Is it hot in here, or is it just me? Regardless, let’s talk temperature sensor! This basic variable is necessary to a variety of processes and segments. In most cases, you need to monitor exact temperatures, and in some, you need precise control. When you learn about temperature sensor from the maintenance point of view, you’ll find that you only need to consider certain points to choose the right sensor for your application.
By now – unless you live under a rock, of course – you’ve probably heard about external temperature sensors. Vendors know that simple solutions sell, so they create complex devices to make temperature monitoring simple. Weird, huh? Anyway, these new devices pull data from your process, like the material and thickness of the pipe, ambient temperature, and more. Then they use special algorithms to calculate the right temperature for the pipes.
Unfortunately, these devices only monitor or control temperature in pipes. Right now we lack similar solutions for other temperature applications. In this article, we’ll identify and discuss the most common temperature sensors on the market. We’ll go over their basic concepts and builds, then give you examples of where you can use them.
If you have more information on a point that we missed or would like to share your experience with a sensor discussed here, feel free to comment below the post.
The most common types
Did you know that nearly every electronic device has a temperature sensor? Take your smartphone, for example. It probably uses a semiconductor-based sensor on its integrated circuits to monitor the temperatures your phone encounters.
You have a ton of different temperature sensors on the market, way too many to talk about in this article. But two in particular really stand out in most process applications, the resistance temperature detector (RTD) and the thermocouple. You’ve probably had contact with both of these temperature sensors at least once in your life.
From the others on the market, we’ll discuss a couple more, the infrared sensor and the bimetallic sensor. They have fewer applications in process automation, but you should know a little about them too. Let’s start with the RTD.
Resistance temperature detector (RTD)
This sensor has a well-earned reputation as one of the most accurate sensors available, providing good accuracy in a variety of applications. Beyond that, it’ll also give you excellent stability and repeatability. How does it do all that? Let’s dig in and find out.
This temperature sensor monitors temperature by detecting resistance in an electrical current. When the temperature changes, the resistance will change in consistent and measurable ways. Therefore, the sensor can translate these shifts into numbers we can read.
When you scale out an RTD, usually the vendor specifies the sensor according its resistance at zero degrees Celsius. On the market, a lot of sensors spec out at 100 ohms. That means that at zero degrees Celsius, the sensor will read a resistance of 100 ohms.
Types of RTDs
Okay, now you know more about RTDs and how they work. But when you scan the market, you see so many different types. How do these differences factor in? Well, let’s start with sensing elements, like platinum, nickel, and copper, the three most commonly used.
Most industries consider platinum the best element for RTDs because it offers stable resistance over a wide range of temperature. Nickel has a more limited range because it doesn’t offer a linear answer after 150 degrees Celsius.
Last but not least, we have copper. This material provides very linear resistance changes throughout the measurement range. However, you can’t use copper over 150 degrees Celsius because the sensor will oxidize.
You can also find different build categories of RTDs, like thin-film, wire-wound, and coil-element., the most common in the industries. For certain applications you need particular sensors, like carbon resistor elements for ultra-low levels of temperature measurement.
Two, three, and four wires
You’ve probably seen this option by now, too. More is better, right? Not so fast! That depends on what you need. Here we have the battle of cost versus accuracy.
When we talk about RTDs, we know the change in resistance indicates a proportional change in temperature value. So far, so good. Now here we have a small secret. A platinum temperature sensor is not completely built with platinum. Usually in a platinum sensor, the sensing element connects to the transmitter using a cable made of a different (cheaper) material, like copper.
Yes, indeed. The cable has a resistance value that can alter the value coming from the sensor element. And here we have the importance of the number of cables. These cables will compensate for the value of the resistance, reducing interference.
Two-wire RTDs won’t have this kind of compensation, so use a two-wire when you only need an approximate value for your application. Most field applications use three-wire RTDs. This kind of sensor uses the Wheatstone bridge circuit to compensate for the resistance shift in your transmitter. And of course, the four-wire RTD will eliminate the most voltage drop in your measurements, reducing its contribution to your error margin.
Now, let’s dive into the thermocouple universe! Industries around the world use this common solution to temperature measurement, but do you know how it works? If you do, bear with me while I explain to the rest of the audience.
A thermocouple uses two different metals to produce the phenomenon called “thermoelectric effect.” That means the sensor generates a voltage when the temperature differs from one end of the thermocouple to the other. The device then translates that voltage into – you guessed it – numbers we can read.
Now, for this kind of sensor you’ll need a reference table to interpret those numbers. The reference table will tell you the temperature depending on the voltage measured by your sensor, and each type of thermocouple on the market uses a different table. So make sure you use the right table for the thermocouple you have.
As I mentioned before, you have a wide range of thermocouples available. They differ in durability, temperature range, chemical resistance, vibration resistance, and compatibility. They also use letters as designations, like type K or R. Let’s check out the details of the most common thermocouples on the market.
