Intake manifold design debate

[Freakingstang]

composite materials are not petro based.  Cummins engines utilize composite pieces on their engines and they are exposed to twice the heat, oil, water than a car engine would.  Some of these pieces have over 20,000 hours one them just in our fleet, from experience.

[Ray III]

The low pressure in a carb venturi (Bernoulli effect) is caused by the high air velocity. It's like blowing across the top of a straw, the operating principle of a paint gun. The more air entering the engine, the higher the velocity past the jets, the more fuel drawn out. The Bernoulli effect is even applied so far as to give airplane wings their lift: the upper surface of a wing is longer than the flat bottom, so air must travel faster over the top to meet up evenly at the trailing edge, and that extra air velocity over the wing means the air under the wing exerts a pressure against it.

That said, the Bernoulli effect has nothing to do with intake manifold design. The carb was just used as an example to illustrate how restricting the airflow results in its higher velocity.

[Aussie XAXB]

This is good conversation.

Steve, do you have info that the Ford valve covers are a higher tech composite? The softening of the valve cover was an official concern I was given. Another contributor, though, would be the acid that forms in the oil from the blowby gasses during combustion. That is the bigger reason than viscosity breakdown of oil to change it every 3,000 miles. The acid that forms in the oil in a reaction with blowby gasses eat at the engine's bearings. If the acid can eat the metal bearings what chance does a plastic or composite valve cover have is how I look at it.

Ray, in reading up on the Bernoulli Principle it said that it is often mistaken as to how planes fly. If the Bernoulli Principle was the reason for flight then whenever a plane flew upside down it would be pulled to the ground.

http://www.mste.uiuc.edu/davea/aviation/bernoulliPrinciple.html


We are going to have to open up a nerd club   


Steve #7

[Ray III]

Airplanes fly upside down IF you aim the nose upward enough to negate the shape of the wing. Aerobatic aircraft have wings with symmetrical cross-secion.

Trust me, airplanes aren't held up by invisible strings.

[TireSmoke]

not to mention the thrust of the engine keeping it up...

[Aussie XAXB]

Trust me, airplanes aren't held up by invisible strings.

Oh yeah, sure, blow it all for me!!

Here is the actual page that has the explanation.

http://www.aa.washington.edu/faculty/eberhardt/lift.htm

I haven't finished reading all of it yet but I learned a lot so far.


Oh, and if planes can fly upside down, counteracting the theorietical lift by whatever means, then the Bernoulli Principle must not apply when the plane is rightside up or else the Bernoulli Principle would be negated in the inverted position and the plane would fall out of the sky.

I have questioned that for a long time but never knew what the answer was. The above link seems to explain it well.


Steve #7

[Ray III]

I started a new thread for the airplane lift debate so this don't get off topic.

http://www.badassmustangs.com/forums/index.php?topic=1980

[Aussie XAXB]

Ah, was wondering why, thanks. 

Steve #7

[Marlin]

I am not an engineer by any means. However I do enjoy talking baout this stuff.

If you can set aside the fact you are a Ford fan for a monent and lets talk GM.

The TPI engines (L98) are similar to the 5.0L. They have very long runners and make lots of torque however they run out of steam around 5K. The LT1 has short runners. It has some of the shortest runners that come on a small block Chevy. They are even shorter than a Victor JR. The LT1 makes lots of torque, has a pretty broad curve, and its lots of fun to drive. It doesn't matter what RPM you at if you press the gas it goes. If you have never driven a LT1 you should. The car sucks but it fun to drive.

The LS1 has longer runners than the LT1 but shorter than the TPI. IT makes lots of HP but suffers in the torque is less. It is a faster car than the LT1 but doesn't seem so. IMHO it isn't as fun to drive.


So I think with a smaller lower evving engine the longer runners are a huge benefit and help improve the cars torque and make the car much easie rto drive and more lilkely that any joe blow who gets in the car will like it. A buddy of mine put of those Hartman plastic box intakes on a Vortech Mustang (It was an A-trim to tell you how long it has been) The car was probaly just as fast but it sucked to drive. It wasn't near as much fun.

So my question is what about a shorter runner intake with a 351 or 347. Something that is naturally going to have more torque. Would they benefit
from a shorter runner? 

[Aussie XAXB]

I'm not an engineer either but I am a Toolmaker and that's where I get my metallurgical background. I also just like to discuss these things and this has been a real good topic with lots of contribution and good insight.

Ironically I am a Mopar fan  I was into Mopars all my life (I'm 40) and only last year got into Fords, so I am doing a lot of learning.

Ok, going back to the begining, as I mentioned, the huge engineering quest was to make the shortest and straightest intake runner for performance street engines. This went on for decades. To me, and this is my opinion based on what I know about physics (that was part of Toolmaker Apprenticeship too), I don't understand how longer, higher friction, more restrictive runner designs got to be a good thing??
The only time making a runner longer previously was a good thing was by adding a carb spacer under the carb in order to get a longer straight shot out of the carburetor to increase velocity. These were never more than an inch at that.....maybe rarely two inches.

When I look at intake manifolds on the engines in the Mercury Mariners we build at our plant all I can see is some new, wet behind the ears engineer just trying to get air to the cylinder heads at all, without regard to manifold design principles. The engineer no longer has to worry about fuel suspension while designing the intake like with a carb because the fuel injectors will be shooting the gas right at the valve, so it looks like as long as the engineer is able to get air to that valve he feels he did his job. It would seem that friction loss that comes with a longer runner (and rough walls like in an aluminum casting) and lots of turns is something that has been forgotten about.

Well, to give you my opinion in response to your question, I will ALWAYS take a shorter runner intake over the modern FI intakes with long, curvy runners on ANY engine.

Oh, here's a point......look into any naturally aspirated drag race car (professional) and see what style of intake is on the engine. What style intake is on a NASCAR car? You can bet it doesn't have twisty runners. There's a reason for that.




Not to say that there aren't any exceptions, but this is what is predominantly our there on V-8's. It's not old school.......its physics.

..........my .02


Steve #7

[Marlin]

I would love to try out an LT1 style intake on a 302.

[tmoss]

The reason the NASCAR and race engines don't is because their power range is ~4,000-10,000 rpm so you don't want long runners - it doesn't work - and you don't run a engine like that on the street much.  I'm 56 years old an I've been moddin cars since before the Muscel car era (just).  The long runners are great for making street power below 6,500 rpm and I have embraced the EFI and longer runners whole heartedly.  I remember driving Z28s and Boss 302s with high duration cams that had NO balls below 3,000 rpm and I much prefer the EFI torque and power band of modern engines.

[Marlin]

Tmoss, The last sentence in your reply was the point I was attempting to make. Well said.

[Aussie XAXB]

Tmoss, I  to your age. 

I still don't think that the intake alone will account that much for the low end of newer engines. Now newer engines have roller cams that get more air into the cylinder with less duration. Cams in the engines in the days of yore did not have that advantage. The difference in the amount of impact that a cam has on an engine is far more significant than an intake. An intake change typically provides only up to about a 10 - 15 HP difference max. Compare this to a cam that can give you 50 hp with only a cam change. Improving breathing will of course produce  a slightly larger improvement to use the cam to its total advantage.

Case in point, the article that I have where they switched from FI on an engine to a carb while it was on the dyno. Power change was insignificant. (Hotrod, Engines, Summer 2004) "High Rollers" article was on different roller cams. They played with a few FI intakes and then went from an Edlebrock FI intake (long runner) to an Edlebrock intake and Demon carb. Across the entire band from 3,000 to 6,000 RPM (results started at 3,000) the difference is 2-3 HP and 2-3 lb/feet of torque, sometimes favoring the carb, sometimes favoring the FI. To me this is negligible. There's Dyno info. Their summary was "Gone are the days of a big torque advantage and a horsepower deficite for the EFI". My own summary to that article is that for me personally the only advantage for the EFI is that you can start the car and go, as opposed to waiting for the engine to warm up with a carb. Otherwise, power is the same for either intake system.

My guess is that if there is a theorietical advantage for the longer runner it is lost in friction over its length and through all the turns, in my opinion.

The engine I am building is a 331 with a roller cam and a carb. We'll see what we get at the rear wheels, though at the flywheel would be nicer.


Steve #7

[tmoss]

I addressed the intake because that was the main thrust of your last post.  I agree about the great changes in cam technology and how it has also added signficantly - just like tire technology, EFI, etc, but your comparison of the carb-to-efi once again uses a 3,000+ rpm power range.  Let's do that and start the dyno at 1,500 rpm with intake runners of similar cross section but different lengths and let's see what happens.  The carb intakes have also improved from days of old.

[tmoss]

Here is an example of a little ole 302 engine with a Performer intake I ported the lower on, used AFR 165 heads, an Ed Curtis custom cam for the street, 65mm TB, 77mm MAF, 1 5/8" long tubes - made 332RWHP and 348RWTQ but the real story is it made 313ft-lbs AVERAGE over the entire 2,000-6,000 rpm band.  Average numbers would have been less with a carb across that same rpm band even though the peak HP may have been higher with a carb.  I wish that dyno would have started at 1,500 rpminstead of 2,000.

