The report comes back clean. Methane, ethane, CO₂, nitrogen, heating value, H₂S, water content, all broken out to the decimal and signed. The numbers look precise enough to build a custody transfer contract on.

But the chromatograph can only measure what reaches it. If the sample that arrived at the lab no longer matched the gas in the line, the analysis is a flawless measurement of the wrong thing. And that failure never shows on the report, which is exactly why it gets trusted.

Everything that decides whether the cylinder holds a true picture of the gas happens in the field, before anyone runs an analysis. The lab cannot fix a sample that went wrong at the tap. It can only measure what it was given.

Gas composition data starts before the lab

A natural gas analysis measures what is in the cylinder, and the cylinder is only useful if what is in it matches what is in the pipe. That match is made or lost at the sampling point.

The heavy ends are usually where it slips. Natural gas holds its heavier hydrocarbons, the C6+ and above, in the vapor phase only as long as pressure and temperature stay up. Let the sample cool, or drop its pressure across a valve while filling the cylinder, and those heavy components begin to condense. The gas that reaches the lab is now lighter than the gas in the line, part of its heavy fraction sitting as liquid in the bottom of the cylinder or left behind at the tap. Every number that follows is skewed, and no step in the lab can put the heavy ends back.

Contamination pushes the composition the opposite direction. Air pulled in through a bad purge adds oxygen and nitrogen the line never carried. Residual liquid or debris in a dirty cylinder carries into the next sample. Moisture picked up along the way shifts the water content and can distort the dew point. What lands in the cylinder is what the lab reports, whether or not it still resembles the gas in the pipe.

Where gas samples lose representativity

Most sampling problems trace to a short list of causes, and the right procedure at the tap heads off all of them.

Phase change from a pressure or temperature drop is the one crews hit most. It comes from pulling a sample too fast, filling across a wide pressure drop, or sampling where the gas is already near its dew point. Controlling how the sample is drawn, so the gas stays in the vapor phase all the way into the cylinder, is what prevents it.

Air intrusion follows a bad purge. A cylinder that was not properly evacuated still holds whatever it carried before, and that residual mixes into the sample. Dirty or wet cylinders do the same with liquids and debris. The sampling location matters just as much: a tap in a dead leg, or in a spot where liquids collect, draws gas that never represented the bulk flow no matter how carefully the cylinder is filled.

What these share is that none of them leave a mark on the analysis. The report looks complete, the lab did its job, and the composition still does not match the gas, because the error was sealed into the cylinder before it left location.

What gas composition data affects

A composition number is not an academic figure. It feeds decisions with money and safety attached, and each one inherits whatever the sample got wrong.

Heating value sets the basis for custody transfer and allocation. A sample that lost its heavy ends undervalues the gas and misprices every volume measured against it. Hydrocarbon dew point depends directly on the heavy fraction a sample is most likely to lose, so a dew point built from a light sample reads lower than reality and leaves the operator blind to where liquids will actually drop out in a line or a plant. Water content carries into the same dew point and corrosion picture. Oxygen, sulfur, H₂S, and mercaptan levels feed corrosion management, treating decisions, and contract compliance, and a sample that picked up oxygen through a bad purge or lost sour components on the way to the lab points those calls in the wrong direction.

Why field sampling and laboratory analysis have to connect

The gap that causes most sampling error is organizational. One party pulls the sample, another analyzes it, and no one owns the whole chain from tap to result.

Whoever takes the sample has to understand what the lab will do with it, and what will corrupt the measurement, to sample in a way that protects the analysis. Treat sampling as a task disconnected from the lab, and the crew at the tap has no way to know that filling too fast just cost the heavy ends, or that a shortcut on the purge just added oxygen the analysis will read as real.

FESCO helps close that gap by using trained laboratory personnel for sample acquisition and connecting field sampling procedures with the laboratory analysis that follows. FESCO’s field sampling and testing capabilities cover natural gas, separator-condition hydrocarbon liquids, CO₂, H₂S, mercaptans, oxygen content, water content, and hydrocarbon dew point, feeding analytical services that include routine and extended gas composition, sulfur speciation, and dew point work. Sampling and analysis run as one chain, handled by people who understand both ends.

The chromatograph is only as honest as the sample

Natural gas composition does not start at the chromatograph. It starts at the tap, in how the sample is drawn, what cylinder it goes into, and whether the gas that reaches the lab still matches the gas in the line.

A flawless analysis on a compromised sample gives numbers that are precise and wrong, and precise-and-wrong is the worse failure, because it reads as reliable and gets built on. Getting composition right means protecting the sample from phase change and contamination where it is collected, following procedure that keeps it representative, and running field sampling and laboratory analysis as one connected process instead of two handoffs. Do that, and the composition on the report is a true reading of the gas, which is the only kind worth a decision. Contact FESCO’s laboratory services team to discuss natural gas sampling and analysis for your composition, custody transfer, and emissions requirements.