A drone calibration turntable.

[GMack]

First, nothing I have is level outdoors so I have to dig around for shims under a plywood piece for the IMU Calibration. Got a bit annoying, as did turning the thing around for the two-axis compass alignment.

Decided I needed something portable for the field. A 16" bamboo turntable from "Bed, Bath & Beyond" housewares department seemed appropriate as it had little metal for the lazy susan part for the spinner. Added a bubble level, and some dowels and a 2-1/2" x 1/2" piece of red oak from Home Depot, and three 1/4"-20 2" long Nylon machine screws from bolt house or hobby shop.

You get the idea in the photo below. The three nylon bolts are screwed down for leveling along with the bubble level. Then they can be backed off as the leveling feet to allow the turntable to spin. The turntable now takes care of the horizontal compass part, and putting in the oak piece with the dowels supports the legs for the vertical compass spin.

Fwiw, I tapped the wooden turntable with a 1/4"-20 tap for the nylon bolts to cut down on any metal (Like T-nuts). Then I put a few drops on the tapped wooden threads with CA thing glue to harden the threaded portion, allowed to dry for a couple of hours, and then re-tapped the holes for the nylon screws to clean up the threads. Glued the bubble level from Home Depot down with the CA glue too.

Works pretty well, if I do say so myself.

Mack-

 

[28wins]

Great!

 

[RedHotPoker]

You could calibrate the IMU indoors, where square tables, & stable counter surfaces abound. ;-)
I like to chill down my drone first too... So the kitchen counter is an ideal place for this.

RedHotPoker

 

[matti]

Great.

How accurate is the bubble level compared to a real long level? (Last time I calibrated the IMU by putting the motor hubs level with a glass panel).

I read a compass technical article referenced here yesterday. I was left wondering what practical difference is there to calibrate the compass with exact horzontal and vertical level vs the usual dance?

 

[Mark The Droner]

I'm confused by the OP. The words state IMU calibration but the actions described is a compass calibration.

...
I read a compass technical article referenced here yesterday. I was left wondering what practical difference is there to calibrate the compass with exact horzontal and vertical level vs the usual dance?

My understanding is, one only needs to hold the aircraft within 45 degrees of exact horizontal and vertical axis when rotating.

 

[matti]

I'm confused by the OP. The words state IMU calibration but the actions described is a compass calibration.

My understanding is, one only needs to hold the aircraft within 45 degrees of exact horizontal and vertical axis when rotating.

AFAIK the AC should be level while doing IMU calibration. I always follow it by gimbal calibration.

...I also babbled about compass calibration ... where did you get the "within 45 degrees of exact horizontal/vertical" figure?

...I now found the post which lead me to read some info about the compass technology and how they very carefullly position the compass before calibrating it:

Constant compass error warning?

http://ww1.microchip.com/downloads/en/AppNotes/00996a.pdf

 

[Mark The Droner]

I think I saw it posted on here somewhere and also in a youtube video. The 45 degree thing might not have been mentioned but the explanation was that it was not over-important to hold the aircraft at an exact horizontal or vertical axis - all that was needed was that it was mostly horizontal and mostly vertical when doing the compass calibration.

 

[N017RW]

The technical paper presented is for calibration. Calibration requires known standard(s). We don't actually calibrate the compass, that's fixed. (Notice the paper describes starting at 0/360 deg prior to calibration)

We compensate for the ferrous materials on-board the aircraft. These materials create so-called hard and soft iron distortions.
Rotating the magnetometer along with the on-board materials while measuring sensor response creates a non-circular plot (quadrant diagram) in the X-Y axes. This measurement data is also distorted by the local magnetic field and once recorded allows the software to construct a table correcting or compensating for the local field and iron distortions thus re-circularizing, sort-of-speak, the quadrant diagram of the response provided by the sensor (shown in figure: 9 of the previously linked paper).

This is why DJI has not specified or required such rigid requirements as would be needed for calibration such as maintaining orthogonal leveling, starting at a known heading such as North, and speed and direction of rotation (CW, CCW) for examples.

This also explains why compensation is not need very often. It is MOST critical when adding, removing, or relocating items in or on the quad.

I applaud the creativity with such fixtures and jigs, and if it makes you feel more confident that's great but there's no need to second-guess compensation results obtained with out using one.

 

[Monte55]

I didn't see the need for the turntable to do compass calibration but for the fact the op wanted a level surface for IMU CALIBRATION in the field. I guess that would only be necessary if there was an issue while out flying otherwise doing it at home on a steady level surface would be fine. Now some with physical handicaps might find it very useful as one had posted some time back. I used to calibration my compass first flight of day every time. I haven't done that in awhile but upon powering up my P1, I make sure it's fairly level and steady during boot up process. Works for me.

 

[N017RW]

Good point about assistance for the handicapped.