unfortunately no one has done a test pipe vs h-pipe dyno run (i will be pretty soon actually), so not much detail there. If you look around, most people believe that the benefits of the cross overs are lost on our engines because of the even firing order.
Thats one of those "on paper" rebuttals. People who say that don't understand that every motor develops exhaust harmonics of various orders that can take advantage of crossovers. Its the crossover design and placement thats up for debate.
In the american V8's with the unequal firing order, the cross over helps to equalize the pulses (read: scavenging) and it also helps makes it sound better because of the equalizing of the pulses. Since our engines have the even firing order (bank 1, bank 2, bank 1, bank 2, bank 1, bank 2), supposedly the cross over isn't as helpful.
That seems logical on its surface, but it doesn't explain why inline six motors with two sequential "banks" of cylinders can still benefit from a crossover - if its properly located.
Ultimately, a crossover is no different than a primary merge for two cylinders. The precise location of the merge should be where the heat value of the exhaust charge drops enough to affect velocity. Thats where the "long tube" concept has its roots (that, and the need to exploit the harmonic that travels back up the pipe after each ignition event to boost scavenging, and - believe it or not - tune the intake charge). The header collector is located precisely where additional chamber volume is needed to help boost velocity. When you crossover two collected banks, the balance has to be located where the velocity drops below a certain threshold.
However, I'm not so sure I'm buying that, I'm no fluid dynamics expert, but it still seems that the pulse from each bank could still be improved by the cross over regardless of firing order and a slight scavenging advantage can still be had. Will be it as significant as it would be on a V8? Maybe not. Either way, no matter what you go with, removing the cats will provide some benefit. But obviously adding a CBE (be it dual pipes or a larger single exit) would maximize that benefit.
You're correct on both statements.
Most factory performance models have free flowing mufflers, and simply need a cat delete and perhaps headers to get the most out of them. In the case of the 3.8, the proper way to do it would have been to have two separate bank chambers inside that back box and use a dual inlet. I can't imagine what the inside of that thing looks like - that huge muffler, but a single inlet? If Hyundai wanted to use a single inlet they could have had something half that size on there and gone with a single exit, but of course it wouldn't have sold well in the American market, where people expect dual/quad tips. Not that these cars sold well here to begin with...
Edit: And actually, if you think about it, the H pipe will be ineffective at higher RPMs, so effectively it becomes just test pipes as the exhaust is traveling too fast to take advantage of the cross over.
Thats a good observation, and one of the reasons the rule of thumb has been "H pipe for torque down low, X pipe for hp up high". The H balances pulses at low RPM when volume (air mass) is lower, the X picks up velocity when volume increases. But, while fluid dynamics certainly applies here - and its really the dominant discipline - acoustics also plays a large role. For all the talk of the air mass and how the exhaust can help the flow, you have to be concerned about standing waves and phase changes as well. (Its one of the reasons why I am suspicious of the stamped style X merges, which cross the banks at nearly a 90 degree angle. I like the siamesed X merge a lot better.) If there are
harmonics in the exhaust that can take advantage of an H pipe, it will still work even if you aren't getting significant
flow across the H.
The irony is that the X/H crossover isn't necessary if you design an exhaust strictly according to fluid dynamics. You always merge:
1). when velocity drops enough
2). when pipe diameter needs to increase to manage increased air mass. This holds true whether you're talking about a primary merge or a bank merge.
I think the biggest tradeoff in the exhaust is after that bank merge (primary collection plays its own game, and its one of the most fascinating areas of exhaust design). On paper, the best thing to do would be to merge banks to a single pipe - again, at the exact location where velocity requires it - and then on to the muffler. That presents a whole bunch of other issues, most notably, what do you do when the air mass involved requires a pipe cross section that is too large to work with? You can either flatten the pipe, like a Dr. Gas boom tube, or you split after the merge into smaller pipes. Also, since these are street vehicles that spend most of their time on the lower half of the tach (unless its mine, and then expect to get criticized here for driving at 4k rpm - note the gauge in my avatar), that single large pipe isn't going to give the best benefit at low rpm, where a street car spends most of its time, so, you split after the merge.
And then we end up having these threads..:grin: