A gas chromatograph separates a sample by carrying it through its column on a stream of inert carrier gas, usually helium or hydrogen, supplied from a bottle. That bottle depletes with every analysis, and when it runs empty the analyzer loses its carrier and goes dark, taking the energy measurement with it. Worse, the trouble often starts before the bottle is truly empty: as the supply pressure sags, the carrier flow drifts, retention times shift, and analyses can degrade before they stop altogether. This guide explains how a custody GC's carrier supply depletes, how a slow pressure drop shifts retention times and can spoil analyses before the bottle runs empty and the analyzer goes dark, and how a host trends carrier bottle pressure to schedule changeouts before an unplanned energy-measurement outage.
Carrier Gas Consumption in one line: Monitoring carrier gas consumption on a custody chromatograph matters because the carrier supply, typically a helium or hydrogen bottle, depletes with use, and when it runs out the analyzer stops measuring, halting the energy measurement it feeds. Before the bottle is fully empty, a falling supply pressure can reduce the carrier flow and shift the retention times the analyzer relies on, degrading analyses ahead of the outright loss. Trending the carrier bottle pressure lets operators see the supply approaching empty and schedule a bottle changeout during planned attention rather than suffering an unplanned outage when the carrier runs dry.
A chromatograph cannot analyze anything without carrier gas, because the carrier is what sweeps the injected sample through the column so its components separate and reach the detector in order. Every analysis consumes carrier, so the bottle supplying it draws down steadily over time, and the rate of consumption depends on the flow the method uses and how often the analyzer cycles. A busy custody GC cycling frequently works through its carrier faster than an occasionally used one, so the depletion is continuous and, at a given duty, fairly predictable.
The consequence of that depletion is stark for a custody point: when the carrier bottle empties, the analyzer can no longer run, and the energy measurement it feeds stops. At a custody transfer point the composition and heating value the GC provides are what turn metered volume into billed energy, so losing the analyzer means losing the ability to measure energy properly, which is a serious commercial and operational problem. An unplanned carrier runout therefore does not just inconvenience the instrument; it interrupts the measurement that the transaction depends on.
Because the supply is finite and its loss is costly, the carrier bottle is a consumable that has to be managed proactively rather than replaced reactively when the analyzer fails. The goal is to change the bottle on a schedule, during a planned visit or attention, before it reaches the point of causing trouble, rather than discovering it empty when the GC goes dark. Achieving that requires knowing how much carrier is left, which is what makes the carrier supply pressure a quantity worth watching continuously.
The failure is not a clean switch from working to empty; there is a degradation phase in between. A chromatograph relies on a stable carrier flow to produce consistent separation, and the components come off the column at characteristic times, their retention times, that the analyzer uses to identify each peak. Those retention times depend on the carrier flow being what the method expects. As the carrier bottle pressure falls toward empty, the pressure available to drive the flow drops, the flow can sag below its target, and the retention times begin to shift.
When retention times shift, the analyzer's ability to correctly identify and integrate each peak is undermined, because it expects components at particular times and the peaks are no longer arriving there. This can degrade the quality of an analysis, misassign or mismeasure components, before the carrier is fully gone, which means a slowly emptying bottle can start producing questionable results while it still appears to be running. A custody measurement that looks like it is working but is quietly drifting because of low carrier is arguably worse than a clean outage, because the bad data may be trusted.
This is why watching for the approach to empty matters more than just detecting the empty condition. By the time the bottle is truly dry the analyzer has already been through a period where its results may have been degrading, so the useful action is to intervene while the pressure is still adequate, well before the flow starts to sag. Catching the decline early keeps the analyzer in the pressure range where its flow and retention times are stable, which is the range where its analyses are trustworthy, and avoids the gray zone of degraded results near runout.
The practical tool for managing all of this is the carrier bottle pressure, monitored continuously and trended over time. A bottle's pressure falls as it empties, so the pressure is a direct indicator of how much carrier remains, and its downward trend, combined with the known consumption rate, lets an operator project when the bottle will reach the point of causing trouble. Rather than a single low-pressure alarm at the last moment, the trend gives an advancing forecast of the changeout, turning a surprise into a scheduled task.
This turns carrier management into planned maintenance. Knowing days in advance that a bottle is heading toward the intervention point, an operator can schedule the changeout to coincide with a planned site visit or attention window, order a replacement bottle in time, and swap it before the pressure ever falls into the range where retention times start to drift. The analyzer stays in its stable operating range continuously, and there is no unplanned outage and no period of degraded custody data, which is exactly the outcome the monitoring is meant to secure.
A cloud SCADA platform such as Merobix is well suited to this because it can trend the carrier bottle pressure for each GC, project the approach to the changeout point from the decline, and alarm early enough to schedule a swap rather than react to a runout. Because custody chromatographs are often at remote, unmanned points, having the carrier pressure trended remotely means the operator sees the bottle emptying at a site they are not standing in, and can dispatch a changeout before the analyzer degrades or goes dark. Watching a single trend that forecasts an outage days ahead is a far better position than driving out to a custody point only to find the GC already offline and the energy measurement already lost.
The analyzer can no longer run, because the carrier is what sweeps the sample through the column to separate it, so with no carrier there is no analysis. At a custody point that means the composition and heating value the GC provides stop, and the energy measurement that turns metered volume into billed energy is interrupted. Because that measurement underpins the transaction, an unplanned carrier runout is a serious commercial and operational problem, not just an instrument inconvenience.
A chromatograph needs a stable carrier flow to separate components consistently, and it identifies each peak by its retention time, the characteristic time it takes to come off the column. As the bottle pressure falls, the pressure driving the flow drops, the flow can sag below target, and retention times shift, so the analyzer starts expecting components at times they no longer arrive. This can misidentify or mismeasure components while the analyzer still appears to be running, producing questionable custody data before the outright loss of carrier.
The bottle pressure falls as the carrier empties, so trending it shows how much carrier remains and, with the known consumption rate, projects when the bottle will reach the point of causing trouble. That advancing forecast lets an operator schedule a bottle changeout during a planned visit and swap it before the pressure ever drops into the range where retention times start to drift. The analyzer stays in its stable range continuously, avoiding both an unplanned outage and a period of degraded custody data near runout.
Merobix reads your field devices into a cloud SCADA - the real thing behind these terms, live in days from any browser.