[Z Code 400]

Question,

How much airflow in CFM can a 24" section of 2.50" exhaust tubing support @ 28 in/H20???

Thanks in advance...Robert

[Jack Gifford]

A couple of flow numbers, courtesy of Tom Hand in his VOE article Dec'06 Smoke Signals, all @ 20.3 in.water-

Straight, 2.25" diameter, 25" long pipe - 350 cfm
Stock Gardner GTO muffler- 160 cfm
VOE II muffler (quiet) - 140 cfm
VOE II muffler (loud) - 207 cfm

[goatless]

According to the Pypes site, 2.5" pipe flows 485 CFM.

[Z Code 400]

Thanks!!!

Now, has anyone tested a running engine to see how many cfm are streaming out of a 2.50" pipe at a given rpm???...Robert

[Murf]

Z Code, I haven't tested it myself. I do recall that when Jim Hand tested mufflers years ago, he tested them as single exhaust on his 455. He did this to make the difference more noticeable. This always gave me the impression that most good dual systems would flow plenty for a smaller street engine. I know it's not a very helpful post but it's something that suprised me, the single exhaust thing.
That info used to be somewhere on the net. Maybe someone can post a link. It was in HPP yrs. ago.

Later, Murf

Here it is, http://www.pontiacstreetperformance....p/exhaust.html

[Z Code 400]

I used to have the data, but I cannot seem to locate it now. We still have tons of stuff in storage, but, basically, we found the flow testing of mufflers was a moot point when we actually tested the CFM of a single 2.50" pipe on a running engine, under load, on the Dyno.

As I recall, the 440 CID engine produced about 800 cfm out one side (220 CID) @ around 6000 rpm.

Yet, when fitted with a 2.50" Hemi muffler, the exhaust system showed 0 psi of backpressure in front of the muffler. If you look at the testing done by Jim Hand, even an open 2.50" pipe can't flow that much cfm.

Interesting, don't you think??? Same can be said for head flow...what you are really measuring is head vacuum...not flow...Robert

[Jack Gifford]

Yes, it's interesting. I suspect that the drastic reduction in volume of the cooling exhaust gases is a very important factor.

[Z Code 400]

I dug up these formulas and text and wanted to post them to aid in the discussion of exhaust flow values on a running engine.

The amount of air that passes through an engine will be equal to the engine displacement times the RPM divided by 2. For an engine of 3 liter displacement going at 3000 RPM, the amount of air pumped for minute will be 4500 liters.

Minus the added fuel, the mass of the exhaust gas will be the same as the mass of the intake gas. From the 'ideal gas law' we know that the increase in volume of the exhaust gas will be proportional to the increase in absolute temperature. If we assume an intake temperature of 80° F, and an exhaust temperature of 1800° F, the absolute temperature will be 540 and 2260 deg Rankine, respectively. The volume will therefore increase to 2260/540, or 4.185.

The combustion formula is C8H18 + 12.5 O2 ==> 8 CO2 + 9 H20. For each 114 grams of C8H18, there will be 12.5 moles of oxygen consumed, producing 8 moles of CO2 and 9 moles of H2O. For gas volume purposes, since equal moles of gas produce equal volume, the volume of exhaust gas replacing oxygen will be equal to 17/12.5 = 1.36.

The volume percentage of oxygen in air is roughly 21%. This volume will be removed, and replaced by exhaust gas with a "volume" of (21*1.36) = 28.56%. The resulting post combustion volume is (79% + 28.56% = 107.56%) of the pre- combustion volume -- assuming no temperature increase.

Combining the increase in volume from combustion reactions and thermal expansion, an engine with a 3 liter displacement running at 3000 rpm with the throttle wide open will have an exhaust volume (at 1800 deg F) of 665.4*1.0756 ~~ 715 cubic feet per minute. For the throttle one-third open, the exhaust flow will be 238.6 cfm.

I was somewhat confused by this, but decided to put some of the mathematical formulas to the test.

