An RTD (resistance temperature detector) is a temperature sensor that measures how much a metal’s electrical resistance changes as temperature rises or falls. A small electric current passes through a metal element, typically platinum, and the resulting voltage reveals the element’s resistance, which maps directly to temperature. RTDs are among the most accurate and stable temperature sensors available, with industrial models drifting less than 0.1°C per year.
How an RTD Works
The core principle is straightforward: metals conduct electricity slightly differently at different temperatures. As temperature increases, the atoms in a metal vibrate more, making it harder for electrons to flow. This raises the metal’s electrical resistance in a predictable, nearly linear way. By measuring that resistance precisely, you can calculate the temperature.
In practice, an RTD element receives a small, constant current from a measurement circuit. The circuit reads the voltage drop across the element, uses Ohm’s law to calculate resistance, and converts that resistance into a temperature reading. The relationship between resistance and temperature is characterized by a value called the temperature coefficient, which tells you how much resistance changes per degree. The most common standard is 0.00385 ohms per ohm per degree Celsius, meaning a 100-ohm sensor gains 0.385 ohms for every 1°C increase.
Platinum, Nickel, and Copper Elements
Platinum is the dominant RTD material because it stays chemically stable across a wide range, resists corrosion, and has a nearly linear resistance-to-temperature relationship. Platinum RTDs cover roughly -200°C to 660°C (-328°F to 1220°F), which suits the vast majority of industrial and laboratory applications. The international standard IEC 60751 is built entirely around platinum elements.
Nickel and copper RTDs exist but serve narrower roles. Nickel elements are more sensitive (their resistance changes faster per degree) but are limited to about 300°C and become nonlinear at higher temperatures. Copper elements work well in a moderate range up to around 260°C and are sometimes used in motor winding measurements because copper wire is already present. For most applications, platinum is the default choice.
PT100 vs. PT1000
The two most common platinum RTDs are the PT100 and the PT1000. The number refers to the element’s resistance at 0°C: 100 ohms for a PT100, 1,000 ohms for a PT1000.
PT100 sensors are the traditional industrial workhorse. They’re cost-effective, widely supported by measurement instruments, and perform reliably at higher temperatures. PT1000 sensors, with their tenfold higher baseline resistance, produce a larger voltage signal for the same measurement current. This makes them less affected by the resistance of the connecting wires, which is an advantage in systems with long cable runs or where simplicity matters. PT1000 sensors are common in HVAC systems and applications requiring precise control at lower temperatures.
Wire-Wound vs. Thin-Film Construction
RTD elements come in two main physical forms. Wire-wound RTDs use a fine platinum wire coiled around a glass or ceramic core. They’re the traditional design and offer excellent accuracy, especially at temperature extremes. Glass-core versions handle vibration reasonably well but lose accuracy above about 400°C as the glass properties shift. Ceramic-core versions resist high-temperature drift better but can crack under vibration. Wire-wound elements are roughly 25 mm (about 1 inch) long.
Thin-film RTDs deposit a thin layer of platinum onto a ceramic substrate, then laser-trim the pattern to achieve the correct resistance. They’re dramatically smaller, around 3 mm (about 1/8 inch), and handle vibration extremely well because of their low mass and lack of moving parts. That makes them ideal for compressors, bearings, and other equipment that shakes. The tradeoff is that thin-film elements can drift more at the extremes of their range, below -50°C or above 500°C. For the broad middle of the temperature spectrum, thin-film RTDs are increasingly the standard choice.
Wiring: 2-Wire, 3-Wire, and 4-Wire
Because RTDs measure resistance so precisely, even the small resistance of the copper wires connecting the sensor to the instrument can introduce errors. Different wiring configurations solve this problem to different degrees.
- 2-wire is the simplest and cheapest setup, but the instrument reads the wire resistance as part of the sensor resistance. On short cable runs this error is negligible. On longer runs, it can become significant. Some instruments let you enter the known wire resistance as an offset to compensate.
- 3-wire is the industrial standard. A third wire allows the instrument to measure the lead resistance and subtract it automatically. This works well as long as all three wires are the same length and gauge, which they almost always are in a standard cable. Minor inaccuracies can still creep in if wire resistances aren’t perfectly matched.
- 4-wire provides the highest accuracy by completely eliminating lead wire resistance from the measurement. Two wires carry the excitation current, and a separate pair measures the voltage directly across the sensor element. This is the configuration used in laboratory and precision applications where every fraction of a degree matters.
Accuracy Classes
The IEC 60751 standard defines tolerance classes that tell you how accurate a platinum RTD is at a given temperature. The two most commonly referenced are Class A and Class B.
Class A sensors have a tolerance of ±0.15°C at 0°C (±0.06 ohms) and are rated for use up to 650°C. Class B sensors allow ±0.3°C at 0°C (±0.12 ohms) and cover the full range up to 850°C. In both cases, the tolerance widens as you move further from 0°C. For everyday industrial process control, Class B is usually sufficient. For tighter requirements, like food safety monitoring, pharmaceutical production, or calibration work, Class A or better sensors are worth the investment.
Precision laboratory RTDs can achieve stability as fine as 0.0025°C per year, which is why the National Institute of Standards and Technology uses platinum RTDs as its primary reference instruments for temperatures between -260°C and 630°C.
Self-Heating Error
Every RTD needs a small current flowing through it to measure resistance, and that current generates a tiny amount of heat inside the sensor element. This is called self-heating, and it causes the sensor to read slightly warmer than the actual temperature of its surroundings. The effect is small, typically fractions of a millidegree for precision instruments, but it matters in high-accuracy applications.
The IEC 60751 standard recommends keeping the measurement current low enough that self-heating doesn’t exceed 25% of the sensor’s stated tolerance. In practice, this means using the lowest excitation current your instrument supports, especially in still air or poorly thermally coupled environments where heat can’t dissipate easily. Sensors immersed in flowing liquids dissipate self-heating much more effectively than sensors mounted in stagnant air.
RTDs Compared to Thermocouples and Thermistors
RTDs occupy a middle ground between thermocouples and thermistors. Thermocouples can handle much higher temperatures, up to 1,700°C with noble metal junctions, but they produce a very small voltage signal and require special extension wires plus reference junction compensation. RTDs generate a much larger output signal and use ordinary copper wires, which often makes the total system cost lower despite the sensor itself being more expensive.
Thermistors are far more sensitive than RTDs, with resistance values in the thousands of ohms that change dramatically with temperature. This makes them excellent for detecting tiny temperature shifts in a narrow range. But that useful range is limited, and thermistors are more susceptible to self-heating because of their high resistance. Basic thermistors cost less than RTDs as bare components, but once you factor in protective housings and tighter specifications, the price gap narrows.
If you need broad range and extreme heat tolerance, thermocouples win. If you need high sensitivity in a narrow band, thermistors are the better fit. For accurate, stable, repeatable measurements across the most common industrial and scientific temperature range, RTDs are the standard choice.