This is a concept I have been toying with for a while, yet up until now have not found a practical or even workable way to implement. I have now found a way that it might work, yet I'm unsure if there would be any added efficiency.
In one of my ideas I proposed a radial engine that rotated itself around a symmetrical camshaft, creating a boxer type. I liked this concept because of its potential added efficiency and overall simplicity, but no one else really did. I propose a radial engine that spins around a drive-cam (which is actually a stationary cam. The only thing is, each piston has two connecting rods that are parallel and connect to two seperate lobes (one in front of the engine, one in the rear). The way the connecting rods stay connected to the lobe is by having a groove in the lobe where there are two rollers that roll along it, making sure the cam both pulls the piston down and pushes it back up.
Here's the interesting part: the cam lobes are not identical. The lobe up front is the intake and exhaust cam while the one in back is the compression and combustion cam. Both connecting rods have springs that can be tentioned but not compressed, a necessary feature since when the intake cam is exhausting spent gasses, the compression cam is at its lowest point. The tention springs allow the lobe that is pushing highest to have control of the piston, while the other connecting rod compensates for the difference. The variable part of all this is that the compression lobe is able to be turned up to 90 degrees. Since the engine spins around the cam, this is possible.
When the compression cam is moved counter-clockwise toward the intake cam, it varries the intake "valley", making it smaller. As is is moved clockwise it makes it "deeper". This has the potential to be more efficient since not only does it not require a throttle and not have any wasted energy (from either vaccuum or, in the case of the Atkinson cycle, sucking in the full stroke of air and then pushing some back out), but at idle it also allows for both more combustion and exhaust time. In this extra exhaust time, it would be possible to have an unrestricted exhaust where it could simply flow out of the cylinder because of the fact that the cam position allows for the cylinder to remain down longer. I imagine a 4 valve head with 2 for intake, 1 for free-flowing exhaust, and 1 for forced exhaust. Maybe some sort of sleeve valve design for unrestricted intake is possible, but I haven't really thought that far.
So to clarify, at idle the compression lobe is nearest the intake lobe, creating perhaps less than an inch of intake but still the full stroke (whatever that might be) of combustion. At WOT (wide open throttle), the compression lobe is a full 180° apart from the intake lobe, making a perfectly smooth boxer engine.
Problems? 1.) There is no potential for direct injection since the engine is rotating 2.) Trying to work out the variable valve timing would be a bitch 3.) the entire intake stroke relies on the resistance of the tention springs since those are the only things pulling the piston down, meaning during all other cycles this is just added inefficiency and finally 4.) with variable intake and compression, ultimately leading to variable ignition and timing, it would sure sound weird and probably vibrate pretty badly at any other point than WOT.
I'll post illustrations soon, and sorry it's so long but I haven't posted in a long time and have saved this for a while.-- acurafan07, Jan 27 2008 Illustration http://i210.photobu.../acurafan07/ggg.jpgCrude, but it gets the point across. [acurafan07, Jan 27 2008] There have been many radial engines in the past where the block spun and the crank was fixed (look up the Gnome for one). There were collector rings for the intake and exhaust, and I'm sure fuel injection could be implemented, although not easily.
I'm also not sure I understand your bit about the bearings, since no rotating radial to my knowledge has even been a boxer type, meaning this at idle would not be any less smooth.
And finally, the exhaust does not need to be immediately foreced. I'll try to find a link to a Ducati "6-stroke" engine. My point was that the spare time would allow for some free flow exhaust while the piston is down, but would still have the same forced exhaust at the end (just using less energy).-- acurafan07, Jan 27 2008 The advantages of the Gnome and Le Rhone rotary 2-stroke engines were: 1. High power to weight ratio 2. Small size, as all the cylinders rotated in the same plane, sequentially. 3. Simplicity.
"Rotary" piston engines, as opposed to radial engines, are almost exclusively 2-strokes, due to the lubrication problem inherent in the central rotating crankcase. Most WW1 rotary aircraft engines had a total-loss lubrication system using castor oil. Also, the volume occupied by the engine at rest is for all practical purposes identical to its kinematic envelope when operating.
Creating a roatating "boxer" engine will surely create an engine with a kinematic envelope out of all proportion to its rest size, leading to mounting and cooling issues. Fuel and lubricant flow would also be problematic, although with modern electronic ignition this could be mounted as a self-contained module on the rotating portion of the engine, powered by a small alternator built onto the shaft, analagous to the magneto in a Le Rhone.
The sleeve valves are a not a bad idea, but are greedy of oil (see the Ricardo diesel engine fitted to the British Mark 1 Tank in WW1) but the killer problems are going to be the swept volume of rotation, and the problem of balance. The shaft is going to have to be strongly supported at BOTH ends, and how then do you extract the power ?-- 8th of 7, Jan 27 2008 Supporting the shaft would be difficult yet do-able. Extracting power would probably be fairly easy in comparison, with some sort of frame connected to the block that has a shaft in the center.
You're right that it would drink oil, centrifugal force would not help this engine in the least. But I don't believe I've ever claimed my designs to be practical, or even in most cases better; just different.-- acurafan07, Jan 27 2008 //.... The idea seems overly complicated. UnaBubba, Jan 27 2008 //
This is a hallmark of Acurafan's engine ideas. I have hammered on him ruthlessly especially early on. I would also note however that this is not a reason for to ban the idea or to imply that this is not an acceptable forum for it. If it were then many of our most interesting members would find themselves homeless.-- jhomrighaus, Jan 27 2008 We support [jhomrighaus] in this, since the idea fulfills all the criteria of being on the one hand just about practical given sufficient time and resource, but on the other hand pointless and no better than many existing systems which duplicate its function effectively, thus making it half-baked.-- 8th of 7, Jan 27 2008 Either my attention span is the size of a fish, or you are rambling. A key to getting buns on this site is to kiss (keep it short stupid). Also, a picture usually says a million words, but your picture doesn't look anything like a "Variable Intake Radial Engine" like your title says, but more like a half to full throttle switch with 3 more tubes than you need.-- quantum_flux, Jan 28 2008 I already had an illustration of a 6-cylinder radial; didn't have the time to edit out 3 of the cylinders. I also didn't have time to make more illustrations of different throttle positions, figuring it was pretty self explanitory.-- acurafan07, Jan 28 2008 Sorry, I think my attention span is really short today. I'll try again later.-- quantum_flux, Jan 28 2008 random, halfbakery