance to 100ft (then tie a strong string).

I think tying a string to it is adding an unnecessary, and potentially unpredictable, variable to it. What will it do when / if it hits the end of the string?

They are amazingly easy to fly in GPS mode. Take your time, mind your orientation and you'll be fine.

 

[rockhoundrob]

Still "sitting at the edge of the pool"....

The only purpose of the string is if it tries to fly away....
It will be about 60 feet long and I will just be doing tests (20 ft movements) to make sure the calibrations are ok... up, down... left right, etc...

 

[jumanoc]

Still "sitting at the edge of the pool"....

The only purpose of the string is if it tries to fly away....
It will be about 60 feet long and I will just be doing tests (20 ft movements) to make sure the calibrations are ok... up, down... left right, etc...

You remembered me a "wise man" who used a thin string for avoiding phantom fly away.
Check this. It is the nearest to your idea ? :lol:

[vimeo]64720911[/vimeo]

 

[Meta4]

Not a possible hover problem.
A definite operator problem (over cautious related)

You're over thinking all of this and being over cautious.
Simply follow the instructions rather than trying to be even more safe and introducing factors that affect things in ways you can't imagine. Like your potentially quite dangerous indoor test .. as N017RW explained the Phantom was acting perfectly but your faulty understanding made you think it indicated a problem.

DJI have sold millions of Phantoms.
They are designed by very clever engineers and have been tested by a huge number of fliers before you to iron out any bugs.
They work.
Forget trying to be more safe than the manual.
Forget string
Forget your idea that the Phantom is going to fly away.
Just go and follow the simple instructions.

 

[Marlin009]

What he said.

Jump in the **** pool already.

 

[SteveMann]

+1

 

[Hughie]

What he said.

What he said he said.

 

[rockhoundrob]

You guys are right... when I work on cars, I double check the bolts I tighten. When I am building something and get ready to make a cut, I measure twice!

I am going to go back and redo the calibration and see if I can get the S1 switch to be GPS for position 1, Atti for position 2, and Failsafe for position 3.
If i get these settings, then the first flight will be this weekend.... :?

Once I seem to get the basic controls down, then I will take a look at NAZA-M.

Part of the reason I am over cautious is I believe about 80-90% of crashes and fly-aways are human error. Even if a flyaway is because of signal interference, it still may be that the flier was somewhere he should not have been. I don't want to be a part of that statistic.

Thanks again everyone!

 

[Meta4]

I am going to go back and redo the calibration and see if I can get the S1 switch to be GPS for position 1, Atti for position 2, and Failsafe for position 3.

Once I seem to get the basic controls down, then I will take a look at NAZA-M.

The only way you can configure the S1 that way is under NAZA.
In Phantom mode the S1 and S2 are only used for initiating calibration, resetting homepoint and recovering control from RTH.
Just keep it simple. Use the default Phantom mode to start and you'll have less to worry about and less to be confused by.

Part of the reason I am over cautious is I believe about 80-90% of crashes and fly-aways are human error. Even if a flyaway is because of signal interference, it still may be that the flier was somewhere he should not have been. I don't want to be a part of that statistic.

SIgnal interference may be blamed but is unlikely to cause flyaway. If the Phantom loses signal because interference swamps the control signal, its normal response is to RTH - not fly somewhere else. In any case it is extremely rare.

Just go to a large open area away from trees or buildings, ensure that home point is recorded and have some fun.
It's easier than you imagine.