So a friend told me about the Duke Axial Engine the other day. It’s certainly a very interested idea and does produce some clever solutions to some of the fundamental complexities and disadvantages to the reciprocating internal combustion engine. I always like the idea of novel engine designs, and this one certainly looks like it has potential. Take a look at the video followed by some analysis after the break.
The basic concept of the engine is a 4-stroke, spark ignition, port injected engine. This engine utilises 5 cylinders, a type of slide porting and swash plate style con-rod/crank arrangement. Duke advertises the main advantages as having significantly fewer parts (meaning weight, complexity and cost) than a comparable in-line, poppet valve type engine, but also being more resilient to pre-ignition, and having significantly reduced first and second order vibrations.
The concept shares some commonalities between different engines through history, but combine to make a pretty unusual engine. The cylinders themselves rotate, something similar to a Rotary (not Wankel) engine (animation) but are aligned with the crankshaft as per an Axial Engine, however, there are no valves as per a poppet valve 4-stroke. Valving is instead achieved using ports similar to a 2-stroke ported engine, however the ports are in the face of the head rather than the bore. Duke also use a novel swash plate type mechanism on the crank which they call the ‘Reciprocator’ which reverses the direction of rotation of the crank to that of the cylinders.
Their current, 3rd generation engine, is a 3-litre, 5 cylinder, 215hp prototype which has apparently been successfully tested on a range of fuels. Duke provide a surprising amount of technical data on this page. Due to essentially sharing parts of the cycle, the 5 cylinder engine only requires 3 cylinders worth of ancillaries and hence only has three spark plugs, intake ports, exhaust ports and fuel injectors. There are two power strokes per revolution of the crankshaft and hence the engine has a high torque output.
The engine appears to have a number of potential challenges, primarily due to the difficulties of sealing the sliding ports, and also the complexities involved in the crank arrangement potentially providing wear and higher friction than conventional crank trains. Valve timing and overlap could also potentially pose an issue. One might also expect high reactionary torques on the mounting of the engine due to much higher inertia caused by the rotation of the pistons and bores, however, the counter-rotating crankshaft is designed to cancel these forces.
Duke actually address many of the issues raised, suggesting that they have solutions for the sealing based on oil films, although they do admit a slightly higher oil consumption. Perhaps these issues could be resolved with high temperature PTFE coatings, ceramics or ‘Nano Surfaces’ as apparently used in the Koenigsegg Agera engine. Apparently the overall friction for the engine is currently similar to that of a conventional engine, where it would appear higher crank losses are negated by the removal of the whole valvetrain.
One quite interesting improvement to the engine which would be simpler than on traditional engines would be to create a variable compression ratio engine, which could be achieved by sliding the swashplate arrangement up and down the crankshaft. This would improve even further the fuel flexibility and potentially enable an engine which could be dynamically compression ignition and spark ignition.
There doesn’t appear to be a great deal of information on the engine that is newer than a few years old, but there is a small bit about a slightly downsized duke engine of 1000cc potentially for motorcycle use here: https://rideapart.com/articles/duke-axial-motorcycle-engine