Two-Stroke Marine Diesel Engine Fundamentals

Why the slow-speed two-stroke dominates large propulsion

Three pieces of physics point the same way, and together they explain why almost every large merchant ship built since the 1980s carries a slow-speed two-stroke. A fixed-pitch propeller reaches its best open-water efficiency at low rotational speed, usually 70 to 110 rpm on a large hull, where the blade tips stay clear of cavitation and the wake is well matched. A two-stroke of a given bore turns at roughly half the speed of a four-stroke of the same bore, because it fires every revolution rather than every second one. Put those together and the engine’s natural speed already sits where the propeller wants it, so the shaft can be bolted straight through a thrust bearing to the propeller with no reduction gear.

Removing the gearbox is worth more than it first looks. A large reduction gear loses about 1.5 to 3 percent of shaft power as heat, costs a substantial fraction of the engine price, adds weight high in the engine room, and brings its own alignment and vibration work. Direct drive avoids all of it. The result is an installation that turns close to half the fuel’s chemical energy into shaft work: a modern slow-speed two-stroke reaches specific fuel oil consumption near 160 to 168 g/kWh at its best load point, against roughly 175 to 195 g/kWh for a comparable geared four-stroke plant. For a ship that burns tens of tonnes of fuel a day, that gap decides the freight economics.

The engine also tolerates the cheapest fuel afloat. Its long combustion window burns the slow-igniting heavy fractions of residual fuel, and its separately dosed cylinder oil handles the acid that high-sulfur fuel makes. For heavy fuel oil burners in particular, no other prime mover matches it on cost per unit of propulsion work. That combination, direct drive plus dirty-fuel tolerance plus high efficiency, is why the type owns container, bulk, and tanker propulsion.

How the type differs from a four-stroke

A two-stroke completes intake, compression, combustion, and exhaust in two piston strokes, one crankshaft turn. A four-stroke marine diesel engine needs four strokes, two turns, for the same events, a cycle set out in four-stroke marine diesel engine fundamentals . The mechanical consequences run deep, and they are easy to get wrong if you reason only from the firing frequency.

• Power per unit displacement. Firing every revolution roughly doubles the power a given swept volume can make at the same speed. In service the slow-speed two-stroke gives that advantage back by running at far lower rpm, so its power per liter ends up broadly comparable to a hard-working medium-speed four-stroke. The win is efficiency and fuel tolerance, not raw specific output.
• Mean piston speed. At rated speed a slow-speed two-stroke runs about 8 to 9.5 m/s mean piston speed, close to a medium-speed four-stroke. A 950 mm bore engine with a 3,460 mm stroke at 80 rpm sits near 9.2 m/s. The very long stroke, often more than three times the bore, keeps that speed moderate despite the huge dimensions, so wear and fatigue behave much as they do on smaller engines.
• Firing pressure. Both types run peak pressures of 160 to 200 bar. The two-stroke reaches slightly higher BMEP at its low speed because uniflow scavenging clears the whole cylinder each cycle.
• Crankcase environment. The piston rod and its stuffing box seal the crankcase off from the combustion space, so the crankcase oil sees no acid, no fuel dilution, and no cylinder-oil drainage. This is the reason cylinder oil is a separate system.
• Combustion-side oil consumption. A two-stroke deliberately loses 0.6 to 1.2 g/kWh of high-base-number cylinder oil to neutralize acid and lubricate the liner. A trunk-piston four-stroke loses maybe 0.2 to 0.4 g/kWh past the ring pack, from a single shared oil.

The uniflow cycle, phase by phase

Scavenging and charging

Near bottom dead center the descending piston uncovers a ring of scavenge ports low in the liner. Air in the scavenge receiver, at about 3.5 to 4.5 bar absolute at full load, flows up through the cylinder and pushes the burnt gas out through the open exhaust valve in the head. The ports are angled to give the incoming column a gentle swirl, but swirl in a slow-speed engine is intentionally weak; the high-pressure fuel sprays, not bulk air motion, do the mixing. The uniflow scavenging geometry, ports in and valve out, is what lets the gas move one way and clears the cylinder without short-circuiting fresh air to the exhaust.

