Abstract

The compression-ignition (diesel) engine has been the dominant prime mover of the world merchant fleet for more than a century, and it remains so today: the overwhelming majority of ocean-going tonnage is propelled either by a single, large, low-speed two-stroke engine coupled directly to the propeller, or by one or more medium-speed four-stroke engines driving the shaft through reduction gearing or supplying a diesel-electric plant. This document reviews the marine diesel engine from three complementary standpoints. First, it traces the historical development of marine compression-ignition propulsion from Rudolf Diesel's original patents and the first diesel-driven vessels of the early 1900s, through the establishment of the ocean-going motor ship, to the ultra-long-stroke, electronically controlled engines of the present day. Second, it separates and contrasts the two fundamental engine cycles used at sea — the two-stroke and the four-stroke — describing their construction, scavenging, combustion and mechanical arrangement, and it treats two operational subjects of first-rank practical importance: the running-in of new or overhauled cylinders, and continuous operation on residual heavy fuel oil. It also sets out the ideal thermodynamic (Diesel) cycle common to both engine types and the losses — heat, gas-exchange, combustion and friction — that separate that ideal from the power actually delivered. Third, it classifies the engines by power range as they are actually deployed on commercial ships today, from small high-speed units to the very large low-speed engines exceeding 80 MW. The document closes with a note on nuclear-powered ships, a survey of current developments in dual-fuel and zero-carbon propulsion, and an account of how engine power and brake horsepower are defined, measured and declared under the governing international standard, ISO 3046 in conjunction with the core standard ISO 15550.

1. Introduction

A marine diesel engine is a reciprocating internal-combustion engine in which fuel is ignited not by a spark but by the heat generated when air is compressed to a high pressure and temperature within the cylinder. The principle was conceived and patented by the German engineer Rudolf Diesel in the 1890s, and within a remarkably short period it displaced the reciprocating steam engine as the preferred means of ship propulsion. The reason is thermodynamic and economic: because the diesel cycle operates at a high compression ratio and burns fuel at high pressure, it converts a substantially larger fraction of the fuel's chemical energy into useful work than any competing heat engine. A modern large marine two-stroke engine achieves a brake thermal efficiency in the region of 50 per cent, and with waste-heat recovery the plant efficiency can approach 55 per cent, figures unmatched by any other prime mover in commercial service.

Two features distinguish the marine diesel engine from its counterparts on land. The first is its ability to burn low-grade residual fuel — the heavy, viscous bottom product of petroleum refining — which is considerably cheaper than distillate. The second is the sheer scale of the largest units: the biggest low-speed engines stand as tall as a four-storey building, weigh over two thousand tonnes, and develop more than 100,000 brake horsepower from a single crankshaft turning at roughly 100 revolutions per minute. Between these giants and the compact high-speed units fitted to patrol craft and yachts lies a continuous spectrum of engine sizes, and this document is organised to make sense of that spectrum: by history, by cycle (two-stroke versus four-stroke), by operational practice, and by power range as actually deployed on the commercial fleet.

2. Historical Development, c. 1900 to the Present

2.1 Origins: Rudolf Diesel and the first marine installations

Rudolf Diesel filed his foundational patent for a rational heat engine in 1892 and demonstrated a working prototype at the Maschinenfabrik Augsburg (a predecessor of MAN, and — through a chain of corporate succession — of today's Everllence) in 1897. Diesel's own ambition was a compact, efficient power unit that could compete with the steam engine wherever fuel economy mattered, and marine propulsion was an obvious target. The earliest diesel-powered vessels appeared in the first years of the twentieth century. Small diesel-driven craft operated on European inland waterways from around 1903 — among the frequently cited early examples are the French canal boat Petit-Pierre and the Russian diesel-electric river tanker Vandal, which worked on the Volga. In parallel, the diesel engine proved decisive for the submarine: its lack of a heat signature, its economy, and — crucially — the fact that it consumed no air-borne funnel gases on the surface made it the natural companion to electric propulsion when submerged, and navies adopted diesel-electric submarine propulsion during the first decade of the century.

The true turning point for merchant shipping came in 1912, when the Danish shipbuilder Burmeister & Wain delivered the ocean-going motor ship Selandia to the East Asiatic Company. Selandia, together with her sister vessels, is generally regarded as the first sea-going diesel motor ship of commercial significance: she crossed oceans without the tall funnel and coal bunkers of a steamer, carried her own fuel in far smaller volume, and required a fraction of the engine-room crew. The demonstration was watched closely by the shipping world, and it established Burmeister & Wain, together with the Swiss firm Sulzer and the German firm MAN, as the founding houses of marine diesel engineering — a lineage that survives in the two-stroke designs of today.

