Regarding BMM 150, is automatic calibration possible only?
Or do I have to use ACC Sensor or 6 axes for automatic calibration?
Solved! Go to Solution.
Thanks for your inquiry.
Regarding BMM150 magnetometer calibration, you can refer to the hands-on video from Digikey talking about how to calibrate magnetometer hard-iron and soft-iron at https://www.youtube.com/watch?v=cGI8mrIanpk. It can be both offline calibration and automatic calibration. If your device is only rotating on a leveled surface, then BMM150 magnetometer itself can be calibrated and used to measure the magnetic heading with respect to earth magnetic north pole.
However, if your device can be tilted up and down, or left and right, then you need ACC to compensate the heading. You can find a lot of articles about how to use ACC to work with magnetometer to behave eCompass. The heading accuracy is good only when your device is stationary.
If your device needs to output accurate dynamic heading for 3D fast rotation motion, then you also need gyro to help. This is so called 9-axis sensor fusion. At this time magnetometer can help gyro z axis not to drift away and gyro can help magnetometer be immune to external magnetic interference. There are also a lot of articles for it.
Thank you for your answer.
We are going to use the golf distance meter, and as you mentioned, we are going to use the BMA+BMM150.
If we use two sensors, is the calibration method done by MCU or through Acc Sensor or BMM150 API?
If you use BMA + BMM150, then the BMM150 calibration will need to be done on your MCU side which means you need to develop your own algorithm to calibrate BMM150. In addition, your MCU also needs to use ACC data to do tilt compensation on the heading angle.
Therefore, to keep it simple it is recommended that you use BHI260AP + BMM150 for your golf distance meter project. BHI260AP is a smart sensor that has built-in 32-bit ultra low power floating point MCU. It also has BSX sensor fusion running inside that can automatically calibrate BMM150 and output tilt compensated dynamic heading angle if you choose 9DoF sensor fusion mode.
Thank you for your reply.
It seems that the acceleration sensor should be used because it is considered too expensive to use the BHI260AP.
It is said that the algorithm needs to be developed directly, but is there any way that it can be done by itself when using an Acc sensor?