Compression

As the piston rises it first covers the ports, then the exhaust valve seats. Only after both close does real compression begin, which is why the trapped, or effective, compression ratio is lower than the geometric figure. The charge reaches a compression pressure of roughly 120 to 160 bar and a temperature high enough for diesel auto-ignition, on the order of 700 to 900 K, at top dead center. The effective compression ratio sits around 15 to 21 depending on the design and its scavenge timing.

Combustion and expansion

Just before top dead center the fuel enters through two to five nozzle holes on each injector, fed either by a common-rail accumulator on electronically controlled engines or by cam-driven jerk pumps on the older mechanical designs. Combustion runs in two stages: a short premixed burn of the auto-ignited charge over a few crank degrees, then a longer diffusion burn at the spray edges as more fuel arrives. Peak pressure lands about 6 to 10 degrees after top dead center, and the gas then expands down the working stroke until the exhaust valve lifts, typically 110 to 130 degrees after top dead center.

Exhaust and cycle close

The exhaust valve opens before the ports uncover, so the high-pressure gas blows down under its own pressure into the exhaust manifold first. This pre-blowdown drops cylinder pressure below scavenge pressure, so when the ports open the fresh air flows in rather than the cylinder back-feeding the receiver. Pre-blowdown, then a full one-way purge, then port and valve closing, is the defining sequence of uniflow scavenging and the source of the type’s high volumetric efficiency. The pressure history through all four phases is what a chief engineer reads on the indicator diagram, covered in indicator diagram analysis .

Scavenging systems and the exhaust valve

Why uniflow won

Early two-strokes used cross scavenging (inlet and exhaust ports on opposite sides of the liner, a deflector-crown piston steering the flow) or loop scavenging (both port rows on the same side, the charge looping up and back). Both route intake and exhaust through the liner wall, and both suffer the same defects: fresh air short-circuits straight to the exhaust ports, burnt gas hides in the corners the loop cannot reach, and the port timing is symmetric about bottom dead center because a single crank drives it. Uniflow puts the exhaust in a head valve, so its timing is independent of the ports, the flow never reverses, and the stroke can grow long without the ports fighting each other. By the late 1980s the major builders had abandoned loop and cross scavenging for uniflow, and no production slow-speed engine has used anything else since.

The exhaust valve and its hydraulics

The single exhaust valve is a large poppet, often 300 to 500 mm across the seat, sitting in a water-cooled cage in the head. It cannot use a normal spring at this size and speed, so it opens hydraulically and closes on an air spring. A cam or, on electronic engines, an electrically triggered hydraulic actuator , sends oil to a piston that pushes the valve open; a cushion of compressed air below the valve spindle drives it shut and seats it gently. Many designs spin the valve a few degrees each lift, using small vanes in the exhaust stream, so the seat wears evenly and burns clean. The valve and its seat run hot and see every acid and vanadium attack the fuel carries, which is why exhaust-valve overhaul sets the top-end interval on most engines.

Turbocharging and the scavenge-air system

Every modern slow-speed two-stroke is turbocharged on the constant-pressure principle: all cylinders feed one common exhaust manifold that acts as a plenum, damping the pulses so the turbine sees a steady pressure. One or two large turbochargers, from suppliers such as MAN TCA, ABB, or Mitsubishi MET, take that flow and compress the scavenge air. Constant-pressure charging gives the best efficiency at high load, which is where these engines live, at the cost of weak response and poor low-load boost.

That low-load weakness is covered by auxiliary blowers. Below roughly 30 to 40 percent load, and during starting, exhaust energy cannot spin the turbocharger fast enough to hold scavenge pressure, so electrically driven blowers cut in automatically and force air into the receiver. They stop once the turbocharger sustains the pressure on its own. Between turbocharger and cylinder sits the charge air cooler, a large finned-tube or plate cooler that drops the compressed air from well over 150 degrees C back to about 45 to 55 degrees C. Cooling raises air density, so more mass enters the cylinder, and a water mist catcher downstream strips condensed water before it can wash the liner. The pressure balance across this train, compressor ratio minus cooler loss, sets the scavenge pressure the engine actually gets, related to the turbocharging system covered separately.

Mean effective pressures and the indicator diagram

The indicator diagram, the pressure-volume loop of one cylinder over one cycle, is the working record of engine health, and the mean effective pressures come straight off it. Indicated mean effective pressure (IMEP) is the constant pressure that, acting over the swept volume, would do the same work as the real gas loop. It measures the cylinder’s gas work before any friction.

