Thermocouples often described incorrectly

Thermocouples generate a difference voltage proportional to a temperature difference.

A thermocouple is made two of wires of different metals where the tips are connected (often welded) to produce a good electrical connection. This junction is the temperature measurement point. Figure a shows the incorrect assumption about thermocouples. There is no voltage developed across the junction, although there will be a voltage across the open ends of the wires (V_T).

In Figure b, assume that the thermocouple junction is at temperature T₁ and, for simplicity, assume that the open ends of the wires are both at T₀ = 0°C. A wire whose ends are at different temperatures will produce a small voltage across those ends, called the Seebeck Effect. That voltage depends on the composition of the wire and the temperature difference across the wire’s ends.

The material of a wire affects the voltage across its ends, so wires of two different materials produce two different voltages (V₁ and V₂) for the same temperature difference. Thus, the voltage across the terminals is V_T = V₁ – V₂. The voltage develops across the wires, not across the junction.

Because keeping the wire ends at 0°C is impractical, equipment manufacturers use a circuit to measure the temperature at the ends of the thermocouple wires and electronically compensate for that temperature. It’s called cold-junction compensation (CJC). Even with CJC, the thermocouple wire connections must be as close to each other’s temperature as possible to minimize measurement errors. Most connections to thermocouple measuring equipment, therefore, are mounted on a thermally conductive block.