Modeling and simulation of magnetoresistive sensor systems (Part 1 of 2)

The magnetoresistive effect supports a variety of sensor applications in automobiles. Major applications are to be found in measuring the speed and angle of mechanical systems. That makes magnetic field sensors part of a complex system consisting of electrical, magnetic, and mechanical components. Since all components influence the way in which a system responds, there is a great deal of importance attached to simulating an overall system when planning it and its operation. Modeling and simulating such systems are the focus of this article.

The increasing use of electronics has made a decisive contribution to advances in the automobile. In the future, too, advances will be largely driven by innovative electronic components. Special significance attaches to sensor technology, which is a means of accessing conditions of a vehicle and its surroundings.

A variety of sensor systems are available for such purposes, enabling the measurement of acceleration, temperature or torque, for instance. Sensors for magnetic-field measurement are especially common in the automobile, serving primarily for contactless detection of mechanical variables. Sensors of this kind are usually implemented either by Hall elements or based on the anisotropic magnetoresistive (AMR) effect.

Compared to solutions using the Hall effect, AMR sensors possess a number of advantages, such as less jitter and higher sensitivity. Both can contribute in equal measure to higher accuracy or reduction of overall system costs. In addition to measuring the magnetic field of the earth for electronic compass applications, magnetoresistive sensors are used in particular to determine angle and speed, where the magnetic field indicates the motion or position of a mechanical system. Data of this kind are needed by anti-skid systems, engine and transmission controls.

The mechanical design and choice of field-generating permanent magnets influence in large part the acquisition of measured data. In-depth analysis using simulation is consequently essential before an overall system is implemented to ensure the targeted functionality and cost reductions. System models, generated in the course of lead development and then serving to support product development, can therefore also be an important contribution to solving the kind of questions arising during the design-in phase. Here we look at the modeling and simulation of an overall system in the case of a new speed sensor.

Signal detection
Modern sensor systems consist essentially of two components: an elementary sensor and a signal-processing ASIC (Figure 1). The anisotropic magnetoresistive effect, discovered by the later Lord Kelvin in 1857, has proven especially suitable for detecting magnetic fields.

Fig. 1. AMR sensor systems consist of two packages (note: "Sensorfahne zur Montageunterstutzung" translates as "Sensor tag as mounting support"; " Signalverarbeitungs-IC" is " Signal-processing ASIC")

The starting point is a ferromagnetic material that generally possesses a variety of domains. These Weiss domains, as they are called, differ in the orientation of their internal magnetization. If such a material is deposited in a thin layer, the magnetization vector is in the layer plane. Plus, to a good approximation, it can be assumed that only one domain is present. When such an element is exposed to an external magnetic field, the latter alters the orientation of the internal magnetization vector. If a current also flows through the element, this produces a resistance, Figure 2, which is dependent on the angle between current and magnetization.

Fig. 2. Anisotropic magnetoresistive effect (note: " Magnetisierung" is "Magnetization"; " Strom" is "Current")

The resistance is at a minimum when current and magnetization orientation are at right angles to one another, and maximum when they are parallel. The magnitude of the change in resistance depends on the material. The nature of the ferromagnetic material also determines dependence on temperature. An optimal alloy in terms of a maximum change in resistance of 2.2% and favorable temperature response is an alloy consisting of 81% nickel and 19% iron.

The elementary sensor of all sensor systems from NXP is implemented by such a permalloy. The individual AMR resistances are configured in a Wheatstone bridge circuit to increase the output signal and improve temperature response. This circuit also enables simple trimming during manufacture. Figure 3 shows how the AMR elements are configured on the die. The setup to determine speed consists, for the most part, of two components: an encoder wheel and the sensor system. The encoder wheel can be either active or passive.

Fig. 3. Configuration of AMR elements on die View a full-size image

An active wheel is magnetized and an MR sensor consequently detects the change between north and south poles. In the case of passive wheels, the magnetization is replaced by a tooth structure. As Figure 1 shows, a permanent magnet for field generation is also necessary on the sensor head.

In the following discussion, only passive encoder wheels are looked at, which are notable for very small tolerances. When the sensor symmetrically faces a tooth or a gap between two teeth of a passive wheel, this produces no deflection of the magnetization vector of the AMR elements. Neglecting external noise fields and allowing for the bridge circuit, the output signal achieves a value of zero. If the sensor head is in front of a tooth edge, however, the magnetic input signal reaches an extreme. The result, as a function of the type of change between the tooth/gap or gap/tooth is, to a good approximation, the minimum or maximum of a sinusoidal magnetic input signal.