Brake mean effective pressure (BMEP) applies the same idea to the brake power at the coupling, after friction and pumping losses. It normalizes torque against displacement, so a 350 mm engine and a 950 mm engine compare on one scale. Slow-speed two-strokes run 18 to 22 bar BMEP; the MAN B&W G95 reaches about 21 bar at its L1 rating, and WinGD’s X-series sits in the same band.

The gap between them is friction mean effective pressure (FMEP), the loss to bearing friction, piston-ring drag, and the work of pumping the charge. Mechanical efficiency is BMEP divided by IMEP, and for a slow-speed two-stroke it runs about 0.90 to 0.94, high because the engine turns slowly and carries few bearings for its size.

Two pressures off the same diagram carry most of the daily diagnostic weight: compression pressure, pcomp, with no fuel burning, and peak firing pressure, pmax, at full combustion. Their ratio, pmax over pcomp, sits near 1.2 to 1.4 at full load. A cylinder drifting from the pack on either figure points at a specific fault, and the ratio itself flags injection timing that has crept early or late.

The load diagram and rating margins

Every slow-speed two-stroke is sold against a load diagram : a power-against-speed box the engine may run inside, bounded by four corner points. L1 is the nominal maximum continuous rating, the top right corner and the highest type-approved point. L2 holds the L1 power at lower speed. L3 holds the L1 speed at lower power. L4 is the bottom left corner, the lowest rated speed and power. The buyer selects a specified maximum continuous rating, the SMCR or simply MCR, anywhere inside the L1-L2-L3-L4 box, and the engine is tuned for that point. Choosing an MCR down and to the left, toward L4, derates the engine for a lower SFOC and a slower propeller, which is why long-stroke families exist.

Day-to-day the engine does not run at MCR. The continuous service rating, the CSR, sits at about 75 to 90 percent of MCR, and the difference is deliberate margin. Sea margin, around 15 percent of power, covers the added resistance of weather, waves, and hull fouling between dry-dockings. Engine margin, about 10 percent, keeps the service point clear of MCR so the engine is not run flat out in normal service. Light running margin, roughly 4 to 7 percent expressed in rpm, offsets the propeller’s habit of absorbing more torque at a given speed as the hull fouls and the sea roughens; a fresh, clean ship therefore runs slightly light, with rpm to spare, so the fouled ship still reaches its speed without overloading the engine. Matching the propeller pitch to land the operating point correctly inside this diagram is one of the harder parts of a newbuild specification, and it interacts directly with any slow-steaming plan the operator intends.

Engine architecture

Crosshead against trunk-piston

Every modern slow-speed two-stroke is a crosshead engine . The piston rod bolts to the underside of the piston, runs down through a stuffing box in the diaphragm plate, and meets the crosshead in the running-gear space. The crosshead pin carries the top of the connecting rod, and the crosshead slides on guide bars that take the whole side-thrust component, so the piston itself feels no lateral load and stays centered in the liner. The payoff is the separate cylinder oil: because nothing from the combustion space reaches the crankcase, the cylinder oil can carry a high base number for acid control without ruining the bearing oil.

Older two-strokes, and all medium-speed engines, use a trunk-piston layout instead. There the connecting rod attaches straight to the piston through a gudgeon pin, the piston skirt takes the side thrust against the liner, and one oil serves both bearings and cylinders. It is shorter, lighter, and simpler, but the shared oil forces a compromise the crosshead engine never has to make.

Major components

The structure builds up from the bedplate, a heavy cast or fabricated base that holds the main bearing saddles and bolts to the ship’s seating. Above it sits the frame box, or A-frame, the side structure that carries the guide bars and supports the cylinder block. The cylinder block on top holds the liners, the jacket cooling water, and the scavenge air space. Long tie rods run from the bedplate up through the frame box into the block and are tensioned hydraulically, squeezing the three castings into one stiff column that carries the firing load in compression rather than through the bolts in tension.

Inside each cylinder unit is a liner (a honed cast-iron sleeve, often with bore-cooling drillings near the top), a cylinder head that carries the exhaust valve cage, the fuel injectors, the starting-air valve, and the cooling passages, and a piston with its crown, ring pack, skirt, rod, and crosshead. The cylinder-side wear parts are treated in cylinder liners and pistons , the rotating assembly in crankshaft and main bearings , and the valve gear in camshaft and valve train .

