181
Speaking of real... lemme get abstract for a minute. The other day Eric one of my genius brothers in law said that it seems to make sense for piston engines to have a number of cylinders into which 360 can be divided. You know, into whole numbers. Like, 360 divided by four is 90. Or something like that - the point being that there's 360 degrees in a circle, and those engines will run best which work best with the way a circle is divided.
I argued that the number of degrees in a circle is completely arbitrary, so we started talking about that for a minute until we were ready to pray for dinner or something, and so i've been thinking about that a little more. We decided that the reason 360 is the number of degrees in a circle is that it's divisible into 2, 3, 4, 5, and 6. It's useful to be able to divide a circle up into thirds, sixths, etcetera, without breaking out the decimal point or the slash of fraction-ness. Also, it's easy to get precise enough for most tasks with whole degrees - one three hundred and sixtieth of a circle is a cute little tiny slice. More like a shim.
So now i'm wondering what number of cylinders does make the most sense. One and two cylinders shake a lot. That's why Harleys act like they are trying to shake all their parts loose. Opposed engines like those found in Subarus and most Porsches are smooth, by virtue of being horizontally opposed, having all vibrations canceled out by the other half of the engine. Three cylinder engines are nice and smooth, but are usually puny. The Geo Metro came with a three.
Four cylinder engines work okay, but they are vibration prone. Fives are just weird, since you have to make them inline, and that gets kind of long. Long engines like inline fives and sixes tend to have uneven temperatures and oil distribution, because the coolant and lubricant has to be pumped all over the place, and needs only the slightest provocation to get uneven. But they are smooth, smooth as that cream cheese frosting i made.
When you get into the realm of sixes, eights, tens and twelves, you are looking at a whole lot of frictional and inertial losses because you have so many heavy, moving parts scraping against each other. Racers use more cylinders (one late sixties Formula One car used two horizontally opposed eight cylinder motors stacked on top of each other) to allow them to get more power from smaller - therefore stronger proportionately - parts.
The engineer in me wants to see a large three cylinder engine be produced for a performance hatchback or something. The engine would be smaller and lighter, and have less internal friction and inertia. You'd get perhaps 10% more power from the same displacement! And it would be smoother than a big inline four, and probably be cheaper too! What the crap is wrong with all these people making cars these days???
Right, so it seems to me that the number of cylinders a car has is a complex engineering problem to which there are no simple mathematical solutions.
Oh, and if anyone actually read that whole thing and was even marginally entertained, please comment because i would love to know if there's anyone else in the universe like me who enjoys thinking about this stuff.
I don’t see the disadvantage of an opposed engine. Except, well, they’re only in expensive cars like Porsches and Subarus, so maybe they are pricey to manufacture. But that late sixties Formula One with the 12-cylinder engine… I wonder why that didn’t catch on!
Also, if the pumping is prone to becoming uneven in the long inline engines, wouldn’t it make sense to have multiple pumps along the length of the engine each servicing one or two cylinders? Sure, it would mean changing multiple caches of oil, instead of one large resevoir, but if it solves the uneven-ness, might it be worth it? I actually know nothing of cars, I’m just hypothesizing from the information you put forth, so maybe I don’t get it and this is already the case.
And how are three-cylindres done? Are they in a V form that is inherently imbalanced, or are they usually in-line? If one could place one at each third or a circle, to make a sort of triangle… Wait, that wouldn’t cancel the vibrations still it’s since 1-up 2-down at a time, which makes imbalanced forces… Umm… Yeah, I guess to me it would make most sense to have any three-cylinder be an inline build. Yeah it would be shaky, but I don’t think there’s any possible way to get a cream-cheese-frosting-smooth operation out of a 3-cylinder.
Haha, please let me know if I’m speaking nonsense here. It makes sense in my head…!
