SensorCatalog

PT100 vs PT1000: which platinum RTD should you use?

PT100 and PT1000 follow the exact same IEC 60751 curve and tolerance classes, but PT1000 has ten times the resistance (1,000 Ω vs 100 Ω at 0 °C). That cuts two-wire lead error by 10× (+0.18 °C vs +1.81 °C on 10 m of 0.5 mm² cable), making PT1000 ideal for 2-wire runs and battery devices.

PT100 vs PT1000 on 10 m of 0.5 mm² two-wire copper cable (20 °C cable)

Temperature (°C)PT100 (Ω)PT1000 (Ω)PT100 2-wire error (°C)PT1000 2-wire error (°C)
-10060.26602.56+1.70+0.17
0100.001000.00+1.76+0.18
50119.401193.97+1.79+0.18
100138.511385.05+1.81+0.18
200175.861758.56+1.87+0.19
300212.052120.51+1.93+0.19
400247.092470.92+2.00+0.20
500280.982809.78+2.07+0.21

Resistance and sensitivity per IEC 60751; two-wire cable resistance is 0.688 Ω (2 × 10 m × 0.0172 / 0.5). Both elements share identical tolerance classes AA, A, B and C.

Try it with the calculators

The short answer

PT100 and PT1000 are both industrial platinum resistance thermometers defined by the exact same standard, IEC 60751. Neither element is inherently more accurate than the other: a class A PT100 and a class A PT1000 have the exact same permitted tolerance of ±0.15 °C at 0 °C and ±0.35 °C at 100 °C.

The entire engineering difference comes from one factor of ten: a PT100 is 100 Ω at 0 °C, while a PT1000 is 1,000 Ω at 0 °C. That single scaling factor reduces two-wire lead-resistance error by a factor of ten, lowers excitation current requirements by up to ten times for the same output voltage, and shifts the balance between two-wire simplicity and three- or four-wire industrial instrumentation.

Identical curve, ten times the resistance

Under IEC 60751, both PT100 and PT1000 follow the Callendar–Van Dusen equation with the same temperature coefficient α = 0.00385055 °C⁻¹ and the exact same polynomial constants A, B and C:

  • At 0 °C, PT100 is 100.00 Ω and PT1000 is 1,000.00 Ω.
  • At 100 °C, PT100 is 138.51 Ω and PT1000 is 1,385.06 Ω.
  • At 500 °C, PT100 is 280.98 Ω and PT1000 is 2,809.78 Ω.
  • At −100 °C, PT100 is 60.26 Ω and PT1000 is 602.56 Ω.

Because the curve is scaled strictly by R₀, the resistance slope (sensitivity) also scales by ten: near 0 °C a PT100 changes by 0.3908 Ω/°C, while a PT1000 changes by 3.9083 Ω/°C; at 100 °C the slopes are 0.3793 Ω/°C and 3.7928 Ω/°C. You can inspect any point on either curve with the PT100 calculator, the PT1000 calculator, or the downloadable PT100 resistance table and PT1000 resistance table.

Lead-wire resistance: where PT1000 wins decisively

Every metre of copper cable adds series resistance in proportion to length and inversely with conductor cross-section. In a two-wire RTD circuit, the measuring instrument cannot distinguish lead resistance from sensor resistance, so the full loop resistance of both conductors appears directly as a positive temperature offset.

Take a typical 10 m cable run of 0.5 mm² copper at 20 °C (ρ = 0.0172 Ω·mm²/m). The round-trip resistance of 20 m of wire is 0.688 Ω:

  • On a PT100 at 100 °C (slope 0.3793 Ω/°C), 0.688 Ω shifts the reading by +1.81 °C — more than five times the ±0.35 °C tolerance of a class A element and more than double the ±0.80 °C tolerance of class B. A two-wire PT100 is therefore unusable on any cable longer than a metre or two unless you calibrate out the wire or accept coarse accuracy.
  • On a PT1000 at 100 °C (slope 3.7928 Ω/°C), the exact same 0.688 Ω cable shifts the reading by only +0.18 °C. Even over 25 m of 0.5 mm² cable (1.72 Ω), a two-wire PT1000 shifts by +0.45 °C, which is often acceptable in HVAC, building automation, heat-cost allocation and motor-winding monitoring.

Contact resistance behaves the same way. A dirty terminal block or slip-ring connector adding 0.1 Ω creates a +0.26 °C error on a PT100 but only +0.026 °C on a PT1000. Whenever you must use two wires — thin sensor harnesses in EV motors, long building-automation runs, or two-pin connectors — PT1000 turns a serious wiring error into a small, predictable bias. Use the RTD lead-wire error calculator to check your exact cable gauge, length and ambient cable temperature.