Using my 412 CID engine, I converted the displacement to litres and arrived at the sum of approximately 6.7 liters. (ldp)

6.7 x 6000 (rpm) /2 = 20,100 lpm x 4.185 = 84,118 /28.3 = 2,972 cfm x 1.0756 = 3,197 cfm

If I did this correctly, the 412 CID engine running at 6000 rpm will produce 3,197 cfm of exhaust flow @ WOT @ 6000 rpm. To break this down further, each cylinder would produce roughly 400 cfm into each of the 1-3/4" header tubes.

Knowing that a section of 2.50" exhaust tubing will flow only 485 cfm of vacuum, the practice of relying on flowbench data seems to disregard a lot of variables. Even mufflers that claim to flow "100% of an open pipe" do not even come close to providing adequate flow.

I hope you find this information interesting...Robert


P.S. Please feel free to post corrections to the formula if you locate an error.

[TCSGTO]

Wow,I think I just flashbacked to my Physics classes. The formulas look good and you have really broken everything down. The main reason 2.5" pipes and mufflers seem to be able to support higher HP is by the time the exhaust reaches the mufflers it is probably "only" a few hundred degrees due to temp loss to the atmosphere. I've always had good luck with 3" headpipes into 2.5" mufflers.

Jim Hands' famous exhaust articles documented very little power loss with 2.25" tailpipes behind a larger system/mufflers when compared to 2.5". His reasoning was the cooler/condensed gasses at the back of the car could pass through a smaller pipe unrestricted. Keeping the pipes further away from the engine smaller may help keep velocity up too on milder apps?

[TCSGTO]

Just to add something else in the Dynomax catalog they rate mufflers flow @ 20.3" H20 and give corresponding HP levels they supposedly can support without restriction.

As an exhample the 17288 2.5" 20" case Ultra Flow is rated at 955 CFM and can support 434HP with one muffler or 868HP with a dual system.

The 17748 20" case Super Turbo is rated at 425CFM and can support 193HP single and 386HP in a dual system.

So according to Dynomax doubling flow will more than double the HP capacity.

[Z Code 400]

TCSGTO,

That sheds some new light on the subject. As posted above, the initial flow volume was calculated at 1800°F. There is a corresponding formula for flow variation based on temperature loss, but I couldn't find it.

What would the variation be between flow rated at 20.3 in/H20 and 28 in/H20??? I am sure it wouldn't be significant, but I am very curious about stated flow values vs. the engine's ability to produce pressurized flow in cfm.

Thanks for the input. That's good data to have in the thread...Robert

[Murf]

I guess I'm a dumb butt but what are you wanting to figure out? If a single exhaust on a 455 doesn't slow it down by much, how much more flow will a 412/413/.060 over 400 need?
I think some real testing with a running motor would be most helpful.

Later, Murf

[Z Code 400]

Murf,

Nobody here is a dumb butt....except maybe for me.....:-) I just used the 412 as an example. You could substitute any engine CID configuration as an evaluation standard.

As of yet, I haven't figured out a way to read exhaust CFM with a running engine on the dyno. I know my buddy's shop doesn't have anything like that, so I am restricted to mathematical equations.

However, I do know, through thermal testing, the EGT behind the muffler and extending back towards the tailpipe is between 600° and 700° F respectively. Even taking into account a reduction in thermal expansion of the exhaust gas due to reduced EGT near the tailpipes, the CFM produced by the engine still exceeds the 2.50" pipe's ability to flow.

The fact that Magnaflow rates it mufflers on a different scale is also misleading. For example, if a 2.50" section of straight pipe, without perforations, can only flow 485 cfm, the 2.50" core Magnaflow #17288 20" Muffler cannot magically flow 955 cfm.

The point I am trying to illustrate here is the dramatic variables and inaccuracies with respect to flowbench testing. We know that a given exhaust port in a 6X-4 head will permit about roughly 200 cfm of vacuum to flow through it, but when the engine is running, the port will have to support over 400 cfm of gas flow and the single 2.50" headpipe, following the 3.00" collector, must support 1400 to 1600 cfm per cylinder bank.