[Aussie XAXB]

That is an excellent engine. Is this what you're driving in the warmer months now?

Out of curiosity could you please post the cam spec's?

That is quite a flat torque curve. However, with torque being around 250 at 2,000 RPM and gaining almost 100 more by 3,750 I still do not see where the significant "gain" is in torque at low RPM. Your torque band starts at 3,750 and continues until about 4,600 RPM when it starts to trail off a little. We may have a difference of opnion on what is low or mid RPM but I consider that mid RPM, but only because I only expect to turn an engine up to about 6,000 max.

I don't believe that the engine is in the 1,000 - 2,500 RPM band for very long or at least not enough to make much difference while seriously accelerating, but for everyday driving I would expect to be there much more. In that case any extra torque that can be produced would be beneficial. The one thing that's a bummer though is with your dyno sheet there is nothing else in stone to be able to compare it to. Each of us has our own ideas of what it would be with a carb. I too wish that the dyno sheet in the magazine article I site had lower RPM data.

Let's do that and start the dyno at 1,500 rpm with intake runners of similar cross section but different lengths and let's see what happens.

That would be a perfect comparison. Hopefully one of the magazines will do that someday.


Steve

[tmoss]

Steve, actually, that is one of my porting customers cars.  My car is about finished tuning with a stock 88 SD GT cam, 9.3:1 GT40P heads, 1.7 rockers, ported Explorer intake, 1 5/8" shorty headers and the Explorer 19# injectors.  Last dyno was  SAE 277RWHP/330RWTQ.  I am tweaking stock Ford iron and do not intend to go aftermarket with this  3 season toy - gets stored in winter.  Final dyno will be this spring.

We do differ in our views - that's ok, but please show me a 302-306 carb car dyno that makes 250+ft-lbs @ 2,000 rpm and then decent power up to 5,500-6,000 rpm. 

[Aussie XAXB]

Nice torque on your engine. 

Yup, we do differ, and that's ok. 

I can't show you that with a 320 or one that is .030" over, but I'll let you know in a few months with the 331 I am building.


Steve

[Ray III]

Alright, let's look at it like trying to siphon out a gas tank in the shortest amount of time possible. Using a tube of X inside diameter, and putting the car as high as feasible, do you want to use as short a tube as possible to reduce friction, or do you want to use a long tube to increase the mass of the fluid for more force?

One of the better EFI manifolds I have seen is on the Mitsu/Chrysler 3.0L V6:

From the upper plenum, the runners aim directly into the cylinders, and their length allows a column of air to "weigh" against the intake valve when it is up to speed and therefore has momentum.

Furthermore, if you look at the plenum right in the middle, there is a dividing wall that the entrance to each runner faces:

Every time that an intake valve slams shut against the moving column of air, it creates a pressure spike that travels like a sound wave back up the runner and reflects off this wall (or on a 5.0, the "box" of the upper manifold) and returns back down the runner. These runners are called tuned-length because at a certain rev range the pressure spike will reflect so many times and then arrive back at the intake valve at the very moment it is open and packs the intake charge solidly into the cylinder...

The result of this manifold deisgn:

Note: Slightly modified from stock
A flat torque curve over most of the rev range gives that linear increase in horsepower representative of a well engineered motor. The dyno starts about 2,000 RPM but the torque actually picks up at around 1,500 RPM. Power can be had all the way up to the 6,000 RPM redline.

[tmoss]

Gee, looks like my dyno............

[Aussie XAXB]

A picture is worth a thousand words.

The diagram of the intake "box" really helps. So does the diagram of the Chrysler intake runner. Let me digest that for a while, but it does make it simpler to understand the "packing" effect.



Steve

[Aussie XAXB]

Ok, here is what comes to mind......

Even though there may be the existance of a pressure spike when the intake valve closes that spike travels back to the box as you mentioned. With the 7 other valves opening and closing and the violent motion of air being pulled to this part of the intake and then another as each cylinder demands air, how is that spike utilized? The pressure spike will follow the path of least resistance, and if when it gets to the box it will travel towards whatever valve is open next and just become part of the vacuum going to that valve. I would think that the environment is too unstable and violent for such a theory to work in the reality of what is going on inside that box. I would believe that the action of all 8 valves functioning as they do off the same box causes a canceling of pulses, much like a muffler system is designed.

Here again the best measure would be a heads up dyno comparison between the long runner EFI intake and a straight carb intake. Also the range that you say that the affect is most effective in is a mere 1,500 RPM, since we've been talking about the 1,500 - 3,000 rpm range for which we've both been having trouble finding dyno results for. This, as opposed to the wider range of 3,000 - 6,000 where there is no appreciable performance gain in the EFI system whatsoever, still keeps my bias against these fandangled EFI systems.