2.2 The inter-war and post-war consolidation

Through the 1920s and 1930s the motor ship steadily gained ground on the steamer, particularly for cargo liners and tankers where fuel economy translated directly into commercial advantage. Two structural decisions of this era shaped everything that followed. The first was the divergence between the low-speed two-stroke engine coupled directly to the propeller, favoured for main propulsion of larger vessels, and the medium- and high-speed four-stroke engine used for smaller craft and for auxiliary power. The second was the progressive adoption of turbocharging: by using a turbine driven by the engine's own exhaust gases to compress the intake air, designers forced substantially more air — and therefore more fuel — through a cylinder of a given size, raising the power output of an engine dramatically without a proportional increase in size or weight. Turbocharging, pioneered in the 1920s and refined continuously since, is the single technology most responsible for the power density of the modern marine engine.

After the Second World War the diesel engine became overwhelmingly dominant, and the industry consolidated around a small number of low-speed two-stroke licensors and a larger population of four-stroke builders. A further and lasting economic shift occurred as engines were progressively adapted to burn residual heavy fuel oil rather than the more expensive distillates, cutting operating costs for deep-sea trades.

2.3 The modern era: long stroke, electronic control and scale

From the 1980s onward, three trends defined the state of the art. First, the stroke-to-bore ratio of low-speed two-stroke engines was progressively lengthened — the so-called long-stroke and ultra-long-stroke engines — so that the propeller could be turned more slowly and efficiently, a large slow propeller being hydrodynamically superior to a small fast one. Second, mechanical control of fuel injection and exhaust-valve timing by camshaft gave way to electronic (common-rail) control, in which the injection pressure, timing and profile are governed by a computer and can be optimised continuously across the load range. The Sulzer RT-flex and MAN B&W ME series, introduced around the turn of the millennium, exemplify this camshaft-less, electronically controlled generation. Third, the relentless growth of the container trade drove the low-speed engine to unprecedented scale.

The emblem of that scale is the Wärtsilä-Sulzer RT-flex96C, whose fourteen-cylinder variant entered service in 2006. With a cylinder bore of 960 mm and a piston stroke of 2,500 mm, it develops around 80,080 kW (approximately 107,000–109,000 brake horsepower) at 102 rpm, and remains the reference point for the largest reciprocating engine ever built. Its two-stroke lineage is today held by WinGD (owned by the China State Shipbuilding Corporation), while the parallel MAN B&W two-stroke line is now marketed under the Everllence B&W brand following the 2025 renaming of MAN Energy Solutions. Together with Japan Engine Corporation (the UE engine line), these houses supply essentially all of the world's large low-speed marine engines.

2.4 Timeline of key milestones

Table 1. Principal milestones in the development of marine diesel propulsion.

3. The Two Engine Cycles: Two-Stroke and Four-Stroke

Every marine diesel engine belongs to one of two families defined by the number of piston strokes required to complete one working cycle. The distinction is fundamental, because it governs not only the engine's efficiency and power density but also its physical construction, its speed range, the vessels it suits, and the fuels it can economically burn. This section describes each cycle in turn and then draws the two together in a direct comparison.

3.1 The four-stroke cycle

In the four-stroke engine one working cycle occupies two complete revolutions of the crankshaft, divided into four distinct piston strokes. On the induction stroke the descending piston draws (or, when turbocharged, receives under pressure) a fresh charge of air into the cylinder through the open inlet valve. On the compression stroke, with both valves closed, the rising piston compresses this air to perhaps one-fifteenth to one-twentieth of its original volume, raising its temperature well above the ignition point of the fuel. Near the top of this stroke fuel is injected as a fine spray, ignites spontaneously, and burns; the resulting high-pressure gas drives the piston down on the power (expansion) stroke, which is the only stroke that produces useful work. Finally, on the exhaust stroke the rising piston expels the burnt gas through the open exhaust valve. Because the cycle uses separate, dedicated strokes for gas exchange, the four-stroke engine achieves a clean separation of intake and exhaust, permits a high compression ratio, and runs efficiently over a wide speed range.

Marine four-stroke engines are of the trunk-piston type: the connecting rod acts directly on the piston through a gudgeon pin, exactly as in an automotive engine but on a far larger scale. They run at medium speed (roughly 250 to 1,000 rpm) or high speed (above 1,000 rpm), and because these speeds are far too high to drive a propeller efficiently, they almost always deliver their power through a reduction gearbox, or else drive electrical generators in a diesel-electric plant. Their compactness and favourable power-to-weight ratio make them the natural choice for smaller vessels, for auxiliary generating sets aboard ships of every size, and for applications where several engines must be fitted into a confined space.

3.2 The two-stroke cycle

In the two-stroke engine one working cycle is completed in a single revolution of the crankshaft, so that every downward stroke of the piston is a power stroke. Compression and gas exchange are compressed into the same two strokes that produce work: as the piston approaches the bottom of its travel it uncovers ports in the cylinder wall, through which pressurised air — supplied by turbocharger and, at low load, by auxiliary electric blowers — sweeps into the cylinder and drives the burnt gas out, a process called scavenging. In the modern marine two-stroke this is arranged as uniflow scavenging: fresh air enters through inlet ports near the bottom of the liner and the exhaust leaves through a single large valve in the centre of the cylinder head at the top, so that the gas flows in one direction along the cylinder axis. The piston then rises, closes the ports, compresses the trapped air, and the cycle repeats. Because a power stroke occurs on every revolution, a two-stroke engine of a given cylinder size and speed produces roughly twice the power impulses of a comparable four-stroke.

