This is how I fixed my tilted Horizon. Chilled my Phantom 4 at 39 degrees Fahrenheit for 1 hour then did imu calibration followed by gimbal calibration and now it is 100% perfect and level...

A cold IMU and gimbal calibration has always kept my horizon  in check. ;-)

RedHotPoker

Been there done that, in the frig at 38* for an hour, no joy.
She goes back to the Mother ship Monday.

That makes zero sense on why putting it in the fridge for an hour then doing cold calibrations can possibly be making your gimbal no longer have crooked horizons. What does the aircraft being cold have to do with anything (logically/technically speaking)? Have you ever heard of DJI putting theirs in the fridge for an hour? I call coincidence. My gimbal goes off and levels out on its own schedule, so there's nothing binding it to show a crooked horizon, it just malfunctions sometimes and does. Please explain your logic on why this would work, and please don't say "some electronics work better at 38°!"

What happens if you put it in the freezer?

Will it keep the calibration longer?

FatedFilmsNC Posted at 2016-5-14 14:41
That makes zero sense on why putting it in the fridge for an hour then doing cold calibrations can p ...

In another phantom dedicated forum there are also a lot of posts about cold IMU calibration, and it seems to be a widely known thing amongst phantom pilots. They say that the IMU warmup time is also shorter before you can fly, but I think that's just because the calibration records a working temperature or atleast stores some sort of characteristic to be achieved.. (so, that's cheating , but heck with it if it works )

However I've also yet to stumble upon a scientific reasoning as to why this works (as a better means of calibration and achieving better stable horizon) or have people just taken it like ancient wisdom.. Can't figure it out logically myself either, why it would technically work and am very curious to find out, so all and any input is appreciated.

Alari Posted at 2016-5-15 09:48
In another phantom dedicated forum there are also a lot of posts about cold IMU calibration, and it ...

Lower temperatures may lead to less 'noise' in the circuitry. And perhaps a better calibration of the reference measurements. Certainly in Astronomy imaging the imaging chips are cooled to anything down to about -40C to reduce the noise produced, many produced by Sony.
I've not yet tried the fridge method. My IMU calibration was done at about 18C, craft seems stable but does suffer horizon tilt I think after a yaw turn. This seems to right itself within about 5-10 seconds in hover.

Aardvark Posted at 2016-5-15 11:16
Lower temperatures may lead to less 'noise' in the circuitry. And perhaps a better calibration of  ...

Thank you for a clear logical explanation!

Aardvark Posted at 2016-5-15 10:16
Lower temperatures may lead to less 'noise' in the circuitry. And perhaps a better calibration of  ...

With my first P4 I also noticed that the gimbal not being level was generally after a yaw turn, but I also noticed that if I continued the yaw turn through to 180 degrees the gimbal would be level again, if I continued the yaw turn past 180 it would again go off level but this time in the opposite direction, and it would level off again when I got to a complete 360 turn.

This second P4 doesn't have the gimbal tilt problem quite as bad as my first, but it does still exhibit a tilt sometimes.

I'm keeping this one though and hoping that DJI give us a fix via a firmware update.  I'm the kind of person that when I get into something I end up buying the Pro or semi pro versions of whatever it is I'm doing (for example I started with a Canon 350D and now own a 5D mk III along with a few 'L' lenses) so I just know that in a year or two time I'll be looking at buying an Inspire or better series of Quad, probably with the X5 or X5R camera.  That all depends on how well DJI support this P4 though.

I take my p4 to the North Pole to do the IMU calibration, -55 f the calibration will last forever.