Cylinder head CFM to HP?

[bill jones]

-Superflow says somehing lke 2.2hp per cfm on an 8 cylinder engine so if you had 300 CFM you have 660hp's worth of air, and cut the numbers in half for a 4 cylinder.

[rfl28]

i've used this one HP = cfm x .2575 X # of cylinders....horsepower potential.... it's gets you close... but DARIN MORGAN says it's not as accurate as you get up in the airflow and power numbers.

[devilbrad]

Are the airflow numbers used in the formulas the max flow, or do you use say the .500 lift number?

[bill jones]

-as far as METHANOL is concerned I'd quess that 8% is a realistic number to expect when you know how to tune the fuel system real good.
-I have yet to see a modern engine run slower using methanol when the tuneup was decent between methanol and gas.
-The only engine that I couldn't get to do tricks with methanol was a flathead at Bonneville and it ran exactly the same as it did with gas (and I have learned the reason why).
-Methanol engines should be able to make a little more power than that 8%---maybe around 12% total power over gas----if you have a little more exhaust duration on the cam, larger exhaust tubes, and if you squeeze the compression ratio.
-on our 122ci pinto we run at Bonneville the difference was from about 147mph on gas to 153 on methanol.
-We did not change anything in the engine other than a few more degrees of timing, and larger headers would not help the engine on gas at all but was worth a couple MPH on methanol.
-This was with a stack fuel injection.
-Carbureted engines on methanol is a lot of work to get right.
--------------------------------------------
-And the definition of what you call gas can also mess up the power spread between gas and methanol if you are really sharp on what you run for "RACE" gas.
-Between various blends of the racing gasolines I've seen serious speed differences at Bonneville.
--------------------------------------------------
-as far as the lift versus the cfm and what do you choose for the calculations, I don't think the numbers matter too much as long as you are close to the honest max lift and close to the max flow of YOUR engine.
-One issue though is most people only have head flow and in real life you have the manifold and usually a carburetor or a throttle body to contend with, so if you don't flow the entire system you need to have enough empirical data to know what to expect for flow losses thru the intake.
--------------------------------------------------
-From what I have seen on single 4bbl cast aluminum intake manifold with a carburetor installed you lose about 1% flow for each .100" valve lift.
so that's about a 5% loss in flow at .500" lift.
-If you have a 180 manifold then it's worse than that something like 1.2% loss(or more) per each .100" valve lift.
----------------------------------------------
-If you change from a hydraulic cam to a roller cam that cuts some of the losses, like if you run open headers vs a full long exhaust system.
-------------------------------------------
-So the calculations are to only be taken as basic idea of what to expect---not some exact to the tenth of HP on a chrome plated California dyno.

[devilbrad]

Thank you!

[kdrolt]

Vizard uses

cfm*number

where cfm is the cfm value for an intake port and number is a number close to 2.0. If the engine is a 305, then the number is 1.74 (i.e. 2*305/350); if the engine is a 383 then the number is 2.19 (i.e. 2*383/350).

It's a rule-of-thumb, derived from an assumed conversion of fuel energy to fwhp, and an assumed air:fuel ratio, and an assumed fuel (gasoline), and finally that assumes you are using an optimal (racing) cam that suited for the flow of the heads. If you use a smaller cam, then you won't achieve the hp output that the airflow through the (intake) says you could reach. That's why I said optimal camshaft.

The only problem that I have with it (Vizard's rule-of-thumb) is that it doesn't account for exhaust flow. So what's implied by his formula is some ratio (an unknown ratio) of i/e, and if your head flows worse than Vizard's i/e value, then the formula will over-estimate the engine fhwp, and if your head has a better i/e then the formula under-estimates the fwhp.

IOW if the exhaust flow is poor, then some of the exhaust won't escape the cylinder by itself so the engine has to push it out and/or compress the ehxaust still in the cylinder. Oushing the exhaust out costs power and remnant exhaust product reduces the effective cfm of the next intake event. Power drops.

A better rule-of-thumb would be to include both the effects of air intake cfm, exhaust cfm, cam duration, and cam lift. But then it stop becoming an easy-to-use rule-of-thumb, and more resembles a formula from Heywood or C.F. Taylor. So it's probably better to leave it alone, as a simple but optimum formula, and remember that it represents close to a ceiling on the power you could achieve for a given engine IF everything were optimally chosen.

Edit: HP = cfm x .2575 X 8 = cfm * 2.06, so the equations are the same except that Vizard assumed 8 cylinders.

