Both RTDs and thermocouples turn temperature into an electrical signal, but they fail and shine in opposite places — so the real question is never “which is better”, it is “which fits this point”. This is a selection guide: pick by range, accuracy, speed, and budget. If you just want the plain working-principle difference, our RTD vs thermocouple overview covers that, while the matrix below tells you which to actually specify.

Contents

• RTD vs thermocouple: the short answer
• How each sensor works
• Accuracy and stability compared
• Temperature range compared
• Response time, ruggedness and wiring
• Cost compared
• Decision matrix: which to choose by application
• Featured temperature sensors
• Frequently asked questions

RTD vs thermocouple: the short answer

For most process points below about 600 °C where accuracy and long-term stability matter, choose an RTD (Pt100 or Pt1000). Reach for a thermocouple when you need very high temperature, the fastest response, the most rugged sensor, or the lowest cost per point. Almost every other difference between the two flows from that trade-off, so anchor your decision there first.

How each sensor works

An RTD measures temperature from the resistance of a pure metal element — a platinum Pt100 reads 100 Ω at 0 °C and rises along a precise, repeatable curve. A thermocouple instead generates a small voltage at the junction of two dissimilar metals (the Seebeck effect); that voltage tracks the temperature difference between the measuring junction and a reference (cold) junction. One reads resistance, the other reads millivolts — which is why they need different wiring and signal handling.

Accuracy and stability compared

RTDs are the more accurate and stable choice. A Class A Pt100 is specified to about ±0.15 °C at 0 °C, and good platinum elements drift very little over years. A base-metal thermocouple is typically ±1.0–2.2 °C, or roughly ±0.4–0.75% of reading, and it drifts more as the junction ages, oxidises, or sees thermal cycling. If repeatable, traceable accuracy is the priority — pharma, food processing, lab, custody — the RTD or a standard reference sensor is the safer pick.

Temperature range compared

This is where thermocouples win decisively. Thermocouples cover roughly -250 °C to +1800 °C depending on type — Type K to about 1260 °C, and noble-metal Types S, R and B higher still. RTDs are usually rated -200 °C to +600 °C, with some specials to +850 °C. Above about 600 °C — furnaces, kilns, boilers and heat treatment — a thermocouple is effectively the only practical option.

Response time, ruggedness and wiring

A thermocouple’s tiny junction has very low thermal mass, so it responds faster and tolerates shock and vibration better — useful for transient or rough service. An RTD’s element is more delicate and a little slower, but you trade that for stability. Wiring differs too: an RTD should use a 3- or 4-wire connection so lead resistance is cancelled (a 2-wire RTD adds the cable’s resistance straight onto the reading), while a thermocouple needs matching extension wire and cold-junction compensation. Many points are easiest to standardise by feeding either sensor into a temperature transmitter that outputs 4–20 mA.

Cost compared

Per point, thermocouples are cheaper — commonly 2.5–3× lower installed cost than an equivalent RTD assembly, and their extension wire is inexpensive. RTDs cost more up front but can be the lower lifetime cost where their stability avoids re-calibration and off-spec product. When you have many non-critical points, thermocouples keep the budget down; for a few critical measurements, the RTD premium is easy to justify.

Decision matrix: which to choose by application

Featured temperature sensors

Three Sino-Inst options covering both technologies and the transmitter that conditions either signal:

Pt100 RTD (Thermal Resistance)

Platinum Pt100/Pt1000 RTD assemblies for accurate, stable measurement from -200 to +600 °C, in 3- or 4-wire configurations.

Platinum-Rhodium Thermocouple

Noble-metal Type S/R/B thermocouples for high-temperature furnace, kiln and heat-treatment service up to ~1700 °C.

Temperature Transmitter

Head- or rail-mount transmitter that accepts RTD or thermocouple input and outputs 4–20 mA/HART for clean, long-distance signals.

Frequently asked questions

Which is more accurate, an RTD or a thermocouple?

An RTD. A Class A Pt100 is about ±0.15 °C at 0 °C with very low drift, while a base-metal thermocouple is typically ±1–2.2 °C or ±0.4–0.75% of reading and drifts more over time. For high-accuracy work below ~600 °C the RTD is the better choice.

When should I use a thermocouple instead of an RTD?

Use a thermocouple when the temperature exceeds about 600 °C, when you need the fastest response or the toughest sensor for shock and vibration, or when cost per point matters across many measurements.

What temperature range can each sensor measure?

Thermocouples cover roughly -250 to +1800 °C depending on type; RTDs are usually -200 to +600 °C, with some specials to +850 °C. Above ~600 °C the thermocouple is effectively the only option.

Why does an RTD need 3 or 4 wires?

Because the meter reads resistance, the resistance of the lead wires adds directly to the measurement. A 3- or 4-wire connection lets the instrument cancel that lead resistance; a 2-wire RTD includes cable resistance as an error.

Is an RTD or thermocouple more expensive?

Thermocouples are cheaper per point — often 2.5–3× lower installed cost — and use inexpensive extension wire. RTDs cost more up front but can have lower lifetime cost where their stability avoids re-calibration.

About this article

Written and technically reviewed by the Sino-Inst engineering team — last reviewed 2026-06-02 (AI-assisted drafting). Based on IEC 60751 (Pt100 RTD) and IEC 60584 (thermocouple) tolerance classes plus field selection experience across process plants. Questions? reach our application engineers.

