A prover is the reference every custody meter is measured against, but a reference is only as good as its own calibration. Prover recertification is the periodic re-establishment of the prover's certified base volume, and the renewal of the certificate that says the prover can be trusted. Skip it, or let the certificate lapse, and every meter factor proved against that prover inherits a reference of unknown validity.
Prover Recertification in one line: Prover recertification is the periodic recalibration of a prover, typically by a water draw, to re-establish its certified base volume and renew its certificate. It is done on a defined interval and also whenever a repair, seal replacement, or damage invalidates the existing certificate. Running proves against a prover with an expired or invalid certificate undermines the meter factors it produces.
The whole custody measurement chain rests on traceability: the meter is trusted because it was proved against the prover, and the prover is trusted because its base volume was established by calibration against an even more fundamental reference. That base volume is not permanent. Over time the prover's internal volume can change subtly through wear, corrosion, deposits, or minor dimensional shifts, and its detector switches can shift position, so the volume between them may no longer match the certificate. Recertification re-establishes the true current base volume so the reference stays honest.
The usual method is a water draw, in which a precisely known volume of water is displaced through the prover under controlled conditions to determine the actual volume between the detectors. This is the same class of calibration that established the prover's original certificate, repeated to produce a fresh, current value. The recertification yields an updated base volume and a renewed certificate stating the volume, the conditions, and the traceability behind it, which then becomes the reference for all subsequent proves until the next recertification.
Recertification happens on a defined interval - set by the governing standard, the regulator, or company policy - so that the prover is requalified before enough time has passed for undetected change to matter. The interval exists precisely because change is gradual and would otherwise go unnoticed; a scheduled recertification catches it. Between recertifications the prover is assumed valid, which is a reasonable assumption only if the interval is short enough relative to how fast the prover could realistically change, and only if nothing has happened to the prover that would break that assumption early.
The interval is not the only thing that ends a certificate's validity. Any event that could have changed the prover's base volume invalidates the certificate immediately, regardless of how much of the interval remains. The clearest examples are physical: a repair to the prover, replacement of the sphere or piston seals, work on or repositioning of the detector switches, or any damage to the prover body. Because each of these can alter the volume between the detectors or the way the detection triggers, the old certified volume can no longer be assumed to hold, and the prover must be recertified before it is trusted again.
Seal replacement deserves specific mention because it is routine maintenance that people can underestimate. The sphere or piston seal defines the boundary of the displaced volume, and changing it can shift the effective volume slightly, so a seal change typically triggers recertification even though it feels like ordinary upkeep. The same caution applies to anything that disturbs the detector switches, since the base volume is defined between them and moving a detector redefines the volume directly.
The governing principle is simple to state: if something could have changed the prover's volume or its detection, the certificate is suspect and recertification is required before further proving. This is why maintenance on a prover is treated as a measurement event, not just a mechanical one, and why the recertification record notes what was done and why. Treating a repaired or reseated prover as still valid on its old certificate is the fast path to producing meter factors that rest on a reference nobody has confirmed.
Proving a meter against a prover whose certificate has expired or been invalidated is a serious problem because it silently poisons the result. The meter factor comes out looking normal - it passes its repeatability check and gets applied - but the reference it was compared against is of unproven validity, so the factor's accuracy is unknown. Every volume then billed on that factor carries that hidden risk, and if the lapse is discovered later, the parties may face questions about all the provings done while the prover was out of certification, potentially reopening periods that were thought settled.
The symptom-to-cause-to-action shape here is really about prevention. The symptom you never want to see is a proving report discovered, after the fact, to have been run against an expired prover certificate. The cause is almost always administrative: the recertification came due, or a repair invalidated the certificate, and the prover kept being used because nobody was tracking its status. The fix is to make the prover's certificate status impossible to overlook - to know at any moment when it is due, when it was last done, and whether any event has invalidated it.
A cloud SCADA platform is well suited to that tracking. Merobix can hold the prover's certificate details, count down the calendar toward the next recertification, and flag when the prover is approaching or past its due date, so a lapse is caught before a prove is run rather than after. It can also record maintenance events against the prover, so a seal replacement or repair that should trigger recertification is visible in the same place as the proving history. The platform does not perform the water draw or issue the certificate - that is a certified calibration activity - but by scheduling recertification, surfacing invalidating events, and keeping the current certificate status alongside the proves that depend on it, it ensures no meter is proved against a reference that has quietly gone out of certification.
The usual method is a water draw, in which a precisely known volume of water is displaced through the prover under controlled conditions to determine the actual volume between its detector switches. This re-establishes the prover's current base volume and produces a renewed certificate stating the volume, conditions, and traceability. That updated base volume then becomes the reference for all subsequent proves until the next recertification.
Any event that could have changed the prover's base volume or its detection invalidates the certificate immediately, regardless of remaining interval. That includes repairs to the prover, sphere or piston seal replacement, work on or repositioning of the detector switches, and any physical damage. Because these can alter the volume between the detectors, the prover must be recertified before it is trusted again, even if the change feels like routine maintenance.
The meter factor can look completely normal and pass its repeatability check, yet it rests on a reference whose validity is unconfirmed, so the factor's accuracy is unknown. Every volume billed on that factor then carries that hidden risk, and if the lapse is discovered later the affected provings may have to be revisited, potentially reopening periods thought to be settled. That is why tracking the prover's certificate status and catching a lapse before proving is essential.
Merobix reads your field devices into a cloud SCADA - the real thing behind these terms, live in days from any browser.