I refer back to our SS/a 427 Ford Fairlaine that we used to drive on the street from time to time. We had stepped-tube race headers on it with 2.50" exhaust all the way out the back through glasspack mufflers. When racing, we ran open exhaust, but when driving on the street we 'corked it up' as a matter of course.

We installed brass fitting at the end of the collectors and my partner took vacuum gauge readings at the 7500-7800 rpm upshifts. We saw some fluctuation on the vacuum/pressure gauge up through the gears (often going into a slight vaccum) but the system never went into pressure that we could 'see' on that measuring device.

In theory, the car shouldn't have ran that hard and we should have seen relatively high pressures in the exhaust system, but we did not. It is a matematical fact that a 427 CID engine turning 7500 rpm will produce about 2,087 cfm per cylinder bank and a 2.50" pipe is incapable of flowing more than 485 cfm @ 28 in/H20.

It is these variables, with respect to flow testing ,that continue to make me so skeptical of their accuracy and influence...Robert

[Jack Gifford]

Originally Posted by Z Code 400

TCSGTO, That sheds some new light on the subject...

With your analytical approach- that would be if post#9 is "newer" than post#7.

[Z Code 400]

Sorry Jack,

I missed that post in the shuffle. But you are correct. Sorry about that!!!...Robert

P.S. Here is an interesting article on flow bench testing:

http://www.rehermorrison.com/techTalk/16.htm

[Z Code 400]

I have to admit, I get a kick out of advertising. Today, I was browsing through some catalogs and saw an article titled, 'Magnaflow out-flows Flowmaster.' In the article, Magnaflow claims one of their 2.50" inch mufflers flows a whopping 900+ cfm.

It doesn't matter whose exhaust products you like...Pypes, Magnaflow, Flowmaster or whoever....It's sad when a company will resort to such exaggerations to sell a product when we know from validated testing a 2.50" straight pipe can only flow about 500 cfm...Robert

[lust4speed]

Will using vacuum to gauge exhaust flow give valid results? It works good when you have to suck in mixture, but I'm not so certain that the results are acceptable on the pressure side. So why doesn't velocity make up in part for tubing size? You have an occurance that is forcing the exhaust gasses out and down the header at an extreme rate of speed. As long as the exhaust tubing can maintain the flow at a high speed, shouldn't the flow be the same? In my simplistic terms, let's say that the 2.5" pipe is capable of exhausting the engine sucessfully at 6,000 RPM by maintaining very, very fast flow. The 3" pipe won't do any better, but can exhaust the engine at much less flow speed of the 2.5" pipe so we won't see any performance improvement. We would have to reduce the pipe size even more or increase engine size before a change might be noted. I'm sort of thinking of this like two garden hoses of different size where both can carry the same amount of water as long as the velocity of water in the smaller hose is greater.

[Z Code 400]

Mick,

You bring up a very valid point and it is one reason why I find flow testing so subjective...Robert

[Z Code 400]

Just for clarification….I am not contesting Tom and Jim Hand’s published flow data on mufflers. I do, however, find a major discrepancy with Dynomax’s claim that their 2.50” side-inlet muffler can flow 2,200 cfm @ 20.3 in/H20.

See attached flyer…Robert

[Tom Hand]

Robert,
That is kind of you to mention us but unfortunately the book we wrote does not provide flow numbers for mufflers or pipes. I am sorry if you misunderstood either Dad or I inferred any specific air flow values. We know that is not the best way to compare mufflers and that is why we spent so much time and so much money and expended so much effort testing them on the drag strip and to a lesser extent, on a loaded engine on a dynomometer. What I did say on page 107 of Dad's book in Chapter 8 Exhaust System is "air was blown through a 24-inch section of 2.5-inch pipe and that air flow number was used as the baseline 100% value. The amount the muffler flowed as compared to this pipe's flow is given as a percentage."

Since only a few manufacturers provide the pressure they use when they force air through mufflers, and since all benches are different, of course the numbers should never be compared. That is why we go to such great lengths to create baselines and do back-to-back tests keeping variables constant between tests.

Hope this helps clarify things.
Tom Hand