Request a Quote

KimGuo11

Wu Peng, born in 1980, is a highly respected and accomplished male engineer with extensive experience in the field of automation. With over 20 years of industry experience, Wu has made significant contributions to both academia and engineering projects.

Throughout his career, Wu Peng has participated in numerous national and international engineering projects. Some of his most notable projects include the development of an intelligent control system for oil refineries, the design of a cutting-edge distributed control system for petrochemical plants, and the optimization of control algorithms for natural gas pipelines.

drurylandetheatre.com

Updated 2026-06-01 by the Sino-Inst Engineering Team

A temperature sensor transmitter is worth fitting only when distance or electrical noise would wreck a raw sensor signal. Run a bare Pt100 a few metres to a nearby panel and you rarely need one. Run it 150 metres past variable-speed drives and the reading drifts, picks up hum, and loses accuracy to lead resistance. This guide shows what a temperature transmitter actually does, when you need one instead of a plain sensor, and the 3-wire wiring detail that quietly biases readings if you get it wrong.

Contents

• What is a temperature sensor transmitter?
• How does a temperature transmitter work?
• RTD or thermocouple input — which should you transmit?
• Do you even need a transmitter? A distance and noise decision table
• 2-wire vs 3-wire Pt100 wiring (and the mistake that biases every reading)
• Choosing the output: 4-20 mA, HART, or RS485
• Key specs to match to your process
• Related temperature products
• Frequently asked questions

What Is a Temperature Sensor Transmitter?

A temperature sensor transmitter is a sensor paired with a small electronic transmitter that converts the sensor’s weak, non-linear signal into a strong, standardized output — most often 4–20 mA. The sensor is the part in contact with the process: a thermocouple or a resistance temperature detector (RTD). The transmitter is the electronics that conditions and rescales that signal so a control system can read it reliably over long cable runs.

People mix up three terms and order the wrong part because of it. The sensor produces millivolts or a resistance change. The transmitter turns that into 4–20 mA. A temperature transmitter assembly is the two together, usually with the electronics in a head-mount module that screws into the sensor’s connection head. If you ask a supplier for a “temperature transmitter” but mean a bare thermocouple, you will get the wrong hardware. Decide first whether you need the raw sensor or the sensor-plus-electronics package.

How Does a Temperature Transmitter Work?

The transmitter does four jobs in sequence: isolate, amplify, linearize, and rescale. It isolates the sensor from ground loops, amplifies a signal measured in millivolts or ohms, linearizes it against the sensor’s known curve, and maps the result onto a 4–20 mA loop. With an RTD, a small excitation current is passed through the element and the voltage across it is read on a bridge, then amplified. With a thermocouple, the transmitter also performs cold-junction compensation so the millivolt reading reflects the true process temperature, not the terminal temperature.

The 4–20 mA scaling is the part to internalize. You set a lower range value (LRV) and an upper range value (URV). The transmitter draws 4 mA at the LRV and 20 mA at the URV, linearly in between. Configure a range of 0–100 °C and 4 mA means 0 °C, 12 mA means 50 °C, 20 mA means 100 °C. Because the live zero is 4 mA, a broken wire reads 0 mA and is instantly distinguishable from a real low temperature — one reason current loops beat raw voltage for field work.

RTD or Thermocouple Input — Which Should You Transmit?

The input sensor decides accuracy and range before the transmitter ever touches the signal. A Pt100 RTD is the right default for process temperatures up to about 600 °C: it is accurate, stable, and repeatable. A thermocouple covers the high end and fast transients a wire RTD cannot reach. Match the sensor to the duty, then add a transmitter that accepts that input.

A Pt100 follows IEC 60751: 100 Ω at 0 °C, rising about 0.385 Ω per °C. That predictable slope is why an RTD transmitter can linearize so accurately. Thermocouples deliver tens of microvolts per °C and need cold-junction compensation, so they trade absolute accuracy for range and speed. For most plant signals under 600 °C, a Pt100 into the transmitter is the safer choice; reach for a thermocouple or a miniature thermocouple when temperature or response time forces it. If you are weighing the two, our RTD vs thermocouple comparison goes deeper.

Do You Even Need a Transmitter? A Distance and Noise Decision Table

This is the decision the catalogues skip. A transmitter is not free, so fit it where the signal would otherwise degrade — not by reflex. The deciding factors are cable distance, electrical noise, and whether the sensor is an RTD (where lead resistance matters) or a thermocouple (where extension wire is costly). Use the table to place your install.

A real one from our field files: a food plant ran a Pt100 about 180 m from a steam header to the control room as a bare 3-wire RTD. The reading wandered more than 2 °C and drifted with plant load as drive noise coupled into the leads. Fitting a head-mount 4–20 mA transmitter at the sensor head fixed it — the current loop shrugged off the noise the millivolt-level RTD signal could not. The lesson holds: convert to 4–20 mA at the sensor when the run is long or noisy, and keep the raw sensor wiring short.

2-Wire vs 3-Wire Pt100 Wiring (and the Mistake That Biases Every Reading)

With an RTD, the wires themselves have resistance, and the transmitter cannot tell lead resistance from element resistance unless you wire it to compensate. A 2-wire Pt100 adds the full loop resistance straight onto the reading — every ohm of lead is a few degrees of error, so 2-wire is only honest on very short runs. A 3-wire connection lets the transmitter measure and subtract the lead resistance, which is why it is the industrial standard. A 4-wire connection removes lead effects entirely and is reserved for laboratory and reference work.

