Intake manifold information

[greenberet]

Maybe I missed something here, but has anyone actually bench flowed the 2 OEM intake manifold to show CFMs across the RPM range?

Longer and wider intake runners don't always equal to improved HP/TQ if the application isn't optimized for port velocities.

Good point - I think the NA1 and NA2 intake manifolds are internally identical even though the engines produce different levels of horsepower. The NA2 engine breathes more air and therefore has a higher velocity in the intake runners. I wonder at what horsepower the stock manifold achieves the optimal velocity.

 

[dmscrx]

Sorry, I should have made it clear that I was asking about the OEM stuff.

Ok gotcha

 

[Honcho]

Good point - I think the NA1 and NA2 intake manifolds are internally identical even though the engines produce different levels of horsepower. The NA2 engine breathes more air and therefore has a higher velocity in the intake runners. I wonder at what horsepower the stock manifold achieves the optimal velocity.

According to Kaz, they are identical. For NA2, there is a measurable restriction in the intake because, as you said, the 3.2L moves more air than the 3.0 and Honda did not increase the intake size- likely due to cost. As for NA1, I would bet the optimal velocity is achieved at 270 crank horsepower, since that was the design spec.

 

[BATMANs]

When I did my engine rebuild project, I just couldn't find ANY verifiable information/data that there is ANY increase in flow rates between the 3.0 and 3.2 intake manifolds.

Any of the claims looked like something that came outta the crowds of low budget Honda 4-banger ricers that followed the culture of claiming 5 extra HP from adding stickers on their car, etc.

On that note, is there any verifiable data/information that demonstrates that the larger 3.2 valves actually adds power by itself over the 3.0 valves?

When I replaced the OEM 3.0 valves with the Ferrera valves, I had the option to "upgrade" to the larger and heavier 3.2 valves (yes, people don't think about valve weight much).

I didn't do the upgrade to the larger valves since there was just too many unanswered questions:

1. Will the port velocities slow down for both intake and exhaust (enough scavenging of the exhaust to pass the tough California smog test)?
2. 3.0 vs. 3.2 valve weight delta
3. Are the 3.2 20HP improvements more of a result of the exhaust OEM header upgrade (there are plenty of threads that appear to prove the majority of the 20HP improvement seems to come from the headers when 3.0 guys upgrade to 3.2 OEM headers)?
4. Are the 3.0 NSX heads so well designed that it can support a 3.2, if not a little more with regards to the flow rates? In other words, any valve size upgrades really don't do anything in terms of appreciable power increases, if any.

 

[stuntman]

NSX heads flow VERY well...

 

[BATMANs]

That's what I figured.

 

[RYU]

also, bigger valves matter less in FI than in NA.

i'm actually really torn with my engine overhaul situation right now so looking into all head work, intake, and ITB pros/cons. most of me wants to do a modest NA build, but even a mild NA build has a very high $/hp ratio. i might just go the easy route, keep my existing CARB legal FI setup and drop my compression below 10:1.

decisions.......

 

[dmscrx]

NSX heads flow VERY well...

They do, but there is quite a bit more to gain from proper machining. My cylinder head guy showed me flow numbers and velocities from stock to race port and the increase was tremendous. Even more with an oversized valve. That kind of setup works well on a 3.0 displacement, even better on a 3.2 and so much more on FI. Don't know what the data would be on an oversized valve with stock ports though so I cannot contribute info in that regard, but typically putting an oversized valve on any cylinder head without un-shrouding and port matching it to the valves has a negative gain. Those heads do not flow better than a K series or an H series when all are fully worked tho

- - - Updated - - -

also, bigger valves matter less in FI than in NA.

You're right, but even with my B series head going from a 33mm valve to 34 equals 60BHP at 30psi of boost so every bit counts

 

[scammy]

has anyone just done a ported manifold vs stock to compare, leaving the heads out of the equation to see what gains if any are there. I know we're all waiting on dmscrx for his manifold project but curious if any gains with the stock system cleaned up for more flow.

