What is the math behind:

runner size (other than short = high rpm pwr and long= low rpm pwr)
plenum size (read somewhere that the total volume should be 80% of your displacement... ?)
injector placement (I know the answer to this sort-of.. but I need a refresher... has to do with combustion process timing, correct?)

That's all that I have been wondering... if anyone else has some useful information, please contribute!

-cc


Modified by CarbonCreations at 2:29 PM 11/5/2005

 

1) Runner size is a balance of flow vs. velocity. You must select the proper diameter based on where you want the maximum effieciency to occur. A good rule of thumb is design the runner so that the gasses going through it max out at Mach 0.4 at the maximum airflow requirement of the engine. This creates a runner smaller than many aftermarket manifolds. Runner length is based on harmonics. Most OEM manifolds have long runners to promote harmonics at lower RPMs, which give a boost to lower-RPM torque. The tradeoff is throttle response.

The actual math is very complicated. I would recommend designing manifolds in SolidWorks then using Cosmos to run fluid dynamics simulations.

2) I don't see why having a big plenum or a small one would really make any difference. You just want to assure equal flow to all runners. Smaller plenums are better for turbo cars because it means less volume to compress.

3) Injector placement is less important than most people think. The fact is that on any EFI engine at mid-high RPMs, the injector remains open longer than the valve remains open, so the injector is never intended to spray into the open valve at all. The point of the injector is to meter fuel. You just want to make sure that fuel doesn't puddle inside the runner and cannot run back into the plenum.

 

wow good info i was going to ask the same quesions

where can i get thoughs programs you spoke of??? and what do they cost ...... or is bittorrent an option lol

 

I have heard injector placement can have a decent impact on power. At high RPM, the airflow almost needs to have the injector far away from the valve to make the mixture atomize completely. This injector placement kills low throttle drivability and efficiency though. This is just what I have heard/read though and may just be meaningful if you are looking to get every last HP possible out of a given design.

With a turbo moving 10+ cubic feet of air per second, I really don't see a rather significant change in plenum volume really affecting turbo lag a whole lot. Seems like the only thing to keep a designer away from a very large plenum has only to do with throttle response?

 

if any one can point me twords math for this i would be most greatfull

 

<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote, originally posted by 99_GS-T &raquo;</TD></TR><TR><TD CLASS="quote">This injector placement kills low throttle drivability and efficiency though.</TD></TR></TABLE>

This might be true on a poorly designed manifold. Fact is that TBI and carburators work just as well as MPFI so long as the manifold has balanced flow, good atomization, no fuel puddling, and the engine is well tuned. You might note that extra injectors added pre-throttle on MPFI systems generally suck; this is a product of the poorly designed plenum that almost all MPFI manifolds have. (They just aren't designed to be "wet")

 

Excellent information! My logic behind the larger intake plenum was that there would be more air to push into each runner. Your logic/explanation makes perfect sense... however I'm still uncertain that a larger plenum will rob power. It seems to me that the larger plenum would have more volume (durr lol) at a lower PSI. This could be beneficial though if there was a lot of pressure with limited volume, because the pressure would disperse across the greater mass of air. *shrug* I'm curious how the energy transfer of the air molecules would be affected by low and high volume plenums... too bad none of the drag racers with product proven results are reading this

Another design intent that possibly needs to be considered is runner angle; I know that this makes a huge difference in ITB's, granted that is comparing apples to oranges.

As far as SolidWorks goes... there's websites that have it for ~$180 with a student ID, and even less for the previous version. Only limitation is that it can't be used for commercial development. (and the version I recieved had a watered-down version of Cosmos)

 

The Cosmos version I have sucks too. Basically it will just do failure analysis. Is there an add-on cosmos version that I can find (or that somebody wants to let me have) somewhere? I'm just a hobbyist/student and I’m not looking to produce anything other then personal projects.

Beepy, you make a good point. How well will a normal straight runner intake manifold that is typically fabricated keep fuel suspended and off the wall/floor of the runner under low throttle conditions?

 

Under light load, very well. Under heavy load, not well at all. Think about it this way: The intake valve is effectively open* for about 220° of a full 720° cycle, or about 31% duty. Most everybody talks about injectors where 80-90% duty is acceptable. That means that the injector is only spraying fuel into the moving airstream about 1/3 of the time. The other 2/3, the fuel is just building up near the valve.

This is where sequential-fire EFI and batch-fire EFI gurus butt heads. It is true that spraying fuel into the open valve is preferred, but the truth is that sequential-fire EFI is only more efficient below 30% injector duty... And then it is only a couple % more efficient.

