More and more applications are moving to electrical actuation. This includes items such as electric throttle bodies, electric oil pumps, electric turbochargers, and, especially, electrical propulsion in battery powered EV and HEV. Not only are these changes making the vehicle more efficient, they are cleaner and more reliable. They are also lighter than the previous mechanical / hydraulic solutions, and this reduced weight makes vehicles even more efficient, thereby burning less fossil fuel or travelling further on a single charge.Yet, actuation has little value without accurate, repeatable and reliable sensing. A control unit needs to be able to sense the current position so that it can calculate how to move to the desired position.
There has been a recent revolution in mechanical position sensing with magnetic technology replacing resistive and optical sensors as the preferred method of sensing. Unlike rotary encoders, a magnetic sensor is able to withstand the dust, dirt, grease, vibration and humidity that are present in vehicles (and many industrial applications). Compared to other commonly used angular and linear displacement sensors, magnetic sensors do not wear – another key attribute that ensures long-term repeatability and reliability.
However, as electrical / electronic technology proliferates the vehicle environment begins to pose an ever-increasing challenge to magnetic sensing. For example, the electric motors that power EV and HEV consume large quantities of electric current and consequently produce magnetic fields in the areas around the cables carrying the electric current from the battery or alternator to the motor. To a lesser extent, even the smaller currents required to power electronic power steering (EPS) pumps, open windows, move the sunroof or any one of the electrically actuated devices in the vehicle can create a stray magnetic field.
While these stray magnetic fields will almost certainly impair the accuracy of the sensors they can, in some cases, lead to very large errors in the output – with potentially catastrophic results. While a sunroof that does not close properly is an inconvenience, a brake pedal, gas pedal or steering system that cannot be sensed accurately is potentially life threatening.
In order to address the safety implications the automotive industry has standards that define the requirements for safety of systems that could be affected by stray magnetic fields (and other issues). Among the key standards are ISO26262 that deals with functional safety and the multiple OEM specifications that addresses immunity to magnetic fields.
Traditional Hall effect and magnetoresistive (MR) sensors are sensitive to the stray fields that are found in vehicles as they are designed to measure the magnetic field produced by a nearby magnet that is connected to the item being measured. Since the stray magnetic field produced by electric currents (especially the hundreds of Amperes that power the traction motors) can be large the sensor output can be significantly inaccurate. In the case of rotary sensors, the angle error can be in excess of 10 degrees, which is substantial when many systems like valves or throttle bodies only physically rotate 90 degrees. Apart from the obvious safety issues in steering and braking, the engine control unit (ECU) would simply not be able to manage the engine properly (if at all).
Designers who wish to implement magnetic sensing in vehicles are faced with two primary options. The first relates to shielding the sensing device (and its associated magnet) from the effects of stray magnetic fields. Not only is this potentially complex and challenging, it is also expensive as materials with high magnetic permeability are required. These do not shield so much as absorb and reroute the field, so they can also have an impact on the sensing as the desired magnetic field is modified as well as the stray field. To avoid this, spacing is required that increases the size, weight and cost – none of which is desirable in modern automotive design.
The other primary approach is to use a type of sensor that is intrinsically immune and insensitive to stray magnetic fields. The Gen III Triaxis magnetic sensors from Melexis include an in-built stray field immune mode that allows for a substantial reduction or elimination of the error caused by stray magnetic fields. As such, the Gen III Triaxis sensors can be used in close proximity to current carrying conductors or by other nearby magnets in the vehicle.
The stray field immune mode only requires a simple 4-pole magnet for rotary motion and a simple 2-pole magnet for linear motion. With simple magnetic design, the error due to stray fields is reduced to below 0.4 degrees of angular error, which is an acceptable value for most major vehicle manufacturers. This means that design efforts can be reduced as it is no longer necessary to locate the sensor away from stray magnetic field sources.
Additionally, any shielding that may have been necessary in the past can be eliminated or substantially reduced, resulting in significant system size, weight and cost savings. The stray field immune mode also delivers the same benefits as the legacy mode including non-contact sensing, high levels of EMC robustness, a small package envelope, and the ability to achieve a high functional safety level (including ASIL B or C).