Background
Cross scavenging takes its name from the cross-cylinder flow path of fresh charge: air enters one side of the cylinder near the bottom, deflects upward off the piston crown’s raised deflector, sweeps across the cylinder cover, and descends into the exhaust ports on the opposite side. The flow makes a single inverted-U traverse of the cylinder volume. Unlike loop scavenging, which depends on angled port cuts to create a return loop, cross scavenging relies on the geometric obstruction of the deflector to organise the flow. Unlike uniflow scavenging, which uses an exhaust valve in the cylinder cover, cross scavenging discharges purely through liner ports. There is no valve, no actuator, and no high-temperature seat to maintain.
This mechanical simplicity made cross scavenging the natural choice for the earliest commercial two-stroke marine engines. Burmeister and Wain (B&W), Sulzer, and Werkspoor all built cross-scavenged designs in the 1900s and 1910s, and cross scavenging remained the standard for small and medium two-stroke marine engines through the interwar period. By the 1950s, however, the inefficiencies of the deflector-piston arrangement had become limiting: deflectors ran hot and cracked, scavenging efficiency was poor, and the fuel consumption gap relative to loop-scavenged engines became commercially intolerable.
Cross scavenging is now obsolete in marine main propulsion. It survives in scattered legacy installations, in some opposed-piston designs, and as historical context for understanding the development arc of the marine two-stroke engine.
Mechanical layout
The cross-scavenged cylinder is geometrically simple:
Port arrangement
Two sets of rectangular ports are cut into the cylinder liner near its bottom. One set, on one side, is connected via the scavenge receiver to the air supply (turbocharger, blower, or crankcase pump). The opposite set discharges to the exhaust manifold. Port windows are typically 80 to 200 mm tall, with widths spanning a substantial portion of the liner circumference on each side.
Deflector piston
The piston crown carries a raised wedge or hump on the side adjacent to the scavenge ports. Incoming air strikes this deflector and is deflected upward. The unaerodynamic shape of the deflector is essential to the scheme: without it, fresh air would proceed directly across the cylinder, short-circuit through the exhaust ports, and accomplish little scavenging.
Cylinder cover
The cylinder cover is unobstructed by valves. It carries only the fuel injection gear, indicator cock, relief valve, and (on some designs) starting air valve. This freedom from valves was historically valued because high-temperature valves with materials of the era were prone to failure.
Air supply
Early cross-scavenged engines used crankcase scavenging (the underside of the piston pumping air through transfer ports), engine-driven scroll or piston blowers, or, later, exhaust-driven turbochargers. The lack of a valve in the gas-exchange path simplified pressure-balance design but limited maximum boost.
Gas-exchange physics
The cross-scavenging flow path is geometrically simple but thermodynamically inefficient. Three loss mechanisms dominate:
Short-circuit losses
A significant fraction of fresh charge entering through the scavenge ports proceeds directly across the cylinder and exits through the exhaust ports without interacting with the bulk cylinder gas. The deflector mitigates but does not eliminate this loss. Trapping efficiency in cross-scavenged engines is typically 0.45 to 0.65, the lowest of any two-stroke scheme in marine use.
Mixing losses
Fresh charge that does deflect upward must traverse the cylinder cover and descend toward the exhaust ports. Along this path it mixes with combustion residuals. The result is a relatively well-mixed cylinder content at the end of scavenging, with scavenging efficiency limited to roughly 0.75 to 0.82.
Deflector-induced losses
The deflector itself disrupts combustion at the next compression-and-firing event. The raised deflector creates an asymmetric combustion chamber that promotes uneven flame spread, increases hot-spot temperatures on one side of the cylinder, and degrades thermal efficiency.
Performance characteristics
| Metric | Cross scavenging | Loop scavenging | Uniflow scavenging |
|---|---|---|---|
| Scavenging efficiency | 0.75 to 0.82 | 0.80 to 0.88 | 0.92 to 0.97 |
| Trapping efficiency | 0.45 to 0.65 | 0.50 to 0.70 | 0.65 to 0.85 |
| Achievable BMEP | 7 to 12 bar | 13 to 17 bar | 18 to 21 bar |
| SFOC | 215 to 240 g/kWh | 195 to 215 g/kWh | 165 to 175 g/kWh |
| Mean piston speed | 5.5 to 6.5 m/s | 6.5 to 7.5 m/s | 7.5 to 8.5 m/s |
| Maximum stroke-bore ratio | 1.5 to 2.2 | 2.5 to 3.0 | 3.5 to 4.7 |
The fuel consumption gap between cross and uniflow scavenging is approximately 50 g/kWh, or about 25 percent. Over a 20,000-hour engine life at 12,000 kW average load, this difference compounds to over 12,000 tonnes of fuel.
The deflector and its problems
The deflector piston was the defining feature and limitation of cross scavenging.
Thermal stress on the deflector
The raised crown projection received heat from combustion on its upper surface and from the scavenge gas wash on its lower edges. Without effective internal cooling, deflector temperatures could reach values that softened the cast iron or alloy steel used. Deflector cracking, deformation, and eventual failure were common.
Liner wear
The asymmetry of the deflector promoted uneven piston ring loading. Rings tilted slightly under combustion pressure, with the tilt direction influenced by the deflector’s offset combustion centre. Liner wear was often noticeably higher on one side, requiring more frequent re-boring and liner replacement than later loop or uniflow designs.
Piston ring sealing
The deflector projected above the piston ring belt, exposing rings to direct combustion radiation around the deflector edges. Rings were often subject to localised heating, leading to ring sticking and accelerated wear at the deflector base.