The mistake we see most: a 3-wire RTD wired with two leads on one terminal where the compensation loop expects matched, separate leads — or three leads of different lengths or gauges. The transmitter then subtracts the wrong lead resistance and gives you a stable, believable, systematically wrong reading. It will not alarm; it will just be off by a degree or two forever. Use three identical conductors, land them exactly per the transmitter’s terminal diagram, and verify with a known-temperature bath after commissioning. A confident wrong number is worse than an obvious fault.

Choosing the Output: 4-20 mA, HART, or RS485

Output is a separate decision from input, and the default is 4–20 mA. It is simple, robust, and understood by every controller. Add HART when you want remote configuration, multi-variable data, or diagnostics layered on the same two wires — useful for a field-mounted HART transmitter you would rather range from the control room than on a ladder. Choose RS485/Modbus when you are connecting many points digitally and want values without analog-to-digital conversion at the controller. Avoid 0–10 V output for field runs: voltage drops over cable resistance and gives you the same long-line error a current loop was designed to defeat.

Key Specs to Match to Your Process

• Measuring range — set the LRV/URV to span your process, not the sensor’s full limits, so resolution lands where you need it.
• Accuracy — a good head-mount transmitter adds about ±0.1% of span; the sensor class usually dominates total error.
• Ambient temperature — the electronics, not the tip, must survive the head environment; typical rating is −40 to +85 °C.
• Loop load — confirm supply voltage drives 20 mA through your total loop resistance (cable plus input resistor).
• Isolation — galvanic isolation between input, output, and power blocks the ground loops that plague long runs.

Related Temperature Products

SI-SBW Field-Mounted HART Temperature Transmitter

Head-mount transmitter for RTD or thermocouple input with 4–20 mA + HART output. Ranges and tags configurable remotely — ideal for long, noisy cable runs in process plants.

Industrial Thermocouple

Type K, S, and assembly thermocouples for furnaces, kilns, and high-temperature lines. Pair with a transmitter when the run is long or many points feed one DCS.

SI-DTM Digital Thermometer / Transmitter

Integrated digital thermometer with local display and 4–20 mA output. Reads temperature at the point and transmits it, where an operator also needs an on-the-spot value.

Frequently Asked Questions

What is the difference between a temperature sensor and a temperature transmitter?

The sensor — an RTD or thermocouple — is the element in contact with the process and produces a weak resistance or millivolt signal. The transmitter is the electronics that conditions, linearizes, and converts that signal into a standardized 4–20 mA or digital output. A “temperature sensor transmitter” is the two combined as one assembly.

When do I need a temperature transmitter instead of just the sensor?

Fit a transmitter when the cable run is long, electrical noise is high, or you are standardizing many points to 4–20 mA for a DCS. For an RTD within about 10 m of a clean panel, a 3-wire sensor straight to the input card is fine. Past roughly 50 m, or near variable-speed drives, convert to 4–20 mA at the sensor head.

Why is 3-wire Pt100 wiring recommended?

A 3-wire connection lets the transmitter measure and subtract the cable’s lead resistance, which a 2-wire connection adds directly onto the reading as error. Use three identical conductors landed exactly per the terminal diagram; mismatched or miswired leads cause a stable but systematically wrong reading.

What does 4-20 mA mean on a temperature transmitter?

The transmitter outputs 4 mA at your lower range value and 20 mA at your upper range value, scaling linearly between. For a 0–100 °C range, 4 mA is 0 °C and 20 mA is 100 °C. The 4 mA live zero also means a broken wire reads 0 mA, distinguishing a fault from a genuine low temperature.

Should I choose 4-20 mA or HART output?

Use plain 4–20 mA for simple, robust analog control. Choose HART when you want to configure ranges, read diagnostics, or pull multi-variable data remotely over the same two wires. RS485/Modbus suits many digital points; avoid 0–10 V on long field runs because cable resistance drops the voltage and adds error.

About this article

Written and technically reviewed by the Sino-Inst engineering team — last reviewed 2026-06-01 (AI-assisted drafting). Based on IEC 60751 RTD tolerances and 4–20 mA loop practice, plus field experience installing temperature transmitters on long, electrically noisy cable runs. Questions? Reach our application engineers.

Request a Quote

Tell us your sensor type, temperature range, cable distance, and output. A Sino-Inst engineer will recommend the right temperature sensor transmitter and wiring for your install.

Request a Quote

Updated May 30, 2026

A furnace thermocouple problem is almost always one of three things: the sensor itself, the compensating cable or its polarity, or the cold junction. You can pin down which one in under 20 minutes with a multimeter — before you ever decide whether to replace the element or change the grade. This guide gives you the field triage first, then the selection logic for picking the right thermocouple type for your furnace atmosphere and temperature. It is written for industrial process, heat-treatment, and kiln furnaces; if you are chasing a dead pilot light on a residential gas furnace, that part is a single-purpose flame-sensing thermocouple and the fix is normally a like-for-like replacement, not a grade decision.

Contents

• How do you know if a furnace thermocouple is bad?
• Type K vs N vs S vs B: which thermocouple for which furnace?
• Why is my furnace thermocouple reading low?
• How to test a furnace thermocouple with a multimeter
• Compensating cable and the reversed-polarity trap
• Protection sheath and placement inside the furnace
• Related thermocouple products
• Frequently asked questions

How do you know if a furnace thermocouple is bad?