-- Curt Blakeslee (Email) (URL) - 01 July '05 - 00:40Wow, where to begin? One thing you guys are forgetting is that the engines you are talking about are probably 4 stroke engines. In a 2 stroke, where every downward stroke is a power stroke, you could make smooth triples and smooth 90 degree 4 cylinder engines. In 4 stroke engines, where only every other downward stroke is a power stroke, it is difficult to do this. Let me describe it in firing orders: Imagine a 4 cylinder, 4 stroke engine with a crankshaft designed in such a way that each cylinder fires 90 degress after the last one. Now imagine how that would work—#1 fires, 90 degrees later, #2 fires, 90 degrees later #3 fires, 90 degress later number 4 fires. So far so good, right? Except now you have 4 cylinders that each have to intake a fresh charge of fuel/air mixture. #1 intakes, 90 degrees later #2 intakes and so on. Power-power-power-power, intake-intake-intake-intake, power-power-power-power, intake-intake-intake-intake. In the real world, what ends up happening is that you get a real longitudinal rocking motion from the engine. If you ever have the chance to observe a 3 cylinder in-line while it is running, you’ll see this in action. There’s really no good way to build a 3 cylinder without having each cylinder fire 120 degrees apart. Curt’s idea of the triangular 3 cylinder is nothign new — some of the early aircraft engines were just that — 3 cylinder radial engines, where the cylinders were arranged around the crankcase like spokes in a wheel. The design worked well, but you still had the lag between fire-fire-fire and intake-intake-intake.
A word on HD engines—the real reason they rock so bad is because the cylinders are firing 45 degrees apart. This is why they sound the way they do at idle, “pop-pop, pop-pop, pop-pop”. The original reason for this was because Harley and Davidson wanted to build a bike that was as narrow as possible, (the old ones are REALLY skinny), and still had 2 cylinders. The frame they were using lent itself to placing the cylinders at a 45 degree angle from each other. In the early ’80’s, Honda came out with a very smooth running v-twin that still placed the cylinder at a pretty narrow angle, but used a 2 throw crankshaft that fooled the engine into thinking it was an opposed twin, (a la BMW).
I could go on and on and on about this if you realy want me to. Write to me and I will.
Talk to you-uns later,
Andrew
-- Andrew John Julius Bernard Willem Koens (Email) - 01 July '05 - 10:44I’m pretty sure horiontally opposed engines are more expensive to produce, and the biggest benefits are that they are short, and the vibrations do balance out. The vibrations i’m talking about have more to do with the way the pistons move in any engine, not the firing sequence. Like, while a piston is at the top of its stroke, it’s going to whack up toward the head faster than it will move away from it while it’s at the bottom of its stroke. This is because the wrist pin (where the piston connects to the connecting rod) is nearer to the crankshaft bearing (the point around which the crankshaft rotates). So even the most carefully balanced four cylinder engine will vibrate at a frequency twice the engine rpm, even when not doing anything but spinning. There have been many engine designs which combat this problem, like counter-rotating weighted shafts. I read somewhere that Porsche’s big four cylinder (maybe three liters or larger?) would have been impossible were it not for the balance shafts. Even big inline threes would probably need one balance shaft to keep the rocking motions under control. Maybe threes seem smoother because a rocking motion at the frequency of engine RPM is less obnoxious than a up-down vibration at twice the RPM.
Even a 2… uh… throw crankshaft in a v-twin like Honda’s older bikes (they moved very backwards to having the connecting rods on a common throw) would have had some vibration. But at least they would go ba-ba-ba-ba-ba instead of baba-baba-baba. Silly Honda.
So, Andrew, how come the only two cylinder 2 strokes i’ve heard of (like, the Yamaha Banshee) have the pistons going up and down at the same time? Is it because they didn’t want to seperate the crankcases for each cylinder?
Finally, long engines can work great, but you’ve got the added complexity of much more plumbing for the coolant and oil, and any dirt, leaks, or any problem will be made that much worse.
-- juanito (Email) (URL) - 01 July '05 - 16:04I rememebr a preacher talking about 360 – or was it Isaac Asimov, that it is kind of a divine number. . .I will research that for you. I READ IT ALL, and was mildly interested – mostly bcz you are my SON!!!
hugs
-- mummu - 01 July '05 - 18:18I read it all too…even all the comments, but i don’t know if my interest rates quite up there with “mildly interested.” :)
-- karen - 01 July '05 - 21:56