Combustion chamber
The non-circular, asymmetric combustion volume created by the deflector limited fuel injection pattern optimization. Flame travel distances varied substantially across the chamber, and combustion was rarely uniform.
These problems were, in the early years, lived with as the price of the scheme’s mechanical simplicity. As materials, heat-treatment, and manufacturing improved, however, the burden of designing around the deflector became greater than the burden of adding an exhaust valve or angled scavenge ports. The transition began.
Historical development
1900s and 1910s
Burmeister and Wain delivered the first commercial two-stroke marine diesel engines using cross scavenging. These were simple, low-rated, slow-speed engines installed primarily in cargo ships and tankers. Sulzer, Werkspoor, and other manufacturers followed similar designs.
1920s and 1930s
Cross scavenging dominated. Stroke-bore ratios remained around 1.3 to 1.6. BMEP values were below 7 bar. SFOC values were typically 230 to 250 g/kWh. The scheme served well at the rotational speeds (90 to 130 rpm), powers (up to about 5,000 kW), and fuel quality (then mostly distillate fuels) of the era.
Late 1940s
Loop scavenging, particularly in its Schnuerle and angled-port forms, began displacing cross scavenging. Sulzer’s RD series and B&W’s K series adopted loop scavenging as standard.
1950s and 1960s
Cross scavenging was relegated to smaller and older designs. The largest commercial new-build cross-scavenged engines were below 10,000 kW. Doxford continued building opposed-piston engines, which were uniflow by their geometry, into the 1980s.
1970s onward
Cross scavenging effectively disappeared from new-build orders. Existing engines continued in service, with progressive scrapping over decades. Today only a handful of cross-scavenged engines remain in commercial operation, mostly on small coastal vessels in less stringent emissions regimes.
Why cross scavenging persisted as long as it did
Despite its inefficiencies, cross scavenging persisted for several decades after better alternatives existed. Three reasons account for the lag:
Materials and manufacturing
Reliable hydraulic exhaust valves required materials and manufacturing capabilities that did not become routine until the 1960s. High-temperature alloy castings, hardened seat inserts, and reliable hydraulic actuators were all expensive in early decades.
Cost of fuel
Through much of the cross-scavenging era, fuel was inexpensive relative to engine capital cost. The SFOC penalty was considered a tolerable trade for mechanical simplicity. Only after the oil price rises of the 1970s did fuel-consumption considerations begin dominating engine selection.
Power demand
Cross scavenging was adequate for the powers actually demanded in early- and mid-century shipping. The shift to higher-powered engines (driven by container shipping, larger tankers, and the demand for higher service speeds) outran cross scavenging’s capability and forced the move to better schemes.
Operational considerations for legacy installations
For owners and operators of remaining cross-scavenged engines, the principal concerns are:
Spare parts availability
Most major manufacturers no longer produce cross-scavenged engine spares as standard catalogue items. Pistons, deflectors, liner sections, and ring sets are often only available through specialist suppliers or through scavenging from decommissioned ships. This drives up maintenance cost and can complicate planning.
Fuel adaptation
Cross-scavenged engines were designed for fuels of an earlier era, sometimes with sulphur content, ash content, and viscosity quite different from modern compliant fuels. Operating these engines on modern very-low-sulphur fuels (VLSFO) or marine gas oil (MGO) sometimes requires re-tuning of injection equipment and may produce performance changes the operator should anticipate.
Emissions
Cross-scavenged engines do not naturally meet IMO Tier II NOx limits and certainly do not meet Tier III. Where in service in non-ECA areas, no specific action is required by the engine; in ECAs and SECAs, the ship may be unable to operate or may need exhaust gas treatment.
Re-engining decisions
For ships with cross-scavenged engines, owners typically face a choice between continued operation with rising maintenance and emissions burden, or re-engining with a modern uniflow design. The re-engining cost is comparable to a substantial fraction of vessel value but recovers fuel savings of 20 to 30 percent and brings the ship into emissions compliance.
Modelling and analysis
Cross-scavenging dynamics can be analysed with the same one-dimensional and CFD techniques applied to other gas-exchange schemes. Modelling the deflector geometry adds complication: the asymmetric flow field requires three-dimensional CFD for accurate prediction, and the moving deflector boundary makes the CFD mesh more complex than for uniflow. For research purposes, cross-scavenging models remain useful when designing retrofits or when investigating the gas dynamics of opposed-piston engines that share some flow features with cross-scavenging.
Related Calculators
- Scavenging Efficiency Calculator
- Trapping Efficiency Calculator
- Specific Fuel Oil Consumption Calculator
- Brake Mean Effective Pressure Calculator
- Mean Piston Speed Calculator
See also
- Loop Scavenging Versus Uniflow Scavenging in Two-Stroke Engines
- Uniflow Scavenging in Two-Stroke Marine Engines
- Two-Stroke Marine Diesel Engine Fundamentals
- Crosshead Diesel Engine Architecture Overview
References
- Sher, E. (1990). “Scavenging the Two-Stroke Engine,” Progress in Energy and Combustion Science, 16(2).
- Heywood, J. B. (2018). Internal Combustion Engine Fundamentals (2nd ed.). McGraw-Hill.
- Woodyard, D. (2009). Pounder’s Marine Diesel Engines and Gas Turbines (9th ed.). Butterworth-Heinemann.
- Harrington, R. L. (Ed.). (1992). Marine Engineering. SNAME.
- Stinson, K. W. (1948). Diesel Engineering Handbook. Diesel Publications.