|
|
|
|
Thread Tools | Display Modes |
03-13-2018, 06:35 AM | #1 |
Scooby Newbie
Member#: 139843
Join Date: Feb 2007
Chapter/Region:
International
Location: in front of an evo
Vehicle:2006 sti 2,5 ppp blue |
What is the right header size for my 700bhp build
Hi. I have a 2.5cdb with 99.75 je 9.2 compression custom pistons , pauter rods, ported and polished 2.5sti avcs heads, supertec +1 valves , springs etc. Power division s2 cams. Brian crower adjustable cam sprockets. Proccess west intake manifold. 75mm boomba racing throttle body, aps725 intercooler and gtx3582r Garrett turbo. I also have a killer b sump that can accommodate el headers.
What I want to know is what is the best internal diameter size on each header to have the best spool and performance. My car has managed 677bhp and 741nm torque with a 39mm internal diameter nel headers. I don't know if I should go el or nel. And what size both on headers and the uppipe?
* Registered users of the site do not see these ads.
|
03-13-2018, 09:30 AM | #2 |
Scooby Specialist
Member#: 9548
Join Date: Aug 2001
Chapter/Region:
RMIC
Location: Colorado Springs
Vehicle:2004 WRX STi 10.154 @139 @ 7,000'DA |
Whatever the Killer b header is.
|
03-13-2018, 02:05 PM | #3 |
Scooby Newbie
Member#: 408701
Join Date: Dec 2014
|
header pipe diameter should be as close to your exhaust port size as possible for smooth transition.
however tubing sizes tend to come in 1/16" or 1/8" increments so you might need to get a size up or down...if that's the case some metal-massaging to again smooth that transition is all it takes. if you want max power, go for equal length, with the longest primaries you can fit, they will help everywhere. UEL is only for the sound and "medium performance". |
03-14-2018, 12:16 AM | #4 |
Scooby Specialist
Member#: 270569
Join Date: Jan 2011
Chapter/Region:
International
Vehicle:'15 Foz XT & GT86 |
|
03-14-2018, 01:58 PM | #5 |
Scooby Newbie
Member#: 139843
Join Date: Feb 2007
Chapter/Region:
International
Location: in front of an evo
Vehicle:2006 sti 2,5 ppp blue |
I thought about it. But the killer b is 1.5" (39mm) in diameter on its primaries. My ports are 46mm. I thought it would be too small.
|
03-14-2018, 04:41 PM | #6 |
Scooby Newbie
Member#: 452601
Join Date: Aug 2016
|
I've been told the exact same thing looking for headers with bigger pipes now
|
03-14-2018, 10:37 PM | #7 | ||
Scooby Guru
Member#: 33782
Join Date: Mar 2003
Chapter/Region:
MAIC
Location: Milton Keynes, UK
Vehicle:96 3MI Racing search FIRST, then PM!!! |
Quote:
Quote:
To the OP, to be honest, most of the Subaru community isn't to this level simply as it isn't needed. KillerBee makes a great header that can survive power. I state that it is has been proven to withstand some serious four digit power figures. Is it the most efficient at those power levels? It could probably benefit form a larger diameter but Chris' goal was to reduce spool time and make more power everywhere, which is accomplishes. To go to a big diameter header to be more efficient at higher RPMs, and big power, will naturally increase lag through decreased exhaust gas velocity and more volume to fill. With a turbo in the system, many of the NA world's rules change a bit. Certain things matter less, as more boost is an easy adjustment for most users. It isn't until you get to the higher echelons of motorsport that people are fighting for those small gains. |
||
03-14-2018, 10:41 PM | #8 |
Scooby Newbie
Member#: 139843
Join Date: Feb 2007
Chapter/Region:
International
Location: in front of an evo
Vehicle:2006 sti 2,5 ppp blue |
The question is how big do we want it exactly? And how long we need the primaries? We don't want too big otherwise the velocity of the exhaust fumes will reduce and we loose performance. Also if we get the length right we make the right back pressure to empty the cylinders after each cycle to fill them again fully withfresh air and fuel. Surely we must have an egg head here to do the math for us. As i am too dumb for things like this.
https://youtu.be/RWTARjxiqlo |
03-14-2018, 10:52 PM | #9 | |
Scooby Newbie
Member#: 139843
Join Date: Feb 2007
Chapter/Region:
International
Location: in front of an evo
Vehicle:2006 sti 2,5 ppp blue |
Quote:
|
|
03-15-2018, 10:07 AM | #10 |
Scooby Guru
Member#: 33782
Join Date: Mar 2003
Chapter/Region:
MAIC
Location: Milton Keynes, UK
Vehicle:96 3MI Racing search FIRST, then PM!!! |
I would have the initial pipe diameter be as close to the port size as you can get but try to avoid being smaller than the port. That's just bad.
