Inlet manifold design for I4 Supercharged engine

[Ponk]

Hi chaps,

I'm looking to design and build my own inlet manifold (single throttle body) for a supercharged 1.6 I4 engine. The key things i've picked up so far are as follows:

- Plenum size roughly 50-70% of displacement

- Raise the runner inlets (ideally with trumpets/air horns) off the bottom of the plenum into a direct path from the TB

- Taper the plenum towards the furthest cylinder to prevent extreme differences (e.g. nearest cylinder running lean and furthest running rich)

- Leave 1"-2" between the outer trumpets and the plenum "ends" to avoid lose of air velocity near the plenum ends.

Can anyone tell me if i'm heading in the right direction? One thing i'm especially curios about is whether the plenum needs to be tapered in a forced induction application or whether that's more for N/A applications.

[SWR Performance]

Ponk said:

Hi chaps,

I'm looking to design and build my own inlet manifold (single throttle body) for a supercharged 1.6 I4 engine. The key things i've picked up so far are as follows:

- Plenum size roughly 50-70% of displacement

The volume needed depends on use, and if this is a supercharger or a turbo?

Ponk said:

- Raise the runner inlets (ideally with trumpets/air horns) off the bottom of the plenum into a direct path from the TB

You DO NOT want any single runner to be directly in line-of-sight of the air out of the TB. If you can look straight down one port if you open the TB, you are asking to have that cylinder run lean at WOT. Air has mass and wants to keep going..

Ponk said:

- Taper the plenum towards the furthest cylinder to prevent extreme differences (e.g. nearest cylinder running lean and furthest running rich)

If you feed the air in from the "front", i.e. at 90º angle to the runners, the rear will often be the leanest.

Ponk said:

- Leave 1"-2" between the outer trumpets and the plenum "ends" to avoid lose of air velocity near the plenum ends.

I'd prefer NO rogue velocities in the plenum.. the more static - and hence higher pressure - the air the better for distribution. And a denser mass for the engine to take a gulp.

Ponk said:

Can anyone tell me if i'm heading in the right direction? One thing i'm especially curios about is whether the plenum needs to be tapered in a forced induction application or whether that's more for N/A applications.

It's the same air, just more dense. More mass at the move = good plenum distribution is the main concern.

Edited by SWR Performance on Thursday 26th January 20:21

[Ponk]

Thanks for your reply. Wish you had seen my thread in August

I gave up on this plan as i've not got the tools or the skills to make a plenum. Shame as it looks fun!

I also found this at a great price. It's from a Citroen C2 VTS and it's what Citroen use on the 185bhp C2R2 rally cars. Perhaps not ideal for a supercharged engine but undoubtedly much better than anything I could make!

[stevieturbo]

There is little reason that wouldnt work very well on a forced induction engine..

The design used by the Audi teams were interesting. And a Cossie intake is also along similar lines.

http://www.bufkinengineering.com/intake%20manifold...

[Boosted LS1]

Take a look at a saab 9000 turbo plenum. It works well enough and is quite simple. I think the plenum volume should be 3 times the volume of a single runner but I can't remember where I read that and it's probably not an absolute rule. Anyway, you won't need anything fancy.

[SWR Performance]

stevieturbo said:

There is little reason that wouldnt work very well on a forced induction engine..

The design used by the Audi teams were interesting. And a Cossie intake is also along similar lines.

http://www.bufkinengineering.com/intake%20manifold...

Only thing I would never build on a custom one would be the silly-long runners.. not enough velocity at just over tickover where that would tune to use any worthwhile wave tuning.

The "diffusor plenums"/ Twin box plenums / Lehmann-style (tons of names for the same thing) as described in the link is the only plenum design I build for boosted app's these days. Nothing else gets close to proper equality of flow per runner on high-hp turbo builds (700bhp 2-liter I4s and up). But I don't use a slit like the other "tuner from Norway" has, that's not always effective enough. Better, yes, but if you're paying £800 and up for a decent one it better WORK.

When a so-called "optimum" plenum sees the need for 5.3% higher DC on the rearmost fuel injector just to not fry a piston on a mild 615hp /823Nm I6 3-liter turbo, you get a steep learning curve.

[Ponk]

stevieturbo said:

There is little reason that wouldnt work very well on a forced induction engine..

