Vehicle: Car: Engine: Rotary
Opposed Piston Twin-Ball Rotary   (+1)  [vote for, against]
Two large ball pistons share a rotating cylinder in a GDi 2-stroke engine.

This rotary engine is essentially a rotating 2-stroke opposed piston engine (it requires a blower to feed it air).

However, the cylinder rotates, and the pistons are ball shaped. Rather than use two crankshafts on opposite sides of the cylinder, a stator cam shape around the engine is used. (See attached pictures). The cam is assymetrical, so the exhaust piston opens before the intake piston.

Since these ball pistons may not seal sufficiently well to provide diesel pressures, gasoline direct injection is used. The central injector can also be used to inject water for cooling/steam power. At maximum power, the engine alternates between injecting fuel and injecting water.

Compared to the famous Ball Piston Engine, this engine is much simpler and doesn't require sliding seals. This engine also provides an incredibly large working volume and can thus be very compact. This engine has an excellent working volume to surface area ratio; comparable to traditional piston engines.

Compared to traditional piston engines, this engine is much simpler and inherently well balanced. It eliminates the crankshaft and counterweights, as well as cams and valves.

The big challenge, of course, is acheiving an adequate seal using a ball shaped piston. This concept can be adapted to using traditional pistons with rollers to engage the outer cam stator, but this solution is boring and lacks "pizazz".
-- IJK, Mar 24 2007

Image 1 http://members.cox..../engine/oballs1.png
Image showing ball pistons at closest approach. [IJK, Mar 24 2007]

Image 2 http://members.cox..../engine/oballs2.png
Image of engine 90 degrees later. [IJK, Mar 24 2007]

Are you providing power to make it rotate from some other source? There seems to be nothing else within the diagrams or description that will produce a rotational effect.
-- the dog's breakfast, Mar 24 2007

The outward movement of the ball pistons causes rotation. Fuel injection occurs when the pistons are at the 12 and 6 o'clock positions. This gas pressure causes them to move outward, which in turn powers rotation until they reach the 2 and 8 o'clock positions. At that point, the exhaust piston lets gas escape.

Just like any other engine, a starter is required to provide initial rotation. From then on, power strokes maintain rotation while inertia keeps the mechanism rotating until the next power stroke.

For continuous power, you can stack two or three of these together within the same stator cam chamber, but this complicates injector design.
-- IJK, Mar 24 2007

I thought this was going to be a position.
-- MaxwellBuchanan, Mar 25 2007

Would need very large flywheel, and also, where does the oil go?
-- the dog's breakfast, Mar 25 2007

Why is a large flywheel necessary? The components themselves act as something of a flywheel already. The cylinder is rotating, of course, and the ball pistons are rolling around in a curved orbit.

For lubrication, it may be best to orient the chamber so that the intake side is on the bottom, where oil can accumulate. One or more sprays of oil can "paint" the balls as they pass by the bottom side with drops of oil. The rolling action of the balls ensures this oil gets evenly distributed.

But really, for an actual practical engine design I don't think ball pistons are ready for prime time. An adaption of this concept using normal pistons and rollers engaging the cam stator would be more practical...but also more boring. These traditional pistons would be lubricated in the traditional way, of course.
-- IJK, Mar 25 2007

[+] then - for the idea.
-- the dog's breakfast, Mar 25 2007

Thanks for the votes!

The AS Four Stroke suffers from fast sliding seals under high pressure (a feature shared by many rotary concepts).

As for what defines "rotary" vs "reciprocating" vs "other"'s sort of fuzzy.

To me, the defining feature of a "rotary" is that all of the major components undergo relatively smooth rotary movement. For each component, there's some central "pivot" point around which the component is always moving clockwise or always moving counterclockwise.

That "pivot" point might not be the same for all components. A lot of rotary concepts have two or more "gears" rotating around offset axes.
-- IJK, Mar 25 2007

random, halfbakery