A gas turbine services manager once chased a run of trips flagged as elevated exhaust temperatures, each one looking like a hot-section problem that any reasonable engineer would start tearing into first. The likely culprit turned out to be a petroleum-based contaminant riding in the gas supply that the installed filters were not configured to catch. Nothing was wrong with the turbine; the fuel was the problem.
That pattern repeats across the fleet more often than operators like to admit. A trip gets logged against the machine, the hot section gets inspected, and the real story is sitting a hundred feet upstream in a knockout vessel, a cold pipe run, or a filter element nobody specified correctly. Walk the problem backward in stages and the fuel-side story usually tells itself.
Stage One: The Gas Leaves the Pipeline Dirtier Than the Spec Sheet Says
Pipeline gas is a moving target. Composition shifts with the source, compressor stations add trace lube oil, and pigging operations stir up solids that sit in low spots for weeks before a pressure change pushes them downstream. The gas showing up at the inlet flange today is not the gas the commissioning report characterized.
That matters because the turbine OEM wrote its fuel specification against clean, dry, superheated gas. Anything the pipeline hands over that falls outside those limits becomes the conditioning system’s problem to solve, not the turbine’s problem to tolerate. Operators who treat the inlet gas as a known quantity instead of a variable one tend to find out the hard way.
Stage Two: Liquids and Solids Slip Past Undersized Separation
The first real line of defense is separation, and it fails in predictable ways. A knockout drum sized for average flow gets overwhelmed during a swing. A filter-coalescer with the wrong element passes aerosols small enough to re-form as liquid downstream. A drain that was supposed to be checked weekly gets checked monthly, then quarterly, then not at all.
A properly sized separation train, usually a two-stage arrangement with a coalescing filter, is what a modern fuel gas conditioning package is built around. The design targets the specific particulate and liquid loads the turbine OEM will not tolerate, and it leaves instrumentation in place so operators actually know when an element is loaded instead of guessing from a differential pressure gauge nobody trends.
Why the fuss over particulate? Erosion is cumulative, and it stays invisible until a trip makes it obvious. Industry troubleshooting work has flagged that high particulate loading in gas supply lines can erode fuel nozzle tips within a single inspection interval, and that most fuel-related trips happen during fuel changeover transitions rather than during steady-state running when everyone is watching.
Stage Three: Pressure Drop Chills the Gas Below Its Dew Point
Even clean, dry gas will drop out liquids if you cool it enough, and the easiest way to cool gas is to drop its pressure. The Joule-Thomson effect across a control valve can pull inlet temperature down by tens of degrees in a single stage. Gas sitting anywhere near its hydrocarbon dew point going into that valve will be sitting below it coming out.
OEM fuel specifications are blunt about this. A common requirement is no liquids at the turbine and a meaningful superheat margin at the control valve inlet, specifically to keep the expansion through the valves from condensing anything in the manifold. Hit the valve with gas that’s already close to saturation and you’ve engineered a two-phase problem into the combustor.
Stage Four: Heating Gets Treated as Optional Until the First Trip
Fuel gas heating is the step most often value-engineered out of a package, usually with the argument that the gas looks warm enough at the inlet. It almost never is after the pressure cut. A performance heater sized against the worst-case inlet temperature, the worst-case pressure letdown, and the heaviest expected hydrocarbon fraction is one of the more reliable ways to keep the gas above its dew point all the way to the manifold.
A few things usually get missed when the heater is sized or re-rated:
- Heaviest hydrocarbon fraction. Dew point is set by the heavy end, not the average composition on the gas chromatograph.
- Worst-case pressure letdown. Sizing against nominal upstream pressure misses the Joule-Thomson swing on a cold morning with a low pipeline header.
- Ambient and startup conditions. A heater that holds superheat at steady state can still fall behind during a cold start or a rapid load change.
Stage Five: The Trip That Finally Gets Logged
By the time the control system sees a problem, the fuel-side chain has usually been breaking down for a while. Nozzle wear shifts the fuel split. A slug of liquid changes the local heat release. Exhaust spreads widen, a thermocouple goes out of band, and the protection logic does its job and takes the unit offline.
The root-cause investigation then has two paths. The easy one blames the turbine and schedules a borescope. The useful one walks back through separation, pressure control, and heating, pulls the gas chromatograph trend, checks the drain log, and asks whether the superheat margin held during the hour before the trip. The useful path is slower, and it’s the one that keeps the next trip from being a repeat of this one.
