Subaru header tests/results

[sesolcSTI]

Here is some good info for you subie owners that are thinking about going with an aftermarket header. Answers questions about power gains, spool up times, etc. Several headers are tested and the write up is very detailed with dyno graphs, etc. The write up is 6mo old or so, so some of you may have seen it already, i just stumbled onto it last night. Enjoy. :mrgreen:

http://www.jscspeed.com/forum/showthread.php?t=33&highlight=header

[Buster]

the guy says it takes longer to spool over the stock mani. Isnt that a draw back. And when they say stage 2 do they also mean with some sort of managment to be able to move that redline. If so does that mean cars like mine with just an intake, DP and UP really would lose out by spending the cash just to have a 10 hp increase at 6800?

[wgknestrick]

This test is flawed in one huge way, it uses an untuned stock ECU. If this test was done with a Utec-ed or Hydra equiped car, the results would mean much more.

Once you start getting significant mods on a Suby, the ECU starts to freak out and changing variables like fueling and timing. Problem is that the ECU doesn't have reasonable limits as to how far it is allowed to change these variables.

So basically they are changing a significant piece of the car, dynoing, and then puting the stock headers back on (which the ECU was tuned for) and showing the difference. The difference is small because they didn't tune for it. Headers have been proven to not only balance EGTs across cylinders, but also decrease them giving your engine a (less detonation prone) flow from the cylinders. This increased saftey factor is what allows headers to make more HP when you tune back to the original safety factor. Basically you can run 1 deg more timing or slightly more boost.

Headers, or a manifold will always produce more HP on ANY car over stock, given the headers don't leak and are of good design.

[sesolcSTI]

It says in the article that the cars tested were stage 2 and stage 4 wrx's. The stage 4 was running the off the shelf turboxs stage 4 map. So it was utec'd. So are you saying that even though they are utec'd it is still unaccurate because they weren't SPECIFICALLY tuned for the manifold? Or maybe you missed the utec part of the article, just wondering. :)

[capaGC8]

it took alot of support to get this comparison together. to say it's flawed is not totally wrong, but you have to think about what you're saying. having someone tune for each header could take up to 1-2 hours a piece. this is an extreme amount of $$ for a good tuner.

if anyone else remembers the thread that was years old for this on nasioc.com, they would value any comparison like this...

btw - headers are either better for low-end or top-end... this comparison was to find which one made the most power in the mid-range (average of both).

[alachua]

The other problem with this test, at least from what I read on the Nasioc Thread, was that they dyno graphs were so small...everything was pretty much jumbled together, using colors that were way too close to know which was which.

-Cliff

[wgknestrick]

I thought I read that they reset every ECU prior to runing the test too, so no learning as well.

I would've liked to have seen a smaller, more detailed test. I think they got lost in the # of brands of headers. Any small leak could provide false results of a header and when you are installing 10 or so different ones, you have to come across leaks that weren't fixed