Flow Bench Test Results

[Jim Allen]

The results are in! And with some very interesting results. First off, this material will be going into a future issue of a magazine (to be mentioned at a later date) but I'm giving you guys the results first because I got the idea from you and you deserve the credit. I take the risk of somebody burning me by using the info somewhere else, so that at least oughta bring me up to Double Whopper status!

Also, in the story I will credit you guys for the mod in whatever way you think is appropriate... and is politically correct for the magazine, of course.

WARNING WILL ROBINSON!

Look close at the two closeup pics to see what happens to the downspout tube at high flow rates, so all of you that have done this may want to trim most the the excess back, leaving just enough to center it in the fenderwell hole. This problem occurred at the maximum flow rate of the housing. It may not occur to a stock truck (see 5.4L theoretical airflow below) but it's probably worth the time to fix it.

The tests were performed by two classes at the University of Northwestern Ohio's High Performance Fuels Class, at their School of High Performance Motorsports, on a Superflow SF-600 flow bench. Each class double checked each result from their class, but each class used a slightly different protocols (both standards in the industry) and I have averaged the results.

NOTE: A 5.4L V8, given 90 percent volumetric efficiency, at 5500 rpm, needs 473 cfm of air. Giving it more than it needs at any given time doesn't necessarily offer any more power. Bear that in mind when reading these results.

Condition/ Average Airflow

1- stock housing & filter, as installed in vehicle- 621 cfm
2- same as #1 but with AEM filter panel- 592.36 cfm
3- stock housing & filter but no silencer or snorkel- 656.83 cfm
4- same as #3 but with AEM panel filter- 632.50 cfm
5- same as #1 but with snorkel tube modification- 597.3
6- same as #5 but with rubber downspout adapter removed- 637.5 cfm
7- same as #5 but with AEM panel filter- 569.51 cfm
8- same as #6 but with AEM filter panel- 619.70 cfm
9- same as #3 but with silencer only- 632.50 cfm
10- Brute Force Intake (for 2004 model)- 733.30 cfm
11- AEM filter only- 1077.25 cfm

You can reach your own conclusion over what this all indicates and I will share mine later if asked

[OneTomcat]

So the bottom line is changing adding or removing anything from the stock air induction is nothing but window dressing?
.....well, it does keep several businesses going

[jimmyv13]

Very interesting results. I did wonder about the flexibility of the downspout as I was installing it the other day. Looks as if I may be going back to the set up. Great job, thanks for the research!!!

[Jackpine]

Yipes! Sure don't like that "pucker" effect!

But now I'm a bit confused. Your tests showed that the stock setup provided ALL the airflow, and then some, that the engine needed at 5500 RPM? More than 473 cfm CANNOT add to power? I can accept this, if the 473 max is correct - in fact, there's no way the engine would take in more than this amount, right?

But, it appears from Bill's dyno testing of an 09, that there IS an HP boost with the snorkel removed, suggesting that the engine IS capable of "sucking" in more air than is available in the stock configuration and mixing more fuel with it - resulting in a bigger "bang".

So, I'm left wondering what is causing the inconsistency. You measured airflow. Bill measured HP. I'd like whatever gives me more miles per gallon of gas.

- Jack

[SubiGT]

I think there are some other factors that go into this power equation besides CFM capability. I would be interested in seeing the pressure differential that the each combination made, as well as any velocity measurements. How many seconds did each combination take to stabilize at those flow numbers? I'll take a fast 569cfm over a relatively slower 623cfm. Time based rate of change is the key to torque(and HP). Also, what was the swept area of each assembly? What volume of air did each hold?

The stock setup might flow as much or more air on a flow bench, but how much work does the engine have to do to get that air from the fenderwell to the MAF sensor? Pumping losses and airflow velocity need to be accounted for.

Know what all those chambers on the sides of the air tube do? Reduce flow noise by reducing velocity.

mass flow rate in=mass flow rate out. mass flow rate=densityXareaXvelocity
for a constant density fluid(air), an increase in area will have a corresponding decrease in velocity.

Looking at the results, I think the leanest running (un-tuned) would be the ones flowing significantly more then the stock combo. The larger numbers(733 and 1077) will trick the MAF by cooling the heated wire too fast, and the fuel/timing tables wont be enough to keep you from running lean. Most of the test intakes flowed within 10% of stock CFM, which the ECU should be able to account for, granted that the you have consistent fuel, a decent state of tune, a clean filter, good spark......


I hope my basic understanding is not too far from the truth

[88Racing]

Okay a great "Thank you" to Jim and the classes doing the bench flow testing.

