It's All in the Cylinder Heads

It is all in the cylinder heads. Port shape, port volume, valve size, valve shrouding. . . They all have an impact on the way the air/fuel mixture enters the cylinder. How does the port flow as the valve is opened? What is the maximum volume the port can flow? You’d think that the larger the port the more it will flow and the more power the cylinder will make. No, and here is where there is a bunch of disinformation and arcane references to “witchcraft” and flatly incorrect statements like “the bigger the better” or just as incorrect “some restriction is good” come from.

First thing to do is to get the air moving. Since this an intake port for an engine the air will begin moving when the valve opens and the piston goes down (For this port we will pretend that the valve is always open). Apparently the maximum pressure drop created by the cylinder descending is 28” of water¹. This isn’t very much. This translates to 2.06” Hg. This means that the port shape is critical to having good flow. I should also say that I don’t see how this is accurate seeing that the manifold vacuum of an engine can easily reach 25” Hg! Of course this is with the throttle plates closed at idle or on deceleration. As you open the throttle plates and put the engine under load the manifold vacuum will easily drop to 10” Hg and at WOT (wide open throttle) I know I have seen manifold vacuum drop to 4” Hg. It has been almost a year since I did any of this testing but I think I have seen vacuum numbers as low as 2” Hg at WOT and heavy load. WOT with high load is when you want maximum power so I guess it makes sense to measure port flow at this pressure drop. But I am getting ahead of myself.

First, let’s consider a port. Let’s think about an imaginary port that is simply a section of pipe. To make things easier we will not even consider how the air enters or leaves the pipe. We also don’t care how long the pipe is. All we care about is its diameter, or more specifically, its cross-sectional area. Let’s look at two pipes. One pipe has an inside diameter (ID) of 1” and the other pipe has an ID of 4”. Now, let’s pick a flow number. I am choosing arbitrary numbers. I don’t know the real numbers for any of this (YET!). Let’s say that we’ve determined that our cylinder needs to be fed 100 cubic feet per minute (CFM). This 100 CFM needs to pass through our straight pipe port.

The formula for determining the velocity for an ideal gas (one without friction, turbulence, etc.) is as follows:



CFM
------------------- = Velocity of flow
Area of opening

So let’s look at the 1” ID pipe.

Area = πr²

Area = π1² = π

Well isn’t that handy.

For this we will just say that π (pi, pronounced “pie”) is 3.14. π is actually an irrational non-repeating decimal that goes on forever. Some folks have calculated it out to millions of places but that’s about as useful to us as the number of angels that can dance on the head of a pin. So the cross sectional area of the 1” ID pipe is 3.14 in². When we convert that to feet we get 0.02180555555ft². We plug that into the formula and we get:

100 CFM
---------------------- = 4585.99 ft/min
0.02180555555ft²

The 1in ID pipe flows 100 CFM at the rate of 4585.99 ft/min.

Let’s look at the 4in ID pipe now.

Area = π4² = 16π = 50.24 in²

Convert 50.24 in² into 0.3488888889ft², plug it in and you get:

100 CFM
-------------------- = 286.62 ft/min
0.3488888889ft²

The 4in ID pipe flows 100 CFM at the rate of 286.62 ft/min.

4585.99 ft/min vs. 286.62 ft/min. That is a pretty big difference in velocity! Of course that is also a big difference in diameter. I chose such different diameters to make the difference clear.

“Big deal. There are two different velocities.”

Yes, it is a big deal when you consider that air has mass and therefore the faster it goes the more momentum it has. It is a pretty simple formula:

P = mv

Or momentum equals mass times velocity. This means that when the air flows into the cylinder from the 1in ID port it will have about 16 times the inertia as the 4in ID port. This ramming effect is helpful in getting air to continue to enter the cylinder even after the piston has started coming up. This improves volumetric efficiency (VE). Improving VE improves torque. Getting good cylinder filling at low RPMs is essential for good low end torque. This is where the misguided statement about “Some restriction is good” comes from.

This means we want the smallest ports possible right? Let’s really get that air moving!
Not exactly. This is where that 28” H₂0 comes into play.

After a fair amount of searching on the internet I was unable to find a formula that approximates the pressure differential required to produce a given volume of flow through a pipe. No doubt such a thing exists but I have been unable to find it. Fortunately it is not critical for us to understand the basics of what is going on.

As we just saw, smaller ports are good. However, we do not want ports too small or else they won’t flow enough (it was here that I was hoping to show the difference in the pressures required for the 1in ID and 4in ID pipes to flow 100 CFM).

“So what do we want?”

We want the smallest ports possible that will still flow the volume we need at 28” H₂0.

“So is that it? Is that all we need to know?”

Of course not! And the truth is that I don’t know what all the things are to pay attention to, but here is one more thing.

Port shape.

We want to port to be shaped to accomplish three things (that I’m aware of). First, we want the shape to match the way the air wants to flow. We don’t want there to be sharp bends where the air piles up or where the inside of the bend is so sharp that the air wants to shoot past and make flow disturbing eddies. Second, we want a shape that sets the air to flow smoothly over the valve and into the cylinder. Third, we want the head to promote swirl into the flow so that the air is swirling around in the cylinder. This is supposed to do lots of good things (increase torque, help the mix burn faster thus reducing the spark advance, improve fuel mileage, I think there are more things but I don’t know).

So that is what I know. Now let’s see what questions I need to answer.

- How much flow do I need for my application?
- How big do the ports need to be in order to do that?
- What influence does stroke have on this?
- Ramming and airspeed are important but how important is the swirl? How much difference will it make on power, torque, fuel efficiency, etc.?
- What other questions should I be asking?

"What about exhaust ports?"

Obviously similar things apply but the flow is mainly in the opposite direction.

"Mainly?"

Yes both intake and exhaust ports can have flow on both directions but this, like exhaust ports, is something that I will need to learn more about.

"Hey! What about valve shrouding?"

I'll talk about that later. It isn't really a part of what takes place in the port.

Well that took a while but it was worth it. I really feel like I have a much more solid grasp of what I know and a better understanding of what I don’t.

Builder

¹The way inches of water is computed can be described easiest by telling you how to make a gauge. Take a board attach a clear tube to it (without puncturing the tube!) so that the tube forms a ‘U’. Then fill the tube until water is halfway up the ‘U’. Then attach one end of the ‘U’ tube to a port connecting to an area with a pressure drop and keep the other end open to the air. Keep the ‘U’ vertical. If there is a pressure drop then one side of the ‘U’ will be higher than the other. Measure the distance from the top of the two water columns. This is how many inches of water the port is pulling. The name of this device is a manometer.