The large marine two-stroke is invariably of the crosshead type, and this is its defining mechanical feature. Between the piston and the connecting rod there is an intermediate member, the crosshead, which slides in guides and takes the sideways (transverse) thrust that would otherwise be transmitted to the piston and cylinder liner. A long piston rod passes through a gas-tight stuffing box in a diaphragm that separates the combustion space above from the crankcase below. This arrangement keeps the dirty products of residual-fuel combustion out of the crankcase lubricating oil, allows the cylinders to be lubricated separately with a dedicated alkaline cylinder oil, and permits the very long strokes that give these engines their efficiency. The two-stroke crosshead engine runs at low speed — typically 60 to 250 rpm — which allows it to be coupled directly to the propeller shaft with no gearbox at all, the single most efficient possible arrangement for a large ship.

3.3 Direct comparison

The two cycles are not competitors for the same duty so much as complementary solutions optimised for different parts of the market. The following table summarises the principal engineering and operational differences.

Table 2. Principal differences between the low-speed two-stroke and the medium/high-speed four-stroke marine diesel engine.

4. The Ideal Diesel Cycle and Engine Losses

Before the real behaviour of an engine can be assessed, it is useful to establish the thermodynamic ideal against which it is measured. Both the two-stroke and the four-stroke engine are, in principle, realisations of the same thermodynamic cycle — the compression-ignition or Diesel cycle — and the number of strokes affects only how the cylinder is charged and cleared, not the high-pressure part of the cycle where work is produced. This section sets out that ideal cycle, shows how the real engine departs from it, and identifies the losses that separate the ideal efficiency from the power actually delivered at the coupling.

4.1 The ideal air-standard Diesel cycle

The classical idealisation is the air-standard Diesel cycle, which replaces the complexity of real combustion and gas exchange with a fixed mass of air behaving as a perfect gas and passing through four internally reversible processes:

• 1 → 2 Isentropic compression. The piston compresses the trapped air with no heat exchange with the surroundings; pressure and temperature rise steeply.

• 2 → 3 Constant-pressure heat addition. Fuel is injected and burned so that heat is added while the piston begins to descend, holding the pressure constant. This constant-pressure heat addition is the defining feature that distinguishes the Diesel cycle from the constant-volume (Otto) cycle of the petrol engine.

• 3 → 4 Isentropic expansion. The high-pressure gas expands and drives the piston down, doing work; this is the power stroke.

• 4 → 1 Constant-volume heat rejection. At the end of expansion the remaining heat is rejected at constant volume, returning the air to its initial state. In the real engine this idealised step represents the exhaust blowdown and the replacement of burnt gas by a fresh charge.

The thermal efficiency of the ideal cycle depends only on the compression ratio r (= V₁/V₂), the cut-off ratio r_c (= V₃/V₂, a measure of how far into the expansion combustion continues), and the ratio of specific heats γ of the working gas:

η = 1 − 1 ⁄ r^γ−1 · [ ( r_c^γ − 1 ) ⁄ ( γ ( r_c − 1 ) ) ]

Two conclusions follow directly and matter for marine practice. First, efficiency rises steadily with the compression ratio, which is why marine diesel engines use very high compression ratios. Second, for a given compression ratio the efficiency falls slightly as the cut-off ratio increases — that is, as more fuel is burned and the load rises — because a larger share of the heat is then added later in the expansion, where it can do less work. The large low-speed marine engine, with its high compression ratio and long expansion stroke, therefore sits close to the most favourable region of this relationship, and this is one reason it is the most efficient prime mover in commercial service.

Drawn on temperature–entropy axes (Figure 2), the cycle shows the heat added and the heat rejected directly as areas beneath the two heat-transfer processes. The enclosed area is the net work, and the ratio of that enclosed area to the area under the heat-addition line is the thermal efficiency — a purely visual statement of the same result given by the formula above.

4.2 The dual (mixed) cycle and real engines

The pure Diesel cycle assumes that all the heat is added at constant pressure, but in a real engine combustion begins almost instantaneously at close to constant volume and only then continues at roughly constant pressure as the piston moves away from top dead centre. A more faithful idealisation, the dual or mixed cycle (also called the Seiliger cycle), therefore splits the heat addition into a constant-volume part followed by a constant-pressure part. Slow-speed engines, in which combustion is spread over a longer period, are well represented by the constant-pressure Diesel cycle; high-speed engines, in which combustion is more rapid, lie closer to the constant-volume (Otto) limit. The dual cycle spans the range between the two. In every case the essential point stands: the thermodynamic ideal is shared by the two-stroke and the four-stroke alike, and it is in the gas exchange — not the high-pressure loop — that the two engine types diverge.

