TMAG5231: TMAG5231A1CQDMRR - Odd results from TIMSS linear analysis when 1.6x0.8mm cylindrical magnet is 2mm on either side of sensor

Steven,

The typical thresholds for this device are at +/-1.8 mT (BOP) and +/-1.2 mT (BRP).  If you look closely at the Z field component, there is a slight negative value which must be large enough to trigger the BOP.  However, as the magnet passes by, the field quickly returns to positive.  The result is that the field strength in the Z direction will pass BRP before continuing on to become very large in the positive direction.  

This happens due to the path of the magnetic field.  In your starting position, the field is directed from the top of the package towards the PCB (negative with respect to the sensor), but as the magnet gets closer to the sensor the field direction eventually reversed, with the field vector directed from the PCB towards the top of the package. 

During the transition period the sensor will briefly release, and due to symmetry we will observe this on both sides of the magnet. 

To eliminate this effect I would increase the range between the magnet and sensor and/or select a device with a higher operating threshold.

Thanks,

Scott