http://www.stangtv.com/news/jon-kaase-explains-header-design-on-engine-masters-winner/
I've seen this engine in person, at Performance Race Industry, the year he won. The pictures don't do it justice. The article reveals, he tried a ton of different revisions, and ended up with this insane looking header design. Obviously, it's a dyno cell so the packaging was very ... liberal. The point of this article, is that he tuned it for a 4,000 rpm power band, and this was his end result. On a 409" engine, he made 722hp and 666lb/ft with a pump gas compression of 11.5:1!!!
BACK TO TOPIC:
One of the big reasons, in my opinion, for the lack of power being produced after headers are cut, welded, and bolted onto the car, is the considerations (or lack thereof) to the rest of the system. The headers are a scavenging device, and there are quite a few variables that go into the design of power boost.
Job #1 is to evacuate the cylinder of all spent air/fuel. The headers can be tuned at length (literally and figuratively), by the use of pulse waves. The sound energy is modified by heat, diameter of piping, locations of "change" which reflect the waves back into the other direction (collector locations, h-pipe and x-pipe locations), and finally length of pipe.
If Job#1 is efficient, there are multiple benefits.
Parasitic loss is reduced. The piston does not have to force as much of the exhaust out of the cylinder, as it's doing it on its own.
Reversion is reduced. The sound waves are timed outside of the parameters of the intake valve being opened, and have minimal opportunity for the intake charge (ambient pressure) to be overpowered by the exhaust charge (100's of PSIa).
Intake Charge has less fill resistance. This is due to less exhaust in the cylinder when the intake valves open.
This is the area that sets up the biggest benefit. The charge will have and maintain a higher velocity as it enters the cylinder, and will be able to fit more of the air/fuel mixture in the cylinder. Obviously, the power produced, has a direct relation to the cylinder pressure, which has a direct relation to the air fuel quantity.
The headers alone, wont maximize these benefits. Because they're going to end up being long tube, equal length, you need a mid pipe, a cross over point, and a final exhaust pipe section. They all play together, in tuning the exhaust pulses. But even the best headers in the world dont mean a damn thing, if the ecu isn't re-tuned. The factory air/fuel, is tuned AFTER the stock system was developed, and is generally, the maximized version of the shitty stock components, design limited by the company treasury department.
The fuel is timed for the stock sytem's achieved velocities, and the spark is timed for the pulses observed within the restricted system. So it is very easy to say the headers are at fault, or a waste, without understanding the system in it's entirety. It is this very reason, why people think "back pressure" is necessary to horsepower, when instead, the ecu is trying to create events in the wrong place at the wrong time. The ECU is, in other words, expecting the pulse to have to travel through the shitty terrain of a manifold log, through 2 cats, a resonator, a muffler, a couple 55 gallon drums, another muffler, a u turn, a couple more 55 gallon drums, and a resonator, for good measure, before the "factory planned" event takes place.
Long story short: Make long tube headers, and use consideration of cam timing, intake charge, rpm, cubic inches, exhaust temperature, length of tubing per section (primary runners, collectors, pipes, mufflers, etc). Once the design is there, have it tuned to take advantage of it's newly efficient scavenging system.
-Doomsday
P.S. There's no aftermarket support, because there's not a large enough market for it. Most Mercedes owners are businessmen flashing their bravado, while on cell phones working their next deal. No time for banging gears around a mountain road. Money is to be made. And lunch is to be eaten. Off their sexy receptionist's fake tits.