Types of Thermometers

Types of Thermometers

There are different types of thermometers. Examples are: mercury in glass thermometer, resistance thermometer, and thermocouple thermometer. All of them are based on the same idea: A property of the material changes with temperature, and we tell the temperature by measuring this property. Such a property is called a thermometric property.

For the mercury in glass thermometer, it is the length of the mercury thread that changes. For the resistance thermometer, it is the resistance of a piece of wire that changes. For the thermocouple thermometer, it is the voltage produced that changes.

I have previously described the glass thermometer. The resistance thermometer works by using an electrical circuit to measure the resistance of a piece of wire, usually platinum. The resistances can be measured at ice point and at steam point. Using these two resistances, we can plot two points on a graph of resistance versus temperature, and draw a straight line through the two points. If we now place the wire at an unknown temperature and measure a new resistance, we can then use the graph to find this temperature.

The procedure for using the resistance thermometer seems to be more involved that for the glass thermometer. In fact, they are based on the same idea. For the glass thermometer, we have markings next to the glass tube to read off the temperature. Instead of using these markings, we can measure the length of the mercury thread at ice and steam points, and plot two points on a graph of length against temperature. Then we can use this graph for new temperatures, as with the resistance thermometer. The procedure for using thermometric property to measure temperature, whether it is using the graph or making the markings, is called calibration.

A thermocouple thermometer is made up of two pieces of wire of different metals, e.g. iron and copper. They are electrically connected to each other at both ends. One of the wires have to be cut and a voltmeter inserted. The voltmeter would measure a voltage. This voltage depends on the difference in temperatures at the two junctions. One end is placed at a fixed temperature, such as room temperature. The other end is placed at the temperature we want to measure. As with the resistance thermometer, we need to do calibration first. We can place this end ice and in steam to measure the voltages, then plot two points on a voltage-temperature graph. If we draw a straight line through the two points, we can now using this graph to measure unknown temperatures. Just put the junction where we need the temperature, read the voltage, and check the graph.

You may wonder why we bother with resistance or thermocouple, when it is so much easier to use a glass thermometer. Indeed, we would only use them if we cannot use the glass thermometer. This happens if we need to measure temperatures that are well below ice point when mercury could freeze, or way above steam point when mercury could vaporise.

How do we choose between resistance and thermocouple then? I would go for thermocouple as a first choice, because the circuit needed to measure the resistance accurately can be fairly complex, compared to just inserting a voltmeter in the thermocouple. The thermocouple also responses a lot more quickly, because only the point at the junction has to be warmed (or cooled) to the temperature required. Resistance thermometer is much slower, because the whole piece of wire has to be warmed.

So when do we use the resistance thermometer? We use it when we need to know the temperature more accurately. For one time measurements, a resistance thermometer made from platinum is only slightly more accurate than a thermocouple. The thermocouple readings, however, can change by a few degrees Celsius after a year, whereas the resistance reading would remain nearly the same.

The Physics Notes