4.3 From the ideal to the real: the indicator diagram

The real behaviour of the cylinder is captured by the indicator diagram — the closed loop traced by cylinder pressure against volume through an actual cycle, historically recorded by the mechanical engine indicator (described in the section on power measurement) and today by electronic in-cylinder pressure sensors. The real loop departs from the ideal in several ways: the sharp corners of the ideal cycle are rounded because combustion and heat release take finite time; the peak pressure is limited by mechanical strength; expansion is not truly adiabatic because heat is lost to the walls; and the exhaust valve or ports open before the piston reaches the bottom of its stroke, so that pressure falls early in a process called blowdown. Most importantly for the comparison between the two engine types, the real diagram includes a gas-exchange loop at the bottom, and it is here that the two-stroke and the four-stroke differ fundamentally.

In the four-stroke engine, gas exchange occupies two dedicated piston strokes — an exhaust stroke that expels the burnt gas and an intake stroke that draws in the fresh charge. On the indicator diagram these appear as a separate loop below the main power loop: the pumping loop. In a naturally aspirated engine this loop represents net work done by the piston on the gas, and is therefore a genuine loss; in a turbocharged engine the boost pressure can exceed the exhaust back-pressure, so the loop may become positive and return a little work. In the two-stroke engine there are no dedicated gas-exchange strokes at all. Scavenging takes place around bottom dead centre, and the work of driving fresh air through the cylinder is done not by the piston but by an external turbocharger and, at low load, by auxiliary blowers. The two-stroke therefore has essentially no piston pumping loop — but, as the next section explains, it pays for gas exchange in a different currency.

4.4 The principal losses

The gap between the ideal cycle's efficiency and the brake efficiency actually delivered at the coupling is the sum of several distinct losses, which fall into four main groups.

• Heat losses. Real compression and expansion are not adiabatic. A substantial fraction of the fuel's energy is carried away by the cooling water and lubricating oil from the cylinder walls, piston and head, and a still larger fraction leaves as sensible heat in the hot exhaust gas. In a large engine the exhaust typically removes on the order of a quarter to a third of the fuel energy and the cooling systems a further sixth or so; together they are by far the largest departure from the ideal. Modern plants recover part of the exhaust heat in waste-heat-recovery boilers and power turbines, reclaiming some of this loss.

• Combustion and timing losses. Combustion takes finite time and is never perfect: some fuel burns late in the expansion or incompletely, the real heat release does not follow the idealised constant-pressure line, and chemical dissociation at peak temperature absorbs energy. Injection timing is always a compromise, and any departure from the optimum costs efficiency.

• Gas-exchange (pumping) losses. Work must be spent moving gas into and out of the cylinder. In the four-stroke this appears as the piston pumping loop of Figure 3; in the two-stroke it appears instead as the power consumed by the turbocharger and scavenge blowers and, uniquely, as scavenging losses — fresh air short-circuiting straight to the exhaust, and incomplete clearing of burnt gas — which reduce the mass of clean air trapped for the next cycle.

• Friction and mechanical losses. The difference between the power developed in the cylinders (indicated power) and the power delivered at the coupling (brake power) is absorbed by friction — principally at the bearings, between the piston rings and skirt and the liner, and in the valve gear — and by engine-driven auxiliaries such as the cooling and lubricating-oil pumps. The ratio of brake power to indicated power is the mechanical efficiency, treated further in the section on power measurement.

4.5 How the losses differ between the two engine types

Because the two cycles handle gas exchange so differently, their loss balances differ in characteristic ways, and these differences help explain why each type dominates its own part of the market.

Table 3. Characteristic differences in the loss balance of the four-stroke and two-stroke marine diesel engine.

The very large low-speed two-stroke reaches the highest brake efficiency of any prime mover because several of these factors act together in its favour: an extremely high compression and expansion ratio extracts more work from each kilogram of fuel; the crosshead construction removes side thrust from the piston and so lowers friction; the low rotational speed keeps friction and pumping losses small; and turbocharging with waste-heat recovery reclaims part of the exhaust loss. The four-stroke accepts a slightly larger share of pumping and friction loss in exchange for its compactness and layout flexibility, which is why it prevails wherever those qualities matter more than the last point of efficiency.

5. Running-In of Marine Diesel Engines

Running-in (also called bedding-in or breaking-in) is the controlled, progressive loading of an engine's cylinders when they are new or have just been overhauled, and it is one of the most consequential operational procedures in the life of a marine engine. Its purpose is to allow the sliding surfaces — principally the piston rings and the cylinder liner — to conform to one another at the microscopic scale before the engine is subjected to full load. A freshly machined liner carries a deliberate crosshatch of fine honing marks that retain oil, and the piston rings, however carefully manufactured, do not initially make perfect contact around their full circumference. Under gentle, gradually increasing load these surfaces wear together until a continuous, gas-tight, oil-retaining contact is established across the whole ring face.

