Knowledge / Ship systems

Marine-fuel conditioning: compatibility and cleanliness from tank to engine

Fuel identity, compatibility, cleaning, viscosity control and evidence from bunker delivery to the engine inlet.

On this page

Marine-fuel conditioning is the controlled preparation of fuel between delivery and combustion. It links tank management, separation, filtration, heating or cooling, measurement and engine requirements. A fuel can meet its delivery specification yet need careful treatment before use; fuel that burns successfully can still be unsuitable under an applicable emissions requirement. This article develops the engineering reasoning needed to keep those questions separate. It covers conventional liquid-fuel arrangements in general terms, not a universal operating procedure, contractual fuel assessment or approval for a particular alternative fuel.

Four different questions about the same fuel

Ask whether the delivered fuel matches the purchase specification, whether its use meets applicable regulation, whether the installed system can handle it and whether its condition at the engine inlet meets the maker’s requirements. These questions concern different boundaries. A laboratory result from a bunker sample cannot describe every change subsequently caused by tank contamination, blending, water ingress or poor treatment. Likewise, the engine inlet is not the correct point for reconstructing every property of the delivered batch.

For sulphur compliance, IMO’s sulphur-limit explanation identifies the MARPOL Annex VI framework, including the global 0.50% m/m limit and 0.10% m/m in applicable emission control areas, with approved equivalent arrangements addressed separately. These regulatory values are not cleanliness limits. Current geographical scope, implementation dates and the ship’s approved compliance method must be checked for the voyage. A separator does not turn high-sulphur fuel into compliant low-sulphur fuel.

Preserve batch identity through the tanks

A storage plan should show which batch is in each tank, the remaining quantity, relevant test status and intended route to the consumers. The transfer system may connect nominally separate tanks through common lines, returns or mixing arrangements. These interfaces deserve attention because segregation on a tank list is ineffective if operation quietly combines the contents elsewhere. Trace the physical path, including where recirculated fuel returns.

An engineering consequence is that usable inventory can be smaller than the total fuel on board. A batch awaiting assessment, an unsuitable blend or inaccessible tank contents may not be available for the next operating period. Fuel planning should therefore distinguish quantity from readiness. The same distinction helps prevent an apparent reserve from disappearing when a treatment restriction or tank problem is discovered close to departure.

Stability and compatibility are different properties

Stability concerns a fuel’s ability to retain its relevant components in a usable condition under the specified assessment conditions. Compatibility concerns the behaviour of fuels when mixed. Two individually acceptable fuels can produce an unacceptable mixture. CIMAC’s stability and compatibility guidance discusses these distinctions and the limits of test interpretation. It is industry guidance, not a statutory substitute for an approved fuel specification or maker’s instructions.

Avoid inferring compatibility solely from similar viscosity, colour or sulphur content. A useful operational question is what evidence supports the intended mixing ratio and conditions. A result for one blend ratio should not be stretched automatically to every possible ratio. Where uncertainty remains, preserve segregation and seek the appropriate technical assessment. Once incompatible fuel has formed sludge throughout a system, the problem extends beyond the contents of the original storage tank.

Follow the contaminants rather than only the flow

Water and particles can move through settling, centrifugal separation and filtration with different efficiencies. Settling offers time and a separation distance; a centrifuge increases the effective separating acceleration; a filter retains material according to its actual construction and rating. These mechanisms are complementary. A final filter should not be assumed to replace inadequate upstream separation, and a clean-looking liquid does not prove that damaging fine particles are absent.

CIMAC’s 2024 fuel-cleaning-system guideline connects fuel properties, system design and operating practice. Its scope includes commercial ships using relevant ISO 8217 fuels and both two- and four-stroke engines. Use it as engineering guidance while obtaining the applicable separator and engine requirements. The article’s following balance calculations illustrate reasoning; they do not establish a universal treatment efficiency.

A worked cleanliness balance

Assume a hypothetical, well-mixed feed contains 50 mg/kg of a measured particulate indicator. Suppose one treatment stage removes 80% of that indicator by mass, and a second stage removes 60% of what remains. The first outlet is 10 mg/kg and the second is 4 mg/kg. Combined removal is 92%, not 140%, because each percentage acts on a different incoming quantity. The arithmetic assumes no contamination is introduced between the stages.

At an assumed fuel mass flow of 2,000 kg/h, the incoming indicator mass is 100 g/h and the final outgoing mass is 8 g/h. The resulting 92 g/h difference must be consistent with a physical removal mechanism and its waste route. This does not predict real catalytic-fine removal: particle size, fuel properties, equipment condition and sampling uncertainty matter. Actual engine-inlet acceptance must use the applicable maker’s criterion, not the illustrative 4 mg/kg result.

