A flare gas recovery (FGR) system captures gas that would otherwise be flared and compresses it back into the fuel or sales system. Because the amount of gas arriving at the flare header swings wildly, from a trickle during steady operation to a flood during an upset, a single fixed-speed compressor cannot follow the load well. Compressor sequencing is the control scheme that decides which compressors run, at what loading, and when to fall back to the flare, so that recovery is maximized without letting the header pressure run away. It is the brain that turns a set of machines into a system that quietly minimizes flaring.
FGR Compressor Sequencing in one line: A flare gas recovery compressor sequencing scheme is the control logic that stages one or more recovery compressors to hold the flare header at a slight positive pressure, adding and shedding capacity as the recovered gas rate rises and falls. It follows the load with lead-lag staging, routes recovered gas to fuel or sales, and bypasses excess gas to the flare only when compression capacity is exceeded.
The controlled variable in an FGR system is the flare header pressure, held at a small positive value just above the flare seal so gas flows toward the compressors instead of to the flare tip, but not so high that it risks pushing back into the sources feeding the header. The compressors act as the outlet for the header: when they pull more gas than arrives, header pressure falls; when they pull less than arrives, header pressure rises. Sequencing is therefore fundamentally a load-following exercise, matching total compressor suction to the incoming flare gas rate in real time.
A single variable-speed or liquid ring compressor can handle modest swings by throttling suction or spilling back a recycle to keep the header at setpoint. Liquid ring compressors are common in FGR service because they tolerate the wet, dirty, variable gas the flare header delivers and can run against a swinging load. But turndown is limited, and one machine cannot both idle gently on a trickle of purge gas and swallow a large process upset. That is why most systems that matter are built around multiple machines with staging logic between them.
The scheme has to reject noise as well as follow trend. Flare header rate is inherently choppy, so the sequencing logic uses deadbands, minimum run and rest timers, and rate-based decisions rather than reacting to every pressure blip. The goal is a header that hovers at setpoint with the smallest number of machines running at efficient loading, not one that hunts compressors on and off with each fluctuation.
With more than one compressor, sequencing follows a lead-lag pattern. A lead machine runs continuously and modulates to hold header pressure; when it reaches its capacity and header pressure keeps rising, a lag machine is staged in to add capacity, and it drops out again when the load falls back. Rotating which machine is designated lead spreads run hours evenly and lets one unit be taken offline for maintenance while the others cover the load. The staging thresholds are set with hysteresis so a lag machine is not started and stopped repeatedly around a single load point.
No practical FGR system is sized to compress the absolute worst-case flare event, because that machine would idle almost all the time. The last line in the sequence is therefore a bypass to the flare: when incoming gas exceeds total installed compression capacity and header pressure climbs past a defined limit, excess gas is allowed to pass to the flare and burn, exactly as an unrecovered system would. This is a deliberate, safe fallback, not a failure. The design intent is that steady-state and routine swings are fully recovered, while genuine large upsets still have the flare as the pressure relief path.
Sequencing also governs where the recovered gas goes on the discharge side. Recovered gas is typically directed first to plant fuel, since displacing purchased fuel gas is the most reliable use, and any surplus beyond fuel demand is routed to the sales or reinjection system if quality and pressure allow. The control scheme coordinates suction staging with this discharge routing so that recovered gas has somewhere to go, because a compressor with no discharge outlet is as useless as one that never starts.
Effective sequencing needs a single point of coordination that sees the whole system, and that is where a SCADA platform earns its place. It brings together the flare header pressure, the incoming flare gas flow, each compressor's status, loading, suction and discharge pressures, and the downstream fuel and sales demand. Only with all of that in one view can the logic, or the operator supervising it, decide correctly whether to load the lead machine harder, stage in the lag, or accept a bypass to flare.
For the widely distributed and often unmanned sites where FGR is deployed, cloud SCADA turns this coordination into something operators can watch and prove from anywhere. Merobix historizes the header pressure and the recovered-versus-flared split so a site can quantify how much gas it actually kept out of the flare over a shift, a day, or a month. That record matters both for the economics, since recovered gas has real value, and increasingly for emissions and regulatory reporting, where minimizing flaring is a documented obligation rather than a nicety.
The trended data also drives tuning and reliability. Frequent short bypass events point to compression that is undersized for the routine load or to staging thresholds set too conservatively; frequent lag-machine cycling points to hysteresis that is too tight; a lead machine that never reaches full load points to a recycle valve leaking or a suction restriction. By reviewing how the sequence behaved against real header events, operators can adjust setpoints, timers, and staging order to push more gas into recovery and less into the flare without risking the header pressure control that keeps the whole system safe.
It controls the flare header pressure, held at a small positive value so gas flows toward the recovery compressors rather than to the flare tip. The compressors are the outlet for the header, so the sequencing logic adds or sheds compressor capacity to keep suction matched to the incoming flare gas rate and hold that header pressure at setpoint.
No FGR system is economically sized to compress the largest possible flare event, because such a machine would sit idle almost all the time. When incoming gas exceeds total compression capacity and header pressure rises past a set limit, excess gas is deliberately allowed to pass to the flare and burn. This is a designed, safe fallback that keeps the flare as the pressure relief path for large upsets while routine gas is still recovered.
A lead compressor runs continuously and modulates to hold header pressure. When it reaches capacity and pressure keeps rising, a lag compressor is staged in to add capacity, then drops back out when the load falls. Staging thresholds use hysteresis and timers to avoid rapid cycling, and rotating which unit is lead spreads run hours and allows maintenance without stopping recovery.
Merobix reads your field devices into a cloud SCADA - the real thing behind these terms, live in days from any browser.