So sorry I can't post a link to the Ford GT forums right now but there is evidence over there that they are getting the most bang for the buck with the OE setup. The only differences that they have found into boosting the air flow is changing out the intake, TB, and porting everything.

Let me tell you one thing you thought we are serious about our trucks. OMG they pick things apart over there to the paranoid level.

Okay now why I mentioned the other forum is this: the basic deduction of why more air through the stock set up?

More surface area of material used for filtration. Take an old stock panel apart and a aftermarket one and see for yourself. Not unless they are using a more porous material.

Found another indepent company whose research concluded that a lot of cai's were worthless unless they were S+B, Banks, and any other company that uses a poly type tube over a metal one. Their reasoning is the metal ones are acting as heat sinks and the hot air build up causes HP reduction and more unstable air.

Whats the reasoning on fast cfms and slow cfms? I thought a cfm was a cfm unless its reading was affected by humidity and or elevation and or barametric pressures?

I also have another theory going about the reduction into the fenderwell but haven't gotton all the evidence on it yet.

Lars

[Jim Allen]

I agree with SubiGt that airflow numbers alone are almost never the end of the story. There are many subtleties, not all of which are clear in our "quick-n-dirty" testing (meaning a few flow bench tests and a coupla dyno tests).

SubiGt:To answer some of his questions, the flowbench had to be really cranked up to keep up with all these systems. The instructor noted that this stock setup was one of the best flowing OE setups he has seen.

The results are corrected to 28" H20, but tested at 20" or 10 ". The one class tested at 20 and the second at 10. exceptions were a few situations where the assembly had too much flow, such as the Brute Force and the stock housing without the snorkel or silencer, which were tested at 15" by the first class. Most people agree that the closer you are to 28" the more accurate the results and in this case, we were able to note the "pucker" in the Downspout Connector at 20", which was not apparent at 10".

While we did not measure the time it took to stabilize the flow, it seemed very fast... just about instantaneous. Velocity can be roughly calculated from the actual CFM, can't it, but I don't know if that's an accurate way to do it.

It would be interesting to run the modified air tube with the silencer chambers removed. I'm not willing to modify mine, but if someone wants to let me borrow one, I'll get it tested. Would be really interesting to find a way to test the system plugged into the fender.

88Racing: I'm not convinced (willing to be, but have not as yet) that there are notable differences in power between metal and poly tubes. In "Forum Poster" type racing, of course, every fractional difference is hotly debated. I've discussed this with several engineers and they say, yeah, there is a tiny difference but the air is moving to fast to pick up much heat from a smooth tube. In the OE, plastic is used more as a weight saver and production expedient.

Also 88Racing: I not clear on what you mean by "more air thru the stock setup" ? Yes the stock Motorcraft air filter seemed to flow a little more than the AEM panel. Was that what you mean? Or was it that the stock setup was very close to the modified system. Note that in #6, we removed the rubber downspout adapter and that's when the modified system exceeded the stock by a little.


JackandJanet: The theoretical airflow number I gave is calculated. The formula is :


CFM= RPM x DISPLACEMENT /3456 x VOLUMETRIC EFFICIENCY

I used 5500 rpm as a reasonable maximum and was generous in giving the fairly free-breathing modular a 90 VE. 85 percent would be more real world but would drop the CFM result. I gave it .90 to factor in some other mods that would increase airflow a little, such as a free flow exhaust and a program. At 100 percent VE, flow would be 525 CFM, still under what the stock system flows.

5500 x 330/3456 x .90= 472.656 CFM

My conclusion is that the stock intake is pretty darn good and can support some extra horsepower. A CAI, as SubiGt noted, tends to lean the mixture a little and that's where most of the power is. Thing is, your Edge or Grypon does the same thing and unless you need the extra airflow, I don't think the two things "stack" all that well. Bill has to essentially richen the mixture with a CAI because the combo may lean it too much, thus negating a big part of the advantage the CAI offers. Bottom line, IMHO,if the stock intake system has the airflow, you may not need the CAI when running a mild program.

I almost forgot: Remember that this is a throttled engine. Everything we are discussing is relevant only at WOT. The throttle plate is the main restriction in any engine (how much time do we spend at WOT??) and beyond that it's the intake valve!

[88Racing]

Jim,

I know the use of the GT data is relatively different to the truck.

1. The location of the motor

2. The location of the intake(air pick up)

3. The stock air intake itself.

Their motors are in a mid position in the vehicles and thus if a metal type intake is used they are finding it to build up heat.

