Turbo Manifold Design
 

Regarding the turbo manifold, I am not finding much empirical testing on runner lengths.

The whole commonly stated notion that "long runners take longer to fill up" doesn't make sense to me and sounds like bullshit.

Here's some math that confirms my thoughts:

If, say at 3000RPM my turbo/motor is flowing a piddly 100CFM, that means my motor flows 1.6 cubic feet per second. If I have 24" runners that are 1.375" ID, that means that my individual runner volume is 143 cubic inches. Assuming an equal length runner manifold, that means my total manifold volume is 143*4= 572 cubic inches or .33 cubic feet.

If my throttle plate is closed, and I slam it open, going from let's say, zeroCFM up to 100CFM instantly, that means the manifold would "fill" with air in 0.2 seconds.

This is, of course, discounting the path from the intake to the turbo manifold, but given that is the path to the manifold is a constant between a long and short runner design, it doesn't matter.

Given the negligible time it takes to push air through a manifold, one thing that doesn't make sense to me, is assuming the same runner diameter, bends, and collector merge angle, why short runners are said to have better spool up than a longer manifold. Since short runner resonance improves high RPM airflow in the intake side, why wouldn't it be the same on the exhaust side?

I think response is whats affected not spoolup.
Though I've seen the begi tests that show the log vs the tubular with the tubular netting a little more power a little higher in the powerband.

If I were you I'd go with a ramhorn like JayL and a couple others had. Or a sidewinder. Most big power cars have long runners. There's a whole thread on this already. Search n00b

Go do a search for any picture of an older F1 turbo setup and you'll find long runners. If that doesn't say anything, end yourself. For us specifically, I'd imagine there is a bit of a logistical issue pertaining to long runner length.

Yeah I think the biggest issue with long runners and sexy manifolds with turbo's hanging off them like 2' away from the engine is our engines vibrate like dildos and something like that would crack to all hell unless you brace/support the shit out of it with like 3 braces/cranes.

The long vs short runners thing has got more to do with pulse tuning than "time it takes to fill".

+1 This I think.

F1 car engines are smoof as butta when they rev and probably also have very little motion in their motor mounts. And ours are (as mentioned) like angry electric dildos with worn gears.

I wonder if some of the length in F1 manifolds has to do with the RPMs of their powerband as well.

another purpose of ultra long runners is to have a very smooth collector to minimize turbulance.

for all intensive purposes, equal length wont change much. my last year manifold had cyl #1 and 2 at ~4" and 3 and 4 at ~10"... the spool and ultimate power was the same as many other cars of equal boost and such with very similar spool characteristics.

What you're missing is that you don't just have to fill the manifold with air - you have to PRESSURIZE it to get the turbo to do any work. If you expand the manifold volume by 30-40%, you're going to see a drop in response. Even if it's only 0.3 seconds, that's significant enough to feel. The long-runner car may not spool up any quicker, but a talented driver on a road course would be able to tell the difference between a small manifold and a big manifold.

Long vs. short is on our to-do list of back-to-back dyno tests - we're going to take dyno and road data with manifold pressure, turbine inlet pressure, and TPS, so we'll be able to quantify exactly what works best where. Give me like 6-8 weeks.

I've gone from the log manifold to the AF original manifold and saw very little change in lag. It's there, but it makes more output at lower RPM, and a welcomed change.

There was some debate here previously about keeping the runner diameter smaller to maintain flow velocity as well. Smaller overall volume and higher velocity would probably be a noticeable difference. As I recall, the primary runner diameter most on here are using is significantly larger area than the area of the diameter of the exhaust ports themselves meaning the exhaust gasses slow down and lose velocity upon entering the manifold. If the exhaust port is the choke point, going larger in the primaries in the manifold shouldn't help overall flow but might reduce the effectiveness and the speed of the pulses.

For the power levels Fae is looking for, he needs the larger runner.

small diameter runners start to choke around ~450whp on all different 4 cyl platforms from my research.

I'll donate $5 to Tim to create:

narrow tube absurdflow mani
long tube absurdflow mani
narrow long tube absurdflow mani

(make it easy, make them straight tubes--thin wall = cheap for testing only, turbo can stick out hood.)

Your runners look longish- 12"?

At what RPM do you hit 9psi in 5th?

I don't know... still tuning and beta testing. maybe 3800-4000?

As long as turbo not hit block? :bowrofl:

Check out Maximum Boost by Corky Bell.
Chapter 6: Intake Manifold
Good stuff in there on manifold design.
"Turbo applications will generally find best results with long runners, which provide a broad, flat torque curve at low speeds, while the turbo keeps the top end strong."
The plenum volume should be a function of engine displacement---in general 50-70%.
Size of the throttle body is also important.
The shape of the runners should start like velocity-stack shapes inside the plenum and decrease narrower to match with the ports on the head. This effects the amount of air that actually gets into the cylinder on the intake cycle when the engine is not under boost. The inside of the runners and cyl head should be finished sanded to aprox 120-150 grit for correct laminar flow.
Now, that being said, instead of a log shape, why wouldn't you want to consider a shape more like one side of an intercooler, tapered, so you get consistent flow onto each cylinder?
Just my 2 cents.
Cheers,
-JB

Sorry, turbo exhaust manifold, not intake manifold.

My bad, that speed reading course ain't work'n out so well. :loser: