1 2
chaparral
chaparral GRM+ Memberand HalfDork
10/11/12 1:39 p.m.

Larger cylinders have less surface area for a given volume. This should save fuel.

Fewer cylinders mean fewer components. This should save money.

Their balance isn't inferior to Fours.

So why don't we see big three-cylinder engines? GM could've easily had a 2.1 liter 3, making about 150 horsepower, when they built the Atlas line.

RossD
RossD UberDork
10/11/12 1:42 p.m.
chaparral wrote: Larger cylinders have less surface area for a given volume. This should save fuel.

What don't I know about this statement? Apparently everything. Please enlighten me.

DeadSkunk
DeadSkunk Dork
10/11/12 1:49 p.m.

Volume increases as the square of the piston radius, while cylinder circumference increases linearly. If you double the piston diameter you have twice the circumference to create friction, but the cylinder volume quadrupled.So,in theory,the friction per cubic inch (I just invented a new measureable) decreases as cylinder volume increases. It's an interesting question, but I don't pretend to know the answer.

Flight Service
Flight Service UltraDork
10/11/12 1:49 p.m.

equations dude, equations

Circumference (what will touch the sides of the cylinder walls) is C=pi*D

Area (which is what is the area the fuel is exploded over) is A=pi R^2

Over a 4 diameter the Area grows faster than the circumference, thus you get more bang on the piston face than you have surface touching the cylinder walls.

Total power available for use=total power made - losses in the system.

His theory is to minimize the losses column.

chaparral
chaparral GRM+ Memberand HalfDork
10/11/12 1:50 p.m.

Volume of an engine cylinder: pi((bore^2 )/ 4 )stroke For a roughly "square" cylinder, bore roughly equals stroke, so v = (pi/4) * bore^3

Surface area of the piston: pibore^2/4 Surface area of the cylinder itself: piborestroke; for a roughly square engine that's pibore^2.

As the bore increases, the surface area increases with its square, and the volume with its cube.

mazdeuce
mazdeuce HalfDork
10/11/12 1:58 p.m.

In the whole system, from the valves to the tires, is cylinder friction really that big of a deal?

Conquest351
Conquest351 SuperDork
10/11/12 2:00 p.m.

I was following untill ya'll decided to do maffs. I'll read and act like I know though.

oldtin
oldtin SuperDork
10/11/12 2:02 p.m.

new BMW turbo triple 178hp/199 ft/lbs

pres589
pres589 SuperDork
10/11/12 2:09 p.m.

This is one of the reasons why I posed that a 1.0 Liter Suzuki Swift triple with a turbo would do better on fuel vs. the Mazda 1.8BP in a Miata with identical gearing and enough boost to get the Suzuki up to the same power level.

I also wonder if the BP combustion chamber and exhaust aren't all that hot but that's a separate issue.

RossD
RossD UberDork
10/11/12 2:22 p.m.

I thought you were eluding to something about fuel atomization or cylinder wall temperature in a larger cylinder diameter that would promote better fuel economy. Not just piston to cylinder wall friction.

tuna55
tuna55 UltraDork
10/11/12 2:26 p.m.

Most hyper car type engines have lots and lots of cylinders. I've always wondered why. I think I know the reason, and it's more than just bragging rights.

Firstly, flame can only propagate across the surface of the piston so fast. This counts especially for high RPM. I know for big inch V8 types that 4.5" is considered somewhat of a rule of thumb for max piston diameter. Smaller piston area allows the flame to fully propagate very quickly, possible requiring less ignition timing and reducing losses at that end.

In addition, you need a cylinder head that can feed and relieve this massive cylinder. The bigger the individual cylinder, the bigger the demands on the intake and exhaust runners (and for that matter the manifolds as well). I suspect it's a bit of a compromise (surprise!) and some schemes tend to work better with high cylinder count while others seem to work better with bigger cylinders.

Sky_Render
Sky_Render HalfDork
10/11/12 2:33 p.m.

3-cylinder engines have more free moments of the 1st and 2nd order, although they have no free forces of either order. They also have an ignition interval of 240 degrees. In other words, they are difficult to balance.

Hypercars have more cylinders because, generally, more cylinders equates to a shorter period of crank rotation between ignitions and thus smoother power delivery. Furthermore, an I6 (and thus a V12, which is simply two I6s with a common crank) are perfectly balanced and have no free forces or moments.

Now, on to piston size. As piston size increases, so necessarily does its mass. As a piston's mass increases, so does the force exerted on the connecting rod and thus crankshaft at TDC and BDC. This is one of the reasons big diesel engines have such low redlines. (For instance, the 16L Volvo I6 has a redline of only 2,100 RPM.) So, one cannot simply increase piston size without taking other things into account.

EDIT: Here are two articles that go fairly in-depth in to piston speed, especially as it relates to connecting rod length:

http://www.motoiq.com/magazine_articles/id/1996/does-length-matter.aspx

http://www.motoiq.com/magazine_articles/id/2041/does-length-matter-part-deux.aspx

WARNING: High math content; not for the faint of heart.