[Mike Rogers]

Here are some test i did on busch north heads in 2000

Chevy 18 deg high port.Castings #363 cnc ported by weld tech
2.150 int. and 1.625 ex. on a 4.155 bore
Super flow 300 at 28"

The first test is with no intake

.100 67.8
.200 143.1
.300 225.3
.400 284.5
.500 329.8
.600 344.9
.700 336.5
The second test is with GM intake 18 deg high port

.100 63.5
.200 135.0
.300 206.4
.400 255.4
.500 287.6
.600 301.0
.700 298.0

The third with wilson intake 18 deg high port edelbrock 2955

.100 63.5
.200 135.0
.300 207.9
.400 260.1
.500 289.9
.600 305.7
.700 305.7

When i first dynoed the engine it made 500 hp @6800
When i put it back on it made 537 hp@ 7200

[Krooser]

Any new thoughts on this old thread?

I am finally planning to assemble my rebuilt/repaired 388 SBM engine for my dirt car.

My question is these formulas dont take into account fuel type, camshaft specs, compression, etc.

I have a bit more compression than before 14-1 vs. 13-1. Two FT cam choices from Mike Jones and Howards. And I have a bit more room to square off the intake ports (W2) for more flow (maybe 315-320 cfm).

I hope to get to 650hp... I get 608 to 616 on these calculators. I know HP is not the end all in dirt track racing but I would like to be competitive in the slow heat races!

A half day of dyno time is in the works.

Thoughts?

[cgarb]

I think that is an aggressive goal. If you make 650hp with a flat tappet cam and 388 inches you have done your homework. Not saying it's not possible or anything because I'm sure it has been done before. My 421 Chevrolet made 651hp at 7000 peak. My heads peak flow was at .800" and was 324cfm if I remember correctly. I have a solid roller and a 950 Methanol carb. Looking back I could make a few changes that would net me some more power also. I hope once I can afford an aftermarket block I can make a piston change and a cam change and get closer to 700hp.

[Krooser]

This engine dyno'd at 575 back in the mid 90's when it was first built.

We all see numbers tossed around over the years....some honest some tweaked a bit. The GM spec engines I see a lot of are 362's with a Brodix head and are said to be mid-600 or so on alky.

I have never used one of these online calculators before... just curious on how one can calculate HP without fuel, carb and other inputs.

[novafornow]

just remember, these are POTENTIAL HORSEPOWER formulas. Just thought I would throw that out.

[PRH]

I used to do some head and dyno work for an engine shop that had a few customers they built motors for some dirt cars.
358”, flat top piston(but no CR rule), Dart Iron heads, roller cams.
Several of the bigger name shops claimed solidly over 600hp from their builds.
Nothing I tested was ever close to that(no matter who built it)........ yet several won track championships against those “600+hp” builds.

It’s just one of those racing dynos/flow benches things.

I think that is an aggressive goal. If you make 650hp with a flat tappet cam and 388 inches you have done your homework.

I agree.

[BlitzA64]

Don't worry about the number, depends on the shop just pay attention to ALL details and work on improving the combination. Also try to use a shop that you can go back to if you need or wish to so it is apples to apples.

I built a 360 SBC that has rule restraints, after 3 times on the dyno, 3 cams, a couple carbs, a myriad of spacers and anything else I could think of I improved baseline by 13hp over as built off the engine stand on a 600+ engine. A lot of work and money to push the limits of a semi restricted combination but I learned a lot, no student loans involved

P.S. The last time on the pump I lost that 13hp with a different cam going just a little further. A very small window where it is happiest, everything else is within 2-3hp "so far." Another flog coming with 1 more idea to try. Glutton for punishment but I really enjoy trying and learning. Good Luck

[PRH]

Edit: HP = cfm x .2575 X 8 = cfm * 2.06, so the equations are the same except that Vizard assumed 8 cylinders.

I remember the formula in the manual that came with my old SF-110 as:
Flow@10” x .43 x (number of cylinders), which on an 8 cylinder engine works out to about 2.057hp/cfm.

In the above formula, it looks like the .2575 replaces the .430 for use with flow @28” instead of 10”.

I will say this........ that formula is actually pretty easy to beat if the combo is right and the numbers “make sense”.

If you have tons of flow, and a very small displacement....... and the predicted peak rpm is in the stratosphere....... you’ll likely fall short.
Same as if the build doesn’t have enough cam or compression to use up the heads.

But if the combo makes sense...... like say, 300cfm heads on a 451” motor with plenty of compression and cam....... you can beat the theoretical ceiling of 617hp without needing to resort to anything exotic.

[digger]

People are better spent focusing effort on a simulation programs rather than a rule of thumb. We are not interested in potential, we are interested in what is achieved. You will learn so much from simulations such that flow never comes into your vocabulary In many ways it is better than your average dyno data as you cant see whats happening at the average dyno as they usually dont record pressures etc