Gimbal Tilted Horizon DJI Repair

[mercdave]

Hey guys!

So i've been scouring the DJI forums, and multiple Phantompilot threads in regards to a tilted gimbal fix, and yes the search function was my friend.
I've done numerous calibrations of both the IMU, gimbal, firmware refresh. At times when i fly the P4 i think the issue is fixed, but that quickly changes within the same flight.

Anyways, long story short. I've contacted DJI Support and they are putting in an RMA request for me. I was just hoping to hear some experiences that some of you may have had regarding the same issue! Has anyone sent in their P4 for the same problem? What was the outcome? Repair/Replacement.. etc. Hopefully a bunch of Happy customers?

Any feedback is appreciated.
Cheers!

 

[mercdave]

To add, when I first purchased it the issue was less predominate, but I started to notice the horizon being off more and more. I had to start correcting in the air with gimbal roll adjustments. I'll upload a couple of successful shots vs one that I took this morning after a fresh firmware reflash and calibration.

Sent from my iPhone using PhantomPilots

 

[Loomster]

Hey guys!

So i've been scouring the DJI forums, and multiple Phantompilot threads in regards to a tilted gimbal fix, and yes the search function was my friend.
I've done numerous calibrations of both the IMU, gimbal, firmware refresh. At times when i fly the P4 i think the issue is fixed, but that quickly changes within the same flight.

Anyways, long story short. I've contacted DJI Support and they are putting in an RMA request for me. I was just hoping to hear some experiences that some of you may have had regarding the same issue! Has anyone sent in their P4 for the same problem? What was the outcome? Repair/Replacement.. etc. Hopefully a bunch of Happy customers?

Any feedback is appreciated.
Cheers!

There is a fix that was posted a week or so ago.

 

[fastfed]

a fix? my only fix on my P3P was RMA

 

[SteveN76]

I have that issue on my p3a. The odd thing as I rotate the bird the tilted horizon corrects itself or it tilts the other way. Not sure if that's normal but if I have a shot I want and it's tilted I just correct the roll with the adjustment. I'm curious if yours is the same way.


Sent from my iPhone using PhantomPilots

 

[WilsonFlyer]

There is a fix that was posted a week or so ago.

There is no user fix. This has been proven time and again here on the forum and others too.

 

[Loomster]

↑
I corrected my tilted horizon by doing the following: With your DJI Go app on, Turn on your controller then your Phantom. Click and hold the bottom right C2 button on the bottom of your controller and turn the camera setting dial wheel on the front right of your controller and it will adjust your gimbal horizon to the left or right. Done!!!



Wow this actually worked like a charm. Thanks for the advice I feel so relieved

#19LMT0286, Sep 7, 2016

 

[Loomster]

I copied and pasted this from another thread....maybe I misread it but it seemed to work.

 

[mercdave]

I've tried different suggestions & re-calibrated everything over and over. I don't believe there is a permanent user fix. Well that that truly lasts.

Sent it in to DJI on Friday, Ill post updates & results as they come!

 

[Peter Galbavy]

I have observed that my aircraft hovered, in minimal to no wind, as an angle. I rotated it 90 degrees all the way around the the tilt remained aligned to the ground not the aircraft. The gimbal, in this instance remained (visually) level with what I considered level and not the ground. I have also noticed after some normal flights the gimbal becomes tilted but I then didn't yet have time to repeat the "hover and look" tests. I have photos which I posted on the DJI official forums so will not bore you here.

The aircraft tilt may have been cause by a very light wind but I am pretty sure there wasn't one - it went away after an IMU calibration.

My "fix" tends to be to recalibrate the gimbal when back to a level surface. It then works again for a while.

 

[mikej]

I used Loomster's method this weekend. Placed my p3p on my level desk and books (laying flat) as background for the camera; adjusted gimbal by press and hold the C2 button while turning the camera thumb dial slowly several times until the dji GO app grid and book lines matched. I then went for a test flight 80 ft +/- up with 360 pan and left/right, up/down movement of craft. The photos and video were perfectly aligned with the horizon. Process of adjusting took about 2 minutes.

Sent from my SM-N920R7 using PhantomPilots mobile app

 

[dennis6112]

I used Loomster's method this weekend. Placed my p3p on my level desk and books (laying flat) as background for the camera; adjusted gimbal by press and hold the C2 button while turning the camera thumb dial slowly several times until the dji GO app grid and book lines matched. I then went for a test flight 80 ft +/- up with 360 pan and left/right, up/down movement of craft. The photos and video were perfectly aligned with the horizon. Process of adjusting took about 2 minutes.

