The idea in one ratio
Every calibration is a comparison, and the comparison is only as good as the reference. If you verify a ±0.04 mA tolerance with a calibrator that is itself uncertain by ±0.03 mA, a large share of what you "measure" is actually your reference wobbling. The TUR quantifies this: the bigger the ratio, the smaller the standard's contribution to the pass/fail decision.
Worked example
A transmitter with ±0.25% of span tolerance on a 4–20 mA output (tolerance = 0.0025 × 16 = ±0.04 mA), checked with a loop calibrator specified at ±0.008 mA at the test point:
- TUR = 0.04 ÷ 0.008 = 5 : 1 — comfortably above 4:1 ✔
- The worst standard that still meets 4:1: 0.04 ÷ 4 = ±0.01 mA — useful when selecting or hiring test equipment
Field notes
- Evaluate at the test point, not full scale. Calibrator specs are usually "% of reading + % of range" — compute the ± value at your actual test value before taking the ratio. Our error converter helps.
- Below 3:1, guard-band: tighten acceptance limits by the standard's uncertainty (a simple guard band is acceptance = tolerance − U_standard) to keep false-accept risk in check.
- The ratio silently decays: your calibrator's uncertainty is only valid while ITS calibration is current. An overdue reference makes every downstream TUR claim fiction.
- Document it: auditors (ISO 9001, 17025, IATF) increasingly ask how test equipment adequacy was decided — a recorded TUR per procedure is the clean answer.
Frequently asked questions
What is TUR in calibration?
Test Uncertainty Ratio: the ratio of the unit-under-test tolerance to the uncertainty of the calibration standard. A TUR of 4:1 means the reference is four times better than the tolerance being verified.
Why is 4:1 the standard TUR?
At 4:1 the standard contributes little enough uncertainty that the risk of falsely passing an out-of-tolerance instrument (or failing a good one) stays acceptably low without complex guard-banding. It is the widely adopted benchmark from metrology practice (e.g. ANSI/NCSL Z540).
What if I cannot achieve 4:1?
Ratios down to about 3:1 are commonly accepted; below that you should apply guard bands (tightening the acceptance limits to offset the standard's uncertainty) or use a better reference.
Is TUR the same as TAR?
Closely related: TAR (Test Accuracy Ratio) compares accuracy specifications, while TUR properly compares the tolerance to the calibration process expanded uncertainty. TUR is the more rigorous, preferred metric.
Provided for reference and education. Formal uncertainty budgets per GUM/ISO 17025 involve more than this ratio. See our disclaimer.