At an important receipt or delivery point, one meter is not always enough - not because a single meter cannot measure flow, but because when money rides on the number, someone wants a way to confirm it. That is the role of a check meter installed alongside the primary. This guide distinguishes the primary meter, whose reading is the allocation-of-record, from the check meter that verifies it, explains the rules for which reading governs when the two disagree, and shows how continuous SCADA comparison turns check metering from an occasional spot check into always-on verification.
Primary vs Check Metering in one line: The primary meter is the measurement of record - the meter whose reading is used for custody transfer and allocation. A check meter is a second meter installed to verify the primary by measuring the same flow independently. When the two disagree, the governing rules - set by contract or measurement standard - decide which reading is used, usually the primary unless it is proven to be in error, at which point the check meter or a correction takes over.
The primary meter is the one that counts, in the literal accounting sense. Its reading is the allocation-of-record - the volume used to transfer custody at a sale point, to allocate production among owners, or to settle a delivery on a pipeline. Everything downstream in the accounting chain flows from the primary's number, which is why the primary is typically the more carefully specified, calibrated, and proved of any meters at the point. When people speak of the custody meter or the allocation meter at a location, they mean the primary.
A check meter serves a different purpose entirely: it exists to verify the primary rather than to be the number. Installed to measure the same stream - often in series with, or immediately adjacent to, the primary - the check meter provides an independent second reading that should closely track the primary if both are healthy. It is not usually the basis of settlement; its job is to give confidence that the primary is right and to raise a flag when it might not be. Because its role is verification, a check meter can sometimes be a different technology than the primary, which is useful precisely because two different measurement principles are unlikely to fail in the same way at the same time.
The value of a check meter appears the moment it disagrees with the primary, and the whole point of having governing rules is to decide what happens next. In normal operation the primary governs - its reading is used for allocation regardless of small differences from the check, because some divergence between any two meters is expected. The check meter simply confirms the primary is within a reasonable band. The interesting case is a divergence large enough to suggest one of the meters is actually wrong, which is when the rules written into the contract or the applicable measurement standard take over.
Those rules generally say that the primary governs unless it can be shown to be in error. If investigation - a proving, an inspection, a review of conditions - establishes that the primary was reading incorrectly, the parties fall back to a defined remedy: the check meter's reading may be substituted for the period, or a correction factor may be applied to the primary, depending on what the agreement specifies. What matters is that the outcome is not improvised in the middle of a dispute; it is decided in advance. A well-written measurement agreement spells out how large a disagreement triggers investigation, how the cause is determined, and precisely which number is used for the affected period, so a meter disagreement becomes a procedure to follow rather than an argument to have.
Historically, comparing a primary against its check was a periodic exercise - someone reads both meters, computes the difference, and confirms they agree closely enough. The weakness is obvious: between checks, the primary could drift and no one would know until the next comparison, by which time it may have mis-measured a large volume. A check meter that is only consulted occasionally catches problems late, and the interval between checks is exactly the window in which an undetected error does its damage. The verification is real but coarse.
Streaming both meters into a SCADA system converts that periodic check into continuous verification. When the primary and check both report to the same platform, the difference between them can be computed on every reading and alarmed the instant it exceeds the allowed band, so a divergence is caught in minutes rather than at the next scheduled comparison. A cloud SCADA platform such as Merobix carries both meter streams into one timestamped record and watches the primary-to-check difference continuously, raising an alarm when the two part ways so the cause can be run down while the flow conditions that produced it are still current. This turns the check meter from a document reviewed after the fact into a live watchdog on the meter of record, catching drift and heading off disputes at receipt and delivery points before a whole period's volumes are affected.
The primary meter is the measurement of record - its reading is used for custody transfer and allocation, so all the downstream accounting depends on it. A check meter is a second meter that measures the same flow independently to verify the primary. The primary is the number that counts; the check exists to confirm the primary is right and to flag when it might not be, so it is usually not the basis of settlement.
Normally the primary governs, because small differences between any two meters are expected. The governing rules - set by contract or measurement standard - decide what happens when the disagreement is large enough to suggest one meter is actually wrong. Typically the primary continues to govern unless it is proven to be in error, at which point a defined remedy applies, such as substituting the check reading or correcting the primary for the affected period.
By comparing the two meters continuously instead of at periodic manual checks. When both streams report to a SCADA platform, the difference is computed on every reading and alarmed the moment it exceeds the allowed band, so drift is caught in minutes rather than at the next scheduled comparison. That turns the check meter from a document reviewed after the fact into a live watchdog on the meter of record, heading off disputes before a whole period's volumes are affected.
Merobix reads your field devices into a cloud SCADA - the real thing behind these terms, live in days from any browser.