Thanks..........this worked for me on the P4 and I found out you can make the adjustment in flight. There are time when I fly high the wind tends to tip the horizon and this is a quick fix.

 

[Fdnyfish]

I just got my P4 back from DJI repair for the same problem. My horizon is now fine. It took 25 calendar days from the day I sent it until the day I received it back

 

[mercdave]

I just got my P4 back from DJI repair for the same problem. My horizon is now fine. It took 25 calendar days from the day I sent it until the day I received it back

Thanks for the info! What did the repair quote say?


Sent from my iPhone using PhantomPilots

 

[Squeamish]

There are time when I fly high the wind tends to tip the horizon and this is a quick fix.

But the entire point of a gimbal is so that externalities like wind causing the aircraft to pitch aren't transmitted to the camera, so that's not really a fix, is it?

 

[Dinerve]

So I was randomly posting in the Mavic release discussion thread as tilted horizons were spotted in initial footage.

Pardon the copy/paste but I think it would be more appropriate here and cba re-writing the whole thing so here goes:

Well most 3D gimbals over light cameras are.

I think it's a non-issue as I see how it works on mine, I will see tilted horizons under strong crosswinds and always in the same tilt angle as the bird.

My theory is that the horizon is mostly corrected from accelerometers values and gravitational pull and the forces resulting from the bird compensating constantly for the wind to stay in place at the birds's scale affect its perception of the vertical by definition gravitational pull, thus tilting the horizon.

Now the Mavic is lighter and the camera being aligned/at the same level with the body and its centre of gravity would mean that the above is even more accentuated in the Mavic's case.

Don't know if any f this makes sense...


Sent from my iPad using PhantomPilots

Oh and not that anyone cares but had to look it up a bit further online as I was trying to illustrate the tilted horizon thing and then we can go back to topic aka the mavic but best way I could find was this video:

The water in the glass is our horizon, if you tilt the glass slowly water will stay leveled, but if you start spinning (similar to the wind pushing and the bird pulling) then this force takes over and the water level (in this case our horizon) starts getting perpendicular to these forces thus tilting from the actual horizon...

Nuff said xD...

 

[Squeamish]

if you start spinning (similar to the wind pushing and the bird pulling)

Those aren't similar at all; wind pushing won't cause any of the same effects of spinning unless it actually causes the bird to spin.

 

[Dinerve]

Those aren't similar at all; wind pushing won't cause any of the same effects of spinning unless it actually causes the bird to spin.

Perhaps they aren't, and it's not spinning I'm necessarily thinking about but more of the swing as if you think about forces applied it makes sense. These things are computed over physics not image processing, would be too slow.

So similarly, fixed point would be the cross section of the 4 arms, Sideway/lateral force the wind, negated by the bird by tilting slightly to maintain position countering the wind=body tilt the gimbal has to compensate for and does most of the time, but sometimes there's so much turbulence up there that I think it is exactly when it happens as can be seen in Attitude indicator in the app (the round thing that shows pitch and bank) when hovering or flying in the same direction.

You would not notice pitch as much as tilt obviously and that's why when flying in winds and changing direction it seem to come and go when changing directions. If manually adjusted when hovering or going in the same direction then it's rock solid.

Never had tilted horizons when the bird is levelled or compensating for its own movement with no or light wind.


Sent from my iPad using PhantomPilots

 

[Erised]

My gimbal is good most of the time then goe's off. The software should adjust gimbal horizontal from the data in the IMU. The IMU should know the exact angle of the craft and make the calculations to compensate and maintain gimbal flat position. Why is it good most of the time then it screws up, this compensation is software driven and I have better results since the last AC update. If I was 100% sure sending it in to dji would fix it I would but I have also read that some sent it in and the new craft was the same.

 

[Dinerve]

My gimbal is good most of the time then goe's off. The software should adjust gimbal horizontal from the data in the IMU. The IMU should know the exact angle of the craft and make the calculations to compensate and maintain gimbal flat position. Why is it good most of the time then it screws up, this compensation is software driven and I have better results since the last AC update. If I was 100% sure sending it in to dji would fix it I would but I have also read that some sent it in and the new craft was the same.

Yup, I agree it's a software issue (besides some extreme situations I've seen where clearly something was very wrong). I think it's a combination though as it must be combining the values given by the IMU but the IMU is subject to the same physics laws.

To me, the best way to really correct this, would be with some sort of image analysis combined with the input from the IMUs.