An automatic sampler builds a composite by taking many small grabs across a transfer, but the intelligence is in when it decides to grab. A proportional-to-flow sampler ties that decision to how much oil has actually passed, not to the clock. This guide explains flow-proportional grab logic, how the flow computer paces the sampler with pulses, and why proportioning the grabs to flow keeps the composite honest when the rate is not steady.
Proportional Sampler in one line: A proportional-to-flow sampler is an automatic sampler that takes each small grab after a fixed increment of volume has passed the meter rather than at fixed time intervals, so the number of grabs tracks the flow. Paced by pulses from the flow computer, it ensures the composite sample represents the batch in proportion to volume even when the flow rate varies.
There are two ways to decide when an automatic sampler fires a grab. Time-proportional logic grabs on a fixed schedule - one grab every so many seconds regardless of flow. Flow-proportional logic grabs after a fixed volume has passed - one grab every so many barrels regardless of how long that took. The difference sounds subtle but it changes what the composite actually represents.
Consider a transfer that runs fast for an hour and then slows to a trickle for another hour. A time-proportional sampler grabs the same number of times in each hour, so the slow, low-volume hour is over-represented in the composite relative to the oil it moved. A flow-proportional sampler grabs in proportion to volume, so the fast hour that moved most of the oil contributes most of the grabs. The composite ends up weighted the way the batch actually flowed.
That weighting is why flow-proportional sampling is the preferred approach for custody crude measurement, where the composite is analyzed for water content and quality that directly affect settlement. Time-proportional sampling can be acceptable when flow is genuinely steady, but real pipeline and LACT flows fluctuate with pumps starting and stopping, so proportioning to flow is the safer default. The methodology for representative sampling is set out in API MPMS Chapter 8.2.
A proportional-to-flow sampler does not measure flow itself; it is paced by the flow computer that is already metering the stream. The flow computer accumulates volume from the meter's pulses and, each time a configured volume increment passes, sends a grab command to the sampler. The sampler responds by extracting one small, fixed-size aliquot from the line and adding it to the composite container.
The grab-per-volume setting is chosen so that a full transfer yields enough grabs to be representative and to fill the container appropriately without overflowing. Set the increment too large and a short batch produces too few grabs to represent it; too small and a long batch overfills the container. Matching the increment to the expected batch size and container capacity is a setup decision, and on variable throughput it may be tuned to keep grab counts in a sensible range.
Because the pacing comes from the flow computer's pulse-driven volume total, the sampler inherits the metering system's own sense of how much has flowed. If the meter, its K-factor, or the pulse signal has a problem, the sampler's pacing is affected too, which is one reason the sampler is treated as part of the integrated measurement system rather than a standalone gadget bolted onto the line.
A proportional-to-flow sampler produces its own diagnostic information: a running grab count, the configured grabs per unit volume, and often a fault or low-container indication. Because the sampler is paced by the flow computer and its grab count is meaningful only against the volume that flowed, those signals are most useful when read alongside the metering data - which is exactly what a SCADA system brings together.
A cloud SCADA platform like Merobix reads those sampler tags from the flow computer or controller - grab count, batch volume, and any sampler alarm - and presents them next to the flow it is proportioned to. That lets an operator confirm the sampler is actually keeping pace with flow across a transfer instead of discovering at ticket time that it stalled halfway through and the composite is not representative. A grab count that has stopped advancing while volume keeps climbing is an obvious remote red flag.
That kind of monitoring matters because a failed sample is a quality dispute waiting to happen. If the composite does not represent the batch - because the sampler jammed, ran out of container capacity, or was paced against a faulty pulse - the water and quality results applied to a settlement are wrong. Watching grab count against flowed volume in real time gives the operator a chance to catch and correct the problem while the batch is still running rather than after the oil is gone.
Flow-proportional sampling takes a grab after each fixed increment of volume passes, so the number of grabs tracks how much oil flowed. Time-proportional sampling takes grabs on a fixed clock schedule regardless of flow. Flow-proportional is preferred for custody measurement because it keeps the composite representative when the flow rate varies.
The flow computer accumulates volume from the meter's pulses and sends a grab command to the sampler each time a configured volume increment has passed. The sampler itself does not measure flow; it simply fires a grab on command. This pacing is what makes the sampling proportional to flow rather than to time.
Enough that the composite genuinely represents the batch and the container fills appropriately, with the grab-per-volume increment set so that expected batch sizes produce a suitable number of grabs. Representative sampling practice under API MPMS Chapter 8.2 guides how the increment is chosen. Too few grabs risks an unrepresentative sample; too many can overfill the container.
This page references the standards, specifications, and official documentation published by the organizations below. Editions, product capabilities, and documentation change over time - confirm current requirements and specifications directly with the source.
Last reviewed: July 27, 2026. Merobix is not affiliated with, endorsed by, or sponsored by these organizations; their names are used only to identify the standards and products discussed.
Merobix reads your field devices into a cloud SCADA - the real thing behind these terms, live in days from any browser.