Run a three-symptom triage before you condemn the sensor. A genuinely failed thermocouple shows one of three signatures: a reading pinned at full-scale or wildly erratic (open circuit), a reading stuck near ambient or close to zero millivolts (short circuit), or a stable but wrong reading that drifts slowly over weeks (sensor aging, contamination, or a cold-junction problem). The first two are sensor or wiring faults you can confirm in minutes. The third is the one that fools people, because the instrument still “works” — it just lies.

Start at the head, not the controller. Disconnect the extension wire at the thermocouple terminal and measure the millivolt output directly across the element with a multimeter on its DC mV range. Compare that reading against the reference table for the grade you have (a Type K thermocouple chart gives you the mV-to-temperature values). If the head reads correctly but the control room reads low, the fault is downstream — in the cable, the cold junction, or the transmitter — not in the thermocouple.

Type K vs N vs S vs B: which thermocouple for which furnace?

Pick the grade by furnace atmosphere and upper temperature, not by what is cheapest on the shelf. Base-metal types (K, N) cover most furnaces below 1260°C; noble-metal types (S, R, B) are for high-temperature and oxidizing duty above that. The trap is atmosphere: a Type K that is perfect in clean air drifts badly in reducing, sulfurous, or alternating atmospheres — the failure mode known as “K-state” short-range ordering.

Tolerances follow IEC 60584-1, where Class 1 is tighter than Class 2 — specify the class your process actually needs rather than defaulting. If your furnace cycles hard, Type N resists drift better than K. For sustained high-temperature work, step up to the noble-metal grades; our guide to the standard platinum-rhodium thermocouple covers the S, R, and B distinctions, and for the extreme end there is the tungsten-rhenium thermocouple for vacuum and inert furnaces past 2000°C.

Why is my furnace thermocouple reading low?

A reading that is low but stable is usually the wiring or the cold junction, not the element. The single most common cause we see in the field is the wrong extension wire: someone replaced a run of compensating cable with ordinary copper wire. Copper does not reproduce the thermocouple’s voltage-temperature curve, so a new thermal junction forms at the splice and the controller reads tens of degrees low. On one annealing furnace, a reading 30 to 50°C below true was traced to exactly this — plain copper instead of Type K compensating cable. Swapping in the correct compensating cable brought it back to zero error.

Work through the low-reading causes in order: extension cable type and polarity first, then cold-junction (reference) compensation in the transmitter or controller, then sheath degradation, and only last the element itself. This order matters because the cable and cold junction are the cheapest to fix and the most likely to be wrong after a maintenance change.

How to test a furnace thermocouple with a multimeter

Yes, a multimeter is enough to confirm an open or short circuit and to sanity-check the output. Set the meter to DC millivolts and measure across the two thermocouple legs at the head with the furnace hot. Three outcomes tell the story:

• Over-range or jumping millivolts — open circuit. The element or a connection has broken.
• Near zero millivolts with a hot furnace — short circuit, often a sheath-to-element short or a pinched cable.
• A sensible millivolt value — convert it with the reference table for your grade and compare to the controller. A gap points downstream to the cold junction or transmitter, which you can cross-check with our temperature converter and 4-20mA output calculator.

A resistance check on a cold element can hint at continuity, but it will not validate calibration — millivolt-against-reference is the meaningful test. If you are weighing the sensor against a resistance device (an RTD) for a rebuild, our RTD vs thermocouple comparison covers where each wins on temperature range and accuracy, and you can browse grades on our thermocouple temperature sensors page.

Compensating cable and the reversed-polarity trap

Reversed polarity is the error that produces a reading that moves the wrong way or sits oddly low, and it is easy to make because thermocouple wire colors vary by standard. A fast field check on Type K: the negative leg (NiAl) is magnetic and the positive leg (NiCr) is only weakly magnetic — a small magnet identifies the legs when the color code is ambiguous. Always carry the polarity through every junction; a single reversed splice in a long compensating run flips the sign of that segment’s contribution.

Use compensating or extension cable that matches the thermocouple grade, keep it away from power cabling to avoid induced noise, and confirm the terminal block and any connectors are the matching alloy. For the K-type details and color codes, see our K-type temperature sensor reference.

Protection sheath and placement inside the furnace

The sheath decides how long the element survives and how fast it responds. Metal sheaths (Inconel, stainless) suit base-metal types in most furnaces; noble-metal elements need a clean ceramic protection tube to keep metal vapors off the platinum, which otherwise contaminates and drifts the junction. Place the tip in the working zone the temperature actually controls — not in a stagnant corner or hard against the wall, where it reads radiant wall temperature instead of process temperature. Insert far enough that conduction down the sheath does not cool the junction. When a noble-metal reading drifts despite a good cable and cold junction, suspect sheath degradation or vapor contamination before the element.

Related thermocouple products

Thermocouple Temperature Sensors

K, N, S, R, B grades with metal or ceramic sheaths for furnace, kiln, and process duty. IEC 60584 Class 1 or 2 tolerance, head or transmitter output.

Platinum-Rhodium Thermocouple

Type S, R, and B noble-metal elements for oxidizing furnaces to 1700°C. Ceramic protection tube to resist metal-vapor contamination and drift.

