I got sick of tryiing to either find somewhere level or grab a bit of wood and use books to level a platform for the IMU calibration so I threw together an adjustable platform to sit the P3 on.

All up it cost about $20.

so simple and nice

Now you'll need a digital spirit level to go with it.....about $80 at Bunnings!! Still, you can use it for other jobs where extreme accuracy is required....

There's a pretty cool app for the iPad called "Measures 2".  Sploodge turned me onto it.  It has a surface level.  I just use it since I have my iPad anyway.  It even has a beeping geiger counter sound that speeds up as the dot gets closer to the middle.  lol

HunterBrooks Posted at 2015-7-28 13:35
There's a pretty cool app for the iPad called "Measures 2".  Sploodge turned me onto it.  It has a s ...

Thought I would give that a try on the board..... Damn..... iPad Air is just a little to big to fit in-between the screw feet.

Ummmm...  How do you know the accelerometer in the ipad is calibrated?  Better to use the traditional spirit level..

Or...  make a flat jig to calibrate the ipad, so you can calibrate your flat table to calibrate the Phantom..  :-)

Simple is better in this case, ipad's are not accurate measuring instruments of anything apart from how much money your kids manage to get you to spend in the app store!

colin.buckle Posted at 2015-7-28 14:45
Thought I would give that a try on the board..... Damn..... iPad Air is just a little to big to fi ...

then use your phone

Phone has a case that would cause it not to be level (i.e. has extrusions in odd places). I'm not going to remove it from the case just to do that so I will stick with the traditional spirit level.

My granite kitchen counter top is perfectly level, so I just use that.

bradneal24 Posted at 2015-7-28 22:35
My granite kitchen counter top is perfectly level, so I just use that.

Agreed. I use that when I am home, but if i'm away out in the bush and want to calibrate after an "issue", trying to find somewhere level can be a real pain.

Oh, no problem. I just take a chunk of the counter top with me. My wife only asks that I leave some behind. :-)

Seriously, though. I have never done a IMU calibration in the field. But you can never have too many tools!

I made one very similar out of HDPE, The level I have is to 0.05 mm and
works great

bradneal24 Posted at 2015-7-29 06:50
Oh, no problem. I just take a chunk of the counter top with me. My wife only asks that I leave some  ...

Have only done it once myself due to a heavier than normal landing

Wow, is that necessary

It seems to be more than an issue of calibrating on a level surface that causes the non-level horizon problems.

One advice that has helped me a lot with IMU calibration is "the refrigerator trick" - or the "cool-down-your-Phantom-in-your-aircon-cooled-car"-trick (for those afraid to put the Phantom in the fridge (personally didn't have condensation problems doing it - but climate and humidity varies).

Basically I've put the (already at room temperature, did not fly before calibrating) P3 in the fridge for 15 minutes or so to get it nicely chilled. Then placed it on a level surface and quickly did the FULL IMU calibration. This seems to make the calibration go well - and last but not least - it completely solves the "IMU warming up" issue that many people have after doing an IMU-calibration with a warm Phantom. The reason is that the Phantom needs to get the IMU up to the temperature it was when it was calibrated before being "Ready to fly".

Colin,

     I like it. Simple and it works.

So, how often is it recommended to run these IMU calibrations. I have only done it once after smacking my P3 into some tree branches and then then obvious hard landing that followed. So is this something that is recommended on a regular basis.  (sorry for the thread highjack)

Do you have to have a GPS signal available to do the IMU calibration? I didn't realize you could do it indoors.

I work at a college biology department. In my buildings we have several 'cold rooms'. They are 20' X 15' rooms or larger with tables and workspaces set up in them already. Instead of getting the P3 cold, then taking it out to do the calibration, would it be OK to do the IMU calibration while it was still IN the cold room? I can easily make a space perfectly level inside one of these rooms and do the calibration without ever raising the temperature. These rooms are not at freezer levels but range from 4 to 18 degrees Celsius. What would be an appropriate temperature to do an IMU calibration at if the drone were to stay in the cold area while being calibrated?

You only need 3 screws.

Phantom Menace Posted at 2015-7-29 21:02
Do you have to have a GPS signal available to do the IMU calibration? I didn't realize you could do  ...

Nope.

The IMU is all about the gyro and accelerator sensors.

IMU =  Inertial Measurement Unit

I don't think the IMU calibration on the P3P involves the compass (magnetometers) since the compass has it's own calibration process.

bradneal24 Posted at 2015-7-29 20:51
So, how often is it recommended to run these IMU calibrations. I have only done it once after smacki ...

Here's when you should calibrate the IMU:

1) Before your first flight
2) After installing new firmware
3) After crashing and/or if your Phantom gets knocked around (e.g. you drop the case)

---

www.PhantomHelp.com
Find the latest must-haves, fixes, mods, how-tos, guides, and accessories for your Phantom.
Phantoms in stock and ready to ship worldwide! (see availability)

liningiv Posted at 2015-7-30 19:03
You only need 3 screws.

Correct..... But being trained in levelling a crane, I just went down the four screw path without thinking.
3 would be much easier for most people though.

I just wanted to give an update on my 'Cold Room' Calibration. I left the P3P in the 4 degree Celsius cold room for about 3 hours, to get it cold all the way through. It was off all this time. Then I turned on my tablet, then turned on my controller, started the app, and got everything ready, then turned on the copter. As fast as I could after it connected, I got to the IMU calibration and started it. It only took a few seconds to get to the right menu and start the calibration. It took 3 or 4 minutes to go through the entire calibration. I also did a gimbal calibration right after the IMU. Then I turned everything off and pulled out the battery on the copter. I brought everything back to a different room that was about 70 degrees Fahrenheit. I let everything sit there with no power until everything slowly returned to room temperature naturally to avoid getting condensation in the equipment. Later that day, I took it out for a flight. It was close to 80 degrees Fahrenheit outside at the time. I spent exactly zero seconds waiting for the aircraft to warm up. It was ready to fly the second it connected to enough satellites. I was the quickest startup I have experienced. It has been like that ever since. It had been taking up to 3 or 4 minutes to warm up prior to this cold calibration. I think being able to leave it in the cold while actually doing the calibration was very useful. It kept it at the lowest possible temperature through the entire process. If you leave it in your air conditioned car, or put it in a fridge, it will warm up considerably when you take it out and get set up to do the calibration compared to this. I highly recommend this method to anyone who has access to a walk-in cooler or a cold room as I had. The operating temperature range of the Phantom 3 is anything between 0 and 40 degrees Celsius, so my 4 degree room was in the range, and at the low end of it, so even on cold days outside (anything above freezing), it won't take any time to warm up the aircraft.

Colin that is great - and I have built one just like yours, so THANKS!

However....here it comes .... after doing many hours of investigating on my own P3's issue of tilted horizon (not drifting - that's something different) and reading TONS of material concerning calibrations of all sorts of flight control perifery, guess what? In my case (maybe ONLY my case), the big square circuit case inside the shell was not exactly level. Now, that does not necessarily mean that the circuits inside the that or any other case are not level, but it was (again, in my case) not in a plane perfectly parallel to the plane of the P3 feet.  I did not change any of this, because my tilt problem has now vanished using another technique, but thought it might be worth mentioning.

When we do a level IMU calibration, what is it we are really trying to level? Like most, I pretty much assume that all the parts are put together in very parallel planes. Lets hope so. Meantime, I'm using your platform whenever I feel the need. So THANKS!

At last, having read this thread I now know what IMU stands for, why it needs calibration and how to do it. Ain't forums useful, thanks guys. From a newbie in England