Intake manifold design resources? 2

[dubc4]

Looking to take on an ambitious project and design my own intake manifold for a project I have coming up.

Has anyone come across any good resources for design criteria/considerations for engine intake manifolds? Mine will be specifically for a forced induction application.

Any links would be very much appreciated.

Thank you!

 

[MikeHalloran]

With a blower, the importance of the intake manifold reduces to mere ductwork. Yeah, there might be 5 or even 10 percent available there, but it will be masked by the 20 ... 100+ percent increase over n/a provided by the blower.

Any number of hotrod sources will show and tell and occasionally prove how to make better manifolds for naturally aspirated engines. ... but they all tend to increase the internal volume and external bulk of the manifold, whereas the usual biggest problem with a blower is getting the blower, intercooler, and ducting within the vehicle's envelope. The intake manifold is therefore reduced to vestigiality in the interest of packaging.

Coupla flanges in the proper place, run tubing between them nicely, and go on to the next challenge.


Mike Halloran
Pembroke Pines, FL, USA

 

[Tmoose]

carb or fuel injection?

 

[BrianPetersen]

... and what type of engine/cylinder configuration.

A lot of intake manifolds nowadays incorporate an air-to-coolant intercooler inside them. Getting intercooling trumps being able to tune the runner lengths when the supercharger is stuffing air into the engine anyhow.

 

[dubc4]

It will be a fuel injected V6 application. Single turbo.

I've researched a few things online and there is obviously great information out there. However, I was hoping for a more comprehensive collection in the form of a good book or two.

Although the duct work is one of the main challenges, I've still read about poor manifold designs causing lean/rich conditions in the different cylinders due to uneven distribution of the intake flow. I'd like to learn about some of the smaller details as well such as the effect of port shape/location on flow efficiency and pressure/velocity drop.

Thanks guys!

 

[BrianPetersen]

If you are using multi-point fuel injection (which you should), the old school wet-flow issues leading to poor cylinder-to-cylinder distribution are nonexistent because the manifold is dry inside and the injector is right at the intake port.

 

[Deividas]

BrianPetersen - but if all injectors injects same amount of fuel, and manifold is poor designed, can't be, that in one runner will be more or less air flow than in other?

 

[MikeHalloran]

The manifold would have to be very badly designed indeed to have that much flow asymmetry. The turbo will stuff in more air than is needed by any cylinder. The manifold acts as a revervoir or accumulator for storing pressurized air between valve openings. ... so the manifold geometry is nowhere near as critical as in a naturally aspirated engine.

Additionally, it's possible that a sufficiently adaptive ECU could detect a weak cylinder just from the engine's instantaneous performance, and compensate by adjusting fuel flow to that cylinder. I don't know what such an ECU is in production now, but it's certainly achievable.



Mike Halloran
Pembroke Pines, FL, USA

 

[gruntguru]

If you have multi point injection, the design aim is a plenum containing near-stagnant air, feeding equal-length runners with bell mouth entry. If the engine has factory MPI try to use the original manifold.

je suis charlie

 

[SomptingGuy]

As mentioned before, once you've made the leap from NA to forced induction, all that's been learned about manifold tuning is essentially obsolete. Still fascinating and clever, but then again, so were carburettors. And mechanical fuel pumps.

Manifold internal flows are still interesting for the purposes of EGR distribution, but I doubt that's a problem here.

I wonder how mean fuel delivery varies from injector to injector and if that variation changes with age?

Steve

 

[PackardV8]

The manifold acts as a reservoir or accumulator for storing pressurized air between valve openings. ... so the manifold geometry is nowhere near as critical as in a naturally aspirated engine.

To expand upon this concept, suppose we have a supercharged V8 engine with flow-restricted side-by-side pairs of intake ports.

I've done some tests by removing the divider between the pairs of intake ports as far as possible. On a flow bench, the increased plenum volume increases flow to each cylinder. That's a huge plus beyond what is available with conventional hand porting.