If an engine is loaded up too quickly, the high points of the ring and liner surfaces carry the entire gas and mechanical load, the local pressure exceeds what the oil film can support, and the metal-to-metal contact tears the surfaces — a failure called scuffing or microseizure. Scuffing destroys the honing pattern, glazes or scores the liner, ruins the ring pack's ability to seal, and can only be corrected by re-honing or renewing components. Correct running-in is therefore not a nicety but the difference between a cylinder that gives a full service life and one that must be opened up again within weeks.

5.1 Practice

• Follow the manufacturer's programme. Each engine builder specifies a running-in schedule — a sequence of load steps held for defined periods over many hours or days — and this schedule, not the operator's judgement, governs the procedure. A running-in programme is mandatory after fitting new piston rings, after re-honing or renewing a liner, and after a major overhaul of the running gear.

• Increase load in steps. Load is raised in stages, each held long enough for the surfaces to stabilise, with the engine returned to a lower load if temperatures or observations suggest distress.

• Increase cylinder lubrication temporarily. On low-speed engines the cylinder-oil feed rate is raised above the normal running figure during running-in to ensure an abundant oil film while the surfaces bed in, then returned to the optimised rate afterwards.

• Monitor closely. Liner-wall and exhaust temperatures, and — where fitted — liner-wall or ring-condition monitoring, are watched throughout; scavenge-port inspections are used to confirm that the ring pack and liner are bedding in evenly and cleanly.

The same principles, in gentler form, apply to medium-speed four-stroke engines after ring or liner renewal, though their smaller components and common-crankcase lubrication make the procedure less protracted than for the large crosshead engine.

6. Operation on Heavy Fuel Oil

A defining commercial feature of the deep-sea marine diesel engine is its ability to burn residual heavy fuel oil (HFO), the dense, high-viscosity fraction that remains after the lighter and more valuable products have been distilled from crude oil. Residual fuel is markedly cheaper than distillate, and on the long voyages of a large ship the fuel bill dominates operating costs; the capacity to run on HFO is therefore central to the economics of ocean shipping. Both low-speed two-stroke engines and larger medium-speed four-stroke engines are routinely designed and operated to burn it. Residual marine fuels are specified internationally under ISO 8217, whose residual grades range up to a kinematic viscosity of 700 cSt at 50 °C.

6.1 Why heavy fuel must be treated

Heavy fuel oil cannot be burned as delivered. It is too viscous to pump or atomise at ambient temperature, and it carries contaminants that are aggressive to the engine: sulphur, which forms corrosive acids on combustion; vanadium and sodium, which together promote high-temperature corrosion of exhaust-valve and turbine surfaces; water; and abrasive catalytic fines (cat fines) — hard particles of aluminium and silicon carried over from the refinery's catalytic cracking process, which cause severe abrasive wear of liners, rings and fuel-injection equipment if not removed. Fuel treatment before combustion is therefore not optional.

6.2 The fuel treatment chain

• Storage and settling. Bunkered fuel is heated in storage and passed to settling tanks, where gross water and heavy sediment separate under gravity and heat.

• Centrifugal separation. The fuel is then cleaned in high-speed centrifugal separators (purifiers and clarifiers), which remove water and — critically — reduce the cat-fine content to a level the engine can tolerate. Separator throughput and temperature are set to maximise this cleaning.

• Heating to injection viscosity. Because injectors require a fuel viscosity of roughly 10–15 cSt for proper atomisation, the heavy fuel is heated — often to around 130–150 °C for the highest-viscosity grades — by a viscosity-controlled final heater immediately before the injection pumps, so that the correct viscosity reaches the injectors regardless of the fuel's grade.

• Fine filtration. Final fine filters protect the injection equipment from any remaining solid particles.

Two further consequences follow from HFO combustion. First, the sulphur in the fuel forms sulphuric acid on the cylinder walls, which must be neutralised by using a cylinder lubricating oil of high base number (BN); the alkalinity of the oil is matched to the sulphur content of the fuel in use. Second, the fuel and its associated systems must be kept hot at all times when in use, so that the fuel remains pumpable; engines that must manoeuvre or lie in port on cleaner fuel change over between heavy and distillate grades under controlled temperature to avoid thermal shock to the injection equipment.

6.3 The sulphur cap and exhaust after-treatment

Since 1 January 2020 the IMO global sulphur cap has limited the sulphur content of marine fuel to 0.50 per cent by mass, and to 0.10 per cent within designated Emission Control Areas (ECAs). Operators comply either by burning low-sulphur fuels (very-low-sulphur fuel oil or distillate) or by continuing to burn high-sulphur HFO while removing the sulphur oxides from the exhaust with an exhaust-gas cleaning system, or scrubber. The economics of the fuel-price differential, the capital cost of a scrubber, and the tightening regime of greenhouse-gas regulation together now shape the choice of fuel for every deep-sea ship.

7. Classification by Power Range in Today's Fleet

On the modern commercial fleet the choice between two-stroke and four-stroke, and the choice of engine size within each family, follows closely from the power required and the type of vessel. It is useful to divide the market into three broad bands — small, large, and very large — recognising that the boundaries are not sharp and that a single ship often combines engines from more than one band (for example a very large two-stroke for propulsion and several medium-speed four-strokes as generating sets). The bands below describe the engine as actually specified for each duty.