 

[Nero Tenebre]

I think 96cxt did a DIY thread on that

 

[RYU]

has anyone just done a ported manifold vs stock to compare, leaving the heads out of the equation to see what gains if any are there. I know we're all waiting on dmscrx for his manifold project but curious if any gains with the stock system cleaned up for more flow.

check threads from l_rao. He's gone deeeeeeep down this rabbit hole.

 

[scammy]

yeah but no before and after comparisons just on the intake manifold? didn't he gut the VVIS too? so not a direct comparison. maybe he will chime in.

 

[RYU]

yeah but no before and after comparisons just on the intake manifold? didn't he gut the VVIS too? so not a direct comparison. maybe he will chime in.

he posted results on flow bence testing different valves and porting.

 

[L_RAO]

yeah but no before and after comparisons just on the intake manifold? didn't he gut the VVIS too? so not a direct comparison. maybe he will chime in.

No flow bench tests on the intake manifold itself. Every person I got quotes from was going to charge me $$$ to build an adapter to mate the intake manifold to the flow bench in order to properly test. Portflow was willing to build the fixture for me for free but I was going to be at the mercy of their schedule (read: maybe they'll get to it next year).

 

[stuntman]

he posted results on flow bence testing different valves and porting.

He didn't post #s or anything too beneficial to the community.


Edit:

I'm not too familiar with the motors but apparently K-series motors can be ported to 342cfms and B-series motors can be ported to 310cfms:

http://tallyimports.com/showthread.php?6828-The-k-series-vs.-the-b-series-h-series&

KA24:


B-Series:


SR20:


Porsche 993 RSR:


BMW E46 M3 motor S54 (great engine):


BMW M54 I-6 (E46 330i):


VW 3.2L VR6:


Infiniti Q45 VH45DE:

 

[greenberet]

has anyone just done a ported manifold vs stock to compare, leaving the heads out of the equation to see what gains if any are there. I know we're all waiting on dmscrx for his manifold project but curious if any gains with the stock system cleaned up for more flow.

I can't remember seeing any independent measurements but Comptech made some claims for a ported OEM manifold in one of their NSX catalogs. See post #23 in this thread.

When I had Comptech carry out work on my engine, I spoke with them about their "High Performance Intake Manifold". Again from their catalog: "Simply send us your original factory intake manifold, less all hoses, electronics, etc. We will then port and polish the manifold to our exacting specifications, utilizing the same equipment used to prepare our race team vehicles." They recommended, if I didn't mind the extra cost, that I send my intake manifold to Extrude Hone instead since that yielded an even better result. So that's what I did but I have no data regarding how it compares to Comptech's hand-porting job.

In their catalog, Comptech also touched on the two kinds of porting options they offered for NSX cylinder heads. "NSX enthusiasts looking to improve overall engine performance should consider the Comptech High-Velocity Cylinder Head option, a component of our highly successful Internal Engine Modification Package ("IEM Package"), the results of which feature an increase of up to 100hp! In developing the High-Velocity Cylinder Head, our race engineers have worked hard to gain additional top-end power without sacrificing low and mid-range torque. We accomplished this by keeping the port as small as possible and removing material in selected areas, with painstaking attention being given to the valve and seat pocket area. Our special design and engineering allows for the port volume to remain small, while increasing air velocity. The results produce a harmonious balance for street and track applications. If you desire more power for racing purposes, check with us regarding our Full Race Porting option (excellent for track use, but low-end drivability is sacrificed)."

 

[96cxt]

Keeping everything else stock and just porting the intake manifold wouldn't show much gains. You would have to port the Throttle body side and have a bigger throttle body to really see the gains on the dyno.
Mine did have the dyno comparison but I also removed the VVIS since I was going boost.

 

[scammy]

I have done all the other supporting mods but have not touched the inlet manifold, so even though i am waiting on what dmscrx is doing i would still like to know what difference a ported inlet would do over a stock one, so if you have ported your heads enlarged the valves (3.2 heads) enlarged the TB (SOS or similar) this is obviously the next item. Also for the benifit of this thread if anyone is doing these mods you would have stock, ported, larger AM intake and then ITB's for choice and accompaning hp figures and obviously costs. I can get a spare inlet so will look at getting that done but having some info on what is best to do in there would be good and leaving the VVIS alone since this is the N/A forum.

