IEC 60751 and ASTM E1137 specify the same Pt100 resistance-temperature polynomial across roughly -200°C to +850°C, but the two documents diverge on tolerance class naming and on numerical tightness at 0°C: IEC 60751 Class A allows ±0.06% of nominal resistance (equivalent to ±0.15°C at 0°C), while ASTM E1137 Grade A is slightly tighter at ±0.05% (±0.13°C) [S1][S2].
Both standards cover the dominant industrial configurations, Pt100, Pt500 and Pt1000, with 100Ω at 0°C by far the most common [S1][S4]. The 2022 revision of IEC 60751 shifted the normative content from numerical tables to the resistance-temperature formulas and extended the marking scheme to encode sensor count, wire configuration, tolerance class and valid temperature range, but it left the four classical classes AA, A, B, and C unchanged in name and tolerance [S5].
What each standard actually covers
IEC 60751 is the international reference for industrial platinum resistance thermometers: it fixes the resistance-temperature relationship through the Callendar-Van Dusen equations, names the tolerance classes AA, A, B, and C, and since the 2008 revision requires the complete thermometer (element, leads, terminations) to meet the class at the connection end, not just the bare resistor [S5][S7]. The 2022 edition added seven technical changes, the operationally important ones being the move to formula-based interpolation and the expanded marking system that must show the number of platinum resistors, wire configuration, tolerance class, and temperature range, in the form "2 x Pt100 / (2/3 B) - F-sp / 3 / -50/+250" [S5].
ASTM E1137 is the US counterpart and uses its own grade letters (typically A and B) with tolerances that are numerically tighter than the IEC equivalent at 0°C, although the two resistance-temperature curves are sufficiently close that a Pt100 calibrated to either standard reads within the other's tolerance band at moderate temperatures [S2][S10]. For higher-accuracy work, ASTM E287 covers the standard platinum resistance thermometer (SPRT) used in calibration laboratories, while E1137 stays focused on the industrial detector [S3].
Tolerance class comparison: IEC 60751 vs ASTM E1137 at 0°C
The cleanest way to see the difference is at the 0°C reference point, where both standards express tolerance as a percentage of nominal resistance. IEC 60751 Class AA sits at ±0.04% (±0.10°C), Class A at ±0.06% (±0.15°C), Class B at ±0.12% (±0.30°C) and Class C at ±0.24% (±0.60°C) [S1][S5][S7][S9]. ASTM E1137 Grade A is roughly ±0.05% (±0.13°C) and Grade B roughly ±0.10% (±0.26°C) at 0°C, sitting between the adjacent IEC classes [S2][S10].
Outside 0°C the tolerance band widens with a polynomial in |t|, so a Class B Pt100 specified to ±0.30°C at 0°C drifts to roughly ±1.35°C at 200°C and to about ±2.55°C at 400°C under the IEC 60751 formula [S2][S7]. Above 650°C the two standards still define the same nominal resistance curve, but the practical operating ceiling for sheath-protected industrial probes usually sits between 600°C and 850°C depending on sheath material and element type [S1][S5][S6].
Which one to call out on a datasheet or purchase order
If the project is global, IEC 60751 is the safer citation: it is recognized across Europe, Asia and most EPC work, and the 2022 edition is the current authoritative version [S5]. The 1983, 2008 and 2022 editions are all still seen on PO line items, and the 2008 change of "complete thermometer must meet the class" remains the version most inspectors enforce today, so a line item that simply says "IEC 60751" is ambiguous until the year is added [S5].
If the project is US-domestic or US-Military, ASTM E1137 gives you slightly tighter Grade A tolerance and a more familiar grade-letter scheme to US procurement staff [S2][S3]. For calibration-lab reference work above 0°C, ASTM E287 (SPRT) is the relevant document rather than E1137 [S3]. In practice, dual-marked Pt100 elements that conform to both standards are widely available because the underlying resistance-temperature curve is essentially identical and the tolerance bands overlap [S10]. For related background on how the Pt100 element is wired into the rest of a measurement loop, see the load cell signal path and indicator essentials piece, which covers the same current-excitation / voltage-sense architecture at a different transducer type. A separate reference to laboratory DC resistor behaviour under IEC 60477-1:2022 is useful when validating the readout side of an RTD bridge against a decade box. The Wikipedia-style overview on the Pt100 element lists the same IEC 60751 tolerance ladder in plainer language.
