TIMSPEED

Anyone know the equation? I'm trying to find out how much air a 2", 2¼", and 2½" pipe can flow. (max) Anyone know this equation?

65Z01

Though I can't give that formula you can get the relative flows by taking the ratio of the square of the radii.

For example: (2.25/2)**2=(1.125)**2=1.265, so the 2 1/4" pipe flows 25% better than the 2" pipe.

Of this is for a relatively well behaved fluid flow.

TIMSPEED

Thanks Jim, that formula DOES help me, because that means if I can find somewhere, what a specific pipe size flows, I can go foreward/backward and find what I need to. :cheers:

mikey

You're asking a question that has answers which are relative to other pieces of information you have left out..

You want an answer of volume per time which requires length, width and height.. 3 dimensions and a time unit of measurement. If you know a diameter, then you have two dimensions.. You're missing the third dimension and time..

If you know air velocity, that would give you a length per minute which you could multiply by your cross section and get your volume per minute..

I doubt I'm helping you much, but you might need to re-word your question more specifically..

Wadoka

He just needs area and velocity for a rough number. Pi (d^2)/2 is square feet, then velocity is in ft/s or ft/min, so cfm = A*V.

Now we have to calculate the volume of air your engine can pump per unit time.

Let's say we have a 4-stroke engine with 350 cid. Now things get hairy. We have to estimate the volumetric efficiency of the engine. A lot of people use 70% Ve for IC engines.

So, every two revs, your engine is pumping out (350)(0.7) = 245 cubic inches of combustion products. 1728 cubic inches is one cubic foot, so we're pumping about 0.142 cubic feet every two revs.

What's your max RPM? I'll pick 6200, that's about where an LT1 auto goes to fuel cutoff. You can substitute. 6200 revs/minute times 0.142 ft3/2 revs yields 440 cfm. A 383 at 0.7 Ve would suck down 687 cfm. Turbo and Supercharging boost the Ve into new realms.

You can see where people can overdo things like TB size and carb size. Their 'mouths' can be larger than their 'stomach'. But them we spend a *lot* of time and effort improving the Ve of our engines, with porting and polishing, headers, exhaust, all the goodies.

So anyway, a factory 350 with 0.7 Ve can only use 440 cfm. You can calculate the area you've got (dual-throat stock 48mm TB, f'r example...)

From here you can calculate a velocity needed for a certain size opening to get your cfm...

Given a stock 48mm TB...

48/25.4 = 1.890 inches = 0.157 feet. Area is Pi (0.158/2)^2 times two...yielding 0.039 square feet. We want to cram 440 Ft3/min into the engine... Units are in ft/min...Vdot/A...440 ft ft ft/min / 0.039 ft ft...

11,282 ft/min or 188 ft/s. Hm, that air has to move as fast as a pellet from an air rifle at 6200 rpm on a stock 350.

I'm stuck here guys. I can't figure out how to estimate a reasonable, achievable intake velocity. I can say that I need 188 ft/s with a 48mm TB, but how do I determine if I can achieve 188 ft/s flow?

Anyway, bigger TB throat = slower intake air velocity...

CentralCoaster

It's not simple really. Mostly empirical data, and that why they use flow benches. The VE of your engine changes everytime you mess with the flow paths or anything else.

Also, there's really no such thing as "max flow" since you can always increase flow by increasing pressure at the inlet.

CentralCoaster

For this to be true, you're assuming that velocity is the same in all three pipes, which it wouldn't be. The velocity would be much lower in the larger pipe.

PurpleC4

We have a program at work called "Einstein"....... it will calculate any volumetric gas flow, be it air, nitrogen, helium, oxygen, hydrogen, CO2....

Actually, if you are striving for a specific, I can probably run the calcs for you. You do realize though that you must provide a list of gas type, temps, pressure, pipe diameter, etc. :eek:

~ Purp

Wadoka

I found this link.

If you have all of the values and some paper, this will tell you when you need to upsize.

The assumptions made are typical engineering fudge factors (k parameters). The formulas are for naturally-aspirated engines only! Atmospheric pressure (Pamb) and temperature (Tamb), R (ideal gas constant with proper units) C(sub)D is another fudge factor called the "Throttle discharge coefficient"... get out your HP calculators and begin!


http://www.simcar.com/literature/sae.../throtflow.htm http://www.simcar.com/literature/sae...menclature.htm



Nice site... I'm going to order the textbook, item 19.

~Matt


[Modified by Matt Black, 8:57 PM 5/10/2003]

CorvetteZ51Racer

Not true if the upstream pressure is ambient...once you reach a sonic nozzle condition, the ONLY way to increase air flow is to increase the area, as you cannot physically move air through an orifice faster than the speed of sound with an upstream pressure of 1 atmosphere.

Someone mentioned using a 70% VE for a street motor....WHY? A healthy motor is running at or near 100% VE at torque peak, though they are much lower away from torque peak. I know you weren't talking about race motors, but very healthy street/strip and race motors are making over 100% VE. If you're a race engine builder and you're not making 105% VE or better, you're likely fired. If you're building in Winston Cup, you've got to go higher. F1 engines are running 140+% VE.

