A sampling procedure is the workflow an operator follows to capture a small quantity of process fluid that truly represents the larger stream it came from. Getting a sample that means something is harder than it sounds: the sample point must be purged, the right container chosen, the material handled so it does not change, and the sample labeled and delivered to the lab with an unbroken chain of custody. A sloppy sample gives a confident lab result about the wrong material. This guide covers the procedure itself and how continuous analyzer data in SCADA reduces how often manual samples must be pulled.
Sampling Procedure in one line: A sampling procedure is the standardized workflow for obtaining a representative sample of a process fluid: purging the sample point of stagnant material, selecting and preparing a suitable container, drawing and handling the sample so it does not change, and labeling it with the information the lab and the chain of custody require. The aim is a sample whose analysis reflects the actual stream, not the sample point.
The governing goal of any sampling procedure is representativeness: the analysis is only useful if the small sample matches the composition of the stream it was drawn from. The single biggest threat to that is stagnant material sitting in the sample point, dead legs, and lines, which no longer looks like the flowing product. That is why the procedure begins with purging, drawing and discarding enough fluid to clear the stale material so what is finally captured is fresh, live product from the main stream.
Container selection is the next decision, and it depends on the fluid and the analysis. A volatile liquid or a live crude that will flash off light ends needs a container that holds pressure or is filled and sealed to prevent loss, because a sample that vents its light components on the way to the lab reports as heavier than the real product. A stable liquid may go into an open container, but even then cleanliness matters, since residue from a previous sample or contamination from the container itself becomes part of the result. Choosing and preparing the right container is part of getting the sample, not an afterthought.
How the sample is drawn and handled after capture decides whether the representativeness survives the trip to the lab. The operator fills the container in a way appropriate to the fluid, avoids introducing air or contamination, seals it, and protects it from temperature swings or agitation that could change it. A sample that is correct at the sample point but mishandled afterward, left in the sun, partly emptied, or cross-contaminated, arrives at the lab no longer representing the stream, and the analysis, however precise, describes something that no longer exists.
A perfect sample is worthless if no one can be sure what it is or where it came from, so labeling is a core step of the procedure, not a formality. The label ties the sample to a specific point, time, and stream, and typically records who took it and under what conditions. When a lab result comes back, that label is what lets an engineer connect the analysis to the right meter, tank, or line and to the moment the sample was drawn, so the result can be acted on correctly.
For samples that support custody transfer or compliance, chain of custody documents the sample's handling from the field to the lab, showing who had it and when, so the result can be defended if it is ever challenged. A break in that chain, a sample that sat unaccounted for or changed hands without a record, opens the door to doubt about whether the analyzed material is really the sample that was taken. The paperwork exists so that a number with money or a regulation riding on it cannot be dismissed as untraceable.
This documentation discipline is why sampling procedures are written down and followed the same way each time. Consistent purging, consistent containers, consistent labeling, and consistent handoffs make results comparable over time and defensible under scrutiny. When a sample gives a surprising result, a well-documented procedure lets a reviewer distinguish a real change in the process from a sampling mistake, which is exactly the question that matters when the number is disputed.
Manual sampling is labor, and it is only a snapshot: each sample describes the stream at one instant, so between samples the process is unmeasured and a shift in quality can go unseen until the next sample happens to catch it. Online analyzers change that by measuring properties continuously in the line, so the stream is watched all the time rather than sampled occasionally. Where an analyzer covers a property, the reason for pulling frequent manual samples of that property shrinks to verification.
A cloud SCADA platform such as Merobix brings analyzer output into a continuous, time-stamped record an engineer can watch from a browser. Instead of waiting days for a lab result on a sample taken this morning, the operator sees quality trending in real time and is alerted when it moves out of range. That continuous view means manual samples can be reserved for calibrating and confirming the analyzer and for the analyses no online instrument performs, rather than being the primary way quality is checked.
The manual sampling procedure does not disappear, and for many custody and compliance analyses a physical sample is still required. But the frequency and the anxiety around it drop when a validated analyzer is watching the stream between samples. The procedure remains the reference for how a correct sample is taken, and the SCADA-borne analyzer data both reduces how often that procedure must be run and provides the continuous context that tells an engineer whether a lab result on a sample still reflects the stream today.
Fluid sitting in the sample point, dead legs, and connecting lines is stagnant and no longer matches the flowing stream. Purging draws off and discards that stale material so the captured sample is fresh, live product from the main line. Skipping the purge produces a sample of the sample point rather than of the process, and the lab result describes the wrong material.
A representative sample matches the composition of the stream it was drawn from at the point and time of sampling. Achieving that requires purging stale material, choosing a container that prevents the fluid from changing, and handling the sample so it does not lose light ends, gain contamination, or otherwise alter before analysis. Without representativeness, even a perfectly precise lab analysis describes something other than the actual process.
Not entirely. Analyzers measure properties continuously and greatly reduce how often manual samples are needed, but manual samples are still used to verify and calibrate the analyzer and for analyses no online instrument performs, including many custody and compliance requirements. The practical effect is that manual sampling becomes less frequent and more targeted while the analyzer watches the stream in between.
Safety & engineering notice. This article is general educational information, not site-specific engineering, safety, or legal advice, and it does not reflect any particular facility. Standards and regulations (for example OSHA, API, IEC, ISO, NFPA, NIST, and NERC CIP requirements) change and vary by edition, jurisdiction, and application. SCADA and remote monitoring cannot verify physical isolation, atmosphere, lockout/tagout, permit status, or a safe go/no-go decision. Qualified personnel must perform site-specific engineering, hazard analysis, and safety review, and confirm current requirements with the authority having jurisdiction, before acting.
Merobix reads your field devices into a cloud SCADA - the real thing behind these terms, live in days from any browser.