An RTD measures temperature by measuring resistance, which creates a problem the moment you run wires to it: the wires have resistance of their own, and the instrument cannot tell the difference between resistance in the platinum element and resistance in the copper leads. The wiring configuration - 2, 3, or 4 wires - is how that problem is solved. Choosing the right one is not a formality; on a long cable run from a wellsite RTD back to an RTU, the wrong choice can add a real, standing error to every temperature the loop ever reads.
3-Wire vs 4-Wire RTD in one line: 2-wire, 3-wire, and 4-wire RTD configurations differ in how they handle the resistance of the lead wires, which the instrument would otherwise add to the element resistance and read as extra temperature. A 2-wire RTD offers no compensation, a 3-wire RTD cancels most lead resistance by assuming the legs are matched, and a 4-wire RTD eliminates it entirely using separate current and sense wires.
An RTD transmitter works by passing a small, known current through the platinum element and measuring the voltage that develops across it; that voltage reveals the resistance, and the resistance reveals the temperature. The catch is that the same current also flows through the copper wires connecting the element to the transmitter, and those wires drop voltage too. Unless the wiring is arranged to account for it, the transmitter reads element resistance plus lead resistance, and it converts that inflated total into a temperature that is too high.
The size of the error depends on cable length, wire gauge, and the RTD's nominal resistance. A short jumper of heavy wire adds a negligible amount, but a long, thin field run can add ohms - and because a PT100 changes only about 0.385 ohms per degree, even one ohm of uncompensated lead resistance shows up as roughly two and a half degrees of standing offset. That is why the same physical element can read correctly on a bench and consistently high once it is wired hundreds of feet back to a panel.
The wiring configuration exists purely to attack this lead-resistance error. It does nothing for the accuracy of the element itself; it is entirely about making sure the transmitter measures the resistance of the platinum and not the resistance of the copper leading up to it. Understanding that framing makes the tradeoffs between two, three, and four wires straightforward.
A 2-wire RTD is the simplest and the least accurate. The element sits at the end of two wires, and the transmitter has no way to separate lead resistance from element resistance, so the full cable resistance is baked into the reading as error. Two-wire is acceptable only where the leads are very short or the accuracy requirement is loose, such as a coarse local indication a few feet from the element. It is generally a poor choice for anything run out to a field enclosure.
The 3-wire RTD is the workhorse of industrial temperature measurement and the usual default for field transmitters. It adds a third wire so the transmitter can measure the resistance of one lead leg and subtract it, on the assumption that all three wires are the same length, gauge, and temperature and therefore have equal resistance. This cancels the great majority of lead error with only one extra conductor, which is why the vast majority of process RTDs and RTU inputs are wired this way. Its one weakness is that the compensation is only as good as the assumption that the legs are truly matched.
The 4-wire RTD removes lead resistance entirely rather than assuming it away. It uses two wires to force the excitation current and two separate wires to sense the voltage right across the element - the true four-terminal, or Kelvin, sensing method. Because essentially no current flows in the sense wires, they drop no meaningful voltage, so the transmitter reads the element resistance directly regardless of lead length or mismatch. This is the most accurate configuration and the reason laboratory and calibration-grade measurements use it, at the cost of a fourth conductor and terminal.
For a typical remote oil and gas installation - a separator temperature, a heater outlet, a line temperature feeding an RTU - the practical answer is almost always a 3-wire RTD. It handles the long cable runs common on a wellpad or facility without the standing offset a 2-wire connection would carry, and it uses one fewer conductor than a 4-wire, which matters when multipair field cable is already crowded. Because the compensation relies on matched legs, the field practice is to use the same wire gauge for all three conductors and route them together so they stay at the same temperature.
Reserve 4-wire for the points where accuracy genuinely justifies the extra conductor: fiscal or allocation-related temperatures, tight energy balances, glycol dew-point work, or any measurement feeding a custody calculation where a fraction of a degree changes the number that gets billed. It is also the configuration to reach for when a run is unusually long or the leg-matching assumption behind 3-wire is hard to guarantee. Two-wire should be treated as the exception, used only for short, non-critical local readings.
The wiring choice matters to a cloud monitoring layer because it determines whether the historized temperature is trustworthy at its face value. When Merobix reads a temperature from a PLC or RTU and trends it, the platform is only as accurate as the transmitter feeding it, and that transmitter is only as accurate as its lead compensation. A point that reads a persistent couple of degrees high compared to a nearby reference is a classic signature of a 2-wire connection or a broken compensation leg, and recognizing that pattern from the trend can save a truck roll to chase a temperature that was never a real process shift.
In principle yes, because a 4-wire configuration eliminates lead-wire resistance completely rather than compensating for it under an assumption. In practice a well-installed 3-wire RTD with matched, evenly routed conductors is accurate enough for the vast majority of process measurements, so the 4-wire advantage only pays off where fractions of a degree matter, such as custody or energy-balance temperatures.
You can physically wire a 3-wire RTD to a 2-wire input by ignoring the third wire, but you lose the lead compensation and reintroduce the error the third wire was there to cancel. Wiring a 2-wire element to a 3-wire input generally requires a jumper at the element to create the compensation leg, and if the jumper is at the panel instead of the sensor it defeats the purpose. Always match the sensor configuration to how the transmitter expects to be wired.
An RTD measures temperature as a resistance, so any extra resistance in the wires adds directly to the measured value and reads as false temperature. A thermocouple instead generates a small voltage, and its accuracy is not sensitive to lead resistance in the same way - its wiring concern is reference-junction compensation, not lead resistance. That fundamental difference is why RTD wiring configuration is such a central topic.
Merobix reads your field devices into a cloud SCADA - the real thing behind these terms, live in days from any browser.