K-Type Armored Thermocouple

Flexible mineral-insulated K-type to 1260°C for furnaces that cycle. Fast response, bendable sheath, matching compensating cable available.

Frequently asked questions

How do I know if my furnace thermocouple is bad?

Measure the millivolt output at the head with a multimeter. An over-range or jumping value is an open circuit; a near-zero value with a hot furnace is a short; a stable but wrong value points to the cable, cold junction, or sheath rather than the element.

What is the difference between a K and N type thermocouple?

Both reach about 1260°C, but Type N (Nicrosil-Nisil) resists oxidation and thermal-cycling drift better than Type K, making it the more stable choice for furnaces that cycle frequently.

Why is my furnace thermocouple reading low?

The usual cause is the wrong extension wire — ordinary copper instead of matching compensating cable — or a cold-junction compensation error. Check cable type and polarity first, then the transmitter, before suspecting the element.

Can you test a thermocouple with a multimeter?

Yes. Use the DC millivolt range across the two legs and compare the value to the reference table for your grade. The multimeter confirms open and short circuits and lets you sanity-check the reading against the controller.

How do I check thermocouple polarity?

On Type K, the negative (NiAl) leg is magnetic and the positive (NiCr) leg is only weakly magnetic, so a small magnet identifies the legs when colors are unclear. Carry the correct polarity through every junction in the run.

Which thermocouple is best above 1500°C?

Use a Type B platinum-rhodium element (to about 1700°C) in oxidizing furnaces, or a tungsten-rhenium element for vacuum and inert atmospheres past 2000°C. Both need the right protection tube to avoid contamination.

Is an industrial furnace thermocouple the same as a residential gas furnace thermocouple?

No. A residential gas furnace uses a single fixed flame-sensing thermocouple as a safety device, replaced like-for-like. An industrial process or heat-treatment furnace measures temperature for control, so you choose the grade (K, N, S, R, B), sheath, and compensating cable to match the atmosphere and upper temperature — that selection is what this guide covers.

About this article

Written and technically reviewed by the Sino-Inst engineering team — last reviewed 2026-05-30 (AI-assisted drafting). Based on IEC 60584-1 thermocouple tolerance classes and field experience with industrial furnace temperature loops. Questions? reach our application engineers.

Get help selecting a furnace thermocouple

Tell us your furnace temperature, atmosphere, and mounting, and our engineers will recommend the grade, sheath, and cable. Request a quote below.

Request a Quote

KimGuo11

Wu Peng, born in 1980, is a highly respected and accomplished male engineer with extensive experience in the field of automation. With over 20 years of industry experience, Wu has made significant contributions to both academia and engineering projects.

Throughout his career, Wu Peng has participated in numerous national and international engineering projects. Some of his most notable projects include the development of an intelligent control system for oil refineries, the design of a cutting-edge distributed control system for petrochemical plants, and the optimization of control algorithms for natural gas pipelines.

drurylandetheatre.com

Updated May 29, 2026 · Sino-Inst Engineering Team

A “standard” platinum rhodium thermocouple is not just a hotter thermocouple. It is a reference-grade device — first-class or second-class — whose whole job is traceable accuracy, so a calibration lab can use it to certify the working thermocouples that actually sit in the furnace. Confuse the two and you will burn an expensive reference sensor doing a working sensor’s job.

This guide explains what reference grade means, how first-class and second-class standards differ, how to choose between Type S, R and B by temperature and atmosphere, which standards (IEC 60584 and ITS-90) define them, and the high-temperature failure modes that quietly wreck platinum thermocouples.

Contents

• What Is a Standard Platinum Rhodium Thermocouple?
• First-Class vs Second-Class Standard Thermocouples
• Type S, R and B — What Is the Difference?
• Which Type Should You Use for Your Temperature Range?
• What Standards Govern Platinum Rhodium Thermocouples?
• Do Type B Thermocouples Need Cold-Junction Compensation?
• Common Failure Modes and Field Mistakes
• Standard and Reference Thermocouples from Sino-Inst
• FAQ

What Is a Standard Platinum Rhodium Thermocouple?

A standard platinum rhodium thermocouple is a reference-grade temperature sensor built from high-purity platinum and platinum-rhodium alloy wire, used to calibrate other thermocouples rather than to run a process. Its defining feature is a certified, traceable relationship between temperature and output EMF — not a higher temperature rating. The “standard” in the name means metrological standard, the same word a calibration lab uses.

Platinum-rhodium (noble-metal) thermocouples earn this role because platinum is chemically stable, melts only at 1768 °C, and produces a smooth, repeatable EMF curve. That is also why base-metal types like K and J are never used as references — they drift far too quickly at high temperature. If you are still deciding between noble-metal and base-metal sensors at all, start with our thermocouple and RTD comparison and the Type K reference chart before committing to platinum.

First-Class vs Second-Class Standard Thermocouples

Reference platinum thermocouples come in two accuracy grades: first-class (primary) and second-class (secondary) standards. A first-class standard is the more accurate reference and is normally kept in the lab to certify second-class standards; the second-class standard is the everyday working reference taken to the furnace to check process thermocouples. Sino-Inst designates these WRPB-1 and WRPB-2.

The practical rule: never put a first-class standard into continuous furnace duty. Each hour at 1500 °C ages the wire and shifts its certified curve. Keep the primary reference clean and lightly used, and let the second-class standard absorb the wear of routine comparison checks.

Type S, R and B — What Is the Difference?

