Don Terrill’s Speed-Talk

But here's the kicker - the big port Edelbrock head that produced this flow curve had a port volume of 382 cc and a mean c/section of 3.99 Sq. ins and a C/section at the manifold face of a whopping 4.35 Sq.ins. The latter being bigger than the valve area - which is not good!

By contrast the 'weird port' head is 322 cc's for a mean c/section of 3.48 Sq. ins. and a c/section at the manifold face of 3.9. That's a whole lot smaller than the conventional Edelbrock port.

Are these two heads using the same valve size?

GARY C wrote: ↑Fri Oct 19, 2018 10:28 pm

But here's the kicker - the big port Edelbrock head that produced this flow curve had a port volume of 382 cc and a mean c/section of 3.99 Sq. ins and a C/section at the manifold face of a whopping 4.35 Sq.ins. The latter being bigger than the valve area - which is not good!

By contrast the 'weird port' head is 322 cc's for a mean c/section of 3.48 Sq. ins. and a c/section at the manifold face of 3.9. That's a whole lot smaller than the conventional Edelbrock port.

Are these two heads using the same valve size?

Gary,
Yes 2.3/1.88

So we have two cylinder heads That flow a similar amount of air but one with a 382 cc port and the other with 322. Obviously one is going to havbe more velocity than the other.
Let's look at a comparison of the mean port velocity and see how these to stack out.

The graph below does just that. At this point we are looking at a comparison of the intake only - that's the blue curves. The thin blue line is the 382 cc port and the thick on the 322 cc port.

Pretty big difference eh!

OK now lets take a look at what we have here.

If you refer to the text books that we had back when I was a student you will find that it was commonly quoted by such folk as Taylor,Ricardo, Hayward,Etc (in other words some of the greats in engine science), that peak power occurred at 300 ft/sec and peak torque 240 ft/sec. Well for their day (1940's to maybe 1960's) these were good numbers to go with for a typical 2 valve engine of the day.

AS it happens they are not too far off what we see for the mean velocity in a typical modern 2 valve engine. As you can imagine many factors conspire to inch this 'magic number up somewhat. Cutting the engines internal friction and parasitic losses are one such factor. The amount of internal losses on a 1960 race engine and what we see now even for a serious street motor are night and day. However there is one factor that has brought up the peak power port velocity of the modern 2 valve engine dramatically and that is the flow bench.

When Stan and I wrote the IOP flow bench program we needed to categorize the velocity so that it applied to what happened in the real world. That proved a whole lot simpler that might be expected. What you see in the graph is a curve that can be used to establish if the port is too big, too small or just right.

For the best combination of flow and velocity our test head needs to have between 300 and about 330 ft.sec to get the best balance between flow and velocity. As you can see the small port head has almost 300 ft/sec at full lift. Pretty good but still not quite good enough.

So how well does the small port head work. Here I will re-post the power curve for the 87 octane 511 I did with Terry Walters in Roanoke. Look for the output graph in my next post. As you can see it was kick butt build - radius seats and all!.

DV

I guess that the velocity is the velocity at 28" on a flowbench and not on a running engine?

Erland

Taylor used a flow rig to determine Mach Index. I think some were pressurized and others used a vacuum or vacuum/low-pressure reservoir.

Erland Cox wrote: ↑Sun Oct 21, 2018 2:44 pm I guess that the velocity is the velocity at 28" on a flowbench and not on a running engine?

Erland

As it happens these numbers are very similar. The velocity shown in the graph is that on the bench at 28 inches. The intake velocity we are shooting for on a 2 valve engine like a BBC (and for that matter a SBC and Ford) as shown on the bench has shown, by yno testing, to be 300 -310 as measured on a 28 inch bench test and then calculated by IOP from the mean CSA and the CFM.

I alluded to the fact that the 240 for peak torque and 300 for peak hp were the standard used back in the 60's. I will post some velocity numbers to show how different design heads vary from those two numbers.
DV

At what people do you start looking at velocity along the path ? The velocity you mention is based on the average cross section but there are infinite possibilities depending on runner entry CSA, min CSA, throat CSA etc ?

David Vizard wrote: ↑Sun Oct 21, 2018 7:52 pm

Erland Cox wrote: ↑Sun Oct 21, 2018 2:44 pm I guess that the velocity is the velocity at 28" on a flowbench and not on a running engine?

Erland

As it happens these numbers are very similar. The velocity shown in the graph is that on the bench at 28 inches. The intake velocity we are shooting for on a 2 valve engine like a BBC (and for that matter a SBC and Ford) as shown on the bench has shown, by yno testing, to be 300 -310 as measured on a 28 inch bench test and then calculated by IOP from the mean CSA and the CFM.

I alluded to the fact that the 240 for peak torque and 300 for peak hp were the standard used back in the 60's. I will post some velocity numbers to show how different design heads vary from those two numbers.
DV

Do you mean that the numbers on the bench at 28 CFM are similar to those on arunning engine at the rpm of max hp?
I always thought that running engine velocities where more than double the flow bench velocities.

Erland

Erland Cox wrote: ↑Mon Oct 22, 2018 4:22 pm

David Vizard wrote: ↑Sun Oct 21, 2018 7:52 pm

Erland Cox wrote: ↑Sun Oct 21, 2018 2:44 pm I guess that the velocity is the velocity at 28" on a flowbench and not on a running engine?

