I have been spending some time looking at different online calculators for intake manifolds, and reading as much as I can about runner length, taper, and plenum design. I also recently stumbled into access to a stratasys printer for cost of materials so prototyping parts with a high degree of accuracy is doable.

I wanted to start this thread to pull information and opinions from the community about what a naturally aspirated intake manifold should look like. I think more air intake development based around the throttle body size and plenum size would also be a good thing to look into.

My current thoughts on the working parameters look something like this:

Peak torque at 6000 RPM?
Stockish intake ports
Stock vvt cam
Racing beat header
Skunk2 TB

The big design considerations are:

*Which harmonic wave to use - I think comparing second and third waves, regardless of packaging would be interesting.
*Runner length
*Runner Taper
*Plenum Volume

I am planning the part as a multi piece carbon design with a aluminum flanges. That would allow for almost any geometry.

I have used several of the calculators available online and they tend to give really different results. Ultimately, things like runner length will need fine tuning at small increments on the dyno but it would be good to have a starting place that makes sense.

I took measurement of the intake ports on the VVT head and got a measurement of 98mm port depth (taken from center edge of opening to valve) and 1285 square mm (1.99 square inches) of port area.

 

i have been doing a ton of digging on custom N/A manifolds as i am still looking for someone to make me one.

What i have gathered

The air has to have a straight shot behind the TB (I expect most gains to come from this alone from other projects iv done with NA8 manifolds)
plenum should be about 1.7 liters or so.
something like 11-12.5 inch runners
runners about 50-52 mm id
the plenum should have a 3.5 degree taper

 

50-52mm ID runners has to be too large. The stock throttle body is 57mm ID.

Twibs, can you share how you came to those numbers? Like what cam duration and target peak torque?

 

seems crazy but it would taper down to the port size on the head.

squaretop manifolds have 50mm ports at the top. not so crazy anymore?

iv been looking at manifolds and other peoples thoughts and designs i havent punched anything into a calculator because i cant even find a proper flange for the head so theres a slight roadblock.

i concluded that the squaretop isnt that bad in the grand sceme of things and that the only improvements to be had on it are a straightened throat slightly more volume and velocity stacks.

 

Also most of us arent and wont ever run cams with a ton of duration >272 because the choices are slim and most people with the exception of a few want some type of idle.

if anything, calculate for msm cam and/or tomei cams. that would cover the majority of peoples setups


just calculated it out using msm and or tomei cams looks like 12.35 inch at the fourth wave at 6000 rpm using a 50mm runner and an edc of 493. pm me for more maths!

correct me if im wrong anyone!

 

I'm out of my depth trying to work the math on this one, but it sounds like your trying to build what basically amounts to an evolution of the square top. Seems like starting with those dimensions and a straight shot behind the TB would be an advantageous place to start.

The Honda manifolds are cool, but something with the correct runner length is much more appealing to me.

 

I PMd you for your math. 12.35" includes port depth or is that starting from the head flange? 50mm runner diameter includes the bellmouth or not? I have a squaretop and a VTCS here, I will open both of them to take measurements as useful reference points. My understanding is that throttle body size should affect plenum size since it will fill at a different rate. I would prefer to ignore the stock throttle body and only work around using a 64mm TB.

I think for testing I would want to get a design within a 3" runner length range, and then make a 3d printed testing part where the runners are split near the flange and use a slip fit/tongue and groove type joint with interchangeable spacers. That would allow for quick A/B comparisons without unbolting anything.

I could knock out a polygon model in 3dsmax in about an hour but I want to do this in solidworks for doing CFD and potentially machining things and I am a lot slower in solidworks.

I should also find a way to measure intake manifold temps near the head. Maybe a good excuse to get an IR thermometer. I think you would need a slightly more specialized epoxy system to handle how much heat the block would radiate into the carbon. Pro-set makes a cost effective resin/hardener combination that will do 286F with post cure.

 

That would be great but it would be complex and you would likely run into issues keeping the damn thing together through a dyno session. mazda already did a great job on the manifolds, IMO you would be better off trying to develop a new upper half for the squaretop

 

Yes, This and bell mouth the openings.

The same could be done with any of the manifolds.

 

Using the math in that PDF you forwarded me, tuning for the third wave, with a runner diameter of 50mm, and a 242 duration cam, peak torque at 6000 RPM I got a runner length of 13.56" (after subtracting 98mm for the port depth). I think that could be packaged with straight runners after moving some stuff around in the engine bay.

Edit: Checking with a tape measure, looks doable but the plenum would be really close to the shock mount, washer bottle would need to be moved, and it might require NA headlight removal to run the intake tubing. Alternatively, nothing says the TB cant be on the back side of the engine bay and the tubing run under the car from there.

The fabrication plan involves doing each manifold in 5 parts. 2 halves for the plenum, and each runner on its own. It would only be about 20% more work to do two sets of runner lengths, one for third wave, one for fourth wave. I am thinking the runners would all use the same mold so that would only require one more small mold.

 

A new upper manifold/plenum is an interesting idea. It would be very easy to change or swap for smog purposes and it sounds like it would be much easier to make. Packaging might be a challenge though.

 

Not at all...

https://www.roskoracing.com/cgi-bin/...on&key=300-035

This guy does some similar work on the Honda H22 intake manifolds with success. To give a more visual idea.

 

Another more extreme example.


H22 cnc'd plenum spacer and custom intakes - Honda-Tech

 

I just opened my squaretop up to measure the runners. I got a measurement of about 12" with 11.5" on one half. A replacement upper half with a different plenum size and good bellmounts would be doable, but not necessarily less work than a full IM. Definately easier to install though. You would lose almost all of the ~8lb of weight savings that would be possible with a carbon IM. On the plus side, the system as I imagine it would be modular so the same plenum pieces could be use for a half manifold as a full one. This would probably be a good starting route.

 

Sheet metal upper may be the way to go.

 

I think sheet metal is a poor choice for a plenum. It's hard to get smooth compound curves.

 

I'm not sure if 6000rpm is the right target rpm to calculate for. Look where maximum power happens on a naturally aspirated Miata.

 

I am open to suggestions for a different peak torque RPM.

 

I think your taking light of what it will require to build a durable manifold for a reasonable price.

The flat top runner length is not far off of your ideal size and length and is certainly a great starting point. My least favorite thing about the Mazda manifolds is getting to the lower bolts 😤

 

That depends HIGHLY on cam selection. The MSM/ExIntake is very mild and likely would only be hurt by moving up peak TQ in the manifold.

Has anyone ever taken the time to measure the 99-00 manifold plenum volume with and without secondaries open?