Each has their own benefit in terms of accuracy or cost.
Thermocouples are mechanically simple. They are formed by the junction of 2 different metals plus another reference junction.
A single junction by itself generates no voltage. The thermocouple generates a voltage by COMPARSION with the reference junction. So please note that the entire system (i.e. the junction, the cable and the reference junction) all take part in measuring temperature. (Today the reference junction is normally electronic, rather than an actual 2nd thermocouple.)
And because a junction of dissimilar metals is used to generate a voltage, every metal junction we add to the circuit will affect it. Hence a copper wire or ‘standard’ plug/socket connector if added will impact on accuracy.
Thermocouples are rugged. It’s possible to weld a thermocouple directly to a metal sheath! Thermocouples can be a simple bead for fast response and measurements such as air temperature, but more commonly they are mounted in stainless (or other) sheaths for protection of themselves and/or the product they are measuring.
Thermocouples come in many different types, based on the selection of dissimilar metals and can be used to measure up to a few thousand degrees in temperature. In New Zealand industry, Type-K thermocouples are the most common. These are made by a junction of Nickel/Chrome and Nickel/Aluminium.
Homershams manufacture or supply thermocouples of many types including K, J, T, S & R types. Please contact us for further data. One surprising thing about thermocouples is their colour codes which are usually determined by the country where the wire is manufactured. Note that in some cases the red cable is negative!
As mentioned above, thermocouples are manufactured in many different types. Normally your application will determine which thermocouple you must use.
The maximum temperature of a K-type thermocouple is around 1250 °C
If you’re working with molten steel or metal hardening, then the more exotic types S, R or B are used. These will handle up to 1750 °C.
Type T (for example) has a narrower temperature range than others (-200 to +370 °C) but also has a higher accuracy.
Accuracy of measurement by thermocouple is affected by a number of factors. E.g. the type of thermocouples used (as above), the grade of the cable (standard or high tolerance) and the environment it works in.
Predominantly Homershams stock Special Tolerance, high quality thermocouple wire. By way of example Standard Tolerance type K thermocouple wire has a specification of ± 2.2 °C or 0.75 % of reading (whichever is greater), while Special Tolerance is ± 1.1 °C or 0.4 % of reading (whichever is greater).
Good thermocouple wire makes a difference especially if running thermocouple wire through high temperature environments where using a good wire can be critical.
On the other hand in a passive environment one can use thermocouple EXTENSION wire which is lower cost, in room temperature environs.
The other way of dealing with long cables is to use a thermocouple transmitter, which converts the mV signal to a higher voltage or current signal. This is explained in more detail later in this article.
Homershams offer a large number of insulation materials on thermocouple wires. Predominantly the selection of insulation is made based on an environmental basis
Please refer to our web site or contact our sales team for options.
A bare thermocouple bead works well as an air temperature measurement device but usually we use a thermocouple bead encased in a metal sleeve (e.g. Stainless steel). This protects both the bead from damage and the product it’s inserted into from contamination.
Homershams and their partners manufacture custom probes to suit customer requirements.
RTD’s (Resistive Temperature Devices) are very popular in industry as they are linear, repeatable and accurate sensors. They consist of a rare metal wire wound on a ceramic former. The resistance of the wire changes with temperature in an almost exactly linear fashion.
The only disadvantage of them is cost (rare metals are not cheap!) and the fact that they are relatively fragile as the ceramic former may crack if dropped. RTD’s are also susceptible to chemical attack so are almost always supplied in a protective sheath. The other limiting factor with RTD’s is maximum temperature. Typically RTD’s are used up to about 500 °C preferably less than 400 °C. The most common RTD type used in NZ are Pt100, being Platinum wire with a resistance of 100 ohms at 0 °C.
Thermistors are a type of resistance thermometer, but they work based on a semiconductor junction. They are like a transistor. Most commonly they have a negative coeffienct of temperature (NTC) such that resistance reduces with increased temperature.
They are faster in response than a RTD, but slightly slower than a Thermocouple. They operate accurately in a narrow temperature band. typically −90 °C to 130 °C1.
Thermocouple wire can run several metres, and a Pt100 a few 10’s of metres, so why use a transmitter - why not just run the temperature sensor direct to the controller etc?
Primarily the answers are cost, reliability and noise.
Thermocouple wire is expensive. As mentioned above, one cannot use copper wire to connect a thermocouple. Furthermore the thermocouple junction, the wire and the connection to the sensing device as well as the entire run of thermocouple wire and the environment it passes through may impact on accuracy.
With a Pt100 we may run simple copper wire, so the situation is a bit simpler, but consider that the signal running through the copper is very low. This makes it very sensitive to external interference and noise. This is also true of the thermocouple.
A transmitter may be the solution.
In a series circuit the current is always the same no matter where you measure, unlike a parallel one. If a series circuit is created with a 24 volt voltage source, the transmitter and receiver, any wire-effects will be voltage related, not current. E.g. if we create a circuit where 4 mA represents 0 % and 20 mA represents 100 %, this will be the current throughout the loop where ever we measure it (and will be unaffected by cable resistance or other impedances).
Now we have a much larger, always dependable current flowing through our cables compared to the tiny resistance (or mV) signal, and we have much better resilience against external noise.
e.g. The transmitter is effectively an electronic ‘resistor’ of between 6,000 and 12,000 ohms (24 v/6000 ohms = 20mA, 24v/12,000 ohms = 4mA) producing this linear resistance from the non-linear mV output of the thermocouple.
For help and advice at any time contact our sales team.