Factory Acceptance Testing of a BOG compressor should establish much more than whether the machine runs and reaches an acceptance point. For the owner, the real value of FAT is to understand how the compressor behaves across its operating envelope, how its individual stages and auxiliary systems respond, and what reference data should follow the machine into commissioning and sea service.
A BOG compressor is not an isolated machine. It is part of a larger gas system, connected to upstream BOG conditions and downstream consumers, piping, valves, pressure-control functions and safety systems. A useful FAT therefore starts with understanding what the compressor will actually be expected to do on board.
Passing the FAT proves an acceptance condition. A good FAT should also establish how the compressor is expected to behave as part of the vessel's gas system.
1. Start with the vessel operating envelope — not only the compressor datasheet
Before defining FAT points, the intended onboard operating philosophy should be understood. The compressor discharge does not terminate at the FAT test boundary in service. It feeds a downstream system with its own pressure losses, control settings, consumers and protection philosophy.
The FAT preparation should therefore consider the expected suction and discharge conditions, required capacity range, downstream pressure requirement, recycle philosophy, consumer demand and the intended operating modes of the vessel.
This becomes particularly important where two compressors may operate individually or in parallel, or where the system can move between part load, full load and recycle operation. Knowing the downstream design and operating logic helps establish a realistic frame for the FAT and provides a much stronger reference for later operation at sea.
2. Establish a performance fingerprint — not only a guaranteed point
Capacity and discharge pressure are important acceptance parameters, but they do not describe the complete machine.
At selected operating points, the FAT should capture a consistent set of parameters that can later be compared with commissioning and service data. Depending on the compressor design, these may include capacity, speed, suction and discharge pressure, individual stage pressures, temperatures, power demand, cooling-water conditions, vibration and crankcase behaviour.
The objective is to establish the first performance fingerprint of the compressor before it accumulates service hours.
A single value tells us whether one requirement was met. A fingerprint tells us how the machine achieved it.
3. Look inside the compressor — stage by stage
For a multistage compressor, total suction and final discharge conditions are not enough. The behaviour of the individual stages can reveal information that disappears when only the overall compression ratio is considered.
Stage suction and discharge pressures, pressure ratios, suction and discharge temperatures and interstage cooling performance should therefore be reviewed together.
An unexpected temperature or pressure distribution does not automatically prove a defect. But it identifies where the next question should be asked.
Do not only ask whether final discharge pressure is correct. Ask whether the compression process that produced it is behaving as expected.
4. Cooler performance deserves its own baseline
Intercoolers and aftercoolers directly influence compressor performance. Their behaviour should therefore be recorded during FAT rather than treated only as auxiliary equipment.
Cooling-medium inlet and outlet temperatures, gas inlet and outlet temperatures and relevant pressure losses provide a useful baseline. Where available, cooling-medium flow should also be recorded.
This information becomes especially valuable later. Even compressors from the same production series may contain cooler units with small manufacturing differences. Recording their initial performance allows the owner to recognise these differences from the beginning rather than discovering them only after commissioning.
There is also a practical shipyard consideration. The vessel's water-glycol or cooling-water system may contain contamination if initial flushing and filtration have not been performed effectively. If a cooler has narrow internal passages — particularly a compact or printed-type construction — contamination can progressively reduce heat-transfer performance or increase pressure loss.
A good FAT baseline therefore provides something against which later onboard cooler behaviour can be compared.
5. AIR FAT and actual BOG operation are not the same condition
Where the compressor FAT is performed with air or another substitute medium, the owner should understand how the test results relate to the compressor's later operation with actual BOG.
Gas properties influence density, mass flow, compression behaviour, discharge temperature and power demand. FAT values obtained with air should therefore not simply be copied into the vessel's expected BOG operating condition without understanding the applicable conversion or performance basis.
The FAT should make that relationship visible. The purpose is not to make air behave like BOG, but to establish which FAT parameters can be used later as meaningful references and how they should be interpreted.
6. Recycle operation is part of the performance envelope
A compressor does not spend its entire life at one stable design point. Changes in BOG generation, consumer demand and downstream pressure can move the machine through different loading conditions.
The recycle or capacity-control system should therefore be observed dynamically. The FAT should show how the compressor responds when demand is reduced, when recycle increases, when the machine returns toward load and when operating conditions stabilise again.
Useful observations include pressure response, valve behaviour, temperatures, vibration, power demand and the stability of the control loop.
A stable full-load point does not prove stable control during transition.
7. Transients often reveal more than steady-state operation
Steady-state measurements are necessary, but some of the most useful observations appear while the machine is changing condition.
Start-up, loading, unloading, recycle transitions and shutdown should be observed as complete sequences. The question is not only whether each individual valve or controller operates, but whether the complete machine responds in a stable and predictable way.
These transitions also provide useful reference information for later commissioning, when the compressor interacts with the real vessel automation and gas consumers.
8. Vibration should be trended, not reduced to one acceptance number
A vibration value below an acceptance limit is important, but the pattern across operating conditions is often more informative.
Measurements should be associated with the corresponding compressor load, speed, stage pressures and temperatures. Where practical, the same measurement positions should be retained for later commissioning and service checks.
