Gimbal drift after 3.08?

[Fourblade]

I've just had my first flight after upgrading to 3.08 etc and all was well except that I immediately noticed that when it's in the air if you turn/yaw the Phantom, then when you let go of the stick the Phantom stops but the gimbal/camera can carry on drifting ever so slightly right or left for a few seconds and it can do this even after a few seconds of it being still without any yaw!
I'm sure it didn't do this before yesterdays upgrade and I've just done a bench test with the props off and even then the gimbal is definitely drifting on it's own without any obvious reason?
Anyone else notice this?

 

[Phantomdee]

Hi Yes i noticed this when i was testing on the table at home.....
I haven't tested in the air yet, h
ave you had any GPS problems?

Darren

 

[Nathan Carter]

I've noticed this on the old firmware! I think it is possibly related to small compass interference... For me it happened on very rare occasion...

 

[Pacific Barbarian]

I know I have a zenmuse, but motors off, on the ground, with left stick left and right as if yawing, my gimbal moves left and right, like I can control it, thought it would get rid of the left leg coming into picture, but it does not. Only noticed this with 3.08.

 

[Fourblade]

I'm certain this slight drift never happened prior to 3.08 I would have noticed for sure!
After a few more flights today and trying both camera pitch settings "upgrade" and "basic" as well as (just in case) doing an advanced IMU calibration, there still can be a tiny drift for no apparent reason.
It does it on the ground with props off and in the air.
The gimbal can creep left or right a very small amount from where it should be solid and still.

I have no explanation for why it is suddenly doing this... but it definitely is and the finger of suspicion is pointing at the new firmware I'm afraid.... :|

 

[RichWest]

Out testing on Friday, I believe I experienced the same thing. I thought it was the wind loading causing the ever so slight yaw correction on the camera. My unit's down for a week, so no good to you at the moment.

 

[aartsf]

Did you together with the update of the assistant sw to 3.08 also update the Zen fw (gimbal/camera) from 1.005 to 1.006 on the upgrade page of the assistant ??
If yes, that may have caused your problem, not the update of the assitant sw as such. See also other threats on this form.
Regards FA

 

[RJ Evans]

Same problem here. Left or Right yaw causes the gimbal to move in the direction of the yaw. My table testing confirms it. I was under the impression that the vertical axis was supposed to be stationary, i.e. no panning movement. Pan should come from the yaw of the bird, not the yaw of the gimbal. :?

 

[Fourblade]

Did you together with the update of the assistant sw to 3.08 also update the Zen fw (gimbal/camera) from 1.005 to 1.006 on the upgrade page of the assistant ??
If yes, that may have caused your problem, not the update of the assitant sw as such. See also other threats on this form.
Regards FA

That sounds like it must be a bug.. all I did was install the updates which the assistant software prompts you to do so. If upgrading to 1.006 has caused this issue then I'm guilty as charged. My crime was to install a flaky update. :roll:

 

[aartsf]

I am afraid you guys are the victum of dji's poor/bad communication and their mistake to ditribute an upgrade for the assistent software to gether with a fw upgrade for the batteries and a fw upgrade with respect to the gimbal/camera, while these 3 upgrade have nothing to do with each other. Actually it depends on the existing configuration of the individual client which of the 3 upgrades would be applicable or not. I am afraid this might again be an not well thought true action of dji !? Hopefully dji take their resonibility and help you guys fix the problems that in my view are caused by their bad communication.

 

[4wd]

Same problem here. Left or Right yaw causes the gimbal to move in the direction of the yaw. My table testing confirms it. I was under the impression that the vertical axis was supposed to be stationary, i.e. no panning movement. Pan should come from the yaw of the bird, not the yaw of the gimbal. :?

It has always done that, this is not something new - it is pre-empting the movement from your yaw input so that the panning motion is smoother.
There is a tendency for a rapid yaw then stop to result in the gimbal carrying on a little too far then jumping back to look straight ahead. I don't think that's new either but admit I hadn't particularly noticed it until 3.04.

 

[Fourblade]

Just to clarify - I'm noticing occasional very slight left/right gimbal movement with no stick input whatsoever - either when the Phantom is in the air or sat on the bench. I just never saw this prior to the last round of software and firmware updates.

 

[RJ Evans]

Same problem here. Left or Right yaw causes the gimbal to move in the direction of the yaw. My table testing confirms it. I was under the impression that the vertical axis was supposed to be stationary, i.e. no panning movement. Pan should come from the yaw of the bird, not the yaw of the gimbal. :?

It has always done that, this is not something new - it is pre-empting the movement from your yaw input so that the panning motion is smoother.
There is a tendency for a rapid yaw then stop to result in the gimbal carrying on a little too far then jumping back to look straight ahead. I don't think that's new either but admit I hadn't particularly noticed it until 3.04.

Thanks for the info. I appreciate it.

 

[RJ Evans]

Just to clarify - I'm noticing occasional very slight left/right gimbal movement with no stick input whatsoever - either when the Phantom is in the air or sat on the bench. I just never saw this prior to the last round of software and firmware updates.

I haven't noticed that particular behavior with my bird. But, I'll keep an eye on it. And, thanks for your clarification.

 

[fastsmiles]

I have noticed a lot of yaw instability in my videos since doing the upgrade, even in low wind conditions.

 

[Nathan Carter]

I believe the gimbal logic is not looking at actual stick input to stabilize the panning/yaw axis, but the info from compass. I only speculate that it is like so, because from engineering standpoint it would be much easier to implement it like that than to use direct stick input and then try to calculate from that. Compass directly tells you how much the craft has rotated. So I'm pretty confident they went with that.
That would explain gimbal movement without stick input. As compass interference happens. Should be easy enough to test with a magnet

 

[RJ Evans]

I believe the gimbal logic is not looking at actual stick input to stabilize the panning/yaw axis, but the info from compass. I only speculate that it is like so, because from engineering standpoint it would be much easier to implement it like that than to use direct stick input and then try to calculate from that. Compass directly tells you how much the craft has rotated. So I'm pretty confident they went with that.
That would explain gimbal movement without stick input. As compass interference happens. Should be easy enough to test with a magnet

Respectfully, I'm not sold on that idea at this point. During my table test I didn't see any movement at all without stick input. I watched it closely, and provided ample time between stick inputs (about 2 minutes). Granted, the bird was stationary on a table top, but it never moved without yaw stick. I haven't had an opportunity to flight test yet. Hopefully I'll have time tomorrow (Thursday) to do so.