What Is an RTD?

Back to Blog

How an RTD Measures Temperature

An RTD exploits a simple physical fact: the electrical resistance of a pure metal rises as it gets hotter. Platinum is the metal of choice because its resistance-versus-temperature curve is highly linear, stable over years, and repeatable between units. A Pt100 element reads 100 ohms at 0 degrees C; a Pt1000 reads 1,000 ohms and is favored where higher resistance reduces lead-wire error. To read the temperature, measurement electronics pass a small, precise current through the element and measure the resulting voltage to derive resistance, then convert that to temperature.

Because the element's resistance is being measured, the resistance of the connecting wires adds directly to the reading as an error - which is why RTD wiring is deliberately designed to cancel it.

2-Wire, 3-Wire, and 4-Wire Wiring

A 2-wire RTD is the simplest but least accurate: lead-wire resistance is added straight into the measurement and cannot be separated out, so it is used only on very short runs or where precision is not critical. A 3-wire RTD, the workhorse of industrial installations, uses a third lead so the electronics can measure and subtract most of the lead-wire resistance. A 4-wire RTD fully separates the current-carrying and voltage-sensing paths, eliminating lead-wire error almost entirely - reserved for laboratory-grade and custody measurement.

In practice, most field RTDs feed a nearby temperature transmitter that performs the lead-wire compensation and outputs a clean 4-20 mA or HART signal, keeping the sensitive resistance measurement short and local.

RTD vs Thermocouple

RTDs win on accuracy, stability, and linearity across the moderate temperature ranges typical of process measurement, which is why they dominate separators, heaters, and metering skids. Thermocouples win on temperature range (they read far hotter), ruggedness, faster response, and lower cost, making them the choice for flares, furnaces, and exhaust. The rule of thumb: reach for an RTD when accuracy matters and temperatures are moderate; reach for a thermocouple when it is very hot or a fast, cheap point is enough.

From Ohms to Degrees: A Worked Example

The IEC 60751 curve fixes the relationship between resistance and temperature, so an ohmmeter reading translates directly. A Pt100 reads 100 ohms at 0 degrees C and about 138.5 ohms at 100 degrees C per the standard table, which works out to roughly 0.385 ohms per degree over that span. So if a bench measurement across a disconnected element shows a value near 110 ohms, the element is sitting somewhere around 26 degrees C - a plausible workshop temperature, which is itself a useful sanity check that you are reading the element and not a fault.

The same arithmetic shows why the wiring configuration matters so much. Suppose each conductor of a long field cable contributes about one ohm - a symbolic figure; the real value depends on gauge and length. In a 2-wire hookup both conductors sit in series with the element, adding about two ohms and pushing the reading roughly five degrees high. A 3-wire circuit cancels most of that error and a 4-wire circuit removes it entirely. Tolerance verification works the same way: the IEC 60751 tolerance classes define how far a given class of element may deviate from the ideal curve, so a resistance check at a known temperature against the published tables is a quick pass/fail test.

Specifying an RTD Assembly

An industrial RTD is rarely a bare element; it is an assembly. The element itself comes wire-wound (excellent stability, more vibration-sensitive) or thin-film (rugged, compact, and now the common industrial default). Around it sit the sheath, whose material and diameter are chosen for the process, the insertion length, a spring-loaded fitting that presses the tip against the bottom of the bore, and a connection head or integral transmitter. For process work the element almost always lives inside a thermowell so it can be replaced without breaking containment.

1. Confirm the process temperature range, including startup and upset excursions.
2. Match sheath material and diameter to the well bore and the process fluid.
3. Set insertion length so the sensitive tip sits in flowing fluid, not in the nozzle.
4. Choose element type, tolerance class, and single or dual element.
5. Pick 3-wire or 4-wire hookup based on the accuracy the point deserves.
6. Decide integral versus remote transmitter and confirm hazardous-area certification.

Two specification choices save future outages. First, dual-element assemblies put two independent elements in one sheath, so a spare is already installed when one fails and the loop can be moved to the good element at the terminal block without pulling the well. Second, choose the tolerance class deliberately: a tighter class costs more and is only worth it where the measurement feeds custody transfer, efficiency calculations, or a tight product-quality limit. For a utility duty on a glycol loop the standard class is usually plenty, because transmitter trim and installation quality dominate the error budget anyway.

Failure Modes and What the Reading Tells You

RTD failures have signatures. A fully open element drives the input to its burnout state - most transmitters can be configured to fail the signal upscale or downscale, and the direction should be chosen as the safe one for that loop by the responsible engineer. Moisture ingress into the sheath or terminal head lowers insulation resistance and typically shunts the element, dragging the reading low and often making it wander with ambient conditions. Corroded terminals or a failing lead splice add series resistance and push the reading high. A value that jumps in steps points at an intermittent connection, frequently in a flexible conduit fitting or at the terminal block, not at the element itself.

Drift is the quiet failure. Mechanical shock and vibration slowly work-harden wire-wound elements, and repeated thermal cycling shifts thin-film elements; both show up as a growing offset against a reference measurement. The practical answer is periodic comparison - against the second element of a dual assembly, a test-well reading, or a calibrated reference. When a reading is suspect, a structured bench test of the RTD separates element faults from wiring and transmitter issues in a few minutes and avoids replacing good hardware on a guess.

Frequently Asked Questions