As for an egg head, I'm one of them. If you knew more about it, there is no ideal exhaust. It comes down to a lot of variables specific to each engine and what you're doing with it. If making peak power is your goal and transients don't matter, then the obvious goal to to make it as efficient as possible. If you want the best spool, scavenging be damned and make the tubing as short as you can (see KillerBee). Turbine size, location, cam size, exhaust port flow data, etc, etc, etc....there isn't a simple answer to this. Even in racing (NASCAR and Ford GT) we did 1D calculations, backed into CAD, and validated on the engine dyno. Obviously the NASCAR stuff this was more important. In the GT, everything is BOP, so we made a good set up and left it alone. No point in spending a ton of time optimizing when IMSA and the FIA would just penalize you for it. Heck, that engine practically ran zero boost at redline by the time we were done with homologation. BOP killed the race engineer's fun. I will say the turbo cars are far less 'header' driven performance than an NA car. They do however still benefit from the same principles. As for your spool issues, that can come from a myriad of other things too...intake manifold size, intercooler volume, intake plumbing volume/length, etc. All of this is also neglecting the tune and shear size of the turbo you've chosen. |
03-15-2018, 12:35 PM | #11 | |
Scooby Newbie
Member#: 408701
Join Date: Dec 2014
|
Quote:
but its a lot of work gluing together 2-3-4-5-6 different sizes of tubing for a few percent more power at X, Y and then Z RPM thru improved scavenging, dunno if OP is chasing max-effort (but his build sheet suggests it). |
|
03-15-2018, 12:45 PM | #12 |
Scooby Guru
Member#: 33782
Join Date: Mar 2003
Chapter/Region:
MAIC
Location: Milton Keynes, UK
Vehicle:96 3MI Racing search FIRST, then PM!!! |
Totally agree that stepped tubing size in a primary (expansion) is a tuning tool, just like the lengths in your expansion pipe spacing (percentage of primary length). Actual step inserts and collector angles are also running tools.
The whole header is really a tuning tool. Sent from my SPH-L720 using NASIOC mobile app |
03-18-2018, 03:38 AM | #13 |
Scooby Newbie
Member#: 139843
Join Date: Feb 2007
Chapter/Region:
International
Location: in front of an evo
Vehicle:2006 sti 2,5 ppp blue |
Thanx for the input everyone. I have decided to sort of replicate the killer b headers to my spec . So it will be 42mm id equal length , shortest possible primaries (as i want quicker spool) and similar primaries geometry. Will also get it ceramic coated to keep the gases flow faster.
|
03-18-2018, 07:34 PM | #14 |
Scooby Guru
Member#: 33782
Join Date: Mar 2003
Chapter/Region:
MAIC
Location: Milton Keynes, UK
Vehicle:96 3MI Racing search FIRST, then PM!!! |
Circa 2010 I calculated and drew up the exhaust manifold for the Turn In Concepts TA car. It was a short 4-1 set up. Did the usual cylindrical firing order into the collector. It was a large primary with a slight expansion on the primaries. Worked really well for the front mount GTX3582R. If I recall boost was full tilt at 4200 RPM on the big cams (272/276) and large volume intake manifold.
Don't recall exhaust sizes and lengths off hand. Heck, the work was done on an old and dead laptop at this point. Knock it out. Use good materials, pay attention to detail, and you should be happy. |
03-18-2018, 10:49 PM | #15 |
NASIOC Vendor
Member#: 198281
Join Date: Dec 2008
Chapter/Region:
MAIC
Location: Virginia
Vehicle:2005 WRX/STi WRB of course |
Have you guys actually tested stepped headers on a turbo application? I would recommend it all day on an N/A car or motorcycle with some decent rev capabilities. On a pressurized manifold you're going to see small gains over a small RPM range, and losses elsewhere. Unless you're making M-A-N-Y sets of headers, or plan on running some simulations on a +$100K simulation program (Nastran, ansys, etc...), you're just throwing darts at a wall.
IMO you're going to benefit most by using a tried and true design and doing everything you can to optimize energy transport (sound, heat, pressure pulse, flow, etc...) from port-to-turbine. If you're concerned with your port being 39mm and our primary being 38mm, that's a super easy one to solve. First, our primaries being welded on both sides are going to be pulled open to fill the flange hole clearance. In other words, they should be a wee bit over 38m. On top of that, we give all the ports a grind and blending finish that will also open them up a wee bit too. Just for argument sake, let's say they aren't. To make them go from 38mm ID to 39mm ID is removing less than .020" from the material thickness. That could be blended in, in such a subtle way, that you wouldn't even see or feel it. S-m-o-o-o-o-t-h. It's arguable, that increasing the secondary from 2" ID to 2 1/8" ID would be a performance benefit.... BUT... only if you're going T4, not T3. T4 is more typically on landspeed cars making +1,100whp. Big primaries are going to make decent torque, but flat elsewhere. Companies calling the 'big tube' headers 'race' headers is more marketing ploy of bigger-is-better than performance enhancement IMO. Why do I hold my ground on the above? While transitions, larger whichever tubes and so on may attribute to gains under a small targeted area of the rev range, the increased losses in the form of sound, heat, pressure pulse, flow, etc, as mentioned above detract from performance in all other areas of the curve. Is our off the shelf Holy Header the end-all-be-all of EJ header design. Heck no! It has to conform to space and fitment constraints. The Low mount header is a pretty decent upgrade about that, and beyond that you are talking a tube chassis car with an open front end, kit car, aircraft or whatnot with a custom space allotment. We might have a used header coming available soon. It's been on the car a while and I don't recall if it's a V2 or V3. If you wanted to make your own setup, it would give you a good starting point to hack-and-tack from. Here's an example of one of the best setups I think we've ever done. This was many years ago, pre-cast collector, but did the job very well. |
Thread Tools | |
Display Modes | |
|
|