rob42
commented
5 years ago The IMU is very sensitive to magnetic disturbances. Rotating it through a figure 8 pattern (eg 360 degrees in every axis) allows the IMU to compensate for local mag fields. I found that even the difference between my desk (with laptop/screen/junk) and a few meters away was significant hence the comment.
You may be right that rotating the vessel is also useful, needs some experimentation I guess. My own issues were about the IMU's location in the vessel, power wires and the compass affects it. Also some-one leaving a tablet or phone nearby ( they have powerful magnetic buttons in them!).
My nav station (catamaran) extends under the saloon seats and I had the IMU in there. Sit on the seat with phone in pocket or leave your tablet on the seat and watch the heading go nuts :-(
@rob42 said:
I do not understand this. What is the expected advantage? If you want to calibrate the device itself, you can do this anywhere. If you want to calibrate the sensor output regarding the surrounding ship, you have to fix the sensor and rotate the whole ship.
Maybe the solution would be to add the device calibration to the ship calibration: First do the "figure 8 pattern" whereever you like. Store this as device calibration. Then mount the device and rotate the ship in circles. This adds the ship calibration. This is how calibration of every commercial fluxgate compass works.