I can now see why it is so tough for people to make good headers that will fit nicely in the stock engine bay. It took me quite a while to get the header to actually even fit in the engine bay before I could even worry about the length and everything of each runner. Props to those who make these things. I was considering making headers but after looking at how tough it would be to do it the way I want I scrapped the idea. Here are some pics of what I made though.. 609mm true equal length header:

 

Why doesn't anyone make a SS header that is the same design as the stock one. Just out and down????

 

What would be the point?

 

that looks sick!.. what program did you use to make it?

 

SolidWorks...


I can finally see how tough it truly is to make long headers fit in our little engine bays.

 

If someone wants to use a CNC Mandrel bender you could crap some of em out pretty nicely but I dont even know if a CNC bender would be able to make some of the bends. If you could though you would be able to get the same length of the long tube header designs and still get it to fit right in front keeping the AC, PS and everything else aside from the stock fans.

 

I was going to say something about that. With that many bends it seems the price would be slightly above a full race setup.

Then you would have to take into account bracing and strength. Right after the power gains come reliability. Looks nice though and I'm sure that it could make some good numbers (I'm not a fluid dynamics expert so I don't know for sure)

 

The bracing and strength would be very easy to put in. Once I realized that to make a header a longtube header but still fit in the front without removing anything was as hard as it is I gave up on the idea. I just figured I would post up what I made for the hell of it

If someone wants me to design them something I would most likely be happy to help ya out though.

 

Here is the point...

The fact that you are redirecting the airflow while it is at it's highest velocity, right out of the head, causes the air to loose speed. The reason that the stock header is not equal length is that the two middle runners have a shorter distance to go than the outer. So what if you keep 1 and 4 the stock layout and go with a less drastic arc for 2 and 3? DO you follow what I am saying just add a bump up on 2&3 and leave 1&4 alone.

Wouldn't that work better?

By the way. I love the pics. I only get to do 2D drawings at work. I love the 3D stuff.

 

Actually, the runners don't need to be exactly the same length, they need to be exactly in phase.

While the length and diameter are important for tuning purposes (like optimum rpm range), it is crucial to avoid collisions at the collector, and the only way to do that is to make sure that each exhaust gas pulse takes the same amount of time to get from the exhaust port to the collector. Each bend in the manifold has a different effect on the velocity of the exhaust gas, and as such it changes the effective length of the runner.

Though each of the runners in this particular design is 609mm long, it may take the exhaust a split-second longer to travel through one runner compared to another, and at 8,000 rpm many things only take a split second. If you had a variance of even 0.0002 seconds you could have 2 or 3 exhaust pulses all hitting the collector at once, which would make for some pretty unimpressive dyno sheets.

I am not knocking this design since I have no idea how it would actually perform. It could very well be the best exhaust manifold ever designed for an Evo. It looks really good and seems like a LOT of thought went into it, I just wanted to point out that there are a ridiculous number of facors that go into making an exhaust manifold not suck, and runner length is only one of them.

 

I didnt get around to doing the CFD stuff yet but for some reason all the big guys run long tube headers. There must be a reason.... all logic would suggest the shortest exhaust runners thus mainting as much heat as possible as well as keeping the velocity as high as possible. For some reason this isn't the case. Once I get around to doing the CFD stuff I will and post up what I find. First I need to create a damned turbine housing to use to measure the shaft speeds, shaft TQ, etc.... At least its something to do

And if you wanted to help ensure you dont have exhaust collisions (If for some reason the CFD isn't 100% exact on the measurements) you can always run a divided housing. This would make it almost impossible to have collisions in the collector.

 

Awesome! Hey Trinababe, what program did you use for the design?

 

^^^^ it was already mentioned.



great design.. too bad it'll be costly to produce.

 

Trina, I think we should make one and try it.

 

The cheapest way I can see to make it would be to get a 3d model printed out of the design. Cut the model so each runner is seperate and then get some stainless and start welding it all together. Investment casting would most likely work as well but I dont know anything about that stuff.

Just for the fun of it I am going to make a few other design headers (Short ram, this one, very long runners, skinny and wide, etc) and start actually testing each one to see what the differences truly are regarding making power and holding power.