Cylinder count and configuration

Slow-speed two-strokes are always in line. A V or H form would spread an engine with a stroke of three or four meters into something too wide and too tall to fit a hull, and it would complicate the crosshead guides. Cylinder counts run from 5 to 14, even and odd both offered, and the count is chosen with the bore to hit the target power and speed. The crank-throw spacing is arranged to balance the inertia forces, and shorter engines are the awkward ones: a six-cylinder engine has well-known second-order balance issues and often carries external moment compensators at one or both ends.

Cylinder lubrication

Cylinder oil is metered onto the liner running surface through a ring of quills, timed to the piston’s passage, by an electronically controlled dosing system such as MAN’s Alpha lubricator with adaptive cylinder-oil control (ACC) or the equivalent on other engines. The system does two jobs at once: it lays an oil film for the rings, and it delivers alkaline reserve to neutralize the sulfuric acid that forms when sulfur in the fuel burns and meets condensed water on the liner. Feed rate runs about 0.6 to 1.2 g/kWh, scaled to engine load and to fuel sulfur content, with a mechanical minimum near 0.6 g/kWh below which the film goes too thin.

Base number, the measure of alkaline reserve, is matched to the fuel. Distillate and 0.50 percent VLSFO run on cylinder oils around BN 40; high-sulfur residual fuel behind a scrubber, up to 3.5 percent sulfur, needs BN 70 to 100, and ultra-high-BN oils reach about 140 for the worst fuel. Get the balance wrong and the engine pays for it. Too little feed or base number, or too cold a liner, and cold corrosion eats the liner surface as unneutralized acid attacks the iron. Too much, and calcium deposits build on the piston top land and scuff the liner. Watching liner temperature, scrape-down iron in used-oil analysis, and the feed rate together is routine superintendent work, and the low-load form of that balance is treated in slow steaming and engine cleanliness .

Performance: SFOC, power, and speed

Specific fuel oil consumption is the headline number, the fuel burned per unit of shaft work, quoted in grams per kWh and corrected to the ISO reference conditions and calorific value in ISO 3046 and ISO 15550 . Current production engines reach about 160 to 168 g/kWh near their best load point, typically 70 to 80 percent of MCR when tuned for part-load running. That figure converts directly to brake thermal efficiency, near 50 percent, through the fuel’s net calorific value.

SFOC is only comparable when the reference basis is stated, so any transfer between measured conditions, or between gas mode and diesel mode on a dual-fuel engine, has to be normalized first. The concept and its correction are set out in specific fuel oil consumption .

On size, the current families span cylinder bores from about 260 mm on the smallest legacy engines to 950 mm on the largest in production, the MAN B&W G95 with its 3,460 mm stroke. WinGD’s flagship X92-B runs a 920 mm bore. The largest bore ever built was 960 mm, on the retired Wartsila RTA96C and RT-flex96C. Continuous ratings run from about 3 MW for a small five-cylinder engine to roughly 82 MW for a fourteen-cylinder big-bore unit on an ultra-large container ship. Rated speed at the top of the diagram falls between about 70 and 105 rpm, with the small bores at the fast end and the big bores turning slowest to match a large, slow propeller.

NOx tiers and how two-strokes comply

NOx from marine engines is capped by MARPOL Annex VI Regulation 13, enforced through the NOx Technical Code 2008 adopted as Resolution MEPC.177(58). The limit is a weighted cycle value in grams per kWh, and it depends on the engine’s rated speed n in rpm, with three tiers set by build date. For the middle speed band, 130 rpm up to but not including 2000 rpm, the tiers follow curves in n:

Slow-speed two-strokes turn below 130 rpm, where the curves flatten to fixed values: 17.0 g/kWh for Tier I, 14.4 g/kWh for Tier II, and 3.4 g/kWh for Tier III. Tier II has applied worldwide to engines built on or after 1 January 2011, and most slow-speed engines meet it by tuning alone: injection timing, injection rate shaping, exhaust-valve timing, and charge-air handling. Tier III applies only inside a designated NOx Emission Control Area and only to engines built on or after the area’s start date, 1 January 2016 for the North American and United States Caribbean areas and 1 January 2021 for the Baltic and North Sea areas. Outside a NECA, a Tier III ship runs to Tier II. The NOx Tier I, II, III article carries the full regulatory detail.

