Metrological Limit
Documented interval statistics quantify total measurement error bounds at a ninety-five percent confidence level. An expanded uncertainty budget k=2 combines Type A statistical evaluation with Type B systematic evaluations, multiplying the combined standard uncertainty by a coverage factor of two. This framework establishes defensible error limits for instrument calibration certificates and quality assurance documentation.
Calibration laboratories state expanded uncertainty to allow direct comparison of sensor specifications against process tolerances.
Combined Variance
Individual uncertainty contributors are quantified and combined using root-sum-of-squares calculation rules. Components include primary reference standard uncertainty, sensor repeatability, hysteresis, thermal drift, and power supply variation. Type A evaluations calculate statistical standard deviations from repeated calibration runs, whereas Type B evaluations extract values from manufacturer datasheets, historical drift records, and environmental monitoring devices.
Coverage Interval
Applying a coverage factor of two assumes an underlying normal distribution of measurement errors across test samples. The resulting interval contains ninety-five point four percent of expected measurement outcomes under stable laboratory conditions. Non-normal distributions require modified coverage factors to maintain equivalent statistical confidence levels.
Audit Trail
Accredited calibration procedures require detailed documentation listing every contributor inside the calculation table. Quality auditors review these budgets during periodic facility inspections to confirm metrological traceability to national standards.