Types of thermocouples
Thermocouples have more range of temperature measurement than RTDS and can cost up to three times less. However, if you need high accuracy and stability, then you need to stick with RTDs. If you don’t, then one of these may suit your application.
Built with nickel-chromium and nickel-aluminum, type K rules the roost because of its accuracy, reliability, and flexibility to cover a wide range of applications.
It has a range from -270 to 1260 degrees Celsius, and the extension wire covers 0 to 200 degrees Celsius. It also has an accuracy of +-0.75 percent and special limits of error (SLE) of +-0.4 percent.
Type J uses iron and constantan, and it has a smaller temperature range and shorter lifespan in high temperatures than type K. This temperature sensor grade has a range from -210 to 760 degrees Celsius and extension wires from 0 to 200 degrees Celsius. Standard accuracy hovers around 0.75 percent and SLE around 0.4 percent, like type K.
Type T mostly appears in low temperature measurement. It uses copper and constantan and has a range from -270 to 370 degrees Celsius, with extension wires from 0 to 200 degrees Celsius. The accuracy and SLE fall in the same ballpark as the first two, +-0.75 percent and +-0.4 percent, respectively.
This thermocouple offers better accuracy and signal quality compared to type K, and a good range of temperature measurement as well. Using nickel-chromium and constantan as its materials, this sensor ranges from -270 to 870 degrees Celsius, and the extension cable from 0 to 200 degrees Celsius. Although it has a similar SLE to the other three, it sports an accuracy of +-0.5 percent.
The N has a similar accuracy and temperature range to the K, although it has nicrosil and nisil for its materials, making it more expensive than a K. This grade supports a range from -270 to 1300 degrees Celsius, with the same extension cable as the others, 0 to 200 degrees Celsius. The accuracy is +-0.75 percent and SLE +-0.4 percent.
Type S supports a high temperature range with high accuracy and stability. Built from platinum and 10 percent rhodium, this grade can cover -50 to 1480 degrees Celsius and the extension wire 0 to 200 degrees Celsius. At an accuracy of 0.25 percent and SLE of 0.1 percent, this represents one of the most accurate sensors in our lineup.
The type R also measures high temperatures in different applications. It only differs from the type S in ratio of metals, at 13 percent rhodium instead of 10. This grade goes from -50 to 1480 degrees Celsius, with an accuracy of +-0.25 percent and SLE of 0.1 percent, just like the type S.
You can find plenty of other types of thermocouples on the market, if you want to check out some of the less common varieties.
Construction of the junctions on a thermocouple can also change its functions and features.
Grounded: This common junction style has the sheath and the thermocouple welded together to create one junction at the probe tip. It responds faster to temperature changes than ungrounded but can pick up transient noise on the circuit.
Ungrounded: This junction has mineral insulation, which protects it from transient noise but slows its response time.
Exposed: Welding the thermocouple wires together can allow you to insert the sensor directly into the process, increasing response time. However, this sensor can degrade or corrode quickly.
Ungrounded uncommon: This one has dual sensors insulated from each other by a sheath. It also insulates its elements from each other.
You’ve seen one of these devices in your daily life, even if you didn’t know it. Supermarkets usually have pyrometers to monitor the temperature of their freezers. An infrared sensor detects thermal radiation emitted by equipment or material. This device has the useful characteristic of non-contact, which means you can check temperatures from a distance.
How does it work? Basically, a lens inside the transmitter focuses thermal radiation onto a detector. The detector converts the radiant power to an electrical signal, and the transmitter will show on its display the temperature in the proper units.
Of course, you need to know the emissivity, or how much infrared energy your equipment or material can emit, to figure out the temperature. Therefore, the device has a database of materials and their emissivity values. It also compensates for ambient temperature in its reading.
Metals expand and contract with a change in temperature. Bimetallic devices rely on this property to measure temperature by converting the mechanical displacement into – yup! – numbers you can read.
The temperature sensor consists of a strip with two different metals that expand and contract at different rates when exposed to temperature changes, most commonly steel and copper. Usually built in the form of a spiral tube, the mechanical expansion of the materials result in rotation. One point of the bimetallic system remains stationary, while the other side rotates a pointer to indicate the temperature.
You can find so many more temperature sensor on the market, like silicon diodes, thermistors, and others. But for the daily activities of the instrumentation engineer, the most important are the RTD and thermocouple. Nevertheless, we plan to create more content later to explain other sensors out there.
If you’d like to discuss a specific sensor or share an experience with us, please leave a comment below or contact us at hello@visaya solutions.
Learn more about temperature sensor and transmitter:
TOP 5 temperature transmitters on the market – We’re up, yup yup, for another TOP 5! We did our first TOP 5 on fancy Coriolis flow meters, and we got some good traffic and comments on our social media. Some peeps got a little mixed up, confusing our opinion piece for research, but that happens. We roll with it.