Erland

As it happens these numbers are very similar. The velocity shown in the graph is that on the bench at 28 inches. The intake velocity we are shooting for on a 2 valve engine like a BBC (and for that matter a SBC and Ford) as shown on the bench has shown, by yno testing, to be 300 -310 as measured on a 28 inch bench test and then calculated by IOP from the mean CSA and the CFM.

I alluded to the fact that the 240 for peak torque and 300 for peak hp were the standard used back in the 60's. I will post some velocity numbers to show how different design heads vary from those two numbers.
DV

Do you mean that the numbers on the bench at 28 CFM are similar to those on arunning engine at the rpm of max hp?
I always thought that running engine velocities where more than double the flow bench velocities.

Erland

Wouldn't it be possible to put a probe mid runner connected to a digital manometer to compare a running engine to FB velocitys?

Erland Cox wrote: ↑Mon Oct 22, 2018 4:22 pm

David Vizard wrote: ↑Sun Oct 21, 2018 7:52 pm

Erland Cox wrote: ↑Sun Oct 21, 2018 2:44 pm I guess that the velocity is the velocity at 28" on a flowbench and not on a running engine?

Erland

As it happens these numbers are very similar. The velocity shown in the graph is that on the bench at 28 inches. The intake velocity we are shooting for on a 2 valve engine like a BBC (and for that matter a SBC and Ford) as shown on the bench has shown, by yno testing, to be 300 -310 as measured on a 28 inch bench test and then calculated by IOP from the mean CSA and the CFM.

I alluded to the fact that the 240 for peak torque and 300 for peak hp were the standard used back in the 60's. I will post some velocity numbers to show how different design heads vary from those two numbers.
DV

Do you mean that the numbers on the bench at 28 CFM are similar to those on arunning engine at the rpm of max hp?
I always thought that running engine velocities where more than double the flow bench velocities.

Erland

Statement in bold ^^^ this is going show an average velocity across the entire valve lift curve, its not going to show peaks in velocity, unless i am mis interpreting something here?

cjperformance wrote: ↑Mon Oct 22, 2018 5:41 pm

Erland Cox wrote: ↑Mon Oct 22, 2018 4:22 pm

David Vizard wrote: ↑Sun Oct 21, 2018 7:52 pm

As it happens these numbers are very similar. The velocity shown in the graph is that on the bench at 28 inches. The intake velocity we are shooting for on a 2 valve engine like a BBC (and for that matter a SBC and Ford) as shown on the bench has shown, by yno testing, to be 300 -310 as measured on a 28 inch bench test and then calculated by IOP from the mean CSA and the CFM.

I alluded to the fact that the 240 for peak torque and 300 for peak hp were the standard used back in the 60's. I will post some velocity numbers to show how different design heads vary from those two numbers.
DV

Do you mean that the numbers on the bench at 28 CFM are similar to those on arunning engine at the rpm of max hp?
I always thought that running engine velocities where more than double the flow bench velocities.

Erland

Statement in bold ^^^ this is going show an average velocity across the entire valve lift curve, its not going to show peaks in velocity, unless i am mis interpreting something here?

Check the graph shown about 10 posts back.
DV

^^^^ Thankyou !

To calculate the mean velocity trough an engine I use the formula below.
It is in metric and gives m/sec.
( Displacement if 1 cyl in cc:s X Volumetric efficiency X Rpm) / ( 3000 X port area in sq mm:s ) = Mean port velocity in m/ sec.
With 400 CFM a 598 engine could give max hp at 6800, I am using an engine from a Pipemax file.

( 1225 X 104,7 X 6800) / (3000 X 2245 ) = 129, 5 m/ sec X 3,28 = 424,875 ft/sec.

And that is mean velocity for one complete revolution.
Isn´t peak velocity mean velocity X Pi /2? That is 667,4 fps
Isn´t power limiting peak velocity close to 625 fps?
Pipemax calculates these areas for the engine I used:

311 FPS CSA= 3,105 Highest useable Port velocity ( possible HP loss )
300 FPS CSA= 3,219 Smallest Port CSA ( Hi Velocity FPS • good TQ and HP )
285 FPS CSA= 3,388 Smallest Port CSA ( very good TQ and HP combination )
260 FPS CSA= 3,714 Recommended average Intake Port CSA (very good TQ and HP)
250 FPS CSA= 3,863 Largest recommended average Intake Port CSA ( good HP )
240 FPS CSA= 4,024 Largest recommended average Intake Port CSA (less Peak TQ)
235 FPS CSA= 4,109 Largest recommended Intake Port Gasket Entry area CSA

I usually calculate area and flow needed for the hp I want my engines to make and port from that.
I used to use 120 m/sec at peak power rpm, 394 fps mean and 618 fps max velocity.
Now I use Pipemax.
But there is a big difference between running engine mean velocity and calculated velocity at 28".
276 fps at 28" versus 425 fps in the running engine.

Erland

You stated in an earlier post the efficacy of the radius seat began at 0.600 inch lifts.Is this from observation alone or is there a realtionship between valve size and the required effective lift.

woodaca3680 wrote: ↑Thu Oct 25, 2018 3:28 pm You stated in an earlier post the efficacy of the radius seat began at 0.600 inch lifts.Is this from observation alone or is there a realtionship between valve size and the required effective lift.

Can you refer me back to the post that makes that statement please. If that is what it says then it's a typo.
DV