3 Choosing a Magnetic Implementation

For many designs the most logical place for the magnet is in the moving component that we desire to sense, rather than the sensor which is likely powered through traces and wires routed from some far-off supply. As such, the magnet needs to fit somewhere in the shifter lever and depending on where the magnet is placed, different kinds of magnets can be appropriate. Figure 3-1 shows two possible magnet implementations. On the left is a diametric magnet centered at the fulcrum of the lever, while on the right is an axial magnet situated some distance from the rotation axis.

The goal of the design presented in this application note is to have an output that is linear without aliasing such as the desired signal shown in Figure 3-2. For detecting objects that rotate around an axis, 3D or multi-axis sensing devices are more designed for than single axis sensing devices. Not only do multi-axis sensing devices have flexibility of package orientation, but are also immune to aliasing that is otherwise unavoidable with a single axis sensing device. This is due to the fact that fields perpendicular to each other exhibit behavior analogous to sine and cosine waves. By taking the arctangent of two perpendicular fields, a linear output over a wide range of motion of the magnet can be derived.

For the diametric magnet, field behavior like in Figure 3-3 is expected while for the axial magnet field behavior like in Figure 3-4 is expected.

Due to the shifter in this design rotating and requiring redundant measurement, the TMAG5170D is selected.