This question isn't at all practical, but I was just wondering. Most cars now have more than 2 valves/cylinder. Its my understanding that this allows the valves to cover more of the cylinder cross section (geometry with circles), along with some thermal advantages, etc. Why, though, must the valves be circular? Is the heat transfer problem too difficult for a design like that?

 

Just a few thoughts on the subject. First, two 1" valves have a greater "flow area" than one 2" valve. Smaller, even if multiple, ports maintain a higher port velocity. Smaller valves transfer heat back to the head more easily, because they have a higher contact area to mass ratio. A lighter, smaller, valve requires less valve spring and is less prone to float. I haven't gone into a lot of detail as to "why", but these are some of the reasons.

Circular? Most valves rotate. Many engines have positive rotaters, just to be sure that they do. Try sticking a square or rectangular valve in your valve refacing machine. It's tough doing between the corners.

RACE ON!!!


[Modified by CFI-EFI, 7:58 PM 2/8/2003]

 

So the valves are supposed to rotate? Is that to keep a more constant A/F ratio, or is it just simpler to balance everything that way? I don't know anything about real engines :)

I was actually thinking more along the lines of hemispherical valves - one (almost) covering half the available area, and the other covering the other half. If valve rotation is necessary, then that idea certainly wouldn't work!

 

The round valves have nothing to do with the A/F ratio OR balance. They are not a part of the rotating/recprocating assembly. They rotate for even wear of the valve face and seat.

"Build it and they will come"

 

Like these guys? http://www.coatesengine.com/technology.html

 

That's awesome. What's holding them back from production?

 

That is a very intresting idea. Cam and valves all in one. very very good idea. I wonder how the combustion chambers look though with those round valve shapes. Personally I don't see why a oval shaped valve would not work, or one that even conformed with the shape of the chamber. I don't see what a valve would need to rotate, then again it does cause any harm that a circular one does as it is.

 

Coates has been hawking this system for a good ten years, but nobody's buying.

Take a look at this valve system and imagine what the flow coefficient is with the valves only partially open. Think about how much friction you have with all the "bearing" area of those "***** and sockets". Also, think about how you are going to provide lubrication to those "***** and sockets" and seal them to prevent oil migration to the ports, and gas migration into the engine. There are a lot of real world engineering issues with this system.

If you were around in the thirties (or you have some engine history books in your library) you might remember sleeve valves as they were all the rage back then, but they ended up not working very well - high friction, high wear, short life.

Poppet valves have been around since the dawn of the internal combustion engine and NOBODY has ever come up with a better solution, but keep those cards and letters comin'.

Duke

 

Another factor is valve shrouding. This is an issue if you go too large on a valve size with a combustion chamber design that doesn't address the shrouding. A bigger valve alone doesn't contribute to increased flow. The combustion chamber design has a lot to do with it too.

Another possible reason I can think of to support multiple valves is that the valves each can be smaller and still flow better than a single large valve, and the smaller size would be lighter and tend to reciprocate better, lending it a tendancy to rev better.

Even a head with 1.94/1.55 valves can make excellent power and torque, provided the head flows well, and is well-matched to the rest of the engine.

Dominion Performance used to sell a very interesting multiple valve head for Chevy V8 small blocks. Amazingly, they would still use the pushrod method from the same cam. The intake tracts were designed to put more of a "spin" to the intake charge, so that as the charge enters the cylinder, it swirls as well as tumbles, improving torque. Whereas the more symmetrical intake tract design only tumbles, without the swirl, creating a sort of torque "lag" at lower rpm's. I read test comparisons between the Dominion heads compared with a reputable aftermarket 2.08/1.60 aluminum head design. The Dominions gave up no torque, but above something like 4200 rpm, the engine's power took off and climbed over 100 hp more than the 2-valve heads.

The only problem with the Dominions was their price. You could build an entire engine, fairly trick, for what a pair of Dominion heads cost.
But can you imagine them on a twin turbo set up? Wow.

 

You get more "valve curtain area" with four valves. The smallest flow area throughout most of the inlet and exhaust stroke is at the valves, and this cylindrical area is defined by valve circumference times valve lift. It's total valve circumference that counts, not total valve area and you can cram a lot more valve circumference into a circular plan form with four rather than two valves, although there is some mutual shrouding between two valves. Very large valves can also be shrouded by the walls of the cylinder.

Once a valve is lifted about 25 percent of its diameter flow is determined by overall port flow characteristics. Further valve lift has little effect. Thus, a smaller valve is both lighter and doesn't have to be lifted as far, and curtain area is larger at low lifts with two small valves than with one large valve, so rev potential is higher to suit the higher flow, which means more power at high revs.

Duke

 

more valves equal better flow, and they're round because, yes they turn, also its much easier to get a good seal on something round, thats why ships use circular port holes instead of square windows (useless knowledge)

 

Increased valve lift certainly can improve power, by the way. It does make a difference on some engines if you increase the ratio of your rockers, or go with a cam that has more lift. Unless I'm misunderstanding the statement.