Sensor applications in the automotive electronics field run a wide gamut from pressure sensors for manifold and exhaust systems to magnetic sensors for gear tooth and valve position. Regardless of the specific application, most automotive sensor modules share a common high-level architecture and a set of stringent requirements that make for a challenging design.

A typical automotive sensor module is shown in Figure 1 below. The example shown is for gas tank vapor pressure monitoring implemented with a piezoelectric pressure sensor and a signal conditioner (ZMD31015).

Quite a few sensors come in a bridge arrangement and produce a differential signal with small amplitude. These bridges have random part-to-part variations in sensitivities, offsets, and non-linearities. Signal conditioning is needed to amplify the signal and correct variations to produce the linear, accurate signal the system needs.

Many sensor modules connect to the system (such as the Engine Control Unit (ECU) in Figure 1) via a 3-wire connection. This connection provides power and ground to the module, and the module returns an analog ratiometric representation of the measureand (vapor pressure in this example).

The system receiving the conditioned sensor signal typically has a microcontroller and receives the signal via a single channel of a multiplexed analog-to-digital converter (ADC). The receiving system often provides a terminating resistor (a pull-up in this example) on the signal line.

Extreme environments
The automotive environment is one of the harshest environments for electronics. Many product manufacturers (OEMs) require a specified operational temperature range from -50 to 150C. In addition, the sensor module must be capable of functioning properly with large power supply disturbances and be able to tolerate ESD (electrostatic discharge) strikes in excess of 8kV. Designing electronics capable of tolerating such an environment presents challenges. Aside from the sensor, the main component of a sensor module is the sensor signal conditioner. Here we will focus on design challenges directly related to this component. Figure 2 below provides an architectural overview of a sensor signal conditioner (ZMD31015).

Temperature and power supply
Operating over a wide temperature range ( 50 to 150C), as well as a wide V_DD range (2.7 to 5.5V), requires very robust circuit design and verification. In the analog pre-amplification path, the designer must ensure op amp stability, open-loop gains, and temperature matching of the resistors that set the amplifier gain. In the ADC (which is implemented with a switch capacitor technique), other conflicting challenges arise at the temperature and supply voltage extremes.

Cold temperature and low V_DD levels present the problem of incomplete charge transfer across pass gate switches due to the elevated V_Ts at -50C coupled with the low overhead gate drive at V_DD = 2.7V. High temperatures (150C) present exponentially increasing junction leakages, which manifest as a charge loss off the switch capacitor nodes.

Balancing switch size (device width and length), capacitor size, and operating frequency to satisfy all extremes requires exhaustive parametric analysis across temperature, V_DD level, and process extremes. Last but not least, good device physics understanding also helps limit the breadth of explored solutions to what is most likely to yield good results.

High temperature alone presents problems for EEPROM retention. The sensor signal conditioner contains correction coefficients used by the digital core to correct for offset, span, temperature, and nonlinearity. These coefficients are stored in EEPROM. These mission critical parameters are written into the EEPROM at the time of sensor module calibration. Loss or corruption of data during operation could cause the sensor module to become uncalibrated—producing a valid in-range signal that the system "thinks" is correct. In steering or braking systems, for example, this loss of calibration could lead to catastrophic events.