Gearheadotaku
Gearheadotaku GRM+ Memberand SuperDork
10/11/12 2:37 p.m.

Ford has some 3's coming in the Fiesta. GM had the 3 cyl Sprint/Metro. But yes, those are all 61 cid engines. I'd rather have more cylinders in the same displacement. (More firing events per revolution). It may cost more, but I feel its worth it.

math makes my head hurt

PHeller
PHeller UltraDork
10/11/12 2:41 p.m.
Raze
Raze SuperDork
10/11/12 3:25 p.m.

Wow, all those years of grad school and I finally can use it here. I know two different things that when taken into consideration probably explain most of the reluctance on the part of the manufacturers:

The first part of the answer is a structures answer: larger pistons = greater mass that has to be decelerated and accelerated at the ends of the stroke which when you examine the stress on the rod, the load increases to the square of the engine speed increase. Thus the rod must be made much larger or out of more expensive materials = $$$.

The second part of the answer is a thermodynamics answer: larger volume in an open/close combustion system are highly dependent on time. We all think of ignition as fuel and spark, but it's actually more complex since there is actually a flame front propagation time through the fuel during combustion. This ends up being more susceptible to perturbations in fuel mixture, air delivery, transient changes, spark angle, etc. in simple terms think of knocking or pinging. With a larger volume you have a longer ignition time since the flame has to propagate further. This means your timing may not be optimal to extract the most power on the stroke, actually causing you to loose efficiency in the system...

I'm not saying either of these couldn't be mitigated, just from a historical/technology/cost perspective I see these as the most likely candidates.

spitfirebill
spitfirebill UltraDork
10/11/12 3:31 p.m.

I got nothing to add.

Conquest351
Conquest351 SuperDork
10/11/12 3:34 p.m.

Wouldn't some modern advances like direct injection and multiple spark plugs per cylinder may solve the flame front and ignition problems. Still, something like a 4.5-5.0" diameter 3 cylinder with a turbo on it would make quite a bit of torque (relatively) and pretty decent hp given the size. Would be awesome for some of the smaller gas sipper cars. with a 5.0" bore and 4.0" stroke, you're at 3.9L and 236 cid. if my maffs is correct. You could make it a 4.25" more and 4.25" stroke and make it an even 3.0L 181 cid engine.

Raze
Raze SuperDork
10/11/12 3:40 p.m.
Conquest351 wrote: Wouldn't some modern advances like direct injection and multiple spark plugs per cylinder may solve the flame front and ignition problems. Still, something like a 4.5-5.0" diameter 3 cylinder with a turbo on it would make quite a bit of torque (relatively) and pretty decent hp given the size. Would be awesome for some of the smaller gas sipper cars.

Some modern tech helps with stability and we are heading in this general direction, DI is the first step. Ideally the HCCI (Homogeneous charge compression ignition i.e. diesel ignition of gasoline) would be the bees knees, but transients and timing are incredibly hard to predict. Even with good sensors and fast computers, the biggest problem is still the fuel itself. Diesel is a very stable fuel with a high ignition temp and a wonderful burn characteristic (it's thick and spreads out more uniformly), while gasoline is much more volatile, spreads out unpredictably due to temp variations within the cylinder, it's pretty hard to predict. There are also variations in gasoline which have to be accounted for, in real-time, on-the-fly. Different companies put different additives in which alter burn profiles. This all means you have to have pretty heavy control laws where most cars use tables, again $$$

DeadSkunk
DeadSkunk Dork
10/11/12 3:41 p.m.

Ya. I was going to say all of that ,too, but I thought I would keep it simple.

singleslammer
singleslammer HalfDork
10/11/12 3:47 p.m.

All I know is that my friends Triumph sounds sweet and my 3 piston starlet should haul ass and save gas (see what I did there).

Ranger50
Ranger50 UltraDork
10/11/12 3:50 p.m.

In reply to Raze:

Mmmmmm diesel............

Raze
Raze SuperDork
10/11/12 3:54 p.m.
Ranger50 wrote: In reply to Raze: Mmmmmm diesel............

Amen to that.

carguy123
carguy123 PowerDork
10/11/12 4:00 p.m.

The advantage I see to the larger piston is that it gives you to be able to pack in multiple valves more easily which is another one of those more modern technologies that might allow you to gain back some of the losses from the weight.

BTW My math card expired a long time ago. I think I still have an outstanding fine too. They don't issue arrest warrants for that do they?

RossD
RossD UberDork
10/11/12 4:26 p.m.
Raze wrote: Wow, all those years of grad school and I finally can use it here. I know two different things that when taken into consideration probably explain most of the reluctance on the part of the manufacturers: [...]

That's the kind of information I was looking for. Thanks.

Ranger50
Ranger50 UltraDork
10/11/12 4:36 p.m.

In reply to carguy123:

Then you could end up with this:

1 2

You'll need to log in to post.

Our Preferred Partners
WEi24UOOzxadu8Emy1FvW3CoUJGT1iC3DhaMKRfe1LeWnFomPJJmJTsrYs4c9hsi