Maintaining a constant temperature around precision resistors is integral to maintaining the stability of a circuit. Even small fluctuations in the temperature of a resistor can cause errors in a circuit due to noise. The relationship between temperature and resistance, resistor self-heating, is well known. However the EMF effect, where a voltage is generated from a resistor, is less known, certainly when it comes to the world beyond that of analogue circuit designers.
Thermal EMF, also known as the Seebeck effect, occurs when two different materials, operating at different temperatures, are joined together. The temperature variation between the two conductors causes a voltage to be produced, or EMF (thermal electromotive force), which can be attributed to electrons flowing from the hot material to the cooler one. The resulting voltage is dependant on the materials joined and can vary from a couple of microvolts per degrees C (µV/ oC) to hundreds of µV/ oC.
This effect isn’t new, and is frequently utilised to measure temperature, a classic example would be entwining two different types of wire to then use as a sensor. Where it becomes less well documented is its relationship to resistors. If we think about the fact that a wire-wound resistor is essentially a series of wires linked together, a lead to a coil and another lead, all of them made from different materials, then suddenly it is easy to see. Precision resistors, such as Evenohm wirewound ones, can produce a couple of µV/ oC. But it’s not just wirewound resistors that are affected by this phenomenon.
Metal film resistors will produce into the tens of µV/ oC and certain non-precision carbon resistors are capable of generating even more, often creeping up to a few hundred µV/ oC. This leads to a DC voltage being introduced to the circuit that was never intended. Most of the time this voltage is too small to make