In practice, what will be the effect of a log/plenum intake all the way to the valves? Absolutely no port separation from supercharger to the eight intake valves?

Of course, cam design is a factor. At some point increases in duration and overlap will begin to be a problem with that degree of interconnection.

What other plus/minus factors could be at work?

jack vines

 

[dubc4]

Thanks for all of the responses guys.

Yes, it is multi-port fuel injection and it was previously a supercharged application. (GM L67 3800 motor). The supercharger was an M90 roots style blower and was therefore taking the place of the plenum.

I have an empty supercharger case that can be used as a plenum, however, the volume is quite small. I can also replace the intake manifold with one from a NA 3800 engine, but again, the volume will be considerably less than what I read is ideal (ratio of 1.6:1 or more compared to the engine displacement). This is what prompted me to look into the custom plenum as well.

As for the unequal distribution of air in the cylinders... My understanding is that the amount of fuel injected will obviously be a function of the amount of air available as read by the MAF and MAP sensors. There are only one of each of these sensors located along the intake tract and the plenum. Therefore if plenum and runner geometry cannot provide equal distribution of air to each cylinder then you can have rich/lean conditions. Or am I over-thinking this?

Intake manifolds like this:

http://www.marcellamanifolds.net/images/IMG_0906.jpg

are designed such that the incoming air flows into a large volume (expansion) and should slow the velocity of the air such that the front runners are not starved if the air were to otherwise blow past and accumulate towards the rear cylinders.

 

[Tmoose]

SAE paper 720214 "Design Refinement of Induction and Exhaust Systems Using Steady-State Flow Bench Techniques"

The authors seems to consider flow balancing (naturally) quite separate from mixture distribution, but still important at least for normally aspirated passenger car applications.
" Branch Balancing- During a flow program on an intake manifold, several design parameters can be varied to improve
flow to one cylinder at the expense of another. The following discussions (Angle of Turn at the Branch, Flow Path
Length, and Turning the Corner) are useful tools in this procedure,which moves the design toward a goal of equal flow
capability in each carburetor-to-port flow channel. Ultimately, a decision must be made as to whether some
channels should be arbitrarily restricted in the interest of equal flow. In our work, we have chosen equality of flow in
preference to overall maximization. Our rationale was that a middle range of engine speed existed in which equal airflow
was necessary to provide equal compression pressures and knock behavior. This uniformity would then permit the
highest compression ratio for a given fuel, combustion chamber, and spark advance.
In this same range of engine speeds and above, equal compression pressures would promote the subjective sense of
smooth engine operation near wide-open throttle. In surveying the practice of United States engine manufacturers,
we have noted both approaches, sometimes within one manufacturer's engine models. In one instance, the flow
capability of the poorest cylinder was 86% of that for the best cylinder (case study D).
Angle of Turn at the Branch - The angle through which the induction stream must turn from the longitudinal runner to
the transverse branch of a V8 intake manifold exerts a considerable effect on flow capability (Fig. 25)_
Minimizing and varying this angle is one means toward high and uniform flow.
Flow Path Length - Similarly, modest variations in length of the induction stream produce readily discernible effects on
flow. This is another variable whose influence must be considered."

==============
In the earlier SAE paper 670067 "Mark II - 427 GT Engine Induction System" FORD developed the "medium riser" engine to equal the noble but hard to package "high riser" series after simultaneously improving airflow and maintaining good airflow balance for each. 670067 includes similar development work on the SOHC engine.

 

[PackardV8]

My previous question about an all-plenum intake reminded me of the opposite design, a plenum at the top of long ram tubes. When turbos began being applied to drag race cars, one of the first I ever saw fed pressure to a large plenum placed on top of 18" individual runner stacks of the Hilborn injection. It ran OK, but was ungainly and prone to pressure leaks. While they were trying to seal pressure in piping designed to operate under vacuum, someone pointed out the turbo boost overcome any wave/ram function and they could do away with the long tubes. Removing the ram tubes with their 32 hose clamps from the intake track made it much easier to seal and no loss of power.

jack vines

 

[gruntguru]

"ratio of 1.6:1 or more compared to the engine displacement"

I believe that is 1.6 time the displacement of a single cylinder. The stock NA3800 should work fine.

