Note: This article is the next part of our six-part series on sensor simulation in ECU validation.
The fundamental principle of a thermocouple resides in what is called the ‘Seebeck Effect’, which, in its simplest terms, observes that when two dissimilar metals are joined together, a thermoelectric difference in potential exists. That difference is proportional to the temperature at the junction. This voltage potential is typically in the millivolt range and can be calculated using a polynomial expression dependent on the type of metals used.
In a real-world scenario, a thermocouple (such as Type J, composed of Iron and Constantan) is connected to a data acquisition unit (DAQ) in the Electronic Control Unit (ECU). The DAQ unit, with its universal copper-copper connection points, plays a crucial role in the thermocouple's functioning. It's important to note that connecting the Iron/Constantan thermocouple to copper creates another Seebeck effect, known as the Cold Junction, which must be considered in the conversion process.
Figure 1: Diagram of a diagram of a cold junction
The table below illustrates the non-linear relationship between degrees C and millivolts for a Type J thermocouple. Engineers and technicians must be aware of this nonlinearity, a key aspect of thermocouple conversion. And that, despite this non-linearity, it's interesting to note that a temperature change from 0 degrees C to 10 degrees C results in a potential change of 500 microvolts, demonstrating the sensitivity of thermocouples to temperature variations.
In many applications, temperatures must be measured to resolutions closer to 1 degree C rather than 10 degrees C. This means a device that simulates a thermocouple, such as Pickering’s PXI thermocouple simulator series (models 41-760 & 41-761), must be able to source a very accurate low-level signal and be flexible enough to support a wide range of thermocouple types. Furthermore, the cold junction must be considered to simulate a thermocouple accurately since there will be a voltage offset—which translates to an error—where the thermocouple wire is terminated at the measurement device. It is also recommended that an open-circuit thermocouple wire be simulated to check the behavior of the ECU during the design and validation phase.