Three-wire and four-wire connections: when PT100 catches up

Once you move to a three-wire or four-wire connection, the lead-resistance advantage of PT1000 shrinks dramatically.

In a three-wire circuit, lead resistance cancels provided the two current-carrying leads have identical resistance. Only the mismatch between conductors (for example 5 % from unequal terminal torque or splice resistance) remains. On our 10 m, 0.5 mm² cable at 100 °C with 5 % mismatch, a three-wire PT100 error drops from +1.81 °C to just +0.045 °C, and a three-wire PT1000 drops to +0.005 °C. Both are well inside class A tolerance.

In a four-wire (Kelvin) circuit, the sense wires carry negligible current, so lead resistance is eliminated completely for both PT100 and PT1000. In industrial process plants — refineries, chemical reactors, pharmaceutical autoclaves and food pasteurisers — sensors already run to three- or four-wire head-mounted transmitters or PLC RTD cards, where PT100 remains the universal spare part stocked on every shelf.

Self-heating and excitation current

Measuring an RTD requires passing an excitation current I through the element, which dissipates power P = I² · R and warms the sensor above the medium temperature. Here engineers often repeat a half-truth: “PT1000 always has less self-heating.” That is true only when you reduce the excitation current.

Compare the power dissipated at 0 °C:

  • PT100 at 1.0 mA produces a 100 mV signal and dissipates 0.100 mW (0.139 mW at 100 °C).
  • PT1000 at 1.0 mA produces a 1,000 mV signal, but dissipates 1.000 mW (1.385 mW at 100 °C) — ten times more heat! Connecting a PT1000 to a constant 1 mA source designed for a PT100 actually worsens self-heating tenfold.
  • PT1000 at 0.1 mA produces the exact same 100 mV signal as a PT100 at 1 mA, while dissipating only 0.010 mW — ten times less heat and ten times lower battery drain.
  • PT1000 at 0.2 mA gives double the signal voltage (200 mV at 0 °C) while still cutting power to 0.040 mW (0.055 mW at 100 °C), less than half the heat of a PT100 at 1 mA.

In battery-powered IoT nodes, wireless loggers and compact air-temperature probes in still air (where thermal dissipation constants can be as low as 2 to 5 mW/K), running a PT1000 at 0.1 to 0.25 mA cuts both current consumption and self-heating error substantially.

Construction, temperature range and stability

Most PT1000 elements are manufactured as thin-film platinum resistors on a ceramic substrate because winding a fine 1,000 Ω platinum wire coil inside a narrow sheath is difficult and expensive. Thin-film PT1000 elements are mechanically rugged, compact and fast to respond, and they cover standard industrial temperatures up to roughly 400 to 500 °C depending on the manufacturer’s datasheet.

PT100 elements are widely available in both thin-film and wire-wound constructions. Wire-wound PT100 elements achieve the widest certified IEC 60751 class ranges (class AA from −50 to 250 °C, class A from −100 to 450 °C, class B from −196 to 600 °C) and are the standard choice for cryogenic tanks, high-temperature steam lines and calibration standards. Above about 600 °C, neither PT100 nor PT1000 maintains a defined IEC 60751 tolerance class, and a thermocouple vs RTD evaluation usually points to a type K or type N thermocouple.

Electrical noise, cable capacitance and input compatibility

Higher sensor impedance cuts lead-wire error, but it also makes the loop slightly more sensitive to capacitively coupled electromagnetic interference (EMI) and insulation leakage at high temperatures.

  • Insulation resistance at high temperature: Ceramic mineral insulation (MgO) loses resistivity above 400 to 500 °C. A parallel leakage resistance of 1 MΩ across a 281 Ω PT100 shunts only 0.028 % of the reading (~0.1 °C), whereas the same 1 MΩ across a 2,810 Ω PT1000 shunts 0.28 % (~1 °C). For high-temperature MI probes above 400 °C, PT100 holds up better against insulation degradation.
  • PLC and transmitter compatibility: Virtually every industrial temperature transmitter, DCS input card and PLC RTD module supports PT100 natively. Most modern transmitters and PLC modules also support PT1000, but older plant racks, legacy safety relays and intrinsic-safety barriers sometimes accept only PT100. Always check the input module datasheet before standardising on PT1000 in a brownfield plant.
  • HVAC, BMS and automotive controllers: Building-automation controllers, heat meters, EV inverters and charging-plug monitors overwhelmingly favour PT1000 (alongside Ni1000 and 10 kΩ NTC thermistors) because two-wire cables are cheaper to pull and 1,000 Ω inputs simplify microcontroller front ends.