Tier II tuning is a trade against fuel. NOx forms fastest where the flame is hottest, so the levers that cut it, retarded injection timing and a slower heat-release, also pull the peak temperature and pressure down and cost a little efficiency. Builders recover part of that loss with rate-shaped injection and with earlier exhaust-valve closing that traps a denser charge, a Miller-type effect, so a Tier II engine gives up only a few grams per kWh of SFOC against an untuned baseline. The setting is fixed in the certified Technical File, and a chief engineer cannot lawfully re-time the engine away from it.

Three paths reach Tier III on a two-stroke . Exhaust gas recirculation routes a cooled, scrubbed fraction of the exhaust back into the scavenge air, diluting oxygen and lowering peak combustion temperature to cut NOx formation at source. Selective catalytic reduction injects urea into the exhaust ahead of a catalyst, converting NOx to nitrogen and water for an 80 to 95 percent reduction; the SCR reactor sits before the turbine on a high-pressure system or after it on a low-pressure one. The third path is simply running on gas: a lean-burn LNG engine can hold NOx below the Tier III ceiling without any aftertreatment, which is one reason the dual-fuel engines have taken the LNG-fueled fleet.

Fuel modes

Diesel mode on residual fuel

The base fuel is residual heavy fuel oil, capped since January 2020 at 0.50 percent sulfur as VLSFO for ships without a scrubber, or run at up to 3.5 percent with one. The bunker quality and ISO 8217 grades set the specification. The engine accepts residual fuel because its long burn tolerates the slow-igniting heavy ends, its jacket water runs hot at about 85 to 90 degrees C at the outlet to keep the liner above the acid dew point, and its high-base-number cylinder oil neutralizes the acid that remains.

Dual-fuel LNG: ME-GI against X-DF

Two architectures split the LNG market, and they sit at opposite ends of the gas-pressure question. MAN’s ME-GI injects LNG at about 300 bar directly into the combustion chamber near top dead center, ignited by a small pilot of liquid fuel, and burns it on the diesel cycle. Because the gas burns in a diesel-like diffusion flame as it enters, almost none escapes unburned, so methane slip stays below roughly 0.1 g/kWh; the cost is a high-pressure fuel-gas supply system that adds weight and capital.

WinGD’s X-DF admits gas into the scavenge air at low pressure, 6 to 16 bar, mixes it homogeneously, and ignites the lean charge with a pilot injection at top dead center, burning on the Otto cycle. The low-pressure supply is cheaper and lighter, and the lean burn drops NOx below Tier III without aftertreatment, but some premixed gas slips past the exhaust valve before it burns, so first-generation X-DF ran methane slip near 2 to 4 g/kWh. WinGD’s answer is iCER, intelligent Control by Exhaust Recycling, a low-pressure EGR path with water that slows and completes the burn; branded X-DF2.0, it pulls slip toward 1 g/kWh, and later variable-compression work narrows the gap to high-pressure engines further. LNG-mode operation is covered in LNG as marine fuel .

Methanol and ammonia

MAN’s ME-LGIM injects methanol as a high-pressure liquid into the chamber with a pilot for ignition; WinGD’s methanol engine is the X-DF-M. Methanol carries no sulfur and little particulate, which suits the tightening fuel rules, and the fuel handling is simpler than cryogenic LNG. See methanol as marine fuel .

Ammonia is the newer and harder fuel, because it ignites reluctantly, burns slowly, and is toxic. WinGD’s X-DF-A uses high-pressure ammonia injection with a pilot of about 5 percent liquid fuel at full load, and the first commercial unit went to sea in 2025 on a gas carrier. MAN’s two-stroke ammonia engine, developed with Mitsui, has run as a full-scale prototype and is entering commercial service. Both need careful handling of unburned ammonia and nitrous oxide in the exhaust. The fuel side is treated in ammonia as marine fuel . Across every gas and alcohol mode the pattern holds: the main fuel supplies the energy, and a small liquid pilot supplies the ignition, so the engine keeps full diesel-mode capability as a fallback. Injection hardware for all of these is covered in fuel injection systems and common-rail technology .