7.1 Small power: high-speed and smaller medium-speed four-strokes

At the lower end of the range — from a few hundred kilowatts up to roughly 3–4 MW per engine — propulsion is almost exclusively four-stroke. High-speed four-stroke engines (running well above 1,000 rpm) power patrol boats, pilot boats, fast ferries, workboats, fishing vessels, yachts and lifeboats; they are compact, light, quick to respond, and typically burn distillate fuel. Smaller medium-speed four-strokes serve tugs, small coasters, dredgers and inland vessels, and they also provide the ubiquitous auxiliary generating sets found aboard ships of every size. In this band, power-to-weight ratio, compactness and the ability to fit multiple units matter more than the last increment of fuel efficiency, and the two-stroke engine has no place at all.

7.2 Large power: medium-speed four-strokes and smaller low-speed two-strokes

In the middle band — very roughly 3 MW to 25 MW — the two families overlap and genuinely compete. Medium-speed four-stroke engines (running at a few hundred rpm and driving through reduction gears, or arranged as diesel-electric prime movers) dominate ferries, ro-ro vessels, offshore support vessels, dredgers and, very characteristically, cruise ships and other passenger vessels, where a diesel-electric or dual-fuel-electric plant of several medium-speed engines offers flexibility, redundancy and freedom in engine-room layout. Where a single, simple, highly efficient propulsion train is wanted instead, a smaller low-speed two-stroke engine coupled directly to the propeller is chosen, as on many general-cargo ships, handy-size bulk carriers and smaller tankers. The decision turns on the balance between the two-stroke's superior fuel economy and the four-stroke's compactness, layout flexibility and multi-engine redundancy.

7.3 Very large power: low-speed two-strokes

Above roughly 25 MW and extending beyond 80 MW, main propulsion is the exclusive territory of the low-speed two-stroke crosshead engine. Large crude and product tankers, capesize and larger bulk carriers, and the great container ships are propelled by a single such engine turning a single large propeller directly, an arrangement whose efficiency no alternative can match at this scale. The largest of these engines — of which the fourteen-cylinder Wärtsilä-Sulzer RT-flex96C, at about 80 MW and some 107,000 brake horsepower, is the canonical example — represent the summit of reciprocating-engine engineering. Ultra-large container ships and very large crude carriers sit firmly in this band; the engine is chosen for one reason above all others, which is that a slow-turning, direct-coupled two-stroke burning residual fuel offers the lowest possible cost per tonne-mile over an ocean passage.

7.4 Summary of the power bands

Table 4. Practical classification of marine propulsion engines by power band, engine type and vessel class. Band boundaries are indicative and overlap in practice.

8. A Note on Nuclear-Powered Ships

Nuclear propulsion occupies a small but distinctive place in the history of ship power, and it is worth setting alongside the diesel engine because it answers the same question — how to move a large ship — by an entirely different means. A marine nuclear plant does not itself turn the propeller by combustion; instead, a pressurised-water reactor generates heat that raises steam, and that steam drives a turbine, so the nuclear ship is at heart a steam-turbine ship whose boiler is a reactor. The reactor's decisive advantage is energy density: a small charge of nuclear fuel releases enormous energy and needs no air, so a nuclear ship can steam for years without refuelling, produces no combustion emissions, and is unconstrained by fuel-bunker capacity or the location of bunkering ports.

In naval service these advantages are compelling, and nuclear propulsion has been standard for submarines and large aircraft carriers since the United States commissioned the submarine Nautilus in 1955; the ability to remain submerged or on station almost indefinitely is of the highest military value. In civilian merchant service, by contrast, nuclear power has never become economic. A handful of nuclear cargo ships were built as demonstrations — the American NS Savannah (1962), the German Otto Hahn (1968) and the Japanese Mutsu (first criticality 1969) — but all were retired without founding a commercial fleet, defeated by high capital and operating costs, complex regulation, restrictions on port access, and public concern over safety and waste. The one enduring category of civilian nuclear ship is the icebreaker: the Soviet Union and Russia have operated nuclear-powered icebreakers since Lenin in 1959, and the type remains in service today for keeping Arctic sea routes open, a duty for which the reactor's endurance and high sustained power are ideally suited. The nuclear-powered cargo carrier Sevmorput, in service since the late 1980s, is a rare survivor of merchant nuclear propulsion.

Interest in marine nuclear power has revived recently in the context of decarbonisation, because a reactor emits no carbon dioxide in service. Several current concepts propose compact small modular reactors (SMRs), including advanced designs such as molten-salt reactors, for large ocean-going ships. Whether any of these overcomes the long-standing barriers of cost, regulation, security and port acceptance remains an open question, but the idea has returned to serious discussion for the first time in decades.

9. New Developments: Dual-Fuel and Zero-Carbon Propulsion

The dominant force shaping marine engine development today is the drive to decarbonise shipping. In 2023 the International Maritime Organization adopted a revised greenhouse-gas strategy setting the goal of net-zero emissions from international shipping by or around 2050, with indicative checkpoints for 2030 and 2040 and a target for zero- and near-zero-emission fuels to supply a meaningful share of shipping's energy by 2030. To translate these aims into binding measures, the IMO's Marine Environment Protection Committee approved, in April 2025, the so-called Net-Zero Framework — a combination of a goal-based marine fuel standard (progressively tightening the permitted greenhouse-gas intensity of marine fuel) and a global emissions-pricing mechanism, applying to large ocean-going ships. Formal adoption of the framework, originally scheduled for October 2025, was adjourned for further negotiation, so its final form and timing remain under discussion; nonetheless the direction of travel is clear, and it is reshaping engine design.

The engine builders' principal response has been the dual-fuel engine, which can run on a conventional liquid fuel and on a cleaner alternative, switching between them as fuel availability, price and regulation dictate. This fuel flexibility de-risks the transition for shipowners who cannot know today which alternative fuel will ultimately prevail.

9.1 The alternative fuels

• Liquefied natural gas (LNG). The most established alternative, offering lower carbon dioxide, sulphur and particulate emissions than oil. Low-pressure gas engines must manage “methane slip” — unburned methane, itself a potent greenhouse gas — which somewhat offsets the carbon benefit.

• Methanol. A liquid at ambient conditions and therefore relatively easy to store and bunker. Two-stroke methanol engines (the MAN B&W / Everllence B&W ME-LGIM series) have reached full commercial maturity, with a large and growing order book and hundreds of thousands of running hours accumulated; methanol is a leading choice for new container ships and, increasingly, other vessel types.

• Ammonia. Carbon-free at the point of use, and therefore of great interest for deep-sea shipping. Two-stroke ammonia engines moved from test bench to first commercial installations in the mid-2020s; ammonia is toxic and demands careful handling, but it avoids carbon emissions altogether when burned.

• LPG, ethane and hydrogen. LPG and ethane serve specific trades (notably the gas carriers that transport them). Hydrogen, typically via fuel cells or as a blend, is under active development for smaller vessels and as a longer-term option.

9.2 Efficiency and supporting technologies

Alongside new fuels, a suite of technologies raises overall efficiency and cuts emissions from any fuel: waste-heat recovery systems that reclaim energy from exhaust and cooling; shaft generators that produce electrical power from the main engine; hybrid and battery installations for load-levelling and zero-emission port operation; wind-assisted propulsion such as rotor sails; air lubrication that reduces hull friction; and, at an early stage, onboard carbon capture. The regulatory apparatus now includes the Energy Efficiency Existing Ship Index (EEXI) and the Carbon Intensity Indicator (CII), both in force, together with regional measures such as the European Union's inclusion of shipping in its emissions trading scheme and its FuelEU Maritime fuel-intensity rules. The combined effect is that the marine engine of the coming decades will be defined as much by the fuel it burns and the emissions it avoids as by its thermodynamic cycle.

10. How Engine Power and Brake Horsepower Are Calculated

The power of a reciprocating engine is defined and measured at two distinct stations in the machine, and it is essential to distinguish them. Indicated power is the power developed by the burning gas on the piston crowns inside the cylinders; brake power is the useful power actually delivered at the engine's output coupling, after the internal friction of bearings, rings and driven auxiliaries has been subtracted. The ratio of the two is the mechanical efficiency of the engine.

10.1 Indicated power

Indicated power is derived from the pressure the gas exerts on the piston through the working cycle. Historically this pressure was recorded against cylinder volume by a mechanical instrument called an engine indicator, which drew a closed loop — the indicator diagram — whose enclosed area represents the net work done on the piston per cycle. The average pressure obtained by dividing that work by the swept volume is the mean indicated pressure (often written pₘ or IMEP). The indicated power of one cylinder then follows from the elementary relation

P_i = p_m × L × A × n × k

where pₘ is the mean indicated pressure, L the length of the piston stroke, A the area of the piston (bore), n the number of working (power) strokes per unit time, and k the number of cylinders. The product L × A is the swept volume of one cylinder; multiplying by the mean pressure gives work per cycle, and multiplying by the firing rate gives power. Summed over all cylinders this yields the total indicated power of the engine. On modern electronically controlled engines the mechanical indicator has been replaced by electronic in-cylinder pressure sensors, but the quantity computed is the same.