Anyway, to answer the original question, IDEAL pipe flow is area*velocity. If you're trying to calculate intake tract airflows, velocities of ~300 ft/min are typically used. Accepted values range from 280-320 ft/min, so I personally use 300 ft/min as a good average. If you're trying to figure out exhaust head size...don't use this method. On a healthy motor, exhaust velocities at the exhaust port throat are supersonic as soon as the valve comes off the seat (that's the healthy crack you hear when a Top Fuel dragster idles...it's a series of sonic booms every times the exhaust valve opens...and remember the upstream pressure is greater than 1 atmosphere so supersonic velocities are possible).

CentralCoaster

Ok. But is this a consideration on cars? Or do our flows stay well below this flow?

Stock VE on a N/A L98 wouldn't be near 100% would it? even at peak torque? I didn't realize the intake was that efficient. Equal length headers would help raise it I suppose, but at some point too much scavenging with overlap hurts you.

Wadoka

My old IC Engines professor, my ICE textbook, and my "Corvette Fuel Injection" book all use the 70% VE figure for a production engine.

Here we go...Charles Probst, page 25 of "Corvette Fuel Injection" "Most engines run only 70-80% volumetric efficiency, even at Wide Open Throttle, due to friction and flow restrictions."

The maximum VE I've seen for a NA engine is 85%. In my holistic consideration, NA engines can't have greater than 100% VE.

We may be comparing apples to oranges - there's a big difference in the value you get for VE depending on if you want to figure VE for just the intake side, or the whole system.

Want to see a good VE argument? Go look at this engineering forum. Same kinds of questions. I haven't posted yet, because the guys aren't defining their conditions well enough. Nice little tussle going on.

http://www.eng-tips.com/gviewthread....d/71/qid/46583




[Modified by Matt Black, 10:27 PM 5/10/2003]

CorvetteZ51Racer

Yes it is, because when you're trying to size the area of the runners or TB on an intake system for a naturally aspirated engine, the upstream pressure is ambient. The 300 ft/min I mentioned before takes into account the changes in air velocity due to the 1-intake stroke per engine cycle as well as the cam. If you notice, 300 ft/min is "roughly" 1/2 the speed of sound at sea level (average velocity...lowest is theoretically 0, highest is mach 1).

On a stock L98, no it's not near 100%, but more like 85%-90%. My L98 with the heads I ported and 1-5/8" longtubes and an otherwise stock setup runs at roughly 100% VE at peak torque (which, since I haven't mentioned it before, is the point where the engine is running at it's maximum VE) I haven't had the car back on the dyno yet, but the stock ZF clutches seem to be starting to get iffy on cars making ~380+ or so RWTQ, and the stock ZF clutch won't hold my motor with the headwork I did.

Out of curiosity, which ICE book do you use? My texts are Internal Combustion Engine Fundamentals by Heywood and Advanced Engine Technology by Heisler....I don't remember what Heywood or Heisler say about it, but most of the high performance production engines we've looked at have been in the 80-90% VE range. Some higher, some lower.

Hmmm.......ANY decent strip/track N/A motor makes at least 100% VE. It is pretty easy to make at least 100% VE on an N/A motor, but using a combination of exhaust scavenging to evacuate the cylinder which allows more room for fresh air, intake manifold tuning which will produce a pressure wave hitting the intake valve just as the valve is closing, thus putting the cylinder at slightly above atmospheric pressures (i.e. "Tuned Port"), and well-matched cams, heads, exhausts, and intakes in order to minimize restrictions while maximizing global system potential. The reason F1 engines and Winston Cup engines run at such high VEs is because of an event called short-circuiting. During valve overlap with bigger cams and higher RPMS, some of the air charge enters the cylinder and shoots out the exhaust valve before it can close. This is a bad thing because it's wasted energy in the air and fuel that is never burned, but a necessary evil in order to sufficiently evacuate the cylinder on the exhaust side and generate enough momentum in the intake side to pack the cylinder with intake manifold tuning.

I don't know how you would figure out the VE for the just the intake side....it wouldn't make any sense to try to do that without looking at the exhaust side as well.

Wadoka

Can't remember, it was eleven years ago now. It was dark gray and darn heavy. I sold it back to the bookstore.

With your discourse on component tuning...the engine has to be able to move more volume than it displaces, and it does this by allowing some extra flow-through (momentum flow?) during overlap? I can see how this works, and the waste that has to happen.

What is causing the 'free' pressure wave in the intake runners?

I'm too tired...

*********************

Prof. Alfred Gardiner "An internal combustion engine is a machine with insufficient power."

CorvetteZ51Racer

It was probably the Heywood book...it's been around for a long time, and it's a tough read.

Ok, tech answer...LONG

You have air moving down the runner towards the cylinder with some momentum. Ideally, from a resonance standpoint, you'd like the intake valve to close right as the air velocity at the valve reaches zero. At this point, there is no more energy that can be transferred to the cylinder on that cycle (since the air stopped moving), and you haven't robbed any energy from the cylinder (because the airflow has reversed up the runner yet). What this means, however, is that you no have the same air pressure on the manifold side of the valve as you do the cylinder side of the valve. At this instant, the pressure at the valve is higher than in the manifold plenum, so the air in the runner begins to flow back to the plenum trying to equalize the pressure....you end up with a yo-yo effect in each runner of the manifold.

