Please log in.
Before you can vote, you need to register. Please log in or create an account.
Vehicle: Car: Engine: Rotary
Radial/Rotary Engine With Rotary Valves   (+6, -1)  [vote for, against]
Radial engine rotates around crank, opening and closing ports in a sleeve

Radial engines with cylinders that rotate around a crank, also called rotary engines, carry out different cycles during different degrees of their rotation.

Rather than use ring cams, push rods, rockers, springs, and poppet valves to control these phases, you could use a stationary ring around the peripheracy of the cylinders that alternately covers up and exposes the ports for inlet and exhaust during different phases of the cycle.

The crank would be stationary as it is on the gnome engine (see link), but the cylinders would rotate around it. Instead of separate heads and valves for each cylinder, you would have a ring-shaped casing fitted around the outside edge. This would have ports cut for intake and exhaust much like a Wankel engine, as well as a spark plug.

The tops of the cylinder sleeves would seal on the inside edge of this ring. Sealing might be a problem as it is in Wankel type engines, but labrinth seals around the edges should work along with spring-loaded metal strips like in Wankel engines.

This design could more easily take advantage of different combustion chamber shapes and might even provide some kind of supercharging similar to the effects of "inertial supercharging" in Wankel engines or the centrifugal forces present in conventional rotary radial engines.
-- discontinuuity, Jul 12 2005

radial engines http://en.wikipedia...otary_piston_engine
Wikipedia article on radial engines [discontinuuity, Jul 12 2005]

Rotating radial http://www.keveney.com/gnome.html
more like what I was talking about [discontinuuity, Jul 12 2005]

Tri-Dyne http://www.deadbeat...iptoid/tridyne.html
Here's a pure rotary engine with only 3 major moving parts (the central power rotor and two adjacent rotating valves, all in 1:1 sync), from July 1969 issue of Popular Science, if I recall right. Features "4-stroke" combustion cycle, 4HP per cubic inch displacement, and 12,000RPM easily. [Vernon, Jul 12 2005, last modified Jul 25 2005]

Electromagnetic Flow Pump http://www.ventraco...ages/Vcor_rotor.jpg
[reensure, Jul 12 2005]

Constant Volume Combustion Engine http://www.ducati.c...rt=general&artID=11
Six cycle engine uses a cam instead of crank [discontinuuity, Jul 22 2005]

Baked? http://www.floridaa...ryPages/g1m0075.htm
I think someone might've beat me to it with a model airplane engine [discontinuuity, May 18 2007]

The link by [ya33a] http://www.rcvengines.com/
as a link. [baconbrain, Aug 14 2007]

Tri-Dyne http://books.google...tri-dyne%22&f=false
Actual July 1969 article in Popular Science [Vernon, Mar 08 2010]

A vote, positive or negative, is sufficient.

What is the envisioned application for an engine of this design?
-- bristolz, Jul 12 2005


This is just an improvement on the big air cooled WWI airplane engines. It could be used for airplanes, cars, motorcylces, or anything. Like any rotary radial, it would require tons of shrouding and take up a large flat, round space. With planetary gears, clutch, etc. it could even be integrated into the wheel hub of a vehicle. I don't know of any specific applications, but airplanes, motorcylces, and small cars would be best.
-- discontinuuity, Jul 12 2005


I don't think it'd work well for aircraft due the rotational mass, at least aircraft with a low wing loading. It didn't work out well with the Sopwith Camel, anyway.
-- bristolz, Jul 12 2005


I really have trouble making any sense of this one, but I know the last paragraph is wishful thinking.
-- baconbrain, Jul 12 2005


I have a problem with " a stationary ring around the peripheracy of the cylinders that alternately covers up and exposes the ports " because I think that is the novel Idea, and is not a bad notion. It is just so poorly explained that I really don't get it the more I think about it.

I believe that two cycle engines work on a similar valving design.
-- reensure, Jul 12 2005


If I understand this correctly, each cylinder would have a single port for use as both intake and exhaust. As the engine rotates around the crank, "gnome" style, this port would pass by the intake and exhaust manifolds that are cut into the stationary part of the engine.