The three noble-metal divisions differ only in how much rhodium is alloyed into the wire, and that single choice sets their EMF output, stability and usable temperature range. Type S and R use a pure-platinum negative leg; Type B uses rhodium on both legs.

Type S is the classic reference division and is what most first-class standards use, because the original ITS-90 interpolation work was done on Pt-10%Rh. Type R trades a marginal accuracy gain for a slightly stronger signal. Type B exists for one reason: it reaches higher and tolerates oxidizing high-temperature work that would shorten an S or R couple.

Which Type Should You Use for Your Temperature Range?

Match the division to your peak temperature first, then to atmosphere and cost. The decision is usually short.

• Up to ~1300 °C, reference accuracy needed: Type S. It is the established standard, widely supported by ITS-90 calibration, and the safest pick for traceable work.
• Up to ~1400 °C, slightly higher signal wanted: Type R. Useful where the readout resolves EMF poorly and the extra millivolts help.
• 1500 °C and above (glass tank, cement kiln, sintering): Type B. It survives where S and R degrade, and its low room-temperature output is an advantage in hot ambient environments.
• Below ~600 °C: do not use platinum at all — a base-metal Type K or an RTD is cheaper and more accurate in that band. Pair the sensor with a temperature transmitter to convert the small EMF to a 4-20 mA loop signal.

For furnace-specific construction — sheath material, insulator choice and immersion depth — see our furnace thermocouple notes, which cover the mechanical side this article does not.

What Standards Govern Platinum Rhodium Thermocouples?

Two documents matter: IEC 60584 defines the EMF-vs-temperature reference tables and tolerance classes, and ITS-90 (the International Temperature Scale of 1990) defines the fixed-point temperatures used to calibrate the references themselves. Type S, R and B all have standardized tables in IEC 60584-1, so a Type S couple from any compliant maker shares the same nominal curve.

Calibration of a standard thermocouple is done against ITS-90 fixed points — typically the freezing points of metals such as zinc (419.5 °C), aluminium (660.3 °C) and silver (961.8 °C) — and the result is certified under ISO/IEC 17025 accreditation, usually at two or three reference temperatures. That certificate, with its measured deviations, is what separates a true standard thermocouple from an uncertified working couple of the same wire.

Do Type B Thermocouples Need Cold-Junction Compensation?

Type B is the one noble-metal couple that effectively does not need cold-junction compensation in normal ambient conditions, and this trips up technicians who expect it to behave like S or R. Below about 50 °C, a Type B couple produces almost no EMF — its output even dips slightly negative around 21 °C before rising — so a cold junction sitting at room temperature contributes a negligible error.

The field consequence: a Type B couple reading near zero millivolts at startup is usually healthy, not broken. By contrast, Type S and R produce meaningful EMF from room temperature up, so they always require accurate cold-junction compensation — skip it and your reading runs low by tens of degrees. When the EMF feeds a control loop, the compensation lives in the field transmitter or the PID controller, not in the sensor.

Common Failure Modes and Field Mistakes

Most platinum thermocouple failures are contamination and handling problems, not wire defects. Four account for the majority of drift complaints.

• Rhodium volatilization and migration. Above ~1400 °C, rhodium slowly evaporates from the alloy leg and can deposit on the pure-platinum leg, shifting the EMF and reading low over time. This is the main reason a standard couple must be re-certified periodically.
• Reducing-atmosphere contamination. Platinum is poisoned by silicon, phosphorus and metal vapors in reducing or vacuum furnaces. Always sheath the couple in dense alumina (not mullite) and never let it contact a hot metallic protection tube directly.
• Insufficient immersion depth. A platinum couple needs immersion of at least 15–20 times the sheath diameter to avoid stem conduction error. Shallow insertion reads low — a classic false “the furnace is cool” alarm.
• Using the standard as a working sensor. Leaving a certified reference in continuous furnace duty ages it out of tolerance within weeks. Keep references for comparison work and run a dedicated working couple for the process.

Standard and Reference Thermocouples from Sino-Inst

Platinum Rhodium Thermocouple (S / R / B)

Noble-metal couples in Type S, R and B for furnace and reference duty to 1800 °C. First-class (WRPB-1) and second-class (WRPB-2) standard grades available with calibration certificate.

SI-SBW Field-Mounted HART Transmitter

Head-mount transmitter with built-in cold-junction compensation for S/R/B and base-metal couples. 4-20 mA + HART, linearized output for the control loop.

SI-SBW Temperature Transmitter 4-20mA

Compact two-wire transmitter that converts thermocouple EMF to a stable 4-20 mA signal. Pairs with platinum couples for long cable runs without EMF degradation.

FAQ

What is the difference between Type S, R and B platinum rhodium thermocouples?

The difference is rhodium content. Type S uses Pt-10%Rh against pure platinum, Type R uses Pt-13%Rh against pure platinum, and Type B uses Pt-30%Rh against Pt-6%Rh. S and R reach about 1600 °C short-term; B reaches 1800 °C short-term and produces almost no output below 50 °C.

Which platinum thermocouple measures the highest temperature?

Type B measures highest, with a long-term limit near 1600 °C and short-term use to 1800 °C. It is the standard choice for glass melting tanks, cement kilns and sintering furnaces where Type S and R would degrade too quickly.

What does a first-class standard thermocouple mean?