Selection rules for the four IEC classes
Use Class AA (±0.10°C at 0°C) only for laboratory-grade or calibration-grade probes where the element and the lead-resistance budget are tightly controlled, typically as a working reference rather than a field sensor [S1][S7]. Class A (±0.15°C at 0°C) is the typical pick for HVAC, pharmaceutical reactor control, and any loop where 0.15°C matters at the calibration point but where 4-wire hookups are used to keep lead error out of the reading [S1][S2][S7].
Class B (±0.30°C at 0°C) is the workhorse: most thin-film Pt100 elements in DIN 43760 / IEC 60751 form ship to Class B by default, and it is the class most US EPCs will accept when the spec simply says "Pt100, IEC 60751" without further qualification [S4][S5][S7]. Class C (±0.60°C at 0°C) is reserved for cost-sensitive OEM embedded sensing where the absolute accuracy is acceptable and where the application can tolerate a wide interchangeability band [S1][S7].
Construction matters as much as the standard
Wire-wound Pt100 elements, where a platinum coil is wound on a ceramic or glass mandrel, give the best long-term stability and the lowest hysteresis, and they are the construction typically used to reach Class AA and Class A across the full -200°C to +850°C range [S1][S4]. Thin-film Pt100 elements, in which a platinum layer is sputtered or screen-printed onto a ceramic substrate, are smaller, cheaper, faster in thermal response, and dominate the high-volume Class B and Class C market, but their long-term drift is higher than wire-wound equivalents [S1][S4].
For an engineering buyer, the practical decision is therefore: pick a class (AA / A / B / C) based on the tolerance you need at 0°C and the widening of that band with |t|, then pick a construction (wire-wound vs thin-film vs coiled element) based on drift budget, thermal response time, and price [S1][S4]. The same Pt100 curve and the same IEC 60751 / ASTM E1137 numbers apply to either construction, but the achievable in-service accuracy after 12 months in the field will differ, and a 15% allowance on top of the interchangeability value is a reasonable rule of thumb for sensor-only error before installation effects are added [S2].
Wiring and lead-resistance impact on stated tolerance
The tolerance band published in IEC 60751 and ASTM E1137 assumes the element itself is at the calibration point; any lead resistance between the element and the transmitter adds directly to the reading on a 2-wire hookup, which is why 2-wire Pt100 installations are common only where the lead run is short and the tolerance is loose [S1][S2]. A 3-wire hookup cancels most of the lead resistance assuming all three leads are matched, and a 4-wire hookup eliminates the lead-resistance term entirely by separating the current-carrying pair from the voltage-sensing pair [S1].
For a Class A Pt100 specified at ±0.15°C at 0°C, a 2-wire hookup through just 1Ω of lead copper (about 0.385°C of error) blows the entire class budget on its own, so 4-wire is mandatory for AA, common for A, and still recommended for B on long runs [S1][S2]. IEC 60751:2022 formalizes the wire-configuration part of the marking, requiring that the sensor be marked with the configuration used (for example 2-wire, 3-wire, or 4-wire) so the installer cannot misread the class on the label [S5].
Failure modes and common spec errors
Specifying "Pt100, Class A" without naming the standard invites procurement trouble: a Class A element built to a tighter house standard can still be outside the IEC 60751:2022 interchangeability band, and a Class A element built to ASTM E1137 will read ±0.13°C at 0°C rather than the ±0.15°C the buyer may have budgeted for [S2][S5][S10]. The fix is to write the year into the line item, IEC 60751:2022 or ASTM E1137, and to add the construction (wire-wound vs thin-film) and the wire count (2 / 3 / 4) on the same line.
A second recurring error is treating the published tolerance as the field accuracy: the interchangeability band covers the element at calibration only, and installation effects (stem conduction, lead resistance, self-heating from excitation current, thermal contact resistance) routinely add 50% to 100% to the sensor-only error budget if not accounted for [S2]. A third error is mixing the standards within a single loop: putting a 4-wire IEC Class A element behind a 2-wire transmitter configuration will deliver a reading that is outside class even though the element itself is in spec [S1][S2].
Trackable signals to watch for: any further amendment to IEC 60751:2022 that revises the seven technical changes, and any ASTM E1137 revision that changes the Grade A / Grade B percentage at 0°C away from the current ±0.05% / ±0.10% values. Also watch the IEC TC 65 work on transmitter-side accuracy budgets, which increasingly folds the IEC 60751 class into a system-level uncertainty calculation rather than treating it as a stand-alone sensor spec.
Component reference pages worth checking: industrial adhesive, and industrial borescope.