Starting, reversing, and going astern

A slow-speed two-stroke has no clutch and no gearbox, so it starts and reverses itself. Compressed air at about 25 to 30 bar is admitted to the cylinders in firing order through starting-air valves in the heads , turning the engine until enough cylinders fire to run on fuel; a distributor times the air, and interlocks stop the sequence once the engine picks up. Reversing means running the same engine the other way. A camshaft engine shifts its fuel cams and the starting-air distributor to a mirror-image astern profile, so the injection and air timing suit the reversed rotation. An electronically controlled ME or X engine has no camshaft at all: injection, exhaust-valve lift, and starting air are timed by the control system, so reversing is a change of timing map in software rather than a mechanical shift. Either way the ship goes astern by turning the propeller backward, which is why crash-stop and maneuvering procedures are written around the engine itself.

Torsional vibration and the barred speed range

The crankshaft, intermediate shafting, and propeller form one long torsional spring-mass system with natural frequencies of its own. When the firing order excites a mode whose natural frequency falls inside the operating speed range, the shaft twists back and forth at resonance, and the stress can crack a crankshaft or a shaft if the engine dwells there. The builder computes the torsional response for the specific engine, shafting, and propeller, and the class society reviews it against its unified requirements. Where a resonance sits inside the running range, the class imposes a barred speed range: a band of rpm marked red on the tachometer that the engine must accelerate through quickly and never run in continuously. Choosing cylinder count, adding a tuning damper or flywheel, or moving the resonance out of the range is part of the newbuild driveline design, and the barred range that results is a permanent operating limit the watchkeeper respects.

Manufacturers and shipyard licensing

The slow-speed two-stroke business is a design duopoly with a small third player. MAN Energy Solutions, licensor of the MAN B&W engines and the successor to MAN B&W Diesel, holds the larger share; the parent adopted the name Everllence in 2025, though the engines keep the long-established MAN B&W designation. WinGD, Winterthur Gas & Diesel, descends from the Sulzer and Wartsila two-stroke line and is now owned by China State Shipbuilding Corporation. Both are designers and licensors: they engineer the engines and sell the rights, but they do not usually cast the metal.

The engines are built under license at shipyards and engine works in China, Korea, and Japan, names such as CSSC, Hyundai Heavy Industries, Doosan, Hitachi, Mitsui, and Kawasaki. A ship’s engine is therefore a MAN B&W or WinGD design carrying the licensee’s build plate. Mitsubishi builds its own smaller UEC series of slow-speed two-strokes, a limited presence seen mostly on Japanese-built bulk carriers and some specialty tonnage. The wider builder and model landscape sits in marine engine makers .

Operation and maintenance

Acceptance testing

A new engine is shop-tested against the ISO 3046 and ISO 15550 reference conditions before delivery, and the works test proves SFOC at the load points, usually 25, 50, 75, and 100 percent of MCR, the NOx level to the certified Technical File, and the mechanical figures: cylinder pressures, exhaust-temperature spread across cylinders, bearing oil pressures and temperatures, jacket water and charge air conditions. The results become the baseline the ship’s staff compare against for the engine’s life.

Monitoring at sea

The daily record is the cylinder pressure trace, taken with a mechanical indicator on older engines or logged continuously from pressure transducers on modern ones. From it the engineer reads firing pressure, compression pressure, indicated mean pressure, the crank angle of peak pressure, and above all the spread between cylinders. The target is a tight cluster, usually under about 5 percent spread; a cylinder outside it points somewhere specific. A low firing pressure with normal compression suggests an injector or fuel-pump fault; a low compression suggests exhaust-valve leakage or ring blow-by; an early peak suggests injection timing that has crept. Turning that raw pressure data into diagnosis is the work described in combustion analysis and performance monitoring .

Crankcase safety and running-gear monitoring

The crankcase of a large engine holds a hot oil mist that can explode if a bearing overheats and vaporizes oil to an ignitable concentration, so the running gear is watched continuously and the crankcase is built to survive an ignition. Under the IACS unified requirements and SOLAS, an engine above 2,250 kW or with cylinders over 300 mm bore carries an oil-mist detector, or equivalent bearing-temperature monitoring, and crankcase explosion relief valves that vent an overpressure and reseat to keep air out. A rising mist reading or a hot bearing triggers a slowdown before the metal reaches ignition temperature, and the standing order after any such alarm is to stop, stay clear of the crankcase doors, and let the space cool before opening up, because an inrush of air onto a hot spot is what turns a primary explosion into a secondary one.

