Sensing the position and movement of automotive control functions such as gear selectors and turn signal indicators is an important requirement in modern vehicle design. However, sensing movement in 3 axes is not straightforward and conventional sensor technologies present engineers with challenges. These range from physical device size and reliability to power consumption and cost. Recently introduced 3D magnetic sensing technology is helping engineers to address these challenges.
Electro-mechanical switching is a common source of failure in many applications, including vehicles. Contacts can corrode over time or burn out, causing failure and inconvenience for the vehicle owner and potentially damaging the reputation of the car manufacturer. As such, manufacturers prefer solid-state technology, such as switching based on Hall effect detection of magnetic-signals, wherever possible to increase reliability and save space and cost.
Two of the most common things we do when driving are signalling to turn and changing gear. In the past, cars relied on heavy-current wiring harnesses to transmit signals and power around the vehicle. Today, using a turn indicator or moving a gear shift is just as likely to send a high-impedance signal to a central processor as it is to physically switch something.
As vehicle controls become more sophisticated and multi-functional, the trend is towards them moving in more than one plane. Many modern cars with automatic gearboxes now offer sequential control by moving the gear lever into a different plane, thus making the task of sensing position more complex.
Approaches to 3D magnetic sensing
There are several ways to implement 3D position sensing based upon the Hall effect. Where the movement to be sensed has multiple fixed positions (in the case of a gear lever or turn signal), one approach is to add individual Hall sensors for each possible position, resulting in as many as seven sensor elements, allowing the controller to know the position by knowing which sensor was 'live'.
Another method, Hall based as well but using flux concentrators, reduces the number of sensor elements needed by integrating two pairs of orthogonal sensing elements onto a CMOS IC. A ferromagnetic film that enhances the magnetic field and increases sensitivity and signal-to-noise ratios also is deposited onto the IC surface.
Through the use of several subtraction and addition algorithms it is possible to sense the magnetic field components running horizontal to the IC (BX and BY components) as well as the vertical field component (BZ) accurately. These analog voltages are then converted to digital values via an analog-to-digital converter (ADC). The final, absolute, position is then determined by digital signal processing.
While each of these solutions is viable from a technology perspective, they are not ideal in mass-production – especially in automotive applications. Using multiple sensors requires a substantial amount of space and as more components are added reliability will inevitably suffer. The second approach has some merit but the complexities involved in the algorithms and post-processing–not to mention the need to calibrate individual sensors–mean that it is a less-than-ideal solution.
A third alternative utilizes a Hall-based sensor capable of simultaneously determining the x, y, and z coordinates of a magnetic source to build a 3D image of the magnetic field surrounding the sensor (Figure 1 ).
The TLE493D-A1B6 3D sensor integrates vertical and horizontal Hall plates into a single IC, thus detecting the planar and vertical planes in a single operation. As the magnet moves, the sensor immediately recognizes a change in at least one of the magnetic field components.
The sensor architecture consists of the three primary building blocks shown in Figure 2 .
The sensor unit contains the Hall plates and measures the magnetic field in all three directions. Using vertical Hall plates for the planar components allows the sensor to achieve high magnetic reconciliation accuracy (±1%), permitting precise angle measurements.