It would be interesting if you could somehow graph AFR as the intake stroke was occuring. You would see that the AFR spikes when the valve opens, then slowly drops until it reaches the target AFR.

*A common camshaft spec is duration, where a common intake duration would be 260° of crankshaft rotation. However, almost no flow occurs until about 0.06" of lift has occured, so I use the less common 0.06-0.06 duration spec.

 

At high RPM though, higher air velocities are reached because of higher pistons speeds. Then you also have a great deal of interia to the moving air/fuel stream once the valve closes.

I would think both of those issues would help keep fuel well suspended at high RPM?

 

beepy

99_gs-t


where did both of you go to school????


i want to be smart like both of you

lol


no relay where are you guys getting this info from???

 

University of Washington, BS, Mechanical Engineering.

But that doesn't mean I know any more or less. I learned car crap the same way most people have: By working on them and fabricating.

<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote &raquo;</TD></TR><TR><TD CLASS="quote">I would think both of those issues would help keep fuel well suspended at high RPM?</TD></TR></TABLE>

Like I said, a well designed manifold shouldn't have issues with fuel puddling.

 

I'm just a little sophomore at the University of Utah. Well, sophomore in Mechanical Engineering, with about 90 semester hours in mostly math and physics classes. I was an EE student for a while, but I've been working in electronics for the last 5 years and I'm just about sick of it, so I switched to ME two semesters ago. Either way, I haven't had anything more then a Thermodynamics class as far as formal education in fluid dynamics is concerned.

Beepy, staying on the topic of this thread, maybe you and others can elaborate on what EXACTLY a good design is?

I was just commenting on the fact that even though the injectors are open for most of the engine cycle, the higher air inertia might help keep the fuel atomized well? At lower RPM, it seems reasonable to think that the lower mean air velocities would let the fuel drop out of suspension easier then at high RPM.

 

im by no means an engineer but here is somthing i can contribute if you have a bigger plenum on a FI car. it should help because these numbers are purly threoteical. when the valve opens on the intake there will be a pressure drop the smaller the plenum the more drop there would because the precentage of air would be more going into the combustion chamber with a larger plenum there is more compressed air wating to enter the valve and there will be less of a drop in the pressure in the plenum is this correct?

 

<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote, originally posted by cstay &raquo;</TD></TR><TR><TD CLASS="quote">im by no means an engineer but here is somthing i can contribute if you have a bigger plenum on a FI car. it should help because these numbers are purly threoteical. when the valve opens on the intake there will be a pressure drop the smaller the plenum the more drop there would because the precentage of air would be more going into the combustion chamber with a larger plenum there is more compressed air wating to enter the valve and there will be less of a drop in the pressure in the plenum is this correct?</TD></TR></TABLE>

thats what i thought, hence all FI guys using victor X's

 

Well, if you want to get into the nitty gritty of what a plenum is for, consider this situation, which we engineers call a "boundary value problem":

Above someone stated that the intertia of the air in the intake manifold will keep it moving at all times. This cannot physically be true directly adjacent to the valve. Since air cannot pass through a closed valve, it only has one choice when the valve is closed: Stop in place. If more air is moving behind it, the air near the valve will be pushed on, and will compress slightly, until either the valve opens again, or all inertia of the moving air has been transferred into compression.

What really happens is somewhere in between. The valve opens before all air in the runner has stopped, and the intake stroke will pull on it again, adding more inertia until the valve once again closes. This cycle creates a harmonic oscillation inside the runner. The same is occuring in all of the runners in the intake manifold.

This harmonic oscillation is physically identical to sound waves traveling in a tube, such as a musical instrument, and it is actually the cause of the "hum" you hear when you open your throttle. So what length of runner should you use? This can be calculated. Depending on the cross-sectional area of the runner (determined to prevent velocities greater than Mach 0.4), and the airflow requirement of the engine at a specific RPM (the max torque point, hopefully), the ideal runner length is that length where the frequency of the oscillation (which is a function of the RPM of the engine) is a lesser harmonic of the sound harmonic. (So, if the engine is to make maximum torque at 2000 RPM, the runner should be designed to create a frequency some harmonic above this, like 10,000 Hz)

You may have heard something like "This manifold is tuned to the 7th harmonic". What this means is that the length of the runner is such that 7 sound pulses travel the length of the runner in the time a single engine revolution occurs. Basically this means the engine is moving very slow compared to the speed of sound (low Mach number in the manifold) and the engine is tuned for low-RPM operation. Most aftermarket manifolds are tuned for the 4th or 5th harmonic, which means something like Mach 0.2-0.3 in the runner, have shorter runners, and are better for midrange. Something with a very short runner (like many hand fabricated ones) would have to be operated at insanely high speeds in order to set up a harmonic. Most don't have harmonics at all.