This creates a baseline against which gradual mechanical changes can be recognised.
9. Oil-free integrity should be treated as a system function
For an oil-free BOG compressor, keeping the process gas free from lubricating oil is not merely a descriptive feature of the machine. It depends on the integrity of the separation between lubricated mechanical spaces and the gas compression path.
The FAT should therefore consider the relevant sealing, distance-piece, drain, vent and crankcase arrangements together with the associated pressure behaviour and alarms.
Where crankcase pressure or other integrity-related parameters are available, their normal FAT values should be recorded. These values can later become useful diagnostic references during operation.
Oil-free operation should be understood as an operating condition to be protected, monitored and trended.
10. Sister compressors should be compared with each other
For a vessel fitted with two compressors of the same design, FAT creates an additional opportunity.
Do not review each machine only against the acceptance limit. Compare them with each other.
Capacity, specific power, stage pressure ratios, discharge temperatures, cooler performance, vibration, crankcase pressure and oil-free integrity can all be placed side by side.
The results do not have to be identical. Manufacturing tolerances and instrumentation will naturally create differences. In addition, individual components — including coolers — may show small performance differences within the same production series.
That is precisely why the comparison is useful. Recording the parameters during FAT establishes the differences while both machines are still new and before shipboard influences are introduced.
This becomes particularly valuable after installation. A difference in cooler performance observed on board may originate from the machine itself, but it may also result from the vessel system. Contamination of the water-glycol circuit, insufficient initial flushing or filtration, restricted flow or fouling of narrow cooler passages can change the thermal behaviour seen during commissioning.
If the original FAT data are available for both compressors, the owner has a much stronger basis for distinguishing an original machine characteristic from a change introduced during installation or service.
An unexplained difference does not automatically mean defect. It means information — and information recorded early is considerably easier to interpret later.
11. Test the protection philosophy — not only the machinery
A FAT should also demonstrate how the compressor protects itself when conditions move outside the intended operating envelope.
Relevant alarms, shutdowns, permissives and interlocks should be considered as sequences rather than isolated digital signals. Where a controlled test can be performed safely, the response should show what the machine does and what the operator sees.
A correctly operating sensor is only one part of the test. The important question is whether the complete protective response is understandable, timely and consistent with the intended vessel operating philosophy.
12. Remember that the compressor is part of a larger BOG system
The compressor FAT boundary may end at its inlet and outlet flanges. The owner's engineering boundary should not.
The expected conditions downstream of the compressor should be understood before the FAT is finalised. This includes piping pressure losses, design pressures, control-valve settings, pressure-control philosophy, PSV arrangement and the pressure requirements of the final gas consumers.
The same applies to operational logic. Can both compressors run in parallel? Is one normally standby? Can both operate at full load simultaneously? How is load sharing handled? What happens when consumer demand suddenly decreases? Which pressure controller becomes dominant, and where does excess capacity go?
These questions help scale the FAT correctly. They also prevent a common problem: accepting the compressor as an isolated package and only later discovering that its control range, pressure behaviour or operating philosophy does not align cleanly with the vessel system.
The compressor can pass its FAT and the BOG system can still have an integration problem.
13. FAT data should follow the compressor to the vessel
The FAT report should not become a document that is archived after delivery.
The most useful parameters should be carried forward into commissioning and gas trials. The first real BOG operation then becomes an opportunity to compare the installed machine against the factory baseline and the maker's expected AIR-to-BOG correlation.
Where the vessel automation permits, key parameters should remain available for trending. The operator then has a reference for recognising gradual changes in compressor efficiency, cooler performance, stage behaviour, vibration or control stability.
14. The FAT should create questions as well as answers
A useful FAT does not only close acceptance items. It can expose assumptions that deserve to be checked against the final vessel installation.
An unexpected pressure distribution may lead back to a piping calculation. A cooler result may justify checking the vessel cooling-medium flow. A recycle response may raise a question about downstream control settings. A difference between sister compressors may identify a parameter worth following during gas trials.
That is not evidence that the FAT failed. It is evidence that the FAT produced useful engineering information.
A good FAT should reduce uncertainty — and identify clearly where uncertainty still remains.
A BOG compressor FAT should establish a reference, not only an acceptance result.
The owner should leave the factory knowing not only that the compressor can achieve its required duty, but how the machine behaves across its operating envelope.
That means understanding stage behaviour, cooler performance, recycle response, vibration, oil-free integrity and the relationship between the FAT test medium and actual BOG operation.
It also means looking beyond the package boundary. Compressor capability, downstream pressure requirements, PSV and control philosophy, consumer demand and parallel operating logic belong to the same engineering picture.
For sister compressors, the opportunity is even stronger: establish both fingerprints while the machines are new, understand the differences from the beginning, and use those references when the vessel enters service.
Do not only ask: “Did the compressor pass?” Ask: “What have we learned today that will help the crew operate it — and help the owner recognise when its behaviour begins to change?”
Independent owner-side technical judgement for vessel machinery, LNG fuel systems, newbuilding supervision and technical due diligence.