Scientific approach to air filter setup

[Lurkin]

please forgive me for my ignorance.

cfds in finance are contracts for difference.. i assume you mean something else?

sounds like I have some reading to do...

[Mr Revhead]

computer flow dynamics!
ie simulated flow of air/liquid through/around things

[fivebob]

Actually I think it stands for Computational Fluid Dynamics.

Still means the same thing though.

[Mr Revhead]

yeah thats what was in my head... but i couldnt spell it and thought my head was making up words again....

and yeah, fluid, that sounds better

[Crampy]

So the little wings being used to direct flow is trying to keep it laminar?

Ideally you wouldn't want any bends in an intake, but if there was the wings would certainly help, or at least keep the curves less sharp and abrupt.

[Crampy]

http://www.blackclouddiesel.com/FORD/TAG/tag.htm

Are these for real or what? Is it really going to do that much to the intake? I have heard of people using honeycomb sections in their intakes, to help direct air properly over the MAF sensor element. I believe Altezzas suffer when putting pod filters on them, making them run a bit rough due to disrupted air flow through the MAF.

[ChaosAD]

Using pvc is also great in that it can be shaped with a bit of heat so you can make a new snorkel from one piece of 3 or 4 inch thick walled pvc pipe. The ends can be flared out by heating a few centimeters of the pipe and pushing it down over a funnel.

in the book I have here they tested the flow of three different intake pickup designs on a flowbench at 3 inches of water. The article is probably on the net somewhere.

1,
A 215mm long 3" tube cut off square at the end with an ID of 73mm and csa of 41.8cm2 flowed 255cfm.

2,
A 215mm long 3" tube with a straight flare was tested, the flare was 11 degrees and 75mm long. the mouth of the opening was 115mm in diameter giving an intake area of 104cm2. This showed a 25% increase over the straight tube with a cfm of 318 recorded.

3,
The third design was a custom made bellmouth. Again it was 215mm long and 3" in diameter. It had the same 11 degree 75mm long flare as the previous design but had a 22mm tube rolled around the top and brazed on to form a bellmouth. Plastic body filler was used to smooth ut any imperfections and give a smooth radius entry into the tube.
The shape of the design was based on published tests that indicated it would flow very well.

The result for the sophisticated bellmouth was 322cfm, only a 1.2% increase over the second design with the 11 degree flare.

[ChaosAD]

http://www.autospeed.com/cms/A_0542/article.html
Theres a lot of other good articles on autospeed if you search 'intake'.
A lot of the information in the articles re-enforcing what crunchy has mentioned.

http://www.blackclouddiesel.com/FORD/TAG/tag.htm

Are these for real or what? Is it really going to do that much to the intake? I have heard of people using honeycomb sections in their intakes, to help direct air properly over the MAF sensor element. I believe Altezzas suffer when putting pod filters on them, making them run a bit rough due to disrupted air flow through the MAF.

The evo4-6 maf sensor come from factory with a honeycomb section in the front of it, similar to what is partially shown on page 7.
http://www.autoshop101.com/forms/h34.pdf

[Truenotch]

Example of CFD and bellmouth shapes etc
http://www.gt40s.com/forum/gt40-tech-en ... study.html

[fivebob]

Which section flows better, round, oval, square, or rectangular

[steve murch]

Which section flows better, round, oval, square, or rectangular

depends on design,angles,volume,engine flow,cam,rpm etc etc

[fivebob]

Which section flows better, round, oval, square, or rectangular

depends on design,angles,volume,engine flow,cam,rpm etc etc

Yes, but if you optimised each section type for the same engine which one would give the greatest flow, or would there be no difference?

[steve murch]

Which section flows better, round, oval, square, or rectangular

depends on design,angles,volume,engine flow,cam,rpm etc etc

Yes, but if you optimised each section type for the same engine which one would give the greatest flow, or would there be no difference?

thats where real world testing would come from,cfd software would give you an idea but doesn't mean it equates to the same thing in real world conditions.
remember race teams try try try and try even after building in the pc then the engine room.
the greatest flow doesn't happen all the way thru you have to target where you want it the most,you cant turn a 1300 into a big block, you still have to pass the same cfm to get that result.
thus why turbos are a great thing

[fivebob]

There was a series of articles about airbox design by Prof Gordon Blair and Melvin Cahoon published in Race Engine Technology magazine which unfortunately aren't available on the net, but there is one of their later publications available online about the design of bellmouths. http://www.profblairandassociates.com/pdfs/RET_Bellmouth_Sept.pdf(1.7mb)

Actually all of Prof Blair's articles are worth a read, and possibly proof that not everything you read on the net is a waste of time

[Crampy]

I've got some bellmouth intake pipes I bought while in Texas last year.

Is it best to have a super smooth surface of the inside of the pipe, or a slightly rougher surface?

[frost]

rough, about 600 grit i think,

[ChaosAD]

Probably dimpled like a golfball to minimise the boundary layer effect

[fivebob]

Is it best to have a super smooth surface of the inside of the pipe, or a slightly rougher surface?

In theory it depends on what sort of flow your have in the pipes, Laminar flow has less drag, but is prone to separation which greatly increases drag, turbulent flow has more drag but is less prone to separation.

In practical terms it probably doesn't matter a great deal as the layer of air that is actually in contact with the pipe walls is not moving anyway. Polishing things to a mirror finish won't give you any noticable performance improvement, though it's generally accepted that a rough surface is better after the point that fuel enters the system to promote mixing.

[Crampy]

Is it best to have a super smooth surface of the inside of the pipe, or a slightly rougher surface?

In theory it depends on what sort of flow your have in the pipes, Laminar flow has less drag, but is prone to separation which greatly increases drag, turbulent flow has more drag but is less prone to separation.

In practical terms it probably doesn't matter a great deal as the layer of air that is actually in contact with the pipe walls is not moving anyway. Polishing things to a mirror finish won't give you any noticable performance improvement, though it's generally accepted that a rough surface is better after the point that fuel enters the system to promote mixing.

Sweet. It's just these things are plastic and painted with a chrome look paint and are really smooth. I always thought it's best not to have highly polished intakes (I knew about intake ports, but wasn't sure about the rest of the intake system)