A first-class (primary) standard is the most accurate reference thermocouple, certified against ITS-90 fixed points and kept in the lab to calibrate second-class standards. A second-class (secondary) standard is the working reference taken to the field to check process thermocouples. Both are more tightly controlled than ordinary working sensors.

Do Type B thermocouples need cold-junction compensation?

In normal ambient conditions, Type B effectively does not. Its output below 50 °C is so small that a room-temperature cold junction adds negligible error. Type S and R, by contrast, always require accurate cold-junction compensation or they read low by tens of degrees.

What standard defines platinum rhodium thermocouples?

IEC 60584-1 defines the EMF reference tables and tolerance classes for Types S, R and B, and ITS-90 defines the fixed-point temperatures used to calibrate the standards. Certification is carried out under ISO/IEC 17025 accreditation, typically at two or three reference points.

How often should a standard platinum thermocouple be recalibrated?

Recalibration interval depends on use, but heavy high-temperature service drives rhodium volatilization that shifts the curve, so annual recertification is common for working references and longer intervals for lightly used primary standards. Any couple that has seen contamination or thermal shock should be checked before reuse.

How do I choose and source the right standard thermocouple?

Send your peak temperature, furnace atmosphere, required accuracy grade and whether you need a calibration certificate to our Sino-Inst engineering team, or reach our application engineers through the contact page. We will recommend the division and grade and reply within one business day.

Need a first-class or second-class standard platinum rhodium thermocouple, or help matching Type S, R or B to your furnace? Send your temperature range and atmosphere through the form below. Our temperature engineers will respond within one business day with a recommendation and quote.

About This Article

Written and technically reviewed by the Sino-Inst engineering team — last reviewed 2026-05-29 (AI-assisted drafting). Based on IEC 60584-1 and ITS-90, plus field experience with Type S, R and B couples in industrial furnaces. Questions? reach our application engineers.

Request a Quote

KimGuo11

Wu Peng, born in 1980, is a highly respected and accomplished male engineer with extensive experience in the field of automation. With over 20 years of industry experience, Wu has made significant contributions to both academia and engineering projects.

Throughout his career, Wu Peng has participated in numerous national and international engineering projects. Some of his most notable projects include the development of an intelligent control system for oil refineries, the design of a cutting-edge distributed control system for petrochemical plants, and the optimization of control algorithms for natural gas pipelines.

drurylandetheatre.com

Updated 2026-05-31 by the Sino-Inst Engineering Team

A bimetal stemmed thermometer is only as accurate as its calibration and its immersion. Calibrate it to the 32°F (0°C) ice point. Push the stem at least 2 inches into the medium. And match the ASME B40.200 accuracy grade to the tolerance your process needs. Skip any one of those and the dial reads with total confidence and is still wrong. This guide walks through the calibration steps first, then immersion, accuracy, and how to choose the right thermometer.

Contents

• What is a bimetal stemmed thermometer, and how does it work?
• How do you calibrate a bimetal thermometer?
• What temperature should you calibrate to?
• How deep must the stem be immersed?
• How accurate is it? ASME B40.200 Grade A vs B
• How to choose: range, dial, connection, thermowell
• Bimetal vs digital, thermocouple, and RTD
• Featured bimetal thermometers
• Frequently asked questions

What Is a Bimetal Stemmed Thermometer, and How Does It Work?

A bimetal stemmed thermometer measures temperature with a coil of two bonded metals that expand at different rates. As the temperature changes, the coil winds or unwinds and rotates a pointer on the dial. There is no power, no battery, and no electronics — just a mechanical movement reading the medium the stem sits in.

That simplicity is the appeal. These thermometers are rugged, cheap, and give a clear local reading on a pipe, tank, oven, or boiler. The trade-off is that they are mechanical: vibration and shock cause calibration drift, and they react more slowly than an electronic probe. If you need a remote signal for a control system, you want a temperature transmitter instead — but for at-a-glance field readings, the bimetal dial is hard to beat.

How Do You Calibrate a Bimetal Thermometer?

There are two field methods: the ice-point bath and the boiling-point test. The ice point is the one to use day to day, because it is more repeatable and does not depend on your altitude. Both end the same way: you hold the calibration nut under the dial head with a wrench and turn the dial until it reads the reference.

Recalibrate after any drop or hard knock, and check a thermometer used for food or quality records at the start of each shift. If the dial will not hold the reference after adjustment, the movement is worn and the instrument should be replaced rather than nursed along.

What Temperature Should You Calibrate To?

Calibrate to 32°F (0°C) in an ice-water bath, or to 212°F (100°C) in boiling water at sea level. The ice point is fixed everywhere, which is why it is the standard reference. The boiling point is not. Water boils lower as you climb, dropping roughly 1°F for every 500 feet of elevation. A boiling-point calibration done at altitude without correcting will set the dial too high.

Pick the reference closest to your working range. If you check refrigeration or chilled product, the 32°F ice point is right next to your readings. If you run hot process near 200°F, a boiling-point check brackets it better. Either way, verify at a single fixed point you trust rather than guessing against another thermometer that may also be off.

How Deep Must the Stem Be Immersed?

Immerse the stem at least 2 inches (50 mm) in liquid, or at least 4 inches (100 mm) in gas, to reach rated accuracy. Below that, heat conducts up the stem to the ambient air and the reading drifts toward room temperature — a mistake called stem conduction error. It is the single most common reason a correctly calibrated dial still reads wrong in service.

