Thermistor / RTD Calculator THERMISTOR

Temperature ↔ resistance for the common HVAC sensors, both directions.
Temperature ↔ resistance for the common HVAC sensors — 10K Type II / III, Johnson 8.7K, Schneider Type 5, 20K, 3K, Balco, Pt100, Pt1000 — both directions, plus an identification mode if you don't know the type yet. Lookup takes a temperature or resistance on a chosen type and gives the other plus the full R/T table. Identify takes (temp, resistance) pairs measured off an unknown sensor and ranks every type by curve fit.

Inputs are seeded with an example — edit them to your numbers.

Look up by
Defining point
Curve
Looking up
Result
Reading
Range check
TemperatureResistance (Ω)

Enter temperature + resistance pairs read off the unknown sensor. Three or more readings spread across a wide span give the cleanest match.

# Temp (°F) Resistance (Ω)
Likely type
Confidence
Points used

Every standard type, ranked by how far its published curve sits from your readings. Pick a row to open that type in Lookup.

Sensor type Equiv. temp error Resistance error
# Measured temp Measured Ω Curve says Off by

Identify mode compares your readings against the same generated nominal curves Lookup uses — not calibration data. Sensor tolerance (±0.1 to ±1 °C is typical), meter accuracy, and self-heating all shift the numbers. Two points often cannot separate similar curves: a 10K Type II and a 10K Type III both pass through 10 kΩ at 77 °F and only diverge away from it. Spread readings across as wide a temperature span as the job allows, use three or more, and treat a close call as “one of these two” rather than a definitive ID.

About these tables

Where the values come from. Each R/T table is generated from a small set of curve parameters — β and R25 for the NTC thermistors, the parallel shunt for the linearized 10K curves, R0 and a temperature coefficient for the RTDs. The Pt100 and Pt1000 tables come from the IEC 60751 Callendar–Van Dusen polynomial (the international standard). Auditing the parameters is what makes a 500+ cell table trustworthy in one pass, instead of one row at a time.

Sources verified against (2026-05 pass). 10K Type II, 10K Type III, and 10K Type 5 (TAC) — BAPI 10K-2 / 10K-3 / 10K-3(11K) output tables, cross-checked against US Sensor "Curve G", Sontay's Compatibility Chart, and Vector Controls' multi-curve reference. 20K and 3K — Vector Controls and Sontay. 1K Balco — a Schneider EcoStruxure Building Operation resistance chart plus the ACI BALCO datasheet. Pt100 and Pt1000 — IEC 60751:2008 directly, cross-checked against Vector, Sontay, Fluke's table generator, and pt100.de. Where multiple sources disagreed, I consulted with a second tech before settling on a curve.

What's still nominal. The Johnson Controls 10K + 8.7K-shunt curve is the one type without a public R/T table — the canonical TE-6300 Product Bulletin URL redirects to a docs-portal landing page. It's modeled as a 10K Type 2 element in parallel with 8.7 kΩ, which is folklore-confirmed but datasheet-unverified. If you have a known-good JCI sensor with calibration data, that's the better reference.

Use as a reference, not for calibration. A single-β model approximates a Steinhart-Hart curve; the fit is within roughly 1 °F across 32–150 °F (the BAS operating range) and degrades to 2–5 °F of equivalent temperature error at the −40 °F extreme for the bare 10K curves. Pt100 and Pt1000 are governed by the international standard and match it exactly. For calibration-grade work, measure your sensor directly against a known-good reference at the working temperature — don't trust any published table, ours or anyone else's, blindly.

← All tools