Note, that a harmonic is not required for the manifold to work. It simply gives a boost in efficiency at the RPM which the harmonic occurs.

Now that you have designed the runners, what should you so with the plenum?

The plenum is basically an inductor. It will have the most inertia, and will balance the harmonics in the runners and provide, hopefully, equal, unturbulent flow. How big to make it? To solve this, I would model the intake manifold as an inductor with 4 parallel oscillators (the "electrical analogy" method is quite useful in system dynamics), then solve for an acceptable pressure oscillation in the plenum.

Of course, for simplicity, you could go with the old standby of just making it 150% the displacement of the engine.

Okay, so what does a well-tuned manifold do? Take for example the Kia Rio, which I drive. Here is the stock dyno run:



You see how flat that torque curve is? Doesn't make much sense, huh? Well, the reason that it appears so flat is because the intake manifold, which is an odd, long-runner design, is tuned to the 7th harmonic and gets a great boost to effiency just off-idle.




Modified by beepy at 6:09 PM 10/14/2005

 

wow that is nice can you go point me to any material where i can learn more about this tuning for harmonics

 

I wasn't implying the airflow never stopped. I was just saying that in the time frame that all this is going on, the air remains very chaotic in motion in the runner while the valve is closed. Put the injector close to this pile up point and the fuel won't have a lot of time to atomize well with the air. This will cause the fuel to be more likely to fall out of the air stream. Put the injector far from the valve and at high RPM, it still has time to atomize well with the air stream and will be less likely to drop out of suspension. That has how I have heard it explained anyway.

Aside from plenum size, how about creating even airflow around each runner?
http://www.utdsm.org/gallery/v...motor
In that picture, you can see HKS went to a pretty large extent in an effort to equalize airflow at each runner inlet. As I look at it, I think the cone shape is there to make the airspeed nearly constant across the length as it feeds through the 'restrictor' slot feeding into the main plenum. Also notice the stage injection with secondary injectors feeding directly down the throat of the runner. Looks like a near ideal design to me. Of course, to maintain any kind of throttle response with that large plenum setup, the car also uses ITBs.

Resonance tuning can have negative effects as well however. If the intake and exhaust are not well matched on the resonance RPM, you will not get the full impact of either part separately. Match the wrong intake manifold with your exhaust manifold and you could potentially lose a good deal of power. When the resonance tuning is done correctly, it seems like the effects of both are greater then the separate components.

 

Internal combustion Engines &lt;-- text book I has the equations...but unless you're well versed in thermal dynamics, physics, etc I dont think they will do you any good...also Im not sure Honda-tech has all the images for all of the variables hahah.

 

I think you got it the wrong way around.

short runner = more high end hp
long runner= more low rpm tourqe

 

The COSMOS package you are after is COSMOSFloWorks.

 

great thread.

what about having a slight taper on the runners to help with velocity? i'll be putting velocity stacks inside the plenum on my new design that i plan on building this winter.

 

without a doubt; tapering will increase air velocity.

 

WOW Awsome thread!!!

Ok im goin to sound like a n00b but thats cause i am, and am tryin to learn cause i cant afford goin to scoo for this... (long story)

anyways, talkin bout taper and i have thought bout building a IM (or havin one built) that would have bout 10-15" of taper (depending on head flow) and a LARGE plenum... BUT some how put in ITBS...

Now this is just all thought, soo tell me if its BS and aint ever goin to work....

BUT PLS keep goin on with this thread im learnin TONS!!!!!

 

10-15" taper is to much. Any thing more than 7" is a bit much.
Have built and tested intakes on imports for the last 10years now and have built most or impute on the intakes in the import drag scene in Canada. they have ranged from VW's, Hondas B and D series and Toyota's. This one has most of the thing talked about full radius velocity stacks, tapered runners and equal distribution. There are a few thing not done mainly due to cost but has not been an issue. The only one direct comparison was one customer running one brand intake manifold for 2 years then switched to one that we built for his set up it went from 440hp at 27psi to 480hp at 24psi. Far as I know nothing else was changed. Only mentioned this to show that a proper intake made to match the engine set up can yield results