Temperature measurement

Temperature Measurement in Industrial Plants:

RTDs, Thermocouples, and Field-Proven Fault Tracing

Temperature is one of the first process variables any instrumentation technician learns to measure, and it's also one of the most misunderstood when it comes to why a reading goes wrong in the field. This article covers the fundamentals briefly, then goes deep into what actually matters on a running plant: choosing between RTD and thermocouple, and troubleshooting the failures you'll actually encounter ,not the ones in the textbook.

What Is Temperature, and Why It Matters in Industry

Temperature is the degree of hotness or coldness of a substance, driven by the kinetic energy of its molecules — faster molecular movement means higher temperature, slower movement means lower temperature. At absolute zero, molecular movement theoretically stops entirely, which is why 0 Kelvin (-273.15°C) is the lower bound of every temperature scale.
In daily life this matters for comfort and health. In industrial processes, it matters far more directly: most process reactions, material properties, and equipment tolerances are temperature-dependent. In a cement plant specifically, kiln shell temperature, pre heater cyclone temperatures, cooler grate temperatures, and bearing temperatures on rotating equipment are all monitored continuously — not as a formality, but because an undetected temperature excursion can mean coating buildup, refractory damage, bearing seizure, or in the worst case, a safety incident. Uncontrolled temperature is one of the most common root causes of structural deterioration in pipelines and vessels and premature failure of motor and gearbox bearings.

Temperature Scales and Conversion

The four scales you'll encounter are Celsius, Fahrenheit, Kelvin, and Rankine. The conversions worth keeping handy:
Fahrenheit to Celsius: °C = (°F − 32) / 1.8
Celsius to Kelvin: K = °C + 273.15
Celsius to Fahrenheit: °F = (°C × 1.8) + 32

Temperature Sensing Technologies:

A Quick Overview

The four main device categories used industrially are thermometers (filled-system and bimetallic), RTDs, thermocouples, and thermostats. Filled thermometers work by sensing pressure or volume change in a liquid or evaporating fluid as it heats, displayed on a local gauge — still common for local indication on tanks and lines where a transmitter isn't needed. Bimetallic strips use two metals with different thermal expansion coefficients bonded together; as temperature rises, the strip bends toward the metal with the lower expansion coefficient, and that mechanical movement is used directly for switching — this is why you'll find bimetallic elements inside simple thermostats and protective devices rather than in precision process measurement.

For process control and monitoring where accuracy and remote transmission matter, RTDs and thermocouples dominate, and choosing correctly between them is one of the more consequential decisions you'll make when specifying a new temperature point.

RTDs: Construction and Working Principle

A Resistance Temperature Detector measures temperature through the predictable change in electrical resistance of a conductor as its temperature changes. The relationship is close to linear across the working range, which is the main reason RTDs are preferred wherever precision matters.

Materials. Platinum, nickel, copper, and molybdenum are all usable, but platinum dominates industrial use because it offers the best combination of stability, linearity, and resistance to chemical attack across a wide range. A good RTD material needs three things: stability (it shouldn't change its electrical characteristics under the operating environment), high resistivity (so less material is needed, keeping the sensing element small), and linearity (resistance should track temperature as predictably as possible across the measurement span).

Pt100 and Pt1000. The Pt100 is the industry standard — 100 ohms resistance at 0°C, rising by approximately 0.385 ohms per °C, giving roughly a 2.6°C change per ohm. Temperature from resistance is calculated as:

°C = (measured resistance − 100) / 0.385
Pt1000 sensors work identically but read 1000 ohms at 0°C, giving a stronger signal that's less affected by lead wire resistance — worth considering for long cable runs where a Pt100's small resistance change can get lost in wiring resistance error.

Construction types.

Wire-wound RTDs are more accurate and more expensive, with the platinum wire wound around a ceramic or glass core. Thin-film RTDs deposit a platinum or metal-glass slurry film onto a flat ceramic substrate, then laser-trim it for accuracy — these are smaller, cheaper, and respond faster to step changes, but are generally less stable long-term than wire-wound construction. For most general process monitoring, film types are perfectly adequate; for critical, long-term stable measurement points, wire-wound is worth the extra cost.

2-wire, 3-wire, and 4-wire connections.

A 2-wire RTD is the simplest but least accurate configuration — the resistance of the lead wires themselves adds directly to the measured resistance, and this error grows with cable length and ambient temperature swings along the cable run. A 3-wire configuration uses matched-length, matched-resistance lead wires arranged so the Wheatstone bridge measuring circuit cancels out the lead resistance, which is why 3-wire is the standard for most industrial RTD installations. 4-wire RTDs go a step further, using a constant current source and separate sense wires so lead resistance is eliminated almost entirely — this is the configuration to specify for lab-grade or custody-transfer-critical measurements, though it's rarely necessary for routine plant monitoring.