Taken seperately, the two ideas are not unique, but I have never seen them combined in this manner. [+] for the concept.
-- Freefall, Jul 12 2005


Yes, [Freefall] you are right. I'm not sure what the best head design would be, or how the ports would work, but this is the basic idea.
-- discontinuuity, Jul 12 2005


You could also make it a two-stroke depending on how the valves/ports work.
-- discontinuuity, Jul 12 2005


How about zero perturbation at the fuel-air interface <link> or regenerative exhaust ports?
-- reensure, Jul 12 2005


some of the old radial engines had no poppet valves, but instead had cutouts in the head that were covered or uncovered as the cylinder was twisted around its axis by a linkage to a cam. ( I am not explainign this well) anyway, there is no need to have pusrods and rocker arms whirling around at maximum radius, the swivel on the cylinder is at the base of the cylinder and so it is a compact , and I believe, elegant design.
-- maniacal_engineer, Jul 13 2005


//You could also make it a two-stroke depending on how the valves/ports work.// You would have no choice but to make it a two-stroke. This idea works out as a two-stroke with very badly-sealing valves added to it. It doesn't have a cam or any equivalent of one, so it can't be a four-stroke.

The idea is really very poorly written for such a technical subject. The word "peripheracy" doesn't really exist, so its meaning must be guessed at when discussing an engine.

The basic idea seems to be an alteration to a Gnome rotary. That design dates from the Great War, and the engine designers back then tried *everything*. Sleeve valves are not new. Sliding valves are not new, they just didn't work. This idea does not contribute anything to making them work, it just hopes they would.
-- baconbrain, Jul 13 2005


Sorry [baconbrain], I don't think you really understand the idea. The engine has intake and exhaust ports that are covered and uncovered by the motion of the cylinders. It has four strokes, but could be modified with different timing and a transfer phase to make a two-stroke. There are no twisting sleeves; the ring-shaped "head" with its intake and exhaust parts and spark plug, as well as the crank shaft, are all stationary, while the cylinders rotate around like the gnome engine.

Some other features:

Since the different components of the ring-shaped "head" (the exhaust manifold, the intake manifold, and the spark plug) all move on separate ring-shaped sections, functions like variable valve timing and spark advance can be controlled by moving that particular section relative to the stationary crankshaft.

As far as the "supercharging" effect, if you shape the ports correctly, with the motion of the cylinders, the ports would act as scoops to draw in or expell gasses.

I'm sorry if my description was a little vague, so I will answer any more questions you might have.
-- discontinuuity, Jul 14 2005


One mistake: you would have to perform all the four strokes within one rotation of the engine, while the gnome engine requires two full rotations to do the same. I don't think this would be a problem though. Maybe a very large cam shaped like an ellipse would work better than a crank.
-- discontinuuity, Jul 14 2005


Here is the engine in its simplist terms:

Imagine a circle divided into four sections. In section one (9 o clock to 12 o clock), there is an intake port around the outside edge. The entire intake phase happens inside this quadrant.

Section two (12 o clock to 3 o clock) is the compression stage.

Section three (3 o clock to 6 o clock) is the combustion/expansion stage. There is a spark plug near 6 o clock.

Section four (6 o clock to 9 o clock) is the exhaust stage. There is an exhaust port along its outside edge.

Now take a radial engine similar to the gnome, take off the heads, and place it inside this ring.

During a cylce, each cylinder takes in air/fuel mixture, turns 90 degrees, compresses the mixture, turns 90 degrees, burns and expands the gas, turns 90 degrees, expells the gass, and turns 90 degrees once more.
-- discontinuuity, Jul 14 2005


//Sorry [baconbrain], I don't think you really understand the idea.// Very true, which is why I keep objecting to the writing. I understood it well enough to say that it had to be a two-stroke, and now the description changes to make a four-stroke possible. I understood that such a change was needed, BEFORE it was added to the design.

I've been sketching airplane engines for years, reading about them and helping to rebuild them. I once sketched an opposed-piston engine, then started researching for anything similar. I found a cutting-edge NASA project that was working on the same concept, but had missed a few things I had thought of. I kept researching, and found that my entire design had been done before, as a Junkers Jumo engine in World War One.

I am also a technical writer and editor, at times. If I do not understand your writing, your writing is at fault. I am also a mechanical engineer. If I and UnaBubba do not understand your design, your design is at fault. I am also a designer, and know how easy it is to make a wrong assumption. I am not assuming you are wrong, but I am losing confidence that you know what you are doing.

If you are assuming I want you to fail, my writing has been at fault. The 'Bakery is here to help.
-- baconbrain, Jul 14 2005


After some fast Googling . . . Yes, it looks like. Which just goes to show that coming up with a truly original engine is not easy. Learning something new is. Thanks, UnaBubba.
-- baconbrain, Jul 14 2005


Hmm, oddly this gives me an idea. +
-- sartep, Jul 14 2005


You have come across as rather rude, arrogant, and insulting, [baconbrain]. I don't mean to start a flame war, but that's how it seemed to me. I may not have explained my idea clearly, but the way you pointed this out did not help either.

I have added a few edits, so I hope that will clear up any misconceptions. It seems one paragraph got lost the first time I wrote it.
-- discontinuuity, Jul 15 2005


The ports work similarly to those in a Wankel engine, except the casing is circular shaped and there is a radial engine in the middle without heads, instead of a triangular rotor. I beleive you are right that with a normal crank this would have to be a two-stroke, but the oval-shaped cam I mentioned would allow four-stroke motion (see link for an explination of this technology). You could also use some kind of eccentric shaft and planetary gears to make the crank move in a 4-cycle motion.
-- discontinuuity, Jul 22 2005


You definitely don't want this much rotating mass in a motorcycle application: it would be damn near undriveable; all that angular momentum equals straight-line performance and fie on you if you want anything different.

In a car app it could work, but the best layout would be to have it rotating in the horizontal plane. Fixed so it could pitch but not roll, would make the car highly resistant to rolling in sharp maneuvers.
-- elhigh, Jul 22 2005


If it was mounted horizontally in a car, wouldn't it create mountrous torque steer, especially on slippery surfaces? The whole car would twist around like a helicopter with a broken tail rotor.
-- discontinuuity, Jul 25 2005


There are only 2 strokes in any full revolution. Therefore it can only be a 2-stroke.
-- moPuddin, Jul 25 2005


No, if you had a conventional crankshaft it would be a two stroke, but I would use a cam instead (see the link 'constant volume combustion engine') to make it a four-stroke.
-- discontinuuity, Jul 27 2005


Or you make the crank rotate at half the speed but the same direction as the cylinders with some planetary gears. Then the entire engine would move except the peripherial ports.
-- discontinuuity, Aug 07 2005


Interesting progression on the moving-head thing [+]

It would be a two-stroke; but it could be a four-stroke if the annular "head" - or the cylinders - were to oscillate back and forth on the rotational axis at a frequency that corresponds to half engine speed, so that the "valve" bits only pass any given cylinder every second revolution. Or would it? I'll have to go away and try and visualize the resulting twisted-figure-8 motion. No, it would: all the activity would happen around one area of the periphery.

Either way, an obvious application would be a twin-engined aircraft, with or without contra-rotating engines.
-- Ned_Ludd, May 18 2007


So it looks like someone beat me to it. Check out the link for a fully baked model airplane engine.
-- discontinuuity, May 18 2007


That link certainly looks like your idea. I wonder how they did the seals.

Congratulations on a workable idea, and sympathies for not making it first. And thanks for the link to an interesting site.
-- baconbrain, May 18 2007


Hi, Yes the idea exists and here it is, you were all correct in the engine is self turbo / supercharging (to a degree) and the torque is incredible, the engine is very smooth and quiet.

Here is the link http://www.rcvengines.com/

I saw one in a model airplane..

Thanks
-- ya33a, Aug 14 2007


I put that link in as a link up there.
-- baconbrain, Aug 14 2007


That last link is not the same thing. It uses a rotating sleeve valve around a single cylinder, not a radial engine.
-- discontinuuity, Aug 14 2007


this is baked.
-- WcW, Mar 08 2010


Radial and rotary engines aren't the same thing - and in a radial engine the cylinders don't move :O As for the idea I guess it could work
-- DIYMatt, Mar 09 2010


Re. [Vernon]'s link to PopSci July '69 - what a blast from the past! Note the hybrid car article, etc. etc.
-- csea, Mar 09 2010



random, halfbakery