First test intake manifold (lots a pics)

[bolsen]

Any updates?

[TrinaBabe]

Well everything is pretty much done but I am in the process of moving still

Ill try to get the rest done and get a prototype at least printed up. Shouldne take me long, just need some time to do it.

[mitsuorder]

I applaud ya TrinaBabe.

[Torquefanatic]

Very interesting, can't help thinking the thread would be enriched by some practical insight from ShaunSG if he's reading...

[KevinD]

well, i hate to be the devils advocate here, but i don't think you addressed the major problem of having a pressure drop in your manifold quite like you should have. everyone goes to huge lengths to eliminate all the pressure drops up to the manifold and big gains are seen from some (air filter, lower intercooler pipe, larger intercooler, upper intercooler pipe, throttle body), and then you are going to restrict the flow down to a tiny channel between plenums. it seems very counter productive simply to give a good distributed flow to each of the cylinders.

what mass flow rates were you using in CFD?

also, what were the velocities going across the channel?

[TrinaBabe]

On most of the tests I was using pressure based openings; ie 65 psia inlet, 14.696 psia outlets. Using it this way, the avg velocity of all openings (Obviously won't happen because all 4 ports wont ever be open at once) is in the 1100-1300 fps range. The pressure at those openings is around 28-29 psia to each runner.

The pressure drop isnt a big problem when engineered correctly as the actual opening is incredibly large. It is thinner but much longer. Air is just as happy to run in a large rectangle as it would be in a smaller pipe. All the slit in the middle does is help distribute the air. Im not saying there is no pressure drop but in actuality there is less of a pressure drop doing it this way as opposed to just letting the air into a large plenum then going into a smaller runner.

[TrinaBabe]

Here are some stats of the last version tested at 60 lbs/min of air:

Oulet Total Pressure = 20.9725
Inlet Total Pressure = 22.648
Inlet Dynamic Pressure = 0.660818
Hydraulic Loss = 2.53581
Percent Loss = 11.197%

If you subtract the loss from a straight pipe in roughly the same diameter it comes out to be just a touch over a 2psi loss. If you compare that number to a conventional tube style intake manifold you would probably be surprised that the hydraulic loss due to sudden expansion would most likely be greater of a pressure drop than this. If you really want I suppose I could do a test for ya to see the pressure drop of a more conventional style and post the results.

[TrinaBabe]

And to show you what I mean about the pressure drop associated with the slit in the middle here is a graphical representation of it. As you can see.... no pressure drop at the area of the slit (Oh and dont worry, this isn't what the manifold will really look like, just using this one to show the slit):



I will go and run some calculations on a more conventional style intake manifold and show what thier pressure drop is so you can compare.

[TrinaBabe]

Here are the stats from a conventional style:

Oulet Total Pressure = 17.3307
Inlet Total Pressure = 18.0696
Inlet Dynamic Pressure = .629711
Hydraulic Loss = 1.27341
Percent Loss = 7.047%

The manifold I used for this test was technically for a DSM which has a larger throttle body and runners than the Evo one I tested with. If I actually switched the DSM version to use the ports and throttlebody from an Evo it has a worse pressure drop than mine. It is quite apparent that the only reason the DSM manifold has less of a pressure drop is because the throttlebody is bigger. The inlet pressure is only ~18psi on this one vs ~22 on the Evo version. Raising the pressure by 4 psi when only working in the 20psi range obviously will have a drastic effect on the pressure drop. I must say I am pretty pleased with the 11% that I have achieved. I promise you it it much less than a conventional intake manifold for an Evo. Also, if I change the throttlebody size for the Evo manifold the pressure drop will become less (Hence the reason throttle body swaps actually help).

[bolsen]

Very interesting stuff. In the next month, I'll be hitting up the dyno. If you get your ducks in line, I may be willing to do some live testing.

[TrinaBabe]

Some other interesting side note of some stats:

My Evo Design:

Mass flow rate [lb/s] = 1.00004
Velocity [ft/s] = 780.871
Density [lb/ft^3] = 0.0825724

XXX DSM Design:

Mass flow rate [lb/s] = 0.999985
Velocity [ft/s] = 537.145
Density [lb/ft^3] = 0.0778011


Even though there is slightly more pressure drop on my Evo manifold it still outflows by keeping the velocity much higher.

