Phidgets Forums

Just wondering if there is a sweet spot in the number of calibration points one should acquire in order to get the best magnetometer calibration with the MOT0110.

Is it a matter of the more points the better, or not necessarily so? As I get more points, the calibration does not seem to improve more than if I get fewer points.

Thanks for your enlightenment.

Generally more is better, but there is an upper limit to how much additional benefit you gain once you cover most of the sphere or circle, depending on calibration type.

Note that calibration is only really effective if the sensor is moving with the machine/assembly it will be attached to, and will be less effective if it is moving towards or away from steel structures or magnets that could cause interference.

For more information, you can see our guide on Magnetometers

It's much more important to cover the sphere than to have many points. You could get a good calibration with 5 seconds worth of data, except that it may be difficult to rotate the spatial through the entire sphere in that much time. And the sphere doesn't need to be completely filled in - a single rotation in each axis would give sufficient coverage.

-Patrick