My other point about surface area and media densities is that there can be more air flow accumalated through a greater surface area. But on the same hand a cai attemps to get the same air penetration through its filter using a less dense material to make up for the lost surface area.

Lars

[jimmyv13]

Originally Posted by Jim Allen

88Racing: I'm not convinced (willing to be, but have not as yet) that there are notable differences in power between metal and poly tubes. In "Forum Poster" type racing, of course, every fractional difference is hotly debated. I've discussed this with several engineers and they say, yeah, there is a tiny difference but the air is moving to fast to pick up much heat from a smooth tube. In the OE, plastic is used more as a weight saver and production expedient.

I have datalogged many hours of intake air temps with turbo vehicles comparing different intake configurations. One thing I noted, the metal intake tubes almost ALWAYS had a higher intake air temp than plastic intake tubes, in the engine bay or not, regardless of filter or intake routing. Of course this testing was on a turbo vehicle where low IAT's are a vital element in producing big horsepower. The turbo heats up the AIT charge, so a low CAI is critical. The lower the temp of the air going into the turbo, the lower it'll be coming out. You'd be amazed how high the AIT gets on a turbo vehicle as it sits at a light...triples sometimes quadruples in 30 seconds-1 minute. Waiting in line at the strip was excruciating at times as I watched my IAT's climb like a tach. I was one of the guys that pushed his car most of the way up the staging lane.

So, CFM compared to IAT's...which is more important on our F150? I am taking Jim's testing results as the stock intake routing is great for CFM, as he has proven. It is in fact a true CAI, so the IAT can't possibly get any colder short of major modification to the intake charge. I don't think the DIY mod negatively affects the IAT, but I do not have the numbers to prove it...and it's difficult for me to think how it could possibly do so.

Originally Posted by Jim Allen

Also 88Racing: I not clear on what you mean by "more air thru the stock setup" ? Yes the stock Motorcraft air filter seemed to flow a little more than the AEM panel. Was that what you mean? Or was it that the stock setup was very close to the modified system. Note that in #6, we removed the rubber downspout adapter and that's when the modified system exceeded the stock by a little.

I wonder how a K&N compares to the AEM and stock filter? My stock filter needs replacing and I'm curious if I should get a K&N or a motorcraft? I didn't like the flexibility of the downspout connector and it sounds as if it may be detrimental to the flow if it collapses, so something else is going to go there.

[Jackpine]

Originally Posted by Jim Allen

JackandJanet: The theoretical airflow number I gave is calculated. The formula is :


CFM= RPM x DISPLACEMENT /3456 x VOLUMETRIC EFFICIENCY

I used 5500 rpm as a reasonable maximum and was generous in giving the fairly free-breathing modular a 90 VE. 85 percent would be more real world but would drop the CFM result. I gave it .90 to factor in some other mods that would increase airflow a little, such as a free flow exhaust and a program. At 100 percent VE, flow would be 525 CFM, still under what the stock system flows.

5500 x 330/3456 x .90= 472.656 CFM

My conclusion is that the stock intake is pretty darn good and can support some extra horsepower. A CAI, as SubiGt noted, tends to lean the mixture a little and that's where most of the power is. Thing is, your Edge or Grypon does the same thing and unless you need the extra airflow, I don't think the two things "stack" all that well. Bill has to essentially richen the mixture with a CAI because the combo may lean it too much, thus negating a big part of the advantage the CAI offers. Bottom line, IMHO,if the stock intake system has the airflow, you may not need the CAI when running a mild program.

I almost forgot: Remember that this is a throttled engine. Everything we are discussing is relevant only at WOT. The throttle plate is the main restriction in any engine (how much time do we spend at WOT??) and beyond that it's the intake valve!

Jim - I was pretty sure you were getting the "required" airflow number this way and I can't see any other way to get it. I would have used an RPM figure of 5000, since power seems to drop off above that, but this would only reduce the airflow needs.

And, we've already agreed that this mod's effects would only be seen at WOT/high RPM.

But, I'm still left with the fact that Bill has measured higher power with the restriction removed.

I wonder if this is an explanation? The possible air volume requirements may well be supported by the stock intake, but, with a cross-sectional intake area of 3.14 sq inches the flow velocity would have to be higher than the velocity with the restriction removed and a cross-sectional area of 7.07 sq inches. Higher velocity equates to lower static pressure. Lower static pressure available at the intake port, equates to lower absolute compression (relative to sea level pressure), similar to what you'd get running at altitude in the high Rockies. Lower compression means lower power.

But, I sure don't like the way the downspout approach "puckered". I think the inlet HAS to be rigid.

- Jack