Next Project: Ideal throttle body size?

[L_RAO]

Cool - I can't remember ever seeing the volume of the intake manifold mentioned before. Is that the entire volume of the intake plenum (top and bottom halves so basically the volume of air in the entire intake manifold assembly minus the volume in the runners)? If you measured the volume - how did you do it?

Just like in high school - filled it water and poured it into a measuring cup.

The 2.4L is for the entire manifold, upper, lower, and VVIS layer. The biggest gains, as you can imagine, occur when the VVIS is machined out/gutted.

I am also thinking about having a taller/thicker VVIS layer made.

 

[A.S. Motorsport]

I'm searching for the same cutaway of a C32A "Type-II" that's the high revving manifold for a Legend.

It has longer runners compared to a NSX.
Although Legend's being SOHC have much more room between banks

3.5L has the bell mouths on longer thinner runners.

 

[L_RAO]

I'm searching for the same cutaway of a C32A "Type-II" that's the high revving manifold for a Legend.

It has longer runners and real bell mouths compared to a NSX.
Although Legend's being SOHC have much more room between banks

I wouldn't be surprised if Honda designed them both to have the same intake port layout/design - it would be one less gasket for them to design.

 

[A.S. Motorsport]

I wouldn't be surprised if Honda designed them both to have the same intake port layout/design - it would be one less gasket for them to design.

Ports are the same for al C engines gaskets aren't since the cooling system runs different on the C32A/C35A compared to NSX.

I am also thinking about having a taller/thicker VVIS layer made.

Nope not going to work, your coolant pipe will get in the way, a year or two ago I though of doing that on a legend.
After some research (it's on the acuralegend.org forum) the issue of clearance came up.

I ended up resorting to the spacer posted above earlier, it worked nicely.

 

[greenberet]

The 2.4L is for the entire manifold, upper, lower, and VVIS layer.

Awesome. Does 2.4L include the runners? I wonder what volume they have.

The biggest gains, as you can imagine, occur when the VVIS is machined out/gutted.

Did you measure the volume displaced by the guts of the VVIS plate? Is it that much?

 

[L_RAO]

Awesome. Does 2.4L include the runners? I wonder what volume they have.



Did you measure the volume displaced by the guts of the VVIS plate? Is it that much?

Yes, the 2.4L includes the runners. My method was crude - I taped metal sheets to block off the ports and filled up the entire upper chamber.

I wasn't able to measure the difference made by the removal of the VVIS plate. Sorry!

 

[greenberet]

Yes, the 2.4L includes the runners. My method was crude - I taped metal sheets to block off the ports and filled up the entire upper chamber.

OK. So you measured the upper chamber/runners and you measured the lower chamber/VVIS and taken together, the total volume came to about 2.4L in stock form. Is that right?

 

[L_RAO]

OK. So you measured the upper chamber/runners and you measured the lower chamber/VVIS and taken together, the total volume came to about 2.4L in stock form. Is that right?

That is correct.

Joe McCarthy over at the k20a.org board makes a manifold that is better than 30 wHP over anything else out there for the Ks...and it features ~4.0L of plenum volume.

 

[greenberet]

Thanks!

When the S2000 came out, Honda wrote this about how they designed the intake manifold for its inline four cylinder engine:

For the F20C engine, an approach was taken to reduce the volume downstream from the throttle valve, yet the maximum target power could be achieved. Intake port diameter and length were reduced as much as possible. An independent straight port was used for each cylinder. This helped intake pressure to respond quickly to the movement of the throttle valve. As shown in Figure 1.1.6, the intake manifold volume could be reduced to the level of a 1.6-liter engine.​

Ferrari wrote this about how they designed the intake manifold for the F430's V-8:

The intake manifold features straight inlet tracts to the two central plenums which, in turn, have trumpets individually cast at the top of the tracts for each cylinder to ensure optimum air flow to the cylinder heads. A rotating drum - actuated pneumatically by the engine control unit - compensates for variations in the effective volume inside the two plenums to optimise the intake resonance characteristics and therefore maximise the torque curve throughout the rev range.​

