seeking header design help . . .

[mikedc]

I'm making a set of Tri-Y headers. I'm trying to figure out a lesser-of-evils issue for airflow.



Engine: Low/midrange torque is more important than high-RPM horses. Closer to a truck motor than a racecar.



The classic SBC/Corvette "Rams Horn" manifolds are known for flowing quite well compared to smaller iron logs. But my gut reaction to the design is that it must really choke off the inner two cylinders, with the long outer tubes landing on the inner exhaust ports like that. There's basically no primary length on half the cylinders and a significant primary length on the others.

I guess what I'm wondering is, once the possibility of near-equal length tubes is off the table, does it still make things a lot worse to have that big mismatch? As opposed to retaining a couple inches of length on the shorter primaries? The extreme mismatch must be less harmful to airflow than I expect or else the Rams Horns would suck.

[astroracer]

The thing you need to take into consideration is the timing sequence. SBC's normally run 18436572, The only cylinders that may cause a flow problem are 5 # 7 as they fire sequentially. With 5 having such a short run it may actually help pull 7's charge through the pipe.
There is so much going on, dynamically, in an exhaust header that you can't trust just your eyes to "see" what is going on. On a mild, torquey small block the rams horns work almost as well as the shorty or mid-length headers. They will perform well enough up into the higher RPM's that you really won't see much of a difference, power wise. A good set of long tubes will be better over all for low-end torque AND a bit more upper RPM power.
Mark

[mikedc]

Thanks for the response.

I'm considering a Tri-Y setup that pairs off the primaries ideally (1+5, 3+7 on the left, and 2+4, 6+8 on the right). But packaging concerns in a couple of places make it tempting to mimic a Rams Horn pairing style, with one cylinder's long primary virtually passing over the exhaust port of the other cylinder in the pairing.

I'm wondering if that non-ideal pairing/merging design would be so detrimental to airflow that it would override the gains of having the cylinders ideally paired off. (The implication being, that I would be better off with less-ideally paired cylinders but they all get some amount of primary length.) As you say, there is so much dynamically going on that it's hard to know.


My general feeling about (street car) headers is that the tuned length issue is mostly a bunch of race-car malarkey. I mean it won't apply unless the whole header design is pretty close to ideal.

Most of the OEM exhaust manifold designs (cast-iron or tubular) seem to crap all over any concept of matching primary lengths. It seems like their only concern is getting adequate raw passage volume to get the gases out, period. That, and avoiding really tight/sharp turns. I know the OEM stuff is very compromised for space and HP ranks low on their priority list in general. But their designers aren't stupid. They will try to avoid the worst airflow offenses just for MPG reasons if nothing else.




Is there any good reading material or practical software to simulate this stuff?

I've seen lots of info & talk on header design. But it all revolves around taking a nearly ideal setup and squeezing 3 more HP out of it. There isn't a lot of help to be found for the Tri-Y setup at all, let alone when you start doing compromising stuff with the primaries.

[astroracer]

Look for a book by David Vizard. He does an excellent job of outlining the how's and why's.
here is an article that uses a lot of his research: http://www.enginelabs.com/engine-tec...-up-the-power/
If you Google exhaust design there is a ton of stuff out there.