Metrological traceability under ISO/IEC 17025
An unbroken, documented chain of calibrations back to the SI, each with its own stated uncertainty. What ISO/IEC 17025 actually requires you to show.
Metrological traceability is the property of a measurement result whereby it can be related to a reference — normally the SI unit — through a documented, unbroken chain of calibrations, each contributing to the measurement uncertainty. The definition has three load-bearing words: documented, unbroken, and uncertainty.
What breaks the chain
- A reference whose own calibration is missing, expired beyond its documented interval, or not traceable itself.
- A comparison with no uncertainty evaluated — a 'check' rather than a calibration.
- A link that cannot be produced on request: the documentation is part of the chain.
Practical evidence
For most organisations, traceability is demonstrated by calibration certificates from laboratories accredited to ISO/IEC 17025 by a body within the ILAC mutual recognition arrangement, or from national metrology institutes. The certificate must state the uncertainty and identify the reference standards used.
The chain, link by link
Traceability is easier to check than to define, because every link has to produce the same four things. Follow it upward from the instrument in your hand and you will either reach the SI or find the break.
| Link | Who | Evidence |
|---|---|---|
| Warehouse logger | You | Calibration certificate from your supplier, with uncertainty |
| Supplier's working standard | Calibration laboratory | Accredited certificate, within its interval, covering the range |
| Laboratory's reference standard | National metrology institute or a higher-tier accredited lab | Certificate with uncertainty, traceable to a fixed point |
| ITS-90 fixed points | NMI | Realisation of the kelvin — the top of the chain |
Each link must have a documented calibration, a stated uncertainty, a date within its interval, and coverage of the quantity and range in question. A link missing any one of those breaks the chain, and the break is not repaired by the links above it being excellent.
Traceability is per-quantity and per-range
The commonest misunderstanding is treating traceability as a property of an organisation. It is a property of a result. A laboratory with impeccable traceability for Pt100 calibration from −40 °C to 250 °C has none at all for the humidity channel on the same instrument, and none for a Pt100 used at −60 °C. Check the certificate against the actual measurement, not against the laboratory's reputation.
Checks that are not calibrations
A comparison with no uncertainty evaluated is a check, not a calibration, and a check does not create a link. This distinction catches a lot of in-house arrangements: comparing a working thermometer against a reference and recording that they agreed within 0.2 °C is a useful control, and it is not traceability unless the comparison produced a correction and an uncertainty. Interim checks are valuable precisely as checks — they detect change between calibrations — but they do not extend an interval.
Where chains break in practice
- An in-house reference used to calibrate everything else and never itself sent out. The chain stops at your own bench.
- A certificate that says traceable to national standards without naming the standards or giving an uncertainty. That sentence is not evidence of anything.
- A reference recalibrated late, leaving a period during which everything it supported has an undocumented gap.
- Adjustment at recalibration with no as-found data, which severs the link to every measurement made since the previous certificate.
- An accreditation symbol on a certificate for work outside the accredited scope — legal, common, and not what the symbol implies.
- Calibration at points that do not cover the range of use, so the result is an extrapolation the certificate does not support.
Frequently asked questions
- Is a calibration sticker proof of traceability?
- No. The sticker records that a calibration happened; the certificate with its uncertainty and reference standards is the evidence of traceability.
- Does traceability require an accredited laboratory?
- Not strictly, but an accredited certificate is the accepted way to demonstrate the chain without auditing the laboratory yourself.
- Can a result be traceable without an uncertainty?
- No. Each link contributes to the uncertainty of the result; a link with no uncertainty is not a calibration in the metrological sense.
References
- [1]JCGM 200:2012 — International vocabulary of metrology (VIM), 3rd edition
- [2]ILAC-P10: ILAC Policy on Metrological Traceability of Measurement Results
- [3]ISO/IEC 17025:2017 — General requirements for the competence of testing and calibration laboratories
General technical guidance written against the cited sources. It is not regulatory or legal advice and does not replace the applicable standard, guideline or a qualified reviewer's judgement.
Have a question on this topic?
Ask ValiTrac AI and see the evidence and calculation behind the answer.
Ask ValiTrac AIFree calculator: Uncertainty budget
Combine Type A and Type B components per the GUM: divisors by distribution, sensitivity coefficients, u_c, effective degrees of freedom and U at k = 2.
Open the calculatorMetrology on ValiTrac
Measurement science, traceability, uncertainty and technical evidence.
Ask a metrology questionRelated articles
Measurement uncertainty explained
What an uncertainty statement means, how it is built up, and why a coverage factor matters.
What is CMC in calibration?
Calibration and Measurement Capability is a laboratory's best-case uncertainty, not the accuracy of your instrument.