Perhaps if we can get our hands on dyno results in the way I mentioned I would be swayed. I think that is the only thing that would do it for me, otherwise we will just be expounding our dissertations on each of our own feelings about these systems for a looong time. I wish we could find such dyno results so that there can be no question either way. Its always nice to really know for sure.

Steve

[tmoss]

The benefit comes from mass of the column of air (long runners = more air moving and it doesn't want to stop) that increases pressure on the back of the valve when it is closing - pushes more air into the closing chamber.  The benefit does not come from the reflected pressure wave (that travels back up the runner as the valve hits the seat and the column is stopped short) hitting the plenum back wall and relfecting back down the same runner, that wave is dispered too much to do any good for that runner again. 

[Aussie XAXB]

Ok, that was in response to Ray III's statement......

Every time that an intake valve slams shut against the moving column of air, it creates a pressure spike that travels like a sound wave back up the runner and reflects off this wall (or on a 5.0, the "box" of the upper manifold) and returns back down the runner. These runners are called tuned-length because at a certain rev range the pressure spike will reflect so many times and then arrive back at the intake valve at the very moment it is open and packs the intake charge solidly into the cylinder...

As far as the air above the valve goes and the weight that the air has (14 psi @ sealevel if I remember correctly) the weight of that air is always there whether it is outside or inside the plenum since air is everywhere. It would be the same as running water through a cylinder in an engine that is at the bottom of the ocean. If you crank the engine over it would not matter what the intake design was (long or short runner) the same amount of water pressure would be pushing the water into the cylinder. Air would act in the same manner (from a physics standpoint). The mass of the air in the runner would not be greater than the mass of the air outside of the runner.

This too is where the friction loss would come into play with a longer runner. Friction loss is calculated when laying out an inground sprinkler system (I used to sell parts) and even the type of pipe used (pvc, polyethylene) was a factor. In every situation, no matter the application or field, when a fluid (even air) has to travel through a confinement (pipe, runner, tube) it will be in contact with the inner walls of that confinement and friction will result. In comparing the flow between the same diameter structures of the same material, the shorter structure will have less friction loss due to less contact area.


Steve

[tmoss]

Sounds like your cosidering the static condition of the air - air pressure is different that the weight of the molecules.  Once those molecules are traveling at high speed, they are hard to stop and provide a ram effect.  The amount of friction is small compared to the crosss section of the runner and a slightly rough surface reduces the effects of laminar flow restriction.

[Aussie XAXB]

With laminar flow being non-turbulent flow the rough surface would increase the restriction by causing turbulence, thereby increasing the distance that the air travels as well, would it not?

By the way I am really enjoying the intellectual volley. Been a while since I have had such an exchange of theory. 

Sounds like your cosidering the static condition of the air - air pressure is different that the weight of the molecules.

Yes. An air molecule weighs X much. Stack them up (such as at sea level to the outer reaches of the atmosphere) and then they will exert a pressure that is much higher than the weight of that single molecule. This weight is consistant whether it is resting on a hood or at a throttle opening. The flow of air through the throttle at sea level will be different than the flow of air at 5,000 feet. The ram effect that one would hope for with that long runner is controlled at this point; how much air can get past the venturi in X amount of time. True, the air will flow faster through a venturi than a straight tube, so why not place another venturi or two in the runner to make it flow even faster? Because at some point it will be more of a restriction than an aid. Whatever air that can flow past that first venturi will dictate the amount of air that the cylinder can get, regardless of intake design. The intake can only hope to be less restrictive than the next model.


.........and Bill the cat rocks. I've loved Bloom County since it first hit the papers in the days of yore.


Steve

[tmoss]

The drag from a smooth wall causes laminar flow and a cross section of velocity would look like the tip of a rocket ship with the faster moving molecules in the center.  each layer of molecules drags on the next.  A slightly rough surface causes tumbling in a VERY small layer of air and actually acts like small air bearings which reduces drag and total flow restriction and the velocity cross section looks like a round bullit nose rather than a rocket ship nose.  I have illustrations of the laminar flow examples but can't post them up either.

When the piston creates a vacuum in the intake tract (because the restriction in the runners/TB will not allow flow to keep up wiith piston speed and volume) air that is pulled through a venturi (TB and runners) increases velocity depending on the vaccum level.  The effect of altitude is the number of molecules per unit of area - air is more dense (more molecules) at lower altitudes. 

Bill is my buddy - I have a bill the cat at home.