A few years later, they wrote this about the 458:

The aluminium intake manifold has been lightened by reducing the wall thickness. It has short, almost straight inlet tracts to reduce losses and a system that varies the geometry of the manifold, optimising the volumetric efficiency throughout the rev range. This is achieved by incorporating three pneumatic throttle valves in the central section between the two plenums. The engine mapping provides four different configurations of the valves for optimum torque values at all revs.​

The NSX's good old VVIS system sounds rather simple in comparison but maybe the low volume of the intake manifold was consciously chosen to improve throttle response.

 

[L_RAO]

Thanks!

When the S2000 came out, Honda wrote this about how they designed the intake manifold for its inline four cylinder engine:

For the F20C engine, an approach was taken to reduce the volume downstream from the throttle valve, yet the maximum target power could be achieved. Intake port diameter and length were reduced as much as possible. An independent straight port was used for each cylinder. This helped intake pressure to respond quickly to the movement of the throttle valve. As shown in Figure 1.1.6, the intake manifold volume could be reduced to the level of a 1.6-liter engine.​

Ferrari wrote this about how they designed the intake manifold for the F430's V-8:

The intake manifold features straight inlet tracts to the two central plenums which, in turn, have trumpets individually cast at the top of the tracts for each cylinder to ensure optimum air flow to the cylinder heads. A rotating drum - actuated pneumatically by the engine control unit - compensates for variations in the effective volume inside the two plenums to optimise the intake resonance characteristics and therefore maximise the torque curve throughout the rev range.​

A few years later, they wrote this about the 458:

The aluminium intake manifold has been lightened by reducing the wall thickness. It has short, almost straight inlet tracts to reduce losses and a system that varies the geometry of the manifold, optimising the volumetric efficiency throughout the rev range. This is achieved by incorporating three pneumatic throttle valves in the central section between the two plenums. The engine mapping provides four different configurations of the valves for optimum torque values at all revs.​

The NSX's good old VVIS system sounds rather simple in comparison but maybe the low volume of the intake manifold was consciously chosen to improve throttle response.

To be fair, an I4 intake manifold is going to have much different design considerations that a V6/8.

Joe McCarthy's intake manifold is designed to take advantage of 2 sets of harmonic waves - it's more than I can explain.

The other consideration is the valve angle. Modern F1 is really leading the way in showing what will be used tomorrow - the valve angles are practically vertical (12 degrees at most).

 

[A.S. Motorsport]

a scan of the 458 manifold:

Few other Ferrari remarks:

348 uses 8 runners per side one for each valve (ZR1/LT5 style) (also has a valve between plenums)
355 uses ITB's in the same way the Lotus Esprint did
360 Modena uses a cross flowing VIS (C27A style in working)
458 uses central valves

Califorina uses BMW M style trumpets in the center plenum (curved)


348 (note the runners and valve)

360 (note Lamborghini Gallardo uses the exact same layout)

 

[greenberet]

To be fair, an I4 intake manifold is going to have much different design considerations that a V6/8.

I fully agree. As far as I understand it, separating or combining the intake pulses makes sense in V-6’s, V-8’s, etc. but not in inline fours. That’s probably the main reason why four cylinder Hondas still run intake manifolds that only have one plenum to feed all four cylinders. The design goals for the S2000 were interesting because Honda wanted to reduce the manifold’s volume as far as they could without hurting horsepower in order to preserve a sharp throttle response.

Joe McCarthy's intake manifold is designed to take advantage of 2 sets of harmonic waves - it's more than I can explain.

Let me try. Air volumes resonate at a certain frequency and to a lesser extent, at multiples of that frequency as well. If you blow across the top of a bottle, you don't generate a pure sine wave of one frequency. You generate some harmonics at multiples of the base frequency, too. The strength of each successive harmonic decreases so people often say that only the first two resonant frequencies are worth considering. Maybe that's what he meant.