Condition viscosity without chasing temperature alone

Temperature is usually a means of reaching a suitable fuel condition, rather than the ultimate requirement. Different fuels can have different viscosities at the same temperature. A temperature copied from a previous batch may therefore be inappropriate. Consider where the controlling measurement is made, the condition at the engine inlet and the effect of recirculation. The plant must also respect material, pressure and thermal limits throughout the route.

Heating is not always the correct response. Low-viscosity fuels may require a different thermal strategy, and cold-flow behaviour can constrain storage or transfer even before the engine is supplied. Use the actual fuel analysis and equipment instructions to establish the operating envelope. Do not extrapolate a heavy-fuel practice to every distillate or biofuel blend. A stable reading at one point is insufficient when the fuel experiences a materially different temperature elsewhere.

Read filter differential pressure in context

A rise in differential pressure shows a larger pressure drop across the measured section. At comparable flow and viscosity it can indicate increased restriction; interpretation also depends on element condition and valve arrangement. KSB’s head-loss explanation distinguishes flow-dependent loss from the resistance characteristics. A high value can reflect contamination or colder fuel; an unexpectedly low value may reflect lower flow or another configuration change. Interpret the trend against comparable operating conditions. Without that context, an alarm history may be misread as repeated filter failures.

Automatic backflushing changes both the operating pattern and waste production. Monitor whether the frequency is increasing, whether the intended function completes and whether the downstream condition remains satisfactory. Frequent cleaning is a symptom to investigate, not evidence that the system can accept unlimited contamination. Keep sufficient approved replacement elements and know the consequences of taking one filter train out of service. Never normalize an unapproved bypass as the solution to restricted flow.

Tank cleanliness can change the feed suddenly

Sediment stores a history of earlier fuel deliveries. Disturbing it can create a transient contaminant burden that a recent bunker analysis does not represent. MAN’s SL2019-674 on fuel-tank cleaning specifically warns that mobilized deposits can increase catalytic fines reaching the engine. This maker document concerns MAN B&W two-stroke engines and a particular service context; it does not establish a universal limit for all engines.

As an engineering interpretation, assess cleaning as an interface change. Ask where mobilized material will go, whether treatment capacity is sufficient and how the resulting feed will be verified. More mixing can improve the representativeness of one sample while sending more accumulated solids downstream. Tank entry, chemical use and cleaning methods require their own safety controls. This article does not recommend adding chemicals or entering a tank.

Changeover is a transient operation

During fuel changeover, the consumer may receive a changing mixture because lines, mixing vessels and return flows retain fuel. A valve movement is therefore not proof that the engine has received the new fuel. The relevant transit volume and actual circulation arrangement matter. Thermal changes can also affect clearances, leakage and viscosity, so the maker’s approved changeover limits must govern the operation.

For a simple thought experiment, a perfectly mixed 1 m³ vessel receiving and discharging 1 m³/h of a replacement liquid retains about 37% of the original liquid after one hour, under constant-volume ideal-mixing assumptions. This follows the exponential mixing model, not a ship-specific compliance calculation. Real fuel systems may include stratification, multiple volumes and recirculation. The example explains why “one volume has passed” is not necessarily equivalent to complete replacement, and why changeover records need a justified method.

Diagnose with evidence at the correct boundaries

Useful evidence includes the delivered batch sample, storage history, treatment inlet and outlet condition, engine-inlet condition and engine symptoms. These are different observations, not duplicate versions of one test. Define the question before choosing sampling locations. To assess a separator, compare representative conditions around that separator; to investigate an engine problem, also examine what happened after treatment and whether the symptoms have another plausible cause.

Preserve timestamps, batch and tank identifiers, operating conditions and the applicable laboratory method. A sample collected after the plant has changed mode may not represent the period when a symptom occurred. Correlation between a new batch and engine trouble is a reason for investigation, not automatic proof of cause. Quality of evidence matters especially when technical diagnosis also has commercial or contractual consequences.

Scope limits and a practical review

Alternative liquid fuels and blends can introduce additional material-compatibility, storage, microbial, oxidation or water-management questions. Their acceptability cannot be established by calling them “drop-in.” Obtain the fuel-specific approvals and current engine guidance. This article also excludes gas-fuel supply systems and their separate safety architecture. No public article can substitute for the installed system’s approved arrangement and operating instructions.

A useful review ends with a clear chain: identified batch, suitable storage, justified mixing decision, verified treatment, appropriate engine-inlet condition and traceable waste. Common errors are relying on sulphur alone, adding removal percentages, confusing delivery quality with engine-inlet quality and applying one temperature to every fuel. The central lesson is that fuel readiness is created and demonstrated through the whole path from tank to engine.

Sources