A meter factor is a derived number - it comes from comparing what the meter counted against a known reference volume, with several corrections applied along the way. Every input to that comparison carries its own uncertainty, and those uncertainties combine into the uncertainty of the factor itself. Meter factor uncertainty is that rolled-up figure, and it is a distinct, granular contributor to the overall custody-transfer uncertainty budget.
Meter Factor Uncertainty in one line: Meter factor uncertainty is the combined uncertainty in a meter factor derived from a prove, built up from the uncertainties of every input used to compute it. Contributors include the base prover volume, temperature, pressure, density, and pulse counting. These individual uncertainties combine into the factor's uncertainty, which then feeds the wider custody-transfer uncertainty budget.
A meter factor is essentially the ratio of a reference volume to the volume the meter registered over the same displacement, so its uncertainty is assembled from the uncertainty of each quantity in that ratio. The reference side starts with the base prover volume - the certified volume of the prover established at reference conditions during its own calibration. That certificate carries an uncertainty, and it is often one of the larger single contributors, which is why the prover's own recertification is taken so seriously.
Then come the corrections that bring both the reference volume and the metered volume to common conditions. Temperature measurement uncertainty enters because the fluid and the prover steel both change volume with temperature and the corrections depend on measured temperatures. Pressure uncertainty enters similarly through the compressibility and steel corrections. Density or relative-density uncertainty matters wherever mass or standard-volume conversions are involved, since the factor may be derived on a mass or a corrected-volume basis. Each measured input feeds a correction, and each correction inherits the uncertainty of its input.
On the meter side, pulse counting contributes its own uncertainty. The meter's output is a pulse train, and resolving exactly how many pulses correspond to the reference displacement has a finite resolution, especially over short prover passes where the pulse count is small; pulse interpolation is used to reduce this but does not eliminate it. Repeatability of the proving runs is a further contributor, capturing the scatter that remains even when everything is nominally steady. Together these define the shortlist of quantities whose uncertainties have to be gathered before the factor's uncertainty can be computed.
Individual uncertainties do not simply add. Each contributor is expressed as a standard uncertainty in consistent terms, usually as a relative uncertainty so that quantities in different units can be combined, and independent contributions are combined by root-sum-square - squaring each, summing, and taking the square root. This reflects the reality that independent errors are unlikely to all push in the same direction at once, so the combined uncertainty is smaller than a naive sum. The result of that combination is the standard uncertainty of the meter factor.
Because the contributions combine in quadrature, the largest single contributor dominates the total, which is a useful guide for where to spend effort. If the base prover volume uncertainty is much larger than the temperature and pulse-counting contributions, then improving temperature measurement barely moves the total and the attention belongs on the prover certificate. This is the practical value of laying out the components: it shows which input is limiting the factor's confidence, so improvement effort goes where it actually helps rather than where it is easiest.
To state the factor's uncertainty at a stated confidence, the combined standard uncertainty is multiplied by a coverage factor to give an expanded uncertainty. This is the same machinery used across metrology; the scope here is simply narrowed to the meter factor specifically, rather than to a whole measurement system or a single instrument. The specific method, the way each contributor is estimated, and the coverage factor used are set by the applicable measurement standards and the site's uncertainty procedure, which is why this page describes the structure of the calculation rather than asserting particular numbers.
The meter factor is one link in a longer chain, and its uncertainty is one input to the overall custody-transfer uncertainty budget. That wider budget also includes the uncertainty of the live measurement between proves, the flow computer's corrections, and the reference conditions, and it is the total that matters commercially because it bounds how much the measured quantity could differ from the true quantity. Scoping the factor's own uncertainty separately, as this topic does, lets you see how much the proving process itself contributes versus the rest of the system.
In field operations, keeping the inputs to that uncertainty visible is where monitoring helps. The temperature, pressure, and density used in the corrections, and the repeatability of the proving runs, are all quantities a cloud SCADA platform already records. Trending them shows whether the conditions during a prove were as stable as the uncertainty estimate assumed, and flags when, say, a temperature transmitter's scatter has grown enough to enlarge its contribution. That gives an early sign that the factor's uncertainty may be creeping up before it shows in the budget.
Merobix records the measurement inputs and the proving results together and makes them viewable from any browser, so the raw material behind an uncertainty estimate is retrievable and the proving history that feeds the repeatability contribution is preserved. The platform does not perform the certified uncertainty calculation or replace the metrology procedure that establishes the budget - that stays with the standards and the flow computer. What it adds is visibility into the contributing measurements over time, which complements the broader measurement-uncertainty and uncertainty-budget work by keeping the factor-level inputs honest and trended rather than assumed.
The main contributors are the base prover volume uncertainty from the prover's calibration certificate, the temperature and pressure measurements used in the corrections, the density where mass or standard-volume conversion is involved, and pulse counting on the meter output. Proving-run repeatability adds a further contribution. Each measured input feeds a correction that inherits that input's uncertainty, and together they define the factor's overall uncertainty.
Each contributor is expressed as a standard uncertainty in consistent, usually relative, terms, and independent contributions are combined by root-sum-square rather than simple addition. Combining in quadrature reflects that independent errors are unlikely to all align, so the total is smaller than a naive sum and the largest single contributor tends to dominate. A coverage factor is then applied to state the result as an expanded uncertainty at a stated confidence.
The meter factor's uncertainty is one input to the wider custody-transfer uncertainty budget, which also includes the live measurement between proves, the flow computer corrections, and the reference conditions. The overall budget is what bounds how much the measured quantity could differ from the true quantity. Scoping the factor's uncertainty on its own shows how much the proving process itself contributes versus the rest of the measurement system.
Merobix reads your field devices into a cloud SCADA - the real thing behind these terms, live in days from any browser.