I am now in the final stages of design for the manifold so I am hoping to have the mold made sometime next week if all is well. This is where I am just going to put in all the little extras for free. I am open to suggestions on what people would want as well. Now is the time to give me your input on what features you guys would like... this is what I am planning on to start with:

- Almost stock position for the throttlebody (No need to make new pipes or modify yours).

-Stock fuel rail mounts

-Place to easily bolt on your throttle cable

-Aftermarket additional fuel rail mounts with 4 recessed places to drill if you want to add 4 more injectors (I will have the larger spot already in place for the o-rings then the middle holes will just be indented and I will get the right sized drill bit so all you need to do is drill out the smaller hole, get 4 injectors and another fuel rail and your good to go).

-4 bungs ready to be drilled and tapped if you want to add nitrous (Perhaps we will offer this as an option to have us drill and tap them for you).

-Design capable of supporting well over 100psi sustained and make it sturdy enough to withstand short bursts of extremely high pressure (Nitrous backfires, normal backfires, stuck BOVs, etc..)

-Three normal vacuum nipples and one larger one for the brake booster. I will try to get the brake booster nipple as close as possible to make sure the line doesnt look stupid.

-Obviously needs to make power

Some other things I was thinking about it is making it so we weld on the throttlebody flange so you can decide if you want it made for the stock throttlebody or some other one. I havent decided this for sure yet. Otherwise I will make sure it is large enough where you could cut off the throttlebody flange and get a new one welded on.

The runner side flange will be made like stock so you can port match it with whatever head modifications you do. I was going to make it with a few different versions but then realized that many people have custom ported heads in which we could never get the correct sizes and no matter what they would still need to be ported to match the head. This way we will just leave enough room to port it so match any reasonable headwork.

Any other ideas or suggestions are great to have. As I said earlier, I will be hopefully making the first prototype sometime next week. Seems like every week is next week but hopefully this time is real

[bolsen]

Here's my suggestions.

To appeal to the masses and probably make the most $$:
1. Make it exactly like stock, but it better make power.


To appeal to the "others":
1. Don't drill or mount anything that doesn't need to be there. Being that all the aftermarket rails bolt up to the stock location, I don't see a need to move em. (If they do, custom order time.)
2. Hide ALL vaccum / ports / bungs / ANYTHING on the underside out of the line of vision. Better yet, only make one nipple for all small vaccum sources and offer a vaccum distribution block. Booster def needs to be underneath and outta the way.
3. Make it polished.
4. Move the throttle cable. I HATE where it is now.

Cleaner is better. No logos, no fancy stuff, no extra ports that I have to cap off, no nothin! If I want a port, I can drill and tap!!

EDIT:

To appeal to the JDM fan boys:
1. Make it out of carbon fiber.

[EVOlutionary]

This may be a dumb question, and if so a quick, simple answer will suffice. Why do we not make an intake manifold that is like a reversed equal-length exhaust header. Instead of having 4 equal length exhaust runners merging into a collector, why not have the UICP attach to a 2.5"-3" collector and then from that run 4 equal length tubes to the intake ports. Everything will be symmetric as far as separating the one mass of air (UICP) into 4 masses (runners/ports). No need to go through all the hassle of trying to get airflow to be evenly dispersed in an asymmetric manifold.

Is there a simple reason why this will not work? If this is a great idea that no one has thought of, it is now copyrighted and patent pending as of today

EVOlutionary

[TrinaBabe]

Some vehicles do use it but it isn't as great as it sounds (Toyota's do it alot). The reason is because the length of the runners directly correlate into the peak powerband of the manifold. Obviously the longer away from the valve the throttlebody is, the more latency there will be when you slam the throttle open. On top of that, you will still need some sort of a plenum to hold more air so the vehicle doesn't starve on large throttle movements. This will normally be around 50%+ the total displacement of the engine. So a 2.0 would need a 1 liter+ plenum to hold the air. Now make all that fit somewhere under the hood

[EVOgasmIX]

Not to hijack or anything, but why not use four individual throttle bodies for each runner?

You can get some rediculous throttle response that way and just need to tune the size of the runners. That would be badass.

Kinda like this: