Ship propulsion: Science and limits

Why only 30% of the fuel actually pushes the ship

What this tab covers What physically happens in the conversion, step by step, the science behind it, and the hard limits that physics sets on how well it can ever work.

A ship burns fuel oil or LNG in a large piston engine that turns a propeller; about half the fuel energy becomes shaft work and roughly 60% of that becomes thrust, so about 30% ends up pushing the hull. Sails and rotors use wind directly; naval reactors use fission.

The conversion, step by step

Combustion and expansion in a slow-speed two-stroke engine
Chemical (heavy fuel oil, marine diesel or LNG) → Shaft work · about 51-54% (calculated from WinGD X92 fuel use of 156-166 g/kWh at 65-100% load and the 42.7 MJ/kg reference heating value)¹⁴²Propeller
Shaft work → Thrust power · about 52-67% in open water; about 55-60% behind the hull for standard propeller series in a bulk carrier case study⁴Hull
Thrust power → Motion against water friction, waves and air drag · all thrust work is lost to resistance; this sets the fuel needed per tonne-km¹End to end (calculated)
Chemical (fuel) → Thrust power · about 30% (0.51 engine x 0.98 shaft line, an assumed typical value, x 0.59 propeller behind hull)¹⁴⁴

The science

Large ships use slow-speed two-stroke diesels, several storeys tall, coupled straight to the propeller. They are efficient because they compress air very hard and big, slow cylinders lose little heat; a turbocharger reuses exhaust energy. Dual-fuel engines burn LNG injected at high pressure (Diesel cycle) or premixed with air (Otto cycle) (which lets some methane escape unburned: 6.4% on average in low-pressure four-strokes, see methods). A large, slow propeller wastes less energy in the wake. Power rises roughly with the cube of speed, so slow steaming is the cheapest saving. Rotor sails use the Magnus effect: a spinning cylinder in wind feels a sideways force usable as thrust.

Limits set by science

Limit	Value	Why it exists	Best today
Carnot limit on the engine	about 85% (calculated, illustrative)²¹⁴	1 - Tc/Th with an assumed peak gas temperature of 2,000 K and seawater at 300 K: 1 - 300/2,000 = 0.85. Real engines fall far short because heat is added over a range of temperatures, walls are cooled and exhaust leaves hot.	about 51-54% (calculated); in gas mode the Diesel-cycle Everllence G95ME-GI beats the Otto-cycle WinGD X92DF by up to 2.9 points at 75% load
Propeller (actuator disc) limit	2 / (1 + wake speed / ship speed)⁴	The faster the propeller throws water back relative to the ship, the more energy is wasted; bigger, slower propellers get closer but are limited by draught.	open-water efficiency about 0.67 for Wageningen B and MAU series
Wind-assist thrust	two 35 m x 5 m rotors gave peak thrust equal to 7 MW of propeller power (test)⁶⁷	Sail force grows with wind speed squared and sail area, but depends on route and weather, so savings are voyage averages.	25% average saving projected by Norsepower for one North Sea ship; up to 6% on a Vale ore carrier with five Anemoi rotors (company figures)

Sources

Fourth IMO GHG Study 2020: Executive Summary International Maritime Organization, 2020
Comparative analysis of the heat balance results of the selected Tier III-compliant gas-fuelled two-stroke main engines (Korlak), Combustion Engines 193(2) Polish Scientific Society of Combustion Engines, 2023
Design of propeller series optimizing fuel consumption and propeller efficiency (Tadros, Ventura and Guedes Soares), Journal of Marine Science and Engineering 9, 1226 MDPI, 2021
Norsepower highlights Rotor Sail results at The Motorship Propulsion and Future Fuels webinar Norsepower, 2021
Anemoi installs rotor sails on Vale's very large ore carrier Ship Technology, 2024
X92 IMO Tier II/Tier III engine data WinGD, 2026