Jack. Long tubes can help fatten the torque curve on a turbo motor. Wave function still applies on boosted engines - look at any top level turbo competition engine. Wave effect is less important on a street or budget race engine where often the boost can be turned up to whatever limit (usually detonation) the engine has.

je suis charlie

 

[yoshimitsuspeed]

I can't believe there are people on this forum who could think intake design is any less important for forced induction or that any theories are actually significantly different on forced induction.
All engines run on air pressure and pressure differentials. In a sense all motors are boosted. If they didn't have air pressure they wouldn't run. All a turbo or supercharger is doing is multiplying that pressure.
Plenum volume is still important. Runner lengths and diameters will still effect the performance and in very similar ways to NA.
I will agree with gruntgurus last post in that it may not be worth spending a ton of time or money on a street build. In fact for a street build an OEM NA intake will usually be a great choice for a street boosted build. If you are going to take the time and or money to make something then it better be better than stock and that will take some thought and effort.

As for the OP if this motor ever came NA then unless you plan on a massive build pushing the ragged edge it's quite likely the OEM NA manifold would work quite well. Again general manifold function won't change much between NA and boosted.

There are a couple theories among turbo manifold designs. Stock manifolds will usually be tuned for a broad power curve and decent low end. This should also help with your low end spool.
Then you could design and build a mani with tuned induction in mind. A long runner mani with a focus on low end and quick spool. Or you can go shorter with a focus on broad power curve.
Or you can go super short giving up pretty much any effect of resonant tuning and just trying to maximize all out airflow. This will tend to give you more peak RPM power.
Of course these are all very simplified generalizations and there is a lot of finer theory to understand to even start to implement it with some understanding and hope of having the effect you hope for.
Even then there is still some art to it too. Top performance manifolds aren't made off the first design knowing they will work the best. You would make a few designs based off what you think will be best and then you test those designs and then make some variants of the best of those and test and repeat until you are satisfied that you have reached an acceptable level.
As far as general theory, understanding harmonics and sound waves this book is a great start.

http://www.amazon.com/Scientific-Ex...ientific+design+of+exhaust+and+intake+systems

 

[GregLocock]

OK show us the money. Let's see some data showing a tuned pipe system with superior performance to a shortest possible direct plumbing solution, both turbocharged.

The reason I'm suspicious that tuned lengths are less important are (1) the turbo provides rather an odd termination, acoustically, so I doubt it'll reflect the waves especially well and (2) skin friction effects will be more important so the gains from resonant forcing are likely to be masked or outweighed by the losses from the longer ducts.

Note that I suspect equal length runners may be worth having, I'm just dubious about the necessity for tuning their lengths.


Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

http://eng-tips.com/market.cfm?

 

[MikeHalloran]

I'm curious about what pressure/ gas density you'd use to tune said runners.


Mike Halloran
Pembroke Pines, FL, USA

 

[BrianPetersen]

If you are an OEM manufacturer - Compare the cost of the fancy tuned-runner intake manifold to the cost of a log manifold but with the turbo boost pressure set 1 or 2 psi higher. Even if it involves making a different compressor wheel to fully "optimize" it.

If there are space constraints under the hood, compare the net benefit of the space-consuming tuned-runner intake manifold and no turbo, to a log manifold but with the turbo sitting where the rest of the intake manifold was.

YES, the wave action is the same regardless of the absolute pressure. It just might not be economically viable or worthwhile to utilize it.

 

[gruntguru]

Greg. The plenum chamber provides a stagnant termination for the runners.

je suis charlie