Product type
A resistance temperature detector measures temperature from the change in electrical resistance of its sensing element. The phrase describes the detecting component rather than the complete instrument, and that distinction is the practical one: what is supplied under it is the element with its own immediate protection and its connection wires, not a fitted assembly with a process thread, a head and a terminal block.
A resistance temperature detector measures temperature from the change in electrical resistance of its sensing element. The phrase describes the detecting component rather than the complete instrument, and that distinction is the practical one: what is supplied under it is the element with its own immediate protection and its connection wires, not a fitted assembly with a process thread, a head and a terminal block.
That makes it the right purchase in two situations. The first is where the buyer has their own housing: a moulded plastic body, a printed circuit board, a machined metal block, a probe body of their own design. The second is where the sensing element is being bought to be built into something else, or replaced inside a device that was designed around a leaded part, and where a full assembly with a process connection would be the wrong shape and the wrong cost.
Two families sit under the same heading. The metal elements, of which platinum is the usual choice and Pt100 and Pt1000 the familiar designations, have a resistance that rises as they warm and follows a published curve that one element follows much as another. The semiconductor thermistors, particularly the negative temperature coefficient type, have a resistance that falls as they warm and change far more steeply over a narrow span. Both are resistance temperature detectors in the literal sense, and both are supplied as leaded components, but they are not interchangeable in a circuit.
The distinction matters commercially. A buyer who needs a reading that holds its consistency between instruments and can be verified later is buying a platinum element and paying for its curve. A buyer who needs a fine resolution over a narrow span, in a thermostat, a battery pack or a domestic appliance, is buying a thermistor and paying a fraction of the price for it. Selling one as the other is where the engineering goes wrong.
Resistance changes with temperature because the material’s electrical behaviour changes. In a metal, the lattice scatters charge carriers more as it vibrates harder, so resistance rises with temperature in a way that is close to linear over a moderate span and is governed by a single coefficient, around 0.00385 per degree for platinum. In a semiconductor thermistor, the number of available charge carriers changes with temperature, and for a negative coefficient type that effect dominates and the resistance falls steeply: the change per degree at room temperature is several times that of a platinum element.
Steepness is the trade. A steep curve resolves a small temperature increment easily, which is why a thermistor suits a thermostat that has to hold a tight band without an expensive converter. The same steepness makes the curve strongly non-linear over a wide span, so the reading has to be linearised against the actual element rather than against a published standard curve. A platinum element is gentler and more regular, which is what makes its curve standardised and its readings comparable between instruments and over years.
Lead resistance is the error that catching people out is the reason the wiring convention exists. In a two-wire connection the resistance of the copper leads is added to the element’s resistance and appears as a temperature offset that the reading circuit cannot distinguish from the measurement. A three-wire connection cancels the effect of equal leads, which is why it is the usual compromise. A four-wire connection removes lead resistance from the measurement almost entirely. The choice is a measurement decision, and it is made at the same time as the element, not afterwards.
Self-heating is the other error. Any excitation current dissipates power in the element, and the resulting temperature rise depends on how quickly heat can leave through the element’s own body and through whatever it is mounted against. A small element in still air warms more than the same element in a moving liquid, so the permissible excitation current depends on the installation rather than on the element alone. That is also why a dissipation constant, where one is published, is a statement about a stated condition rather than a general property.
Most of the specification is fixed by these decisions rather than by the accuracy class. Each one constrains the next, which is why they are settled in order.
| Decision | Options to fix | What it actually decides |
|---|---|---|
| Element material | Platinum, nickel, copper, or a semiconductor thermistor with a negative or positive coefficient | The shape of the resistance-temperature curve, whether it is standardised, and therefore whether the reading can be reproduced by another instrument or verified against a reference. |
| Nominal resistance | 100 Ω, 1000 Ω and others for platinum; a few kilo-ohms at the reference temperature for a thermistor | The resistance the reading circuit has to resolve and the excitation current it can use. A 1000 Ω element develops four times the signal of a 100 Ω one for the same current, which matters where the lead resistance is significant. |
| Tolerance class | A class or grade that defines the permitted deviation at the reference temperature and across the range | How far the element may sit from the standard curve. Two elements of the same class agree closely; without a stated class, two elements only agree near the reference temperature. |
| Number of conductors | Two, three or four wires, and where the element is a thermistor, two wires essentially always | How much lead resistance remains inside the reading. This is the decision that most often separates a reading that holds up over a long run from one that quietly does not. |
| Element construction | Wire wound on a former, thin film on a substrate, or a moulded or glass-encapsulated bead | Vibration tolerance, how fast the element responds, the size it can be made, and how much of the assembly’s accuracy survives service. |
| Protection and lead | Bare element, glass or resin encapsulation, a metal tube, and the insulation and length of the lead wires | What the element survives in service and how it is fitted into the buyer’s own housing. Lead insulation temperature rating is a separate limit from the element’s range and is often the one that is actually reached first. |
| Factor | How to settle it | Why it decides the answer |
|---|---|---|
| Whether the reading must agree between instruments | Ask whether two points will be compared, whether a reading will be repeated later, or whether the value will be audited. | This is the question that separates the two families. A standardised curve is what makes agreement possible, and a thermistor does not have one at a useful tolerance. |
| The span to be covered | Take the working range and the excursions, then compare it against the linearity of the curve available. | A platinum element covers a wide span with a gentle curve. A thermistor resolves finely over a narrow window and then stops being practical, because the resistance change per degree collapses at the extremes of its range. |
| The resolution actually needed | State the smallest temperature change the application has to show, not the smallest the instrument can display. | Resolution is bought with signal and with converter quality. Where the requirement is coarse, a platinum element and a modest converter serve; where it is fine over a narrow band, the thermistor is the cheaper route to the same result. |
| Distance to the reading circuit | Measure the run and decide the conductor count before the element is ordered. | Lead resistance is added to the measurement in a two-wire circuit and can only be cancelled with more conductors. The wiring decision is made once, at the element, and cannot be recovered later in software. |
| Excitation available | Establish how much current the reading circuit can supply and what power that dissipates in the element. | Too little current and the signal is lost in noise; too much and the element heats itself into a reading error. A low-resistance element is the more forgiving of the two, which is one reason 1000 Ω platinum elements are common. |
| How the element is fitted | Decide whether it is potted, clamped, inserted into a machined pocket or held against a surface. | The thermal path from the element to whatever is being measured is part of the measurement. An element with excellent tolerance and a poor thermal path gives a reading of its own surroundings. |
| Certification and traceability | State whether a certificate is required and what it has to say. | On a leaded element this is a separate purchase in most cases, and it has to be arranged with the element rather than added later, because it attaches to the specific part supplied. |
These are the types an enquiry for a resistance temperature detector usually turns out to be, and the point at which the choice is genuinely decided.
| Nearest alternative | What separates it from a resistance detector element |
|---|---|
| RTD temperature sensors | The same measuring principle supplied as a complete assembly: element in a sheath, with a process connection and a terminal head. Where the requirement is a fitted and connectable probe rather than a component, the assembly is the more complete specification and this page is the wrong one. |
| Thermocouples | Generates a voltage rather than changing a resistance, so it needs no excitation, responds quickly and survives higher temperatures. What it gives up is the consistency between instruments that a standardised resistance curve provides. |
| Temperature transmitters | Reads the element and converts it to a loop output, and does no measuring of its own. Where the element is supplied as a component, the transmitter is usually mounted away from it rather than in a head. |
| Temperature controllers | Reads the element and drives a heater or cooler towards a set point. It is where a leaded element most often ends up, and it is the part that has to be configured for the element’s curve and conductor count. |
| Digital thermometers | Portable and complete, with its own display and battery, intended for spot checks. It contains an element of some kind and is not a substitute for one, because it cannot be built into another device or left in a process. |
The two families share a name and almost nothing else. Decide first whether the reading has to be reproducible between instruments, then choose the element, then choose the conductor count.
The first reply can be a quotation rather than a set of questions if the enquiry carries these. Where a figure is not known, send the duty description and we will confirm what it implies.
Available products
Resistance Temperature Detectors
Stainless steel NTC thermistor probe with a 10 kΩ element at 25°C and a 3950 K B value, supplied with lead wires and plug terminals for control boards and acquisition modules.
Also in Temperature Measurement
Every type we quote for under Temperature Measurement. A type with no link yet is still quoted for - it is described on the Temperature Measurement overview, and gets a page of its own as soon as a model is published against it.
Frequently asked
Literally, yes: it is a detector whose resistance indicates temperature, and the phrase covers it. In common use the abbreviation has come to mean the metal element with a standardised, rising curve, and that is the reading most people have in mind, so it is worth stating which family is wanted rather than relying on the term. The two behave differently, are specified by different figures and cannot be substituted for each other without changing the reading circuit.
Because the conductors have to be attached to the element and brought out together. A three-wire element has three leads by construction, and a two-wire circuit cannot recover the cancelled lead resistance from a three-wire element. Where the run is long or the element resistance is low, a two-wire element will read an offset that no amount of adjustment in the instrument can separate from the measurement.
How far the element is allowed to depart from the standard curve at the reference temperature, and usually how that permitted deviation grows across the range. It does not cover the lead resistance, the self-heating or the thermal path to the medium. Those are the terms that usually make an installed reading worse than the class suggests, and they are the reason a class is a starting point rather than a statement about the installation.
As little as the reading circuit can still resolve without noise, because any current dissipates power in the element and warms it. The temperature rise that results depends on how quickly heat leaves, so the same element tolerates more current in a moving liquid than in still air. Where a dissipation constant is published it describes a stated condition, and the safe figure for an installation has to be worked out from how the element is mounted.
Yes, and that is what a leaded element is for. What the buyer then owns is the thermal path: how the element is clamped or potted, how firmly it touches the surface it is measuring and how much of it is exposed. Those decide the reading more than the element’s own tolerance does, which is why a good element fitted badly measures badly.
Send the specification and we will confirm the matching model, availability and lead time. Minimum order quantity is 1 pc.
Request for quotation
Send the technical detail and we will confirm availability, lead time and terms. In-stock items ship within 48 hours; customized specifications typically take 2-10 days. Minimum order quantity is 1 pc.