Most stems carry a dimple or groove marking the minimum sensing depth; get that mark below the surface. In a thin pipe where you cannot reach 2 inches straight in, use an angled or adjustable-angle thermometer, or a thermowell that lets the stem sit along the flow. Never judge immersion by eye on a short stem in a gas duct, where the 4-inch requirement is easy to underestimate.

How Accurate Is a Bimetal Thermometer? ASME B40.200 Grade A vs B

Industrial bimetal thermometers are specified to ASME B40.200 (formerly B40.3), which sets accuracy as a percent of full scale. Grade A is ±1% and Grade B is ±2%. Because the tolerance is a share of the whole span, a wide range costs you resolution. A 0–500°F Grade A dial is ±5°F, while a tighter 50–300°F dial of the same grade resolves much closer.

The lesson for selection: do not over-buy range. Choose a span that brackets your normal operating temperature with a little headroom, and the same grade will give you a tighter real-world reading. Pairing the grade with proper immersion and a fresh ice-point check is what actually delivers the number on the datasheet.

How to Choose One: Range, Dial Size, Connection, and Thermowell

• Range — bracket your operating temperature; common spans run from -40 to 550°C. Tighter range plus Grade A gives the best resolution.
• Dial size — 1″ to 5″; bigger dials read from farther away on tanks and overhead pipes.
• Stem length — long enough to clear the 2″ liquid / 4″ gas immersion rule at the mounting point.
• Connection — back, bottom, or adjustable-angle so the dial faces the operator.
• Thermowell — required on pressurized or fast-moving process lines so you can remove the thermometer without breaking containment.
• Hygiene — for food and pharma, choose a sanitary, cleanable design and check the ice point daily.

Bimetal vs Digital, Thermocouple, and RTD: When to Use Which

Reach for a bimetal dial when you want a rugged, powerless local reading and roughly ±1% is good enough. Move to an electronic sensor when you need a remote signal, tighter accuracy, or faster response. A thermocouple suits very high temperatures — see our Type K thermocouple chart and the platinum-rhodium types for furnaces and kilns. An RTD wins on accuracy and stability in the moderate range; our guide on RTD vs thermocouple covers the choice. For high-heat applications a furnace thermocouple replaces the dial entirely, and where you need a logged record a paperless recorder captures the trend.

Featured Bimetal Thermometers

Industrial Bimetal Dial Thermometer

Back- or bottom-connect dial thermometer for pipes, tanks, and HVAC. ASME B40.200 Grade A accuracy, ranges to 550°C, 1″–5″ dials.

Adjustable-Angle Bimetal Thermometer

Rotating head so the dial faces the operator on overhead or low pipes, while the stem still meets the immersion depth for rated accuracy.

Remote-Reading Dial Thermometer

Threaded-connection thermometer for HVAC and process where the display sits away from the measuring point. Easy-read dial for hard-to-reach spots.

Frequently Asked Questions

What temperature should a bimetal thermometer be calibrated to?

Calibrate to 32°F (0°C) using an ice-water bath, or to 212°F (100°C) in boiling water at sea level. The ice point is preferred because it is fixed regardless of altitude. Hold the calibration nut and turn the dial to the reference.

How do you calibrate a bimetal stemmed thermometer?

Fill a cup with crushed ice and cold water, wait a minute, then insert the stem past the sensing dimple without touching the sides. After 30 seconds, hold the nut under the dial with a wrench and adjust the pointer to 32°F (0°C).

How deep should the stem be immersed?

At least 2 inches (50 mm) in liquid and 4 inches (100 mm) in gas, with the dimple below the surface. Shallow immersion lets heat escape up the stem — stem conduction error — and pulls the reading toward ambient.

How accurate is a bimetal thermometer?

Per ASME B40.200, Grade A is ±1% of full scale and Grade B is ±2%. Because it is a percent of the whole span, a narrower range gives a tighter reading at the same grade, provided the stem is properly immersed and recently calibrated.

What is the temperature range of a bimetal stemmed thermometer?

Common ranges run from about -40 to 550°C, with many narrower spans available. For higher temperatures a thermocouple is the better choice. Pick the narrowest range that still brackets your operating point for the best resolution.

Bimetal vs digital thermometer — which is better?

A bimetal dial is rugged, needs no power, and is ideal for local readings at roughly ±1%. A digital sensor is better when you need higher accuracy, faster response, or a remote signal to a controller or recorder. Many plants use both.

About this article

Written and technically reviewed by the Sino-Inst engineering team — last reviewed 2026-05-31 (AI-assisted drafting). Based on ASME B40.200 accuracy grades and ice-point / boiling-point calibration practice, plus field experience installing bimetal dial thermometers on process, HVAC, and food lines. Questions? Reach our application engineers.

Request a Quote

Tell us your temperature range, connection, and mounting, and a Sino-Inst engineer will recommend the right bimetal thermometer grade, stem length, and thermowell.

Request a Quote

KimGuo11

Wu Peng, born in 1980, is a highly respected and accomplished male engineer with extensive experience in the field of automation. With over 20 years of industry experience, Wu has made significant contributions to both academia and engineering projects.

Throughout his career, Wu Peng has participated in numerous national and international engineering projects. Some of his most notable projects include the development of an intelligent control system for oil refineries, the design of a cutting-edge distributed control system for petrochemical plants, and the optimization of control algorithms for natural gas pipelines.

drurylandetheatre.com