You have such resonances in the intake runners and those are the most important of the resonances, from what I’ve read. The resonances in the intake plenum are also important, but not quite as much. The further a component is from the intake valves, the less its resonances will boost the engine’s volumetric efficiency. In addition to the resonances, the timing when the intake pulse from one cylinder hits another cylinder impacts power (similar to scavenging in the exhaust stream), the airspeed in the runners has an impact, how sharp of a turn the air needs to make is important, etc. Modern Ferrari engines try to take advantage of those things without having to pay much regard to production costs. Or, perhaps, reliability.

To boost power at every rpm, you could design an NSX intake manifold with straight but variable-length runners, a variable volume plenum, valves to control when the intake pulses from one cylinder are allowed to get to the next, etc. I wonder what the chances of a screw falling out of that would be over time. Probably higher than in the VVIS!

The other consideration is the valve angle. Modern F1 is really leading the way in showing what will be used tomorrow - the valve angles are practically vertical (12 degrees at most).

Yes - to be most efficient, the intake runners should shoot down the axis of the cylinder. With the 60° angle between the intake and exhaust valves in an NSX, that’s just not going to happen unless you design a new cylinder head.

While sticking with the OEM cylinder heads and the OEM intake manifold with its curved runners, FastraxTurbo lost power in the midrange when he deleted the VVIS in a naturally aspirated NSX and got almost no gain up top. If you do a before/after dyno test as FaxtraxTurbo did, it will be interesting whether your results are the same. Deleting the VVIS will banish the possibility of the engine injesting a screw in any case.

 

[greenberet]

a scan of the 458 manifold:

Adnan – thanks for the pictures. It’s fascinating to see all the things Ferrari installed in those manifolds to take advantage of intake pulses and resonances.

 

[RYU]

Adnan – thanks for the pictures. It’s fascinating to see all the things Ferrari installed in those manifolds to take advantage of intake pulses and resonances.

+1

Thank you Adnan!

 

[L_RAO]

Here's a picture of a Ferrari 599 intake manifold sitting at Joe McCarthy's workbench:

 

[A.S. Motorsport]

Here's a picture of a Ferrari 599 intake manifold sitting at Joe McCarthy's workbench:

Typical V12 configuration,

Take a look at a BMW M70/73 V12, it uses the same dual configuration i recall those being crossed like Ferrari, but Jaguar runs them in C shaped with 60 degree runners.
Lamborghini uses straight runners, like the Ferrari's above.
Very cool engine to see is the old V12 24V engines from Lambo, take a look at a old Countach or Miura.
The intake valve is in the center where the spark plug normally sits, the carburetors are placed on top of the valve-covers.
(despite the very odd configuration it's still a Hemi head, even better Lamborghini based there V12 around Honda V12 F1 engines of the early sixties) pic: http://www.huskyclub.com/engdrawDeckHeight2.jpg

That being said M70/72 BMW engines are interesting topic too.. getting carried away here

 

[greenberet]

That being said M70/72 BMW engines are interesting topic too.. getting carried away here

Yeah, it's just too bad that the short side radius of the M70/73 exhaust ports is something like five millimeters.

When Ferrari detuned the Enzo's engine to put it into the 599, I recall that they had to get rid of the straight intake runners because they made the engine too tall to fit under the front hood. So curved runners it was.

 

[A.S. Motorsport]

Yeah, it's just too bad that the short side radius of the M70/73 exhaust ports is something like five millimeters.

When Ferrari detuned the Enzo's engine to put it into the 599, I recall that they had to get rid of the straight intake runners because they made the engine too tall to fit under the front hood. So curved runners it was.

Correct the Ferrari California manifold is also curved like a BMW M manifold (V8/V10)
But if you look at the same engine in a Maserati a cast NSX style manifold is used.

