Why not rotary valves?

[john70t]

A pretty good argument:

[john70t]

And while we are at it:

Why not a dual opposing-piston cylinder?
Gas pressure expands in all directions. Why not double the faces it works against.

[rwest]

Originally Posted by john70t

And while we are at it:

Why not a dual opposing-piston cylinder?
Gas pressure expands in all directions. Why not double the faces it works against.

I’m not the brightest guy here or most places for that matter, but would two moving masses have twice the friction with the same amount of force?

[afterburn 549]

I have spent a lot of time on this subject of interest.
They have been tried, rotating cylinders, and a whole lot of other stuff.
This was all during WW2
YouTube has a bunch of information stored there too.
There are people trying to make these work to this hour.
AFAIK, or understand from porting projects, there is CFM flow and velocity.
There are all sorts of tricks to make common valves work better.
But the velocity is truly a work of art on the short side of the port (the curve.)
Remove that, and you get a slug.
There are engines running with these rotary valves of sorts..
But they dont work well.
As far as opposing pistons, afaik this is in progress and may be working somewhere.
What we really need is a sort of Wankel that works.
There are several versions in progress.

[unclebilly]

Too many moving parts and dynamic seals.

[Reg]

I think old 2 stroke skidoos used them for their high performance engines Rotax.

[john70t]

Originally Posted by rwest

I’m not the brightest guy here or most places for that matter, but would two moving masses have twice the friction with the same amount of force?

Good point. I'm not either. just free-thinking here.

1). That didn't seem to slow down the Porsche/Ferrari tiny 12cyl which make over a thousand hp.
(but they seem to be going with larger and less cylinders with hybrid these days)

2). A mono-block makes more sense than two pieces bolted together with a seal at the highest point of pressure..ie the head gasket. That requires extra bolts running through the block and getting in the way of coolant and oil passages and other structure. All that has to be worked around.

3). Combustion chamber with piston shape is all designed around the poppet valves getting in the way. Plus the swirl of compression gasses. Two opposing masses of air would mix better and naturally create a higher compression ratio. I'd guess. The pistons could be more concave, think like two halves of a circle joining, to receive a higher percentage of trust. Direct injection near the spark plug could guaranty the fire starts, but a weaker mix in the rest to keep it burning. Like the 2-chamber Honda CVCC head back from the late 1970s.

4). As for ring drag...an extra titanium piston lip could protect most pressure from reaching the rings, allowing for low-drag types to be used.

5). Put that in an Atkinson crank arraignment https://enginehoist.net/engine-types-atkinson-cycle-vs-otto-vs-miller/ to get the benefits of a stroker but the crank throw of a high rev'ing race car with coffee can pistons. The end result would be a low rpm powerplant using it's entire burn cycle more efficiently. Hopefully. lol

(disclaimer: I'm not a programmer but watching Silicon Valley and thinking 'middle out' whatever that means)

[Bill Verburg]

The Astron Omega1 is the most interesting candidate I've seen

[TimT]

There are a number of opposing piston engines in production today..

https://en.wikipedia.org/wiki/Opposed-piston_engine


https://en.wikipedia.org/wiki/Fairbanks_Morse_38_8-1/8_diesel_engine

[aschen]

It takes work to compress valve springs but that work is returned to the system, minus friction, when the springs extend. Once the valves start closing the engine springs forward.

I don't think rotary valves are a terrible idea just trading new compromises for ones that are already optimized

[afterburn 549]

Conundrums in dynamics.
I dont want to represent myself as an expert, I am not.
But having messed with valve timing and CR ratios on a stupid Harley long enough to get over 1.25 HP per cube will educate one somewhat.
There is this-
Wth rotary timing, is the problem of speeding up and slowing dwn. A hole is only a slot as the drum starts to rotate, and it will only turn @ 1/2 the crankspeed.
This is not nearly good enough.
It just can't get opened fast enough or closed fast enough either.
The miracle of the cam is the ramps, the events can be sped up and slowed dwn.
This will start the event and stop the event with some gentleness, if you will, and open with the ramp speed and the rocker arm ratio installed.
The cam does not have to move faster; the ramp accomplishes this.

Add MVT (mechanical variable timing) with *computers, these events have now become somewhat infinitely adjustable.
Now- with the addition of pneumatic valve train operation, such as the F1 platforms supposedly use, there is no mechanical timing involved!
Just a computer that says when the event should be, and for how long.
SO a rotary style looks good on paper, but as a piston starts and *stops at TDC, BDC, a tulip type valve can be opened .050 in a couple deg of crank revolution, the rotary just can't do it.
Oh, one might reinvent the wheel and get it to happen with accessories; however, a lumpy cam is far easier.
What we need is no valves, and an engine that is between a turbine and a piston configuration, so torque can be had at low and high RPMs

*There is MVT engines out there with no electronics involved in timing.

* One has to time the LSA, and overlap too