Thermocouples: The Other Major Option

A thermocouple works on an entirely different principle — the Seebeck effect. Two dissimilar metal wires are joined at one end (the measuring junction); when a temperature difference exists between that junction and the reference junction, a small voltage is generated, proportional to the temperature difference. Common types include Type K (chromel-alumel, general purpose, wide range), Type J (iron-constantan, lower cost but limited upper range and prone to oxidation), and Type R/S (platinum-rhodium, for very high temperature applications like kiln burning zone measurement).

RTD vs Thermocouple: Choosing Correctly

This decision comes up constantly when specifying new instrumentation, and getting it wrong means either overpaying for precision you don't need or under-specifying a point that actually needs better accuracy.

RTD

Accuracy
Higher, better long-term stability

Range
Typically -200°C to 650°C (Pt100)

Response time
Slower

Point sensing
Poor — needs a sensing mass

Cost
Higher, more wiring (3/4-wire)

Vibration sensitivity
More sensitive

Thermocouple

Accuracy
Lower, more prone to drift

Range
Very wide — Type R/S usable beyond 1600°C

Response time
Faster

Point sensing
Excellent — very small junction

Cost
Lower, simple 2-wire

Vibration sensitivity
More rugged

In a cement plant, this plays out predictably: RTDs go on bearing temperature monitoring, cooling water lines, and any measurement point where long-term stability and accuracy matter more than response speed. Thermocouples — usually Type R or S — go on kiln burning zone and preheater cyclone measurement where temperatures exceed the RTD's practical range and where the robustness of a thermocouple junction matters more than the last degree of accuracy.

Where RTDs Actually Fail in the Field

This is the part most tutorials skip entirely, and it's the part that actually matters when you're the one troubleshooting a bad reading at 2 AM.
Lead wire compensation errors. If a 3-wire RTD's lead wires aren't genuinely matched in length and resistance — which happens more often than you'd expect after cable repairs or extensions done without matching the third leg — the bridge circuit can't fully cancel the lead resistance, and you get a temperature offset that looks exactly like sensor drift. Before condemning an RTD that's reading consistently high or low, check all three (or four) leads for continuity and approximately equal resistance.

Junction box moisture ingress.

Cement plant environments are hard on any field wiring, and RTD junction boxes are no exception. Moisture ingress at the terminal block introduces a parallel resistance path that corrupts the reading — usually intermittently, which makes it maddening to diagnose because the fault appears and disappears with humidity and temperature. If a reading is erratic rather than steadily offset, check the junction box gasket and cable gland before suspecting the element.

Self-heating error.

The measuring current passed through an RTD to sense its resistance also heats it slightly. This is usually negligible, but on RTDs installed in low-flow or stagnant fluid (as opposed to a fast-moving process stream that carries the heat away), self-heating can introduce a small but real positive bias. If you're chasing a persistent small offset on a low-flow application, this is worth ruling out.

Thermo well response lag.

An RTD or thermocouple installed in a thermowell for process protection reads the thermowell's temperature, not the process fluid's temperature directly — and a heavy-gauge thermowell has real thermal mass, so during fast temperature transients the transmitter output lags the actual process by longer than expected. This isn't a fault, but it needs to be understood before someone chases a "slow-responding transmitter" that's actually just doing what a thermowell of that mass will always do.

Element damage from vibration.

RTDs, especially wire-wound types, are more vibration-sensitive than thermocouples. On equipment with continuous vibration — near crushers, mills, or high-speed rotating machinery — a wire-wound element can develop intermittent opens or resistance shifts over time from fatigue. If a reading is unstable specifically during equipment operation but stable when the equipment is stopped, suspect mechanical fatigue in the element rather than an electronics fault.

Practical Fault-Tracing Checklist

When a temperature reading looks wrong, work through this sequence before condemning the sensor:
Check lead wire continuity and matched resistance (3-wire) or verify the current source connections (4-wire)
Inspect the junction box for moisture, corrosion, or loose terminals
Verify the transmitter's configured sensor type and wiring configuration actually matches what's installed — a 2-wire RTD wired into a 3-wire configured transmitter will read consistently wrong
Compare the reading against a calibrated reference probe at the same physical point, not just at the transmitter output
Check for self-heating bias on low-flow or stagnant applications
If the reading is unstable specifically under vibration or equipment operation, suspect element fatigue

Takeaway

RTDs and thermocouples both do the same fundamental job — converting temperature into an electrical signal — but they fail differently, and knowing which technology you're dealing with tells you where to look first. Most "sensor faults" reported against RTDs in the field are actually wiring, moisture, or configuration issues rather than a failed element, and most thermocouple complaints trace back to junction degradation or reference compensation problems rather than the wire itself. Understanding the mechanism behind the number, not just the number itself, is what turns troubleshooting from guesswork into a quick, confident diagnosis..

industrial instrumentation basics

Search This Blog