 

[greenberet]

... i would still like to know what difference a ported inlet would do over a stock one ...

I'd also like to see some independent measurements. If you want to see a manufacturer's claim, please flip back to page #1 of this thread and read post #23. To quote what Comptech claimed: "On a typical stock setup, you will recognize a 5 hp gain. Combined with the IEM [Internal Engine Modification] Package, the High-Performance [ported and polished OEM, leaving the VVIS in place] Intake Manifold generates a 10 horsepower increase." But again, that's a manufacturer's claim, not an independent measurement.

- - - Updated - - -

As for NA1, I would bet the optimal velocity is achieved at 270 crank horsepower, since that was the design spec.

I agree with you regarding the velocity in the intake runners. Looking at the overall design of the intake manifold and all of the turns it forces the air through, it looks like it might have been designed to maximize the torque of a 3.0 L V-6 while making sure horsepower doesn't exceed the 280 PS limit Japanese manufacturers set at the time.

 

[scammy]

Cool thought i had read that somewhere, now the next thing is what did they do and what did they leave alone so i can get this spare one i have done.
I will do a before and after dyno on the same day as i am going to install my new Link ecu and tune the car with that then put the new inlet on.

 

[96cxt]

All I have left to do is finish up the fuel rail mounts and away we go . The manifolds that will be sold will have another alteration to them that will enhance flow a little better but this is the one I am going to run and test first.

Great job on the manifold! Do you have more pictures of the inside of the intake manifold? Would like to see how you have placed the velocity stacks, how far apart, and the transition to the runners.
Will the production manifold be a 2 or 3 piece design so we can unbolt the plenum and add spacers if needed?
What is the total volume of the plenum.

Thanks for doing this for the NSX community and sharing your progress every step of the way.
Can't wait to see the results.

 

[RYU]

Can anyone speculate what kind of numbers a variable runner length intake setup could yield on our cars? If the carrot at the end of the stick is enticing enough I might take up the challenge in the future

 

[dmscrx]

Great job on the manifold! Do you have more pictures of the inside of the intake manifold? Would like to see how you have placed the velocity stacks, how far apart, and the transition to the runners.
Will the production manifold be a 2 or 3 piece design so we can unbolt the plenum and add spacers if needed?
What is the total volume of the plenum.

Thanks for doing this for the NSX community and sharing your progress every step of the way.
Can't wait to see the results.

I have plenty of pictures but this manifold does not represent the final product so it will have to wait. This is a manifold built to test a certain design principle. If it works well, and the balance of flow between cylinders is good enough then no changes will be made. If (any) changes are to be made then they will be applied quickly for the final production run. The transition of the runners to the cylinder head will be as smooth as the design will allow. Plenum spacers as you can see would require a flange on the plenum and there isn't one so no spacer options on this design. I have probed opinions on a bigger plenum option in the future and so far there is interest. As of now, the plenum is roughly 7L which in forced induction standards would accommodate more than 1000WHP as it sits, but there is room for a 9L option by just altering the design only a couple of inches in certain areas. The bigger plenum design however would not fit under any engine covers so they would need to be removed.

 

[Nero Tenebre]

I don't think there will be any market for the bigger plenum, 1000whp will be plenty and most NSX owners will want the ability to retain the stock cover from what I've seen in other threads.

 

[greenberet]

Can anyone speculate what kind of numbers a variable runner length intake setup could yield on our cars? If the carrot at the end of the stick is enticing enough I might take up the challenge in the future

I think that's going to be really hard to accurately guess at this point. You can optimize the intake manifold for a certain:

• cylinder head geometry (intake port and valve angles - which are the same for all NSXs)
• horsepower level (flow rates and therefore optimal cross sectional areas – which vary based on your build)
• rpm window (resonances and pulse timing that determine at which rpm the engine gets a torque boost - which is a matter of preference)

With a variable runner length you can broaden the rpm window in which the manifold works best.

If you had a manifold optimized for your engine (geometry and horsepower level), you could guess what gains you'd see if you had variable runner lengths as well to broaden the rpm sweet spot. We're still missing the baseline, though: a manifold that's been optimized for a certain horsepower level and rpm window given an NSX's cylinder head geometry