As the air moves back up the runner, a wave front, or pressure wave is created. When it reaches the plenum, it does one of two things: 1) reflects off the plenum and goes back down the runner, or 2) skips across the plenum and down another runner. That pressure wave is going down some runner, and if the length of that runner is optimized, that pressure wave will hit the valve at either intake valve opening (to help improve exhaust scavenging...you're increasing the pressure differential between the intake and exhaust side of the cylinder, thus increasing flow out of the exhaust), or at intake valve closing (trying to pack that last bit of energy into the cylinder, increasing the pressure at which the cylinder and runner equalize). We've now completed the cycle and it begins again.

If you notice, if the engine is held at a constant RPM, you will have a complete wave forming in the intake tract. However, as you can also see, if the runner length is just slightly off of the corresponding engine speed by a few hundred RPM, you no longer have this nice simple wave with pressure fronts hitting the valves at the optimum time.

Instead, you have air still moving towards the valve when it closes if the RPM is higher than the tuned speed, causing air to bounce off of the valve and back up the runner, creating stronger waves which are no longer in harmony with one another. You now have waves interfering with other waves...you have created the scientific definition of chaos.

If the engine is running too slow for the tuned length of the intake and the valve timing, the air will reverse flow and begin heading back up the intake before the valve can close, thus robbing engine power and typically causing the engine to run rich (as the proportionate amount of fuel will not turn back up the intake with the air).

Also, if the engine is running too slow for the valve overlap, there is more pressure inside the cylinder when the intake valve opens than the intake manifold, thus blowing combustion gases up into the intake, creating a wonderful EGR system. You know that nice, rough, slow, inconsistent lope from a heavily cammed engine idling? The "nice lope of a healthy cam" is really a motor choking to death on exhaust gases. If you speed up the idle a bit, you'll notice it still has a hard hitting exhaust note, but it's smooths out and no longer is inconsistent.

Rynda

Timspeed,
Straight of the "Pocket Ref, by Thomas J. Glover"
1) CFM=60VA where V=air velocity in feet per second
A=pipe cross section in square feet

2) V= Square Root of (25000DP/L) where V=air velocity in feet per second
D=Pipe iside diameter in inches
L=length in feet
P=Pressure loss due to air friction
in ounces/square inch
Approximate values of P:
Pide diameter in inches, 10 feet long
Velocity in ft/sec 1 2 4
1 0.004 0.0002 0.0001
2 0.0016 0.0008 0.0004
5 0.0100 0.005 0.0025
10 0.04 0.02 0.01
15 0.09 0.45 0.0225
20 0.16 0.08 0.04
25 0.25 0.125 0.0625
30 0.36 0.18 0.09

In other words it is not a straight forward calculation, and as previously stated in some of the other replies, the equations are based on imperical data. There are probably other equations out there that may appear slighly different.
The problem is that air is a compressible liquid, and these equations are based on steady state flow. Comparing what happens in your engine to the max flow in a straight pipe under the conditions that these type of equations are developed would sharpen your math skills, but that is about all.
For instance; with a highly tuned intake and exhaust system, you can actually overfill the cylinders at the designed RPM, these equations would never tell you that.
If we could actually calculate the air flow in an engine, we could discard flow benches.

Rynda

well, that's the last time I'll put a table in a reply :crazy:

a)The left column of numbers are the Velocity in ft/sec1, 2,5,10,15,25,30
b)The 1,2,4 are the column headers for the pipe diameter in inches, for the 10 ft pipe.
c)The numbers with decimal points are the table entries

CFI-EFI

Just a quick reminder. Simply, VE (volumetric efficiency) is the amount of air admitted to a cylinder in relation to the capacity of that cylinder.
There is no figuring of intake side, or exhaust side, VEs. Intake and/or exhaust modifications can affect the VE, but the VE refers to the conditions in the cylinder, only. If you have 48 cid of air, at the ambient temp, humidity, and barometer in a 50 cid cylinder, you have achieved a VE of 96%.
Proper intake and exhaust runner tuning CAN produce VEs over 100% in a NA engine.

RACE ON!!!
1984 Crossfire
Member NHRA

CorvetteZ51Racer

Sorry to be technical stickler, but it is important to recognize that you will always have 50 CI of gases in a 50 CID cylinder (gases will always fill teh complete volume). The 48 cid of air CFI refers to is 48 cid of air *at standard temperature and pressure*, or the equivalent volume. Now, the composition of those gases aren't necessarily pure air/fuel charge (you'll have *some* EGR in it no matter what you do), and they are never at standard temperature or pressure inside the cylinder when the intake valve closes. This is why there is no equation for mathematically computing the VE of a motor based simply on head flow, intake and exhaust flow, and cam specs...

Wadoka

I see it. It also comes to me that this is must be most useful in high-performance racing engines that are optimized for a pretty narrow rpm range.

I seem to remember that Saab or Volvo was experimenting with an engine that could vary the length of its intake runners...