A custody flow computer needs a heating value to turn a measured volume of gas into an energy quantity, and normally it gets that heating value from a live gas chromatograph analyzing the stream. But analyzers fail, get pulled for maintenance, run calibrations, and lose communication, and the gas keeps flowing the whole time. Heating value substitution is what the flow computer or SCADA does to keep computing energy during those gaps, when the live composition is not available. Rather than stopping or reporting zero energy, it substitutes a heating value from an agreed fallback, and there is an accepted order of preference for where that fallback comes from. The substituted intervals also have to be logged and flagged clearly, because a period of energy computed on a substituted value must be visible for audit and reconciliation rather than blended invisibly into the good data.
Heating value substitution in one line: Heating value substitution is what a flow computer or SCADA does to keep calculating gas energy when the custody analyzer is offline and no live composition is available. Instead of stopping, it substitutes a heating value from an agreed fallback, typically holding the last known good composition, applying a contractual default value, or borrowing the reading from a nearby representative analyzer. Industry measurement practice defines a hierarchy of accepted substitution methods, and each substituted interval must be logged and flagged so the affected energy is visible for audit and reconciliation.
Energy, not volume, is what natural gas is bought and sold on, and energy is the measured volume multiplied by the gas's heating value. The flow computer gets that heating value from the custody gas chromatograph, which analyzes the stream and hands over a composition and the heating value derived from it. When the analyzer is healthy and communicating this is seamless, but the analyzer is a piece of instrumentation with its own failure modes: it runs scheduled calibrations, needs periodic maintenance, can lose carrier gas or sample flow, and can drop its communication link to the flow computer. During any of those events the live heating value simply is not there.
The gas, however, does not stop flowing while the analyzer is down. Volume keeps accumulating through the meter, and the flow computer still has to attribute an energy to that volume, because an interval of flow with no energy assigned would understate the delivered quantity and corrupt the accounting. So the flow computer cannot simply report zero heating value or refuse to compute; it needs some heating value to apply to the volume that flowed while the analyzer was unavailable. Substitution is the mechanism that supplies one, keeping the energy calculation continuous across the outage rather than leaving a hole in the record.
The goal of substitution is to make the substituted energy as close as possible to what the real gas would have measured, using the best available information about what the composition probably was during the gap. Because the substituted value is an estimate rather than a measurement, the method chosen matters, and so does the honesty of recording that an estimate was used. A good substitution scheme both picks the most defensible fallback and marks the interval clearly, so that the accounting stays continuous without pretending the substituted energy is as certain as measured energy.
There is an accepted order of preference for where a substituted heating value should come from, reflecting how likely each source is to represent the real gas. The first and usually preferred choice is to hold the last known good composition, the most recent valid analysis from the same analyzer before it went offline. For a stream whose composition is fairly steady, the gas flowing during a short outage is probably very similar to what was there just before the analyzer failed, so the last good value is often the best estimate of the current gas. This works well for short gaps and stable streams and less well the longer the outage runs and the more the gas could have changed.
A second option is to substitute the reading from a nearby representative analyzer, another custody GC on the same or a closely related stream whose composition should track the affected point. If a neighboring analyzer is measuring essentially the same gas, borrowing its live heating value can be a better estimate than holding an increasingly stale last-good value, especially for a longer outage where the composition may have moved. This depends on there genuinely being a nearby analyzer on comparable gas, which is not always the case, so it is available in some configurations and not others.
The third option is a contractual default heating value, a fixed value agreed in advance and written into the contract to be used when no better live source is available. This is the fallback of last resort in the sense that it does not reflect the current gas at all, only an agreed representative number, but it has the virtue of being unambiguous and pre-negotiated so neither party can dispute which value applied during the outage. Industry measurement practice, in the standards that govern electronic gas measurement, lays out this kind of hierarchy so that operators apply the methods in a defensible order rather than choosing arbitrarily, and so that an auditor can see that the substitution followed accepted practice.
The substitution itself is only half of correct practice; the other half is making every substituted interval visible. When a flow computer applies a substituted heating value, that fact must be recorded so that anyone reviewing the measurement later can see exactly which intervals used real composition and which used a fallback, and which fallback was applied. An interval of substituted energy blended silently into the good data is a problem, because it hides an estimate inside numbers that are otherwise measured, and it undermines the auditability that custody measurement depends on. Flagging the substituted periods keeps the estimate honest and traceable.
This is where a monitoring platform adds real value beyond the flow computer's own logs. A cloud SCADA platform such as Merobix can watch analyzer health across a fleet of meters, detect when an analyzer goes offline or a composition goes stale, and surface the substitution as an event rather than leaving it buried in a single device. Bringing the analyzer status, the substituted intervals, and the heating value actually used into one place lets a measurement team see at a glance which meters are running on substituted values right now and for how long, so that a lingering analyzer outage gets attention before it stretches into a large volume of substituted, and therefore estimated, energy.
The reason this visibility matters commercially is that substituted energy is estimated energy, and the longer it persists the larger the potential difference between the estimate and what real composition would have measured. An outage that lasts a shift on a stable stream may substitute negligible error, but an analyzer left offline for days, especially on a stream whose composition is drifting, can substitute a meaningful energy discrepancy that surfaces later as a reconciliation dispute. Alarming on analyzer downtime and trending how long each meter has been on a substituted value turns a silent accumulation of estimated energy into a prompt to fix the analyzer, and having the substituted intervals clearly flagged in the record is what lets both parties reconcile the affected period fairly when the analyzer comes back.
It substitutes a heating value so it can keep computing energy for the gas that is still flowing, rather than reporting zero. The usual choices, in order of preference, are to hold the last known good composition from the analyzer before it failed, to borrow the reading from a nearby representative analyzer on comparable gas, or to apply a fixed contractual default value. Each substituted interval is then logged and flagged so the estimated energy is visible for audit.
The generally preferred first choice is to hold the last good composition, since for a short outage on a stable stream the gas is probably close to what was there just before the analyzer failed. If a nearby analyzer measures comparable gas, its live reading can be a better estimate for a longer outage. A pre-agreed contractual default is the fallback when no better live source exists. Industry electronic measurement practice defines this hierarchy so substitution follows a defensible order.
Because substituted energy is estimated rather than measured, and blending it silently into the good data hides an estimate inside otherwise measured numbers and undermines auditability. Flagging every substituted interval, and recording which fallback was used, keeps the accounting honest and lets both parties reconcile the affected period fairly when the analyzer returns. The longer an outage lasts, the larger the potential estimation error, so clear flagging is essential for a defensible custody record.
Merobix reads your field devices into a cloud SCADA - the real thing behind these terms, live in days from any browser.