Also a big problem is hood clearance in front engined cars.

What's the downfall of the M70/73 in terms of performance,
Horrible flow and valve positioning making it impossible to port.
primitive M20 Head design makes high rpm's or high comp impossible.

It's so bad that intake and exhaust system modifications have nearly zero effect.
The trick from going to a low efficiency engine to M-Style monster is swapping in M50/M52 head's with minor work your right head bolt's on but then the HUGE issue of the left head needing major custom engineering to get running.

This creates a Mc Laren F1 engine and we all know how quick they are.
I know of only 3 M70/73 engine modified this way it's hugely expensive.

Exact same story applies to the Jaguar V12.

 

[BBVNSX]

Some tuners in japan state that with ECU reprogramation + BBTB they can achieve a great improvement (up to 32 hp ). Is this possible ? if so this is a great HP gain... anyone achieved this kind of numbers with BBTB + ECU tune?

please see:

http://translate.google.pt/translat...ww.rf-yamamoto.com/products_nsx/p_nsx_19.html

or

http://translate.google.pt/translate?hl=pt-PT&sl=ja&tl=en&u=http://www.gt-rom.net/

click on NSX and then GT-ONE/THROTTLE ECU


I thought that BBTB was only good for NA2... will a mild tuned I/H/E NA1 benefit from BBTB?

 

[greenberet]

Kaz tested big bore throttle bodies for NSXs. He mentioned what he found here:

http://www.nsxcb.co.uk/showthread.php?10146-B-line-motorsports&p=94119#post94119

In a nutshell: a bored-out throttle body can increase an NA2’s power but not a mildly-modified NA1’s. If vendors claim differently, I assume it’s just marketing hype.

 

[A.S. Motorsport]

I agree with greenberet, on a NA stock to mild C30A the BBTB isn't worth it.
That gain by RFY will be on a highly tuned car. (see youtube)

 

[sparky]

Throttle body diameter based on CRANK HP
=(6.2*2+((6.2*2)^2+270*10.5853663961039*4*PI())^0.5)/PI()

Throttle body diameter based on REAR WHEEL HP
=(6.2*2+((6.2*2)^2+235*12.1619103274385*4*PI())^0.5)/PI()

using those formulas 74mm TB can theoretically deliver 316whp and 363 engine hp

 

[L_RAO]

In a nutshell: a bored-out throttle body can increase an NA2’s power but not a mildly-modified NA1’s. If vendors claim differently, I assume it’s just marketing hype.

The problem with RFY's claim is that they did not test the BBTB as the lone variable; therefore it cannot be determined how much of the improvement was due to the BBTB, and how much was due to the ECU tune.

In addition, from greenberet's measurements, the claims of 4-5 mm increase in the bore size is suspicious at best, as he was only able to measure a 2.5-3 mm increase in his BBTB. There simply is not enough material to go with a +5 mm oversize.

I agree with greenberet, on a NA stock to mild C30A the BBTB isn't worth it.
That gain by RFY will be on a highly tuned car. (see youtube)

My current set of experiences suggest that the best sequence for an NA build is:

1. I/H/E (I still haven't done the I or E, just the H) :tongue:
2. RDX injectors + ECU tune
3. Larger throttle body and/or ported intake manifold, and perhaps a re-tune? (TBD soon)

 

[Nero Tenebre]

How soon will you be doing the TB and PP manifold? Can you do them independently to test each? I don't think us DBW guys are going to find it easy to do the BBTB so I'd like to know if there are gains with just a PP manifold.

 

[L_RAO]

How soon will you be doing the TB and PP manifold? Can you do them independently to test each? I don't think us DBW guys are going to find it easy to do the BBTB so I'd like to know if there are gains with just a PP manifold.

I cannot, as I had my guy cut off the snout in order to design/fabricate a new flange to accommodate the 74 mm TB.

I will have extra flanges for sale, so if you want to test an aftermarket TB by itself, you can. :smile: