RTD Calculator: Pt100, Pt1000, Wiring, Classes & ADC Design

Sensor and plot range

RTD type

Curve start

°C

Curve end

°C

Excitation modelSignal chain

Excitation mode

Excitation current

Wiring modelReal-world error

RTD connection

Lead resistance / wire

3-wire lead mismatch

Nominal accuracy class

Thermal and ADC assumptionsIntegrity lab

Self-heating coefficient

°C/mW

Permitted self-heating

°C

ADC full-scale reference

ADC resolution

Interactive SVG curve

Voltage versus RTD resistance

Blue is the ideal sensor signal. Amber includes the selected lead model; direct click or drag selects the same point as the keyboard-accessible cursor below.

IEC 60751 · α 0.00385

Ideal sensorWith lead model

Curve cursor72.68 °C

-50 °C100 °C250 °C

Cursor temperature72.68 °C

Sensor resistance128.099 Ω

Output voltage128.149 mV

Voltage sensitivity382.4359 µV/°C

R(T)=R_0[1+AT+BT^2+C(T-100)T^3]

V_{RTD}=I_{EXC}R_{RTD}

\Delta T_{self}=I_{EXC}^2R_{RTD}\,\theta

Measurement integrity

Wiring, heat and conversion error budget

Each contribution is referred to the cursor temperature so that an apparently small cable or ADC value can be compared with the RTD class tolerance.

Design decision

Apparent resistance128.149 Ω

Lead error referred to temperature0.13 °C

Nominal RTD class tolerance±0.66 °C

Estimated self-heating+0.05 °C

Native ADC resolution0.1317 °C/LSB

Additive error screen±0.91 °C

Wiring verdictThe selected 3-wire mismatch consumes less than one quarter of the nominal sensor tolerance at this cursor position.

Nominal signal-chain check passesThe displayed cursor point stays within the selected ADC reference and self-heating budget. This is a design screen, not a calibration result.

The additive screen is intentionally conservative, not a statistical uncertainty calculation. Class accuracy depends on RTD construction and stated temperature range; use the sensor datasheet and a calibrated system for final acceptance.

IEC accuracy classes

What do Class AA, A, B and C mean?

The class is the permitted nominal RTD error in degrees Celsius. In every formula, |t| is the absolute measured temperature in °C.

Class AA±(0.10 + 0.0017 × |t|) °C

Tightest nominal tolerance; use only when the RTD construction and qualified temperature range explicitly support it.

Class A±(0.15 + 0.002 × |t|) °C

High-accuracy option for controlled instrumentation chains with a documented temperature range.

Class B±(0.30 + 0.005 × |t|) °C

Common industrial baseline; the default choice when the sensor specification calls out Class B.

Class C±(0.60 + 0.010 × |t|) °C

Wider permitted error, useful where robustness matters more than fine absolute accuracy.

The formulas are nominal IEC 60751 tolerances, not a guarantee that every physical RTD can use every class across −200 to +850 °C. Check wire-wound versus thin-film construction and the manufacturer’s declared class range.

Excitation-window planner

Can the current meet both resolution and heat limits?

The lower limit reaches the requested direct-ADC temperature step at the weakest RTD slope in the selected range. The upper limit protects the sensor at its highest resistance.

Trade-off solver

Target direct-ADC step

°C/LSB

ADC LSB50.3548 µV

Minimum current for resolution1.391 mA

Maximum current for heat1.135 mA

Balanced suggested current—

Suggested differential gain26.1×

No direct-ADC current windowThe requested direct-ADC step conflicts with the entered self-heating limit. Relax one constraint, select a higher-R₀ RTD, or introduce appropriate analogue gain.

Suggested gain uses 90% of the ideal signal span and assumes the analogue chain removes the RTD offset safely. Verify common-mode range, reference accuracy, noise, filtering and fault conditions separately.

RTD measurement guide

Design an RTD signal chain that preserves accuracy

An RTD does not directly output temperature: its resistance is excited, carried through cabling, converted to voltage, and digitised. A good sensor choice can therefore lose accuracy in its leads, self-heating, or insufficient ADC resolution.

Method and assumptions

The tool applies the nominal IEC 60751 Callendar–Van Dusen relation, with α = 0.00385, to Pt50 through Pt1000 profiles. It then turns resistance into voltage through a current source or ratiometric divider. Comparing ideal and apparent signals converts cable resistance into temperature error at the chosen cursor point.

Inputs to verify

Actual sensor and range

Choose the actual ordered R₀—Pt100, Pt500, and Pt1000 do not generate the same voltage for the same current—then limit the plot to the product’s useful range. The standard calculation covers −200 to +850 °C, but not necessarily the physical construction of your sensor.

Cable and connection

In 2-wire mode, both lead resistances add directly to the RTD. In 3-wire mode, the tool assumes matched compensation leads and shows the residual from their mismatch. In 4-wire mode, the model ideally removes lead error without hiding measurement-electronics errors.

Heating and digitisation

Enter the published self-heating coefficient for the actual installation, not a generic value. Also check ADC reference, bit depth, and chain noise: a theoretical LSB is not usable resolution if noise or offset exceeds it.

Recommended workflow

1Select the platinum type, temperature range, and excitation mode, then move the curve cursor to inspect the critical points in that range.
2Compare lead error, self-heating, and class tolerance at the same temperature; then decide whether 2-wire remains acceptable or 3-/4-wire sensing is necessary.
3Use the excitation window to detect a conflict between direct resolution and heating, then validate gain, noise, references, and faults with the schematic and prototype.

Decision example

With a Pt100 at 1 mA, each cable ohm can represent several degrees depending on temperature. A 2-wire link with 1 Ω per conductor adds about 2 Ω and can consume or exceed Class B tolerance. Properly matched 3-wire sensing reduces this to residual mismatch; 4-wire sensing is preferable when traceable accuracy is required.

Limits to keep in mind

The curve is nominal, not a calibration certificate. It does not model drift, humidity, thermal gradients, variable contact resistance, noise, or front-end nonlinearity/common mode. IEC classes also depend on wire-wound or thin-film construction and the manufacturer’s declared range.

Frequently asked questions

Why choose Pt1000 over Pt100?

At the same current, a Pt1000 provides about ten times the voltage and slope of a Pt100, making ADC resolution easier and lead error relatively smaller. In return, it can constrain excitation current to preserve self-heating and does not replace good interconnection.

Does 3-wire always cancel cable error?

No. It ideally compensates two leads with equal resistance and a front end intended for that topology. Differences in length, gauge, temperature, or contacts leave residual error, which the lab deliberately exposes.

References to consult

Method references: IEC 60751 for the nominal curve and accuracy classes; RTD datasheet for self-heating coefficient, construction, and qualified range; ADC front-end datasheet for noise, offset, and common-mode limits.

Original educational content, reviewed for technical clarity on 20 August 2026. Always verify datasheets, applicable standards, and your design before power-up or manufacture.