ISO 17025 CMC (Calibration & Measurement Capability)

Roll reference-standard, repeatability, resolution, and other contributors into an ILAC P14-style CMC claim for your scope of accreditation.

Inputu_ref = U_ref/2 u_rep = s/√n u_res = r/(2√3) , u_c = √(u_ref² + u_rep² + u_res² + u_other²) CMC = k · u_c

Reference standard uncertainty U (from cert, k=2) (meas. units)Repeatability std dev (best existing device) (meas. units)Number of repeat readingsResolution of best existing device (meas. units)Other standard uncertainty (env., drift, etc.) (meas. units) — optionalCoverage factor k for CMC

Your recent runs (stored only in your browser)

No calculations yet — results land here so you can compare runs.

The engineering

Your CMC is the smallest expanded uncertainty you can legitimately claim on your accredited scope — the uncertainty of calibrating the 'best existing device' available to a customer, not your house artifact on its best day. This card builds the ILAC P14 budget: reference-standard uncertainty (de-rated from its k=2 certificate to standard form), Type A repeatability of the mean of n readings, resolution treated as a rectangular distribution over ±half a digit, and one catch-all line for environment, drift, and connectors. Everything RSSes to a combined standard uncertainty, then expands by k (normally 2 for ~95%).

Two gotchas assessors love. First, the resolution floor: ILAC P14 says your CMC can never be smaller than what the best existing device's resolution supports, so if u(res) dominates, adding averaging won't buy you anything. Second, watch the dominant-contributor line — if one term carries >90% of the variance, the effective degrees of freedom may be low enough that k=2 is optimistic and a Welch-Satterthwaite check (or k=2.1–2.3) is warranted before the claim goes on your scope.

Where this math comes from

The term CMC was born in the 1999 CIPM Mutual Recognition Arrangement, when national metrology institutes needed a common currency to compare 'best capability' claims in the BIPM key-comparison database. Accredited commercial labs, however, kept using the older, fuzzier 'best measurement capability,' and no two accreditation bodies computed it quite the same way — a headache for any lab shipping calibrations across borders.

ILAC P14, first issued in 2010 and drafted jointly with the BIPM, ended the ambiguity: one definition, mandatory inclusion of the best-existing-device repeatability, an explicit resolution floor, and k chosen for ~95% coverage. The math underneath is the ISO Guide to the Expression of Uncertainty in Measurement (GUM, 1993) — the same RSS-and-expand recipe this card runs.

1993ISO / BIPM (GUM working group)Guide to the Expression of Uncertainty in Measurement standardizes the combine-and-expand method.
1999CIPMMutual Recognition Arrangement coins 'CMC' for capability entries in the BIPM key-comparison database.
2005ISO/IEC17025:2005 requires accredited cal labs to state best measurement capability on scope.
2010ILAC / BIPMILAC P14 harmonizes the CMC definition and calculation for accredited laboratories worldwide.
2020ILACP14:09/2020 revision aligns CMC policy with ISO/IEC 17025:2017 decision-rule requirements.

See the full timeline of the math behind every calculator →

Runs entirely in your browser — nothing you enter leaves this page. Your recent runs are stored only on your device.