The other fire risk sits on the air side. Oil and unburned fuel that collect in the scavenge space, wiped down past the piston rings, can ignite on a hot exhaust-gas leak and start a scavenge fire, seen as a local exhaust-temperature rise and a hot patch on the receiver. Keeping the cylinder-oil feed correct, the rings gas-tight, and the under-piston drains clear is the routine defense, and the scavenge space is inspected at each top-end overhaul.

Bearing condition is tracked without opening the engine wherever possible: crankshaft deflection readings taken with a dial gauge between the webs show main-bearing wear-down and hull-induced misalignment, oil analysis flags white-metal and iron in the sump, and the main and crosshead bearing temperatures are logged. These readings, not the calendar, decide when the bottom end is opened.

The thrust bearing and direct-coupled shafting

Because there is no gearbox, the propeller thrust runs straight back through the shafting into the engine, and a single main thrust bearing at the aft end of the crankshaft, or just behind it, transfers that thrust into the ship’s structure. The bearing carries the full ahead or astern thrust of the propeller, tens or hundreds of tonnes on a large ship, through tilting pads riding on an oil film, and its condition sets a hard limit on how much power the plant can push into the water. Alignment of the whole train, crankshaft, intermediate shafts, and the stern tube bearings that carry the propeller shaft, is checked by bearing-load and deflection measurement, because a slow-speed two-stroke bolted rigidly to a flexing hull is unforgiving of a shaft that is not laid straight.

Overhaul intervals

Top-end overhauls, heads, exhaust valves, injectors, and fuel pumps, run about every 12,000 to 18,000 hours on HFO-fired engines and longer on cleaner fuels. Crosshead, main, and bottom-end bearings are opened at roughly 30,000 to 60,000 hours. Liner overhaul, re-honing and ring-groove and port work, falls near 60,000 to 100,000 hours depending on the cylinder-oil regime and fuel quality. The full crankshaft and running-gear inspection is not usually a fixed-interval job but a condition-driven one, triggered by bearing-wear-down readings, crankshaft deflection checks, and oil analysis rather than a calendar.

Limitations and practitioner caveats

The figures here are typical of current production families and should be read as ranges, not guarantees. The single authority for any given engine is its own project guide, shop-test record, and NOx Technical File; bore, stroke, rated power and speed, SFOC, pmax, scavenge pressure, and cylinder-oil feed all shift between builders, marks, and tuning options, and a derated or long-stroke variant can sit well outside the numbers quoted above. Tier III compliance depends on the trading pattern as much as the engine: an EGR or SCR system only earns its Tier III certificate for the areas and dates written into the ship’s paperwork, and gas-mode NOx credit holds only while the engine actually runs on gas. Methane-slip figures move with load and with each engine generation, so a slip number is meaningful only against a stated load point and design version. When the manufacturer data and a general figure disagree, the manufacturer data wins.

Related Articles

• Marine diesel engine : broader marine diesel context and the four-stroke comparison.
• Uniflow scavenging in two-stroke marine engines : the scavenging scheme in detail.
• Indicator diagram analysis : reading the cylinder pressure loop.
• Marine engine turbocharging : constant-pressure charging and the scavenge-air train.
• Marine engine fuel injection systems : injection hardware.
• Marine engine common rail technology : common-rail against cam-driven injection.
• Marine engine cylinder liners and pistons : cylinder-side wear parts.
• Marine engine crankshaft and main bearings : running-gear detail.
• Marine engine camshaft and valve train : valve actuation.
• Marine engine combustion analysis : combustion-quality assessment.
• Marine engine performance monitoring : performance-record practice.
• Specific fuel oil consumption : SFOC concept and correction.
• Marine engine makers : builders and model families.
• Slow steaming : the operating strategy the load diagram supports.
• Heavy fuel oil : the base fuel and its handling.
• Bunker quality and ISO 8217 : fuel grade specifications.
• NOx Tier I, II, III : NOx regulatory framework.
• Selective catalytic reduction : SCR for Tier III compliance.
• LNG as marine fuel : LNG-mode operation.
• Methanol as marine fuel : methanol-mode operation.
• Ammonia as marine fuel : ammonia-mode operation.
• Two-stroke engine future developments : where the slow-speed two-stroke is heading.