I have been analyzing the behavior of the gyroscope thoroughly in order to quantify its drift. I did this by mounting the LSM6DSM onto a step motor and measuring its output. I tried this for two different devices which gave me for one device a constant error of 2.6% and for the other 3.3%. I also used an LSM303AGR magnetometer and found exactly the same over more than 100 tests. A rotation of 360 degrees on the magnetometer indicated 370 degrees on the gyroscope. This is significantly more than the 1% which is stated in the datasheet. Am I missing something or could it be that the scale factor error is far more than the 1% indicated?