The design used by the Audi teams were interesting. And a Cossie intake is also along similar lines.

http://www.bufkinengineering.com/intake%20manifold...

That's an interesting read. The Jenvey turbo plenum is very similar to the Audi inlets. I did look at them but I don't have a spare grand!

I looked at modifying an Audi 20v inlet too but figured chopping it about defeats the point of using a well designed inlet.

[anonymous-user]

SWR Performance said:

Only thing I would never build on a custom one would be the silly-long runners .. not enough velocity at just over tickover where that would tune to use any worthwhile wave tuning.

You might want to consider reading up on "Harmonics". Obviously all the OEM's are getting their plenum designs completely wrong, you could probably tell them how to do it better ;-)


(btw, runner length is independant of mean intake velocity)

[SWR Performance]

Max_Torque said:

You might want to consider reading up on "Harmonics". Obviously all the OEM's are getting their plenum designs completely wrong, you could probably tell them how to do it better ;-)


(btw, runner length is independant of mean intake velocity)

OEM's are looking for Helmholz effects just after tickover for Granny to not stall at every junction and manage emissions better, I do not as 99.999% of my customers couldn't care less... ;-)

[anonymous-user]

SWR Performance said:

OEM's are looking for Helmholz effects just after tickover for Granny to not stall at every junction and manage emissions better, I do not as 99.999% of my customers couldn't care less... ;-)

Are we now? Thanks for letting us know...........


So what you are saying is that even though a long intake tract has no affect on peak power, and is capable of increasing manifold Vol Eff across the speed range in multiples of the fundamental harmonic frequency, you don't use them because your "customers" would preffer a peaky and difficult to drive engine with poor low speed BMEP.

Well, that's your choice.

[SWR Performance]

Max_Torque said:

Are we now? Thanks for letting us know...........

It was a little more of a tongue-in-cheek comment than I think came across to you. Sorry if it was seen as derogatory in any way.

Max_Torque said:

So what you are saying is that even though a long intake tract has no affect on peak power, and is capable of increasing manifold Vol Eff across the speed range in multiples of the fundamental harmonic frequency, you don't use them because your "customers" would preffer a peaky and difficult to drive engine with poor low speed BMEP.

Well, that's your choice.

I am not just throwing the shortest possible runner intake on any turbo build, trust me. If possible they are tuned to the rpm peak the customer wants, second pulse if possible, third pulse if space is at a premium. And peaky and difficult is not what my customers get - unless they ask for it - I have customers laying waste to 911 GT3RS's at AutoCross in something as unlikely as a Miata/MX-5 turbo, and you don't do that with a peaky and difficult engine to drive. It's a 1.9-liter I4 using a GT3076R turbo, so no micro turbo to get response either.

[anonymous-user]

Fair enough ! Sounds like you at least have put a lot of thought and some science into your manifold designs and not just stuck in what ever you saw fit (even if perhaps your first (and probably more throwaway than it perhaps came across ;-) length comment sounded otherwise on a first read.

keep on beating those pesky porkas

[SWR Performance]

Max_Torque said:

Fair enough ! Sounds like you at least have put a lot of thought and some science into your manifold designs and not just stuck in what ever you saw fit (even if perhaps your first (and probably more throwaway than it perhaps came across ;-) length comment sounded otherwise on a first read.

keep on beating those pesky porkas

You will not belive the hour count of making an intake have proper equal distribution I have spent... so there's quite a few hours of R&D in every single one of them, never making something "because it looks good" or "I saw something similar in a magazine"..

[chuntington101]

SWR Performance said:

You will not belive the hour count of making an intake have proper equal distribution I have spent... so there's quite a few hours of R&D in every single one of them, never making something "because it looks good" or "I saw something similar in a magazine"..

Can you post some pics of the intakes you have created? Be really intresting to see them.

[SWR Performance]

chuntington101 said:

Can you post some pics of the intakes you have created? Be really intresting to see them.

I have a few... this one is the venturi section of a Sierra IS-T build of a friend..





The stacks were not done on the first pic, they look like this:



And on engine:



And a couple of a YB one for the polishing crowd..





I have loads more, but won't bore you guys with those...

[Boosted LS1]

Slightly off topic I guess but these are my LS7 TB's.





They're in Sweden having jigs made to allow future machining and to have the shafts and blades fitted etc.

Edited by Boosted LS1 on Sunday 29th January 10:01