Smart sensor emulation speed design of signal conditioning systems

The time to develop sensor-signal conditioning electronics and its associated calibration algorithms can be difficult and time-consuming. Using a hardware sensor emulator can dramatically reduce development time.

Cycling the temperature from cold to hot in an environmental chamber is the main factor that makes calibration so time consuming because it can take hours for the temperature to fully stabilize. For example, a typical three-point calibration (-55°C, 25°C, and 85°C), can take eight hours. A sensor emulator programmed to model a sensor’s temperature drift allows engineers to do a simulated calibration run in only minutes. Thus, during a day, an engineer can do many calibration runs. This allows the engineer to focus on debugging their sensor hardware or developing their calibration algorithm.

Another difficulty in working with “real world” sensors is that they can sometimes require costly specialized test equipment to provide the stimulus (for example, humidity, acceleration, pH, pressure and strain). In many cases, access to this equipment is limited. Using a sensor emulator to debug your software and hardware prior to using the real world equipment greatly improves the chance of success in the lab. In addition, the equipment that provides the real world stimulus is sometimes custom-developed for a specific sensor and as such may be unproven.

For example, a custom mechanical calibration system developed by Contrivance Engineering* applies torque to its sensors. Because sensor electronics were developed in parallel to that of calibration systems, it makes it difficult to debug the sensor electronics without a sensor emulator. Furthermore, the calibration system initially had a problem with mechanical oscillations. Without an emulator, it’s difficult to determine whether the oscillations were from the mechanical system or the electrical system.

Real world sensors can also have non-repeatability issues. This can make it difficult to ascertain whether the errors are due to the sensor, or the capability of the sensor electronics. For example, a pressure sensor can have pressure hysteresis, or temperature hysteresis which can sometimes be compounded with time hysteresis. Since the sensor and electronics are typically connected in a hermetically-sealed case, it is challenging to ascertain if the error is from the sensor or the electronics. A sensor emulator does not have the non-repeatability issues and so a good assessment of the sensor electronics accuracy is possible.