Many engineers, and some instrument companies, think thermocouples produce a small voltage created by the junction of two dissimilar metals. In fact, the temperature difference from one end of a conductor to the other creates the small electromotive force (EMF) that lets us measure temperatures. Thomas Seebeck discovered this effect in 1822, and it bears his name.
But if we use the same metal as a return electrical path, the temperature gradient creates an equal and opposite EMF in the return circuit. The result: a net EMF of zero. If we choose another metal that has a different temperature vs. EMF response, we can measure the difference, as shown in the figure. The junction — a weld or solder joint — simply connects the two metals.
So, contrary to common belief, a thermocouple does not let us measure the temperature at the junction. It produces a voltage that tells us the difference between the temperatures at each end of the circuit.
Because a thermocouple’s EMF arises from a temperature difference, you must know the temperature at one end to determine the temperature at the other. A 0 degrees C “ice bath” can serve as a known reference at the instrument end. Modern instruments provide an isothermal junction block and circuits in the instrument measure the block’s temperatures and adjust the thermocouple’s EMF accordingly. Manufacturers refer to this technique as cold-junction compensation.
Thermocouples comprise standard pairs of conductors, designated with a letter and identified by the colors of the two conductors’ insulation. (See Table 1 in the Web version of this column.) For each thermocouple type, the National Institute of Standards and Technology (NIST) produced a table that relates millivolt measurements — referenced to 0 degrees C — to temperatures. If you use an ice-bath reference, you can simply convert voltage readings to temperatures. Instruments can automatically convert measurements into temperatures or into voltages that directly represent temperatures.
When you use thermocouples, keep the following tips in mind. Measurement equipment must present thermocouples with a high-impedance load. You do not want current to flow and affect the EMF generated in a thermocouple circuit. So, don’t try to use a low-cost handheld meter to measure a millivolt signal from a thermocouple.
Long thermocouple wires can pick up signals from nearby equipment and power lines. To reduce such interfering signals, you may choose to pass thermocouple signals through a low-pass filter to remove power-line noise prior to measuring the signals. Some instruments include such filters.
Thermocouple wires may pass through hot or cold areas without instruments detecting those temperature changes. The EMFs created as a wire enters and leaves such an area sum to zero.
|
|
|
|
|
|
|
|
A thermocouple generates a small voltage that represents the temperature difference from end to end. To determine the actual temperature, you need a known reference temperature, such as an ice bath. |
|
|