Dual-Die Hall Sensors With Redundancy Function Address Safety-Critical Automotive ApplicationsJune 24, 2016 By: Vita Thilo Rubehn, Micronas
The non-contacting measuring principle of Hall-effect sensors in CMOS technology has proven its worth in automobile electronics for many years. For example, relatively simple Hall sensors with switching output detect if a car door is fully closed or not.
On the other hand, Hall sensors with linear output are used to detect the accelerator pedal position. This parameter is needed for the electronic control of the combustion process and must be detected reliably and with ultimate accuracy. The magnetic field measurement using Hall sensors is highly robust, although the measurement result is influenced by the crystal temperature of the silicon semiconductor chip ("die"). To compensate this factor, the junction temperature will be determined first. This will function best if the temperature sensor is integrated in the same die which also houses the Hall plate.
However, this integrated temperature sensor is exposed to production tolerances – with the ultimate effect that the sensor also needs to be checked. Once the semiconductor has been manufactured, a test phase is therefore carried out where a constant external test temperature is selected and compared with the values of the temperature sensor.
A linear function can be determined in order to balance the ambient temperature and the junction temperature. To be on the safe side, this process is usually carried out at two different temperatures. The constant external temperatures allow the simultaneous measurement of the Hall element by applying a highly accurate and well known magnetic field and detecting the response of the Hall sensor.
Since the 1980s, measurement tasks in automobile applications have been successfully solved with the help of Hall sensor ICs. But the requirements in terms of the reliability of the position data obtained have increased tremendously over the last years. For example, it may be relatively non-critical if a sensor fails to detect an open car door, but detecting an incorrect accelerator pedal position may lead to dangerous driving situations. To enhance safety, there is the option of using not just one, but two sensors for the same measurement. The output signals of both sensors can so be compared and error deviations can be detected. The on-board electronics is now capable of responding accordingly.
The Redundancy Principle
This so-called redundancy principle is one of the best known methods for safety-critical sensor applications. Additional sensors may increase the weight of the application and will also require more installation space, but this problem can be solved by increasing the integration density of the sensors. A method commonly used in the semiconductor industry involves the integration of several CMOS semiconductor chips in a single package. However, this is a novel approach in Hall effect sensors.
The two semiconductor chips may be arranged differently inside the package. One option is to mount both dies side by side on the lead frame of the package. The disadvantage of this solution is that the Hall plates of both chips are positioned differently and will therefore not detect the same magnetic field range. This results in errors of measurement between the two chips and leads to problems in the plausibility check of the redundancy function.
Most Read Articles