10.2 Brake power and brake horsepower

Brake power — the figure most often quoted, and the origin of the term “brake horsepower” (BHP) — is the power measured at the crankshaft coupling. The name comes from the historical method of measurement, in which a friction brake (a dynamometer) was applied to the output shaft; today the dynamometer may be hydraulic, electrical or an eddy-current type on a test bed, while aboard ship the delivered power is found from a torsionmeter that measures the twist of the propeller shaft. In every case the principle is the same: brake power is the product of the torque delivered and the rotational speed,

P_b = 2π N T

where T is the measured torque and N the rotational speed. Because brake power is what the engine actually delivers to the propeller or generator, it is the basis for engine rating and for the contractual and regulatory declaration of power. Marine engines are rated at a Maximum Continuous Rating (MCR) — the maximum power the engine may develop continuously — and are normally operated at a Continuous Service (or Normal) Rating of some 85–90 per cent of MCR, which gives an efficiency and reliability margin. Fuel economy is expressed as the Specific Fuel Oil Consumption (SFOC), the mass of fuel burned per unit of brake power per hour (typically grams per kilowatt-hour).

The relationship between the two powers is:

mechanical efficiency = P_b / P_i (typically ~0.85–0.92 at full load)

10.3 The governing industrial standard: ISO 3046 and ISO 15550

Because a declared power figure means nothing without agreed conditions and methods, the power of marine (and other) reciprocating internal-combustion engines is declared and measured under an international standard. The governing document is the ISO 3046 series, “Reciprocating internal combustion engines — Performance”, of which Part 1 (ISO 3046-1) specifies the standard reference conditions, the declarations of power and fuel and lubricating-oil consumption, and the test methods. ISO 3046 applies expressly to engines for land, rail-traction and marine use, and it is used, for example, for the statements of power stamped on an engine's data plate.

ISO 3046-1 does not stand alone: it is a “satellite” standard applied together with the “core” standard ISO 15550, “Internal combustion engines — Determination and method for the measurement of engine power — General requirements”, which defines the underlying reference conditions and measurement methodology. A central purpose of the standard is to correct measured power to a set of standard reference conditions of ambient air temperature, pressure and humidity (and, for marine engines, charge-air coolant temperature), so that engines tested under different site conditions can be compared on a common, agreed basis; the correction formulae and exponents used were derived by CIMAC, the international council on combustion engines. For engines driving electrical generating sets a related standard, ISO 8528, applies. In addition, the classification societies (such as DNV, Lloyd's Register, ABS and Bureau Veritas) impose their own requirements on the survey and approval of marine engines, working in conjunction with the ISO performance standards.

In summary: indicated power is computed from in-cylinder pressure; brake power (BHP) is measured at the coupling as torque × speed; the two differ by the engine's mechanical losses; and the declaration of that power for a marine engine is governed by ISO 3046 in conjunction with the core measurement standard ISO 15550, corrected to standard reference conditions.

11. Concluding Remarks

For more than a century the compression-ignition engine has been the heart of the merchant ship, and the reasons are the same today as in 1912: no other prime mover converts so large a share of a cheap fuel into propulsive work. The two great branches of the family — the slow, direct-coupled, residual-burning two-stroke that drives the world's tankers, bulkers and container ships, and the compact, versatile four-stroke that powers smaller vessels, provides ships' electrical power, and dominates diesel-electric and passenger tonnage — remain complementary rather than competing, each optimised for its own part of the power spectrum, and each realising the same ideal Diesel cycle while paying its losses in a slightly different coin. Nuclear propulsion, for all its endurance, has stayed confined to naval vessels and icebreakers. What is changing, and changing quickly, is the fuel: under the pressure of decarbonisation, the same fundamental engines are being adapted to burn methanol, ammonia, natural gas and hydrogen, so that the marine diesel engine of the near future will keep its essential mechanism while shedding the carbon that has always accompanied it. Throughout, the discipline of measuring and declaring engine power on a common, internationally agreed basis — the province of ISO 3046 and ISO 15550 — remains the quiet foundation on which the specification, sale and regulation of every one of these engines rests.

Sources and Further Reading

This document synthesises well-established marine-engineering knowledge together with current information on standards, regulation and engine developments. The principal current-source references consulted are listed below.

• International Organization for Standardization — ISO 3046-1:2002, Reciprocating internal combustion engines — Performance — Part 1: Declarations of power, fuel and lubricating oil consumptions, and test methods; and ISO 15550, Internal combustion engines — Determination and method for the measurement of engine power — General requirements.

• International Maritime Organization — 2023 IMO Strategy on Reduction of GHG Emissions from Ships (MEPC 80); and the IMO Net-Zero Framework (approved MEPC 83, April 2025; adoption adjourned, October 2025).

• Wärtsilä Corporation and WinGD — technical data on the Wärtsilä-Sulzer RT-flex96C low-speed two-stroke engine (14-cylinder variant, ~80,080 kW at 102 rpm).

• Everllence (formerly MAN Energy Solutions, renamed June 2025) — two-stroke dual-fuel engine programmes: B&W ME-LGIM (methanol) and B&W two-stroke ammonia engine development.

• ISO 8217 — Petroleum products — Fuels (class F) — Specifications of marine fuels (residual and distillate marine fuel grades).

• General marine-engineering references on the thermodynamics of the Diesel and dual (Seiliger) cycles, engine heat balance, and the indicator diagram (standard marine motor-engineering and applied-thermodynamics texts).