A PT100 sensor measures temperature by tracking how the electrical resistance of a platinum element changes as it gets hotter or colder. At exactly 0°C, the platinum element has a resistance of 100 ohms, which is where the “100” in PT100 comes from. As temperature rises, resistance increases in a predictable, nearly linear way. A connected instrument measures that resistance, converts it using a known formula, and displays a temperature reading.
The relationship between platinum and temperature is so stable and repeatable that PT100 sensors are one of the most trusted temperature measurement tools in industrial and laboratory settings, covering a range from roughly -200°C to +850°C depending on construction.
The Core Principle: Resistance Changes With Temperature
Platinum atoms vibrate more as they get hotter. Those vibrations make it harder for electrical current to pass through, which increases the metal’s resistance. This property is incredibly consistent in high-purity platinum, meaning the same temperature always produces the same resistance value. That consistency is what makes platinum the preferred metal for precision temperature sensing over alternatives like nickel or copper.
The standard temperature coefficient for a PT100 is 0.00385 ohms per ohm per degree Celsius. In practical terms, the element’s resistance changes by about 0.39 ohms for every 1°C change in temperature. So at 100°C, a PT100 reads approximately 138.5 ohms. At -100°C, it reads roughly 60.3 ohms. An instrument sends a small, known current through the element, measures the voltage drop across it, calculates resistance from that, and then converts resistance to a temperature value.
How the Math Works
The conversion from resistance to temperature uses an equation called the Callendar-Van Dusen formula. For temperatures above 0°C, it’s relatively simple: resistance equals the base resistance at 0°C multiplied by a polynomial that includes the temperature and two coefficients (commonly labeled A and B). Below 0°C, a third coefficient (C) kicks in to account for platinum’s slightly different behavior at colder temperatures.
You don’t need to solve this equation yourself. Your transmitter, PLC, or data acquisition system has these coefficients built in and handles the conversion automatically. But understanding that the relationship isn’t perfectly linear, and that this equation corrects for that curvature, explains why PT100 sensors maintain accuracy across such a wide range.
Two Main Construction Types
Wire Wound Elements
The traditional design wraps an ultra-fine platinum wire, typically 99.99% pure, around a ceramic or glass core. The winding pattern maximizes the wire’s surface area exposure while keeping the element compact. These sensors handle extreme temperatures, rated from -200°C to +850°C, and deliver the highest accuracy because the platinum is in its purest bulk form with minimal mechanical stress on the wire.
Thin Film Elements
The more modern approach deposits a microscopically thin layer of platinum onto a flat ceramic substrate using vapor deposition, then laser-etches a precise circuit pattern into the film. This manufacturing process is cheaper and more automated, producing smaller, faster-responding sensors. The tradeoff is a narrower operating range, typically -70°C to +500°C. Thin film elements dominate in HVAC, automotive, and general industrial applications where extreme temperatures aren’t involved.
Wiring Configurations and Lead Resistance
Here’s a practical problem: the wires connecting your PT100 element to the instrument also have electrical resistance. Since the instrument measures total circuit resistance, long cable runs or thin wires can add enough extra resistance to throw off your temperature reading. PT100 sensors use three different wiring schemes to deal with this.
A 2-wire connection is the simplest, with one wire on each end of the element. It offers no compensation for lead wire resistance, so it’s only suitable for short cable runs where the added error is acceptable. A 3-wire connection adds a third lead to one side of the element. The instrument assumes both legs have equal resistance, measures the resistance in the third wire, and subtracts it from the total. This cancels out most lead wire error, as long as all three wires are the same length and gauge. It’s the most common configuration in industrial installations.
A 4-wire connection fully eliminates lead wire error. Two wires carry the excitation current through the element, and two separate wires measure the voltage drop directly across the element itself. Because the voltage measurement wires carry virtually no current, their own resistance doesn’t affect the reading. This is the preferred method for laboratory-grade and high-precision measurements.
Accuracy Classes
The international standard IEC 60751 defines tolerance classes that tell you how accurate a PT100 sensor is at any given temperature. The tolerance isn’t a fixed number. It’s a formula that increases with temperature, meaning sensors are most accurate near 0°C and less accurate at extremes.
- Class AA: ±(0.1 + 0.0017 × |t|)°C. At 100°C, that’s ±0.27°C. Valid from -50°C to +250°C for wire wound elements.
- Class A: ±(0.15 + 0.002 × |t|)°C. At 100°C, that’s ±0.35°C. Valid from -100°C to +450°C for wire wound elements.
- Class B: ±(0.3 + 0.005 × |t|)°C. At 100°C, that’s ±0.8°C. Valid from -196°C to +600°C for wire wound elements.
Thin film elements have slightly narrower valid temperature ranges within each class. Class B is the default for most industrial sensors. If you need tighter accuracy, you’ll need to specifically order Class A or AA and pair it with appropriately precise instrumentation.
Self-Heating: A Hidden Error Source
To measure resistance, the instrument pushes a small current through the platinum element. That current generates heat, just like any electrical current flowing through a resistor. This self-heating raises the element’s temperature slightly above the actual process temperature, creating a measurement error.
The effect is small but real. Research shows that the difference between operating at 2 mA and 0.5 mA excitation current produces roughly 0.0075°C of self-heating error. The general guideline is to keep power dissipation in the element below 0.01 watts, which keeps self-heating error under 0.2°C. Most modern transmitters and signal conditioners use excitation currents in the range of 0.5 to 1 mA specifically to minimize this effect. Self-heating becomes more significant when the sensor is mounted in still air rather than a flowing liquid, because air is much worse at carrying away the excess heat.
PT100 vs. PT1000
A PT1000 sensor works on the exact same principle but uses 1,000 ohms as its base resistance at 0°C instead of 100 ohms. This gives it 10 times the sensitivity: 3.9 ohms per degree Celsius compared to 0.39 ohms for a PT100. That higher resistance change per degree makes the signal easier to read and reduces the relative impact of lead wire resistance.
PT1000 sensors are particularly useful in 2-wire installations or applications with long cable runs, where the lead wire resistance would cause unacceptable error in a PT100. If your cable adds 2 ohms of lead resistance, that represents a 2% error on a 100-ohm PT100 but only 0.2% on a 1,000-ohm PT1000. The tradeoff is that PT100 remains far more common in industrial infrastructure, so transmitters, PLCs, and control systems are more universally compatible with PT100 inputs.