When to choose PT100

  • Process plants, oil and gas, food and pharma, and power generation using 3-wire or 4-wire wiring or head-mounted 4–20 mA transmitters.
  • Temperatures below −50 °C or above 400 °C where wire-wound elements and low sensitivity to MgO insulation leakage matter.
  • Legacy PLC, DCS or calibration equipment where PT100 is already the site spare standard.
  • Reference and laboratory probes requiring maximum long-term stability.

When to choose PT1000

  • Two-wire installations where pulling a third or fourth wire adds unacceptable cable or connector cost.
  • HVAC, building automation, district heating meters, solar thermal loops and EV motor/charger temperature sensing.
  • Battery-powered wireless sensors and embedded PCBs where 0.1–0.2 mA excitation saves battery life and cuts self-heating.
  • Direct connection to microcontroller ADCs or low-cost RTD front ends without high-gain amplification.

Common mistakes

Running a PT1000 at 1 mA excitation. A 1 mA source produces 1 mW at 0 °C and 1.39 mW at 100 °C — enough to add 0.3 to 0.7 °C of self-heating error in still air. Drop excitation to 0.1–0.25 mA for PT1000.

Assuming PT1000 is more accurate than PT100. Both share identical IEC 60751 class AA, A, B and C limits. PT1000 reduces wiring error in two-wire circuits; in a properly wired four-wire loop, element accuracy is identical. Check the full chain in the temperature measurement error calculator.

Swapping a PT100 for a PT1000 without reconfiguring the input. Because the resistance is ten times higher, a PT1000 wired to a PT100 input immediately drives the reading to full-scale over-range or sensor-break fault.

Ignoring thin-film temperature limits. Many catalog PT1000 elements are thin-film parts rated to 300 or 400 °C even though the IEC 60751 table continues to 850 °C. Always verify the probe’s own operating and class-tolerance limits on the datasheet.

Frequently asked questions

Is a PT1000 more accurate than a PT100?

The sensing element itself has the exact same accuracy class under IEC 60751: class A allows ±0.15 °C at 0 °C and ±0.35 °C at 100 °C for both PT100 and PT1000. However, in a two-wire circuit a PT1000 produces a much more accurate measurement because copper lead resistance creates one-tenth the temperature error compared with a PT100.

Can I use a PT1000 with two wires?

Yes, that is the main reason engineers choose PT1000. Ten metres of 0.5 mm² copper cable adds 0.688 Ω, which causes only +0.18 °C of error on a PT1000 versus +1.81 °C on a PT100 at 100 °C. For long runs or tight class A budgets, you can still use three- or four-wire PT1000 probes or calibrate out the known lead offset.

Why is PT100 still more common in factories than PT1000?

Industrial process plants standardized on three- and four-wire PT100 probes decades ago, when wire-wound 1,000 Ω platinum elements were impractical to build. Once three or four wires cancel lead resistance, PT100 offers broader wire-wound availability, wider high-temperature tolerance ranges, lower sensitivity to insulation leakage and universal compatibility with existing transmitters and PLC cards.

Does a PT1000 have less self-heating than a PT100?

Only when excited with a smaller current. At the same 1 mA current, a PT1000 dissipates ten times more power than a PT100 (1.0 mW vs 0.1 mW at 0 °C). Because a PT1000 has ten times the resistance, you can lower the current to 0.1 mA for the same 100 mV signal and achieve ten times lower self-heating power (0.01 mW).

Can a PT1000 replace a PT100 on the same transmitter or PLC input?

Not without reconfiguring the input channel. If the transmitter or PLC card supports PT1000 in its configuration menu, you can switch the sensor type and change the probe. If the input is fixed for PT100, a PT1000 will read roughly ten times too high and trigger an open-circuit or over-range fault.

What is the resistance of a PT1000 at 0 °C and 100 °C?

Per IEC 60751, a PT1000 measures exactly 1,000.00 Ω at 0 °C and 1,385.06 Ω at 100 °C, which is ten times the PT100 values of 100.00 Ω and 138.51 Ω. Its average temperature coefficient between 0 and 100 °C is 0.00385055 °C⁻¹.

Should I use a PT1000 or an NTC thermistor in HVAC or battery electronics?

Use a PT1000 when you need IEC 60751 interchangeability across vendors, a wide temperature range above 125 °C, or a nearly linear curve without part-specific firmware tables. Use a 10 kΩ NTC thermistor when lowest component cost and maximum sensitivity around room temperature matter most.

↑ ↓ choose · Enter open · Esc close