355 Exhaust Manifolds (Headers)

I my opinion, unless you are doing all out racing, where you will be in the 8000rpm+ range all the time (i.e. oval racing) it's better to have tri-y headers vs 4-1 headers. The 4-1's make more power at the top end, and well designed tri-y's make more power in the mid range and are a few hp less than the 4-1 at the top. The factory already runs a tri-y header on the 355 with a non sequential pairing. Meaning that the primaries, those are the pipes coming right out of the head, are paired with a cylinder that is not next in the firing order. Using a firing order of 1-3-4-2, cylinders 1-4 and 2-3 are paired. This type of pairing gives decent midrange power and still makes power up top.

The headers on my 348 are paired sequentially, meaning that the primaries are paired with the very next cylinder in the firing order. But there are two different pairings you can do. You can have 1-2 and 3-4 paired, or 1-3 and 2-4 paired (this is how mine are paired). The shop that made my headers (Hytech) found that the 1-3 and 2-4 sequential pairing makes the best all around power, and really shine in the mid range.

Then there are the issues of tube diameter, length, and merge angle. All that stuff can get very complicated. It can deal with supersonic and subsonic air, sound waves, reversion, harmonics, and loads of other things. So I just left all that to the guys who built my headers, as they have done hundreds, if not thousand of hours of dyno and track testing.

 

Actually, the CFD simulations showed that from 5500-9000 RPMs (i.e. top end), when both (all three) header designs have been optimized/tuned for <the same> top end, that both 4-2-1 and 4-1 headers produce exactly the same TQ curve. In fact 4-long-2-short-1 and 4-medium-2-medium-1 also create the same top end--given that they were tuned to top end. They require different tube sizes (by 2mm) lengths and slightly different expansioin chamber sizing, but they do produce the same power at the top end.

Its comming 'on-song' (3000-4500) where the 4-long-2-short-1 beat the 4-1 andalso beat the 4-medium-2-medium-1.

And BTW the header data I can show having the best over-the-whole power-band are 4-long-2-short-1 style.

 

Why do you think I'm running a tri-y. You've seen my headers, so you know.

 

I don't, and I have seen your headers. I was responding to the mid-range header versus the top-end header argument only.

 

At the very top end the 4-1 do make more peak power. Especially when you are spinning the crap out of the engine. But it's still better to have a well designed, tuned tri-y, for the street/track in my opinion.

 

Mitch,

As an engineer this discussion seems to be extracting the last bit of HP/Torque which still does not mean lower laptimes. But that aside now reliable are the CFD computations? What compromises have been made in that programing? Is the science of that so known and easy that it can be applied to the 355 engine vs the SBC (chevy) for example. Maybe there are some fluid dynamics of how the gas comes out of a 355 that is different from a SBC and then the results would be not useable. Example that scares the cr@p out of me is guys flow ferrari heads and start grinding on ferrari heads but using what they know from SBC experience because that is what they know.

 

I don't know the answer to that question. However, the <comercial version of that> simulator I am using has been used to design F1 engines.

The major limitation is that it is a not a 3D CFD simulation. It does have a lot of ancilary "devices" that are used to better simulate the air/gass flow through the various accoutremen. For example, I do specify the join angles of the merge collectors (and those in the Air intake system). The valve sizes, port diameters, port lenghts, cam timing, ramp rates, throttle openings are all modeled. The only thing I cannot really model is the 3rd intake valve running 10 degrees behind the other 2 intake valves.

I worried a lot about this in particular. However, using my own set of headers and the measurements I got from them, and the measured parameters of the F355 air box and velocity stacks, the simulator produces a tourqe curve that looks surprisingly like that measured when a F355 is on a dyno.

The following is the output of the F355 engine simulator when not running the F355 air intake system. This isolates the effect of the header from the rest of the engine componetry.

http://www.freeimagehosting.net/newuploads/vsiao.jpg

Can you determine which of the colored lines is the 4-medium-2-medium-1; the 4-long-2-short-1; the 4-1 header?

 

Which of these headers are you currently running on your 355?

 

I am currently running factory headers with the packing material removed from under the heat shields.

I am getting ready to build a set of the 4-long-2-short-1 headers. While I wait, I am contemplating doing an exhaust that does not incorporate the bypass y-pipe (ala 348, 360).

 

I believe Goth did an exhaust system for his 355 in which he eliminated the secondary bypass pipe and used 360 exhaust manifolds with good results.



Jimmy

 

I was wondering why the companies building aftermarket exhaust systems have not offered a system more like the 348 or 360 for the 355. Are there any real advantages of having that secondary bypass pipe? The cons are increased weight, bypass valve problems, and possible heat close to the air boxes.


Jimmy

 

If you do a google search :[Gothspeed or Goth exhaust] there was a video with some engine compartment shots without the secondary pipe and a test drive.


Jimmy

 

I have a set of 360CS headers posted for sale. http://www.ferrarichat.com/forum/vbclassified.php?do=ad&id=8634

These are brand new parts I purchased from Daniel Pass at Ricambi. I was going to use them for a killer Goth Exhaust but the project never came together and I ended up selling the car to buy a 16M.

Let me know if you have any interest.

Rob

 

Technut was there suppose to be any increases in hp and tq with system?


Jimmy

 

Yes that system worked very well !! It made just over 18 RWHP with factory CS manifolds, full cats and no CELs on the 5.2 motronic ......... knowing more now, I think that bypass delete configuration can make a decent measure more!!

Here is the sound clip vid for that system;

http://www.metacafe.com/watch/2729518/custom_ferrari_355_exhaust_by_gothspeed/

Lately I have been focusing my time on a monster system that is compatible with stock headers ...... but if one also needs manifolds, there are a few ways to delete the bypass. one can do factory CS manifolds, fabspeed 360 manifolds ..... both would need air rails if desired.

I have also been working on a manifold design that would improve low to mid power .... and have the peak gains of an optimised 4-1. It has been in the works for a while ....... and when all is a go ..... it may also get a dyno chronicle.