RTD & Thermocouple Wiring
The two dominant temperature sensors and why their wiring is unlike anything else — an RTD reads resistance, so the leads lie unless you compensate; a thermocouple makes its own millivolts, so the wire itself has to be the right metal.
Safety. This guide is educational reference material, not a work instruction. Electrical work is performed de-energized and verified by qualified personnel under your site’s LOTO procedures, following the device manufacturer’s manual and the authority having jurisdiction. Temperature sensors often sit on hot or pressurized process connections — treat the process side as hazardous until proven otherwise.
Overview
Industrial temperature measurement is dominated by two sensor types, and what makes each one’s wiring special is completely different:
- RTD (resistance temperature detector) — a precision resistor whose resistance tracks temperature (Pt100 = 100 Ω at 0 °C, Pt1000 = 1000 Ω). The instrument measures resistance, so the resistance of the lead wires adds to the reading unless it is compensated. Lead-resistance compensation is the entire wiring story.
- Thermocouple (TC) — two dissimilar metals joined at a measuring junction that generate a small millivolt signal proportional to the difference between that junction and a reference junction. That forces two things: the extension wire must be the matched thermocouple type, and the instrument must perform cold-junction compensation.
Both are low-level signals — resistance or millivolts — so noise, not current, is the enemy, which is why the shielding and separation rules matter as much as the connections. Three terminal groups recur:
- Sensor — the RTD element or TC junction in the process.
- Extension / lead — RTD lead wire (2, 3, or 4 conductors) or matched TC extension wire from the sensor head to the input.
- Input card / transmitter — the RTD/TC analog input module, or a head/rail-mounted transmitter that converts to 4–20 mA.
flowchart LR
SENS[Sensor<br/>RTD element / TC junction] --> LEAD[Lead / extension wire<br/>RTD leads · matched TC wire]
LEAD --> IN[Input card or transmitter<br/>lead comp / cold-junction comp]
This guide covers wiring one RTD or thermocouple point to its input. Terminal designations, the region-specific TC color codes, and card type selection are vendor- and standard-specific — they come from the device manual and the applicable color-code standard, never from a guide.
Before You Start
Have on hand before wiring:
- Sensor type and the matching input. Confirm the sensor — RTD (Pt100 / Pt1000) or thermocouple type (J, K, T, E, N, …) — and that the input card or transmitter is configured for exactly that type. A Type J sensor on a Type K input applies the wrong characteristic and reads a systematic error.
- RTD 2/3/4-wire decision. Decide the lead configuration up front — it determines the conductor count and the accuracy (see Control / Signal Wiring). Retrofitting from 2-wire to 3-wire means pulling another conductor.
- Transmitter at the sensor, or direct to the card? A head/rail transmitter converting to 4–20 mA at the sensor is often the better answer for long runs; a direct raw run to an RTD/TC card is simpler for short in-panel distances.
- Drawings and environment. Loop/instrument diagrams, cable route relative to power and VFD cabling, and process connection details are decided upstream.
Sizing & Protection
There are no conductor-ampacity tables here — these are signal circuits carrying resistance or millivolts, not power. The “sizing” that matters is signal integrity:
- RTD lead resistance is the sizing concern. On a 2-wire RTD the lead resistance adds directly to the element resistance and reads as extra temperature. Heavier gauge and shorter runs cut the error; 3-wire and 4-wire methods compensate it (below).
- TC extension gauge and length. Matched extension wire has a defined loop resistance; long runs and the input impedance affect the millivolt signal. On a long run, converting to 4–20 mA at the sensor sidesteps the problem.
- Noise, not current. Because these are low-level signals, the design target is keeping induced noise below the measurement resolution — there is no overcurrent sizing in the power sense.
Power Wiring
- The loop-powered transmitter option — often the better long-run answer. A head- or rail-mounted temperature transmitter can sit at the sensor and convert RTD resistance or TC millivolts to a robust 4–20 mA signal that shrugs off lead resistance and coupled noise, doing the cold-junction and lead compensation locally. For anything but short in-panel runs this is frequently preferable to running the raw sensor signal all the way to the card. The loop practice — loop-resistance budget, active/passive matching, single-point grounding — is covered in 4–20 mA current loop wiring.
- Torque discipline. Terminal torque and wire ranges are vendor values; record what you used.
Control / Signal Wiring
This is the core. Getting the RTD lead configuration and the thermocouple material/polarity right is the whole job.
RTD — 2-wire vs 3-wire vs 4-wire
| Configuration | How it handles lead resistance | Accuracy | Typical use |
|---|---|---|---|
| 2-wire | No compensation — lead resistance adds directly to the reading as error | Lowest | Short runs, low-accuracy points only |
| 3-wire | A third conductor lets the input measure and subtract the leads, assuming all leads are equal | Good — the industrial standard | The workhorse for most plant RTDs |
| 4-wire | True Kelvin sensing — a separate excitation pair and sense pair cancel lead resistance entirely, regardless of matching | Best | Traceable measurement, long or unequal leads |
- 2-wire — the lead resistance is read as extra temperature, an uncompensated error that grows with cable length. Short runs only.
- 3-wire — the industrial standard: compensates the lead resistance on the assumption that the leads are equal, so keep the three conductors the same gauge and length.
- 4-wire — best accuracy, lead resistance cancels regardless of matching; used where the point must be traceable or the leads are long.
Thermocouple
- Use matched extension wire — always. The extension must be the same thermocouple type as the sensor. Running plain copper wire from a thermocouple creates a second, unintended junction at the transition and injects error — the classic thermocouple mistake.
- Observe polarity, and mind the region. TC polarity matters (the two legs are different metals), and the color codes differ by region — the same color means different things under different code standards, and in several codes the negative leg is the distinguishing color. Consult the applicable color-code standard and the device manual; do not assume.
- Cold-junction compensation happens at the card / transmitter. The instrument measures its reference-junction temperature and corrects for it — but only if the matched extension wire carries the signal unbroken all the way to that point.
- Keep both away from power and VFD cabling. The millivolt/low-resistance signal is highly susceptible to coupled noise; segregation is not optional for a clean reading.
Grounding, Shielding & EMC
Device-specifics here; the deep treatment is owned by the noise & EMC mitigation guide and the shield-landing theory by panel grounding & bonding.
- Shield landed at ONE end only for these low-frequency, low-level signals. The shield drains capacitively coupled noise without becoming a conductor between two grounds; grounding it at both ends across any ground-potential difference carries 50/60 Hz hum straight into the signal. Same single-end policy as other low-frequency analog — the frequency reasoning is owned by grounding & bonding.
- Grounded vs ungrounded TC junction. A grounded junction (bonded to the sheath) responds faster but ties the signal to process ground — a ground-loop source; an ungrounded (isolated) junction is slower but floating, preferred where ground loops or isolation matter. Application-specific.
- Ground loops corrupt millivolt signals. Two ground references at different potentials drive circulating current that appears as an offset or noise on the tiny signal. Single-point grounding, an ungrounded junction, or an isolated input breaks the loop.
Common Mistakes
- Copper extension wire on a thermocouple instead of matched TC wire. The copper-to-TC transition forms an unintended junction; the error tracks the transition-point temperature and defeats the cold-junction compensation.
- Thermocouple polarity reversed. The reading moves the wrong way or shows a large offset — made easier to get wrong by region-varying color codes. Verify against the applicable code, not from memory.
- 2-wire RTD on a long run. Lead resistance adds directly to the reading: a steady positive temperature error that grows with cable length. Use 3-wire (or 4-wire) for anything but a short run.
- RTD/TC cable in the tray with VFD motor cable. PWM noise couples into the low-level signal; the giveaway is a reading that is noisy only while the drive runs. Segregate the routes.
- Shield grounded at both ends. The screen becomes a ground-loop conductor and injects a steady 50/60 Hz hum into the millivolt signal. Land the shield at one end only.
- Mixing TC types, or the wrong card type. A Type J sensor on a Type K card (or mismatched extension) applies the wrong characteristic — a systematic error that looks plausible until it is checked against a reference.
Verification Checks
Before handing the point to the process (evidence-retaining sheets in templates):
- Sensor type confirmed against the card/transmitter configuration (RTD Pt100/Pt1000, or TC type)
- Signal injection — inject a known resistance (RTD decade box) or millivolt value (TC source) at the field terminals and confirm the input reads the corresponding temperature, verifying the card independently of the sensor
- Known-temperature comparison against a reference thermometer or ice-point/bath at a known temperature
- Lead-resistance check (RTD) — leads within the configuration’s tolerance, especially before trusting a 2- or 3-wire point
- TC polarity confirmed against the applicable color-code standard
- Shield landed at one end only; low-level cable segregated from power/VFD routes per the plan
- Terminal torques per the device manual, recorded; completed loop sheet filed
Standards References
- IEC 60751 — industrial platinum RTDs: the standard characteristic (Pt100 etc.) and tolerance classes (category-level).
- IEC 60584 — thermocouples: TC types (J, K, T, E, N, R, S, B), reference tables and tolerances; region color codes per the applicable part (category-level).
- NFPA 79:2024 / IEC 60204-1:2016+AMD1:2021 — machine electrical wiring-practice chapters: conductor identification, routing, and separation of low-level signal from power (chapter-level).
- Device manuals and the applicable TC color-code standard are the authority for terminal designations, color codes, and card configuration.
Related Pages
- 4–20 mA current loop wiring — the transmitter output and where converting at the sensor pays off
- Panel grounding & bonding — shield-landing policy and the jobs of “ground”
- Noise & EMC mitigation — separation classes and victim hardening
- PLC I/O wiring — where the RTD/TC input card lives
- Commissioning templates — loop sheets and signal-injection records
This site is an independent educational reference for industrial automation standards. It paraphrases and organizes guidance in original language; it is not a substitute for the official published standards, professional engineering judgment, or legal review, and content has not been independently verified against current published editions.
Items marked Review pending await technical review against the identified edition. Items marked Planned have no reviewed source material behind them. Do not rely on this site for compliance determinations, safety-critical design decisions, or legal interpretation.