Which header/extractors would you choose?

[70XIN]

Yo,

I am in the wonderful position that i need to make some headers/extractors for my car. I don't wish to reuse the old ones, and i don't want to fork out and actually BUY a set, especially since i can weld .. sorta. I've trawled the interwebs for a LONG while, and i've narrowed down my choices to a small list. I AM limited by the designs i choose, as i don't have a pipe bender, and will be using pre-bent 45/90/180 degree mandrel bends. There are a few *awesome* headers that use very long-radius bends, but they are not an option for me, so please don't post up pictures / 'why dont you do these'.

So anyway, simply say a number (or two, or three) if you prefer certain headers, or even leave comments or criticisms. I need to make up my mind, and trust me, i damn sure can't do it on my own. I'm asigning the headers by numbers, not names, as it's easier that way.

Love Seamus

1. Possibly the easiest to make out of the list. Looks good, but might be close to the washer bottle (JDM) that sits on the strut tower.

http://www.ae86drivingclub.com.au/dcimages/1/6/64986.jpg


2. Very similar to the above (made by the same business), but aiming forward. Looks a little bit more difficult for oil lines (no biggy), but moves the headers away from the washer bottle.

http://www.ae86drivingclub.com.au/dcimages/1/6/64987.jpg


3. Badass as hell, from the looks of it - much harder to make than most of the others.

http://img848.imageshack.us/img848/9849/ae111047.jpg


4. I kinda like them, but i'm on the fence. Something looks weird about them. Possibly difficult to make if i weld the pipes in the wrong order.

http://www.ae86drivingclub.com.au/dcimages/1/6/64989.jpg


5. Cool, but dunno.

http://www.ae86drivingclub.com.au/dcimages/1/6/64990.jpg


6. Similar to number 1 and 2, but runs the pipes 'in the middle' instead of forwards or backwards. Probably the one i'm leaning towards most at the moment.

http://img851.imageshack.us/img851/7140/post17411168079551.jpg

[fantapants]

fwiw mate, i have heard from a few different people over the years that have had the oppurtunity to test such things, that keeping the initial header as straight as possible for as long as possible makes big gains.

but fwiw, i like number 3 :)

[Gunner]

What about these ones, off my/magic mans old set up?

http://www.ae86drivingclub.com.au/dcimages/65005.jpg
http://www.ae86drivingclub.com.au/dcimages/65006.jpg

Supposedly made some good power, I'll try and get some specs if you are curious.

[GOT-20V]

I love number 4, have always had a fascination for that style.

[marvis]

1 and 3.. Rest i'm not a fan.

4 is cool, but they look like they come up heaps.

[Todd]

i like 3 an 5.

[blake.]

Number 3 mate

[R&D Mechanical]

1 and 3

[70XIN]

cheers guys, that gives me a bit to think about :)

[trikzlane]

yo tox, no.1 is my fav, cause its not to complicated to make and the touge range would be awsome, + i kinder fills the empty space on the exhaust side of na 4ag's

[Low Style]

1 and 3

[Vance]

go no.1


2 and 6 are cool also but look like they were designed to get them further away from the master cylinder on a USDM car. 1 is obviously on a USDM car too but its bloody close to the master cylinder so im guessing theyre for a RHD car.

3 are badass as hell but i wouldnt even try to attempt them myself or pay someone else too. they look bloody hard to make

[Sam-Q]

Seamus what experience have you had in making extractors? also is it for looks or power?

[70XIN]

Sam, absolutely none. I've made a couple of turbo manifolds (of the 'ramhorn' variety, as they call them in the states), but this will be my first header.

And a bit of both - i won't deny that i'm a bit of a ricer at heart, but performance is really what matters here. I've talked to various fabricators (in australia, and abroad - including john from jspfab, and a few other hondatech gurus) and they reckon the runner-length of the headers above are fairly decent (long). I can't remember what the length was that burns stainless suggested, but i seem to remember they suggested that pipes would run all the way to the cat haha.

One of the few things i've been told i could do 'better' with a motor of this size/spec is use a slightly larger diameter runners (i'm only going to use 1.5"), but getting 1-3/4" oe 1-7/8" for a reasonable price (in stainless) proved to be very difficult. I'm also using a proper slip-fit collector (like the burns stainless type), that will neck down to 2", and transition back to 2-1/4 or 2-1/2".

My main two concerns are of course, (a) failing, being the first time i've done a header and (b) working out if i need to backpurge the welds, or if headers are a little more lenient than manifolds -> i don't have a setup readily available to me, and CBF setting one up

[Gunner]

My experience has taught me 2 inch after the collector, is too small for a 20v, have to remember that these things move a hell alot of gas in the top end.

When I had my 111, and had access to a dyno on a daily basis, i went through 5 different mid sections. Lowest power across the board was a transition to 2inch from a 2.5 collector, transition to 2.25 and then 2.25 the rest of the way, no centre muffler. made 88kw

Most power at 96kw (std bt 6spd, with cold air feed and exhaust, with only changes to the exhaust) 2.5 collector, transition to 2.25, 2.25 to the cat, 2.5 from the cat, to resonator and 2.5 for the rest of the system, engine was far more responsive, and you really felt the vvti come off up top, enough to start to turn the tyres in the top end of 2nd gear.

Thats what I found any way.

[Gunner]

Also I have seen a few clubman guys, make good power with 3 inch systems (110+), also beau (Magic Man) was running 3" on his n2 replica, and apart from head work that thing was STD.

[xero]

Sam, absolutely none. I've made a couple of turbo manifolds (of the 'ramhorn' variety, as they call them in the states), but this will be my first header.

And a bit of both - i won't deny that i'm a bit of a ricer at heart, but performance is really what matters here. I've talked to various fabricators (in australia, and abroad - including john from jspfab, and a few other hondatech gurus) and they reckon the runner-length of the headers above are fairly decent (long). I can't remember what the length was that burns stainless suggested, but i seem to remember they suggested that pipes would run all the way to the cat haha.

One of the few things i've been told i could do 'better' with a motor of this size/spec is use a slightly larger diameter runners (i'm only going to use 1.5"), but getting 1-3/4" oe 1-7/8" for a reasonable price (in stainless) proved to be very difficult. I'm also using a proper slip-fit collector (like the burns stainless type), that will neck down to 2", and transition back to 2-1/4 or 2-1/2".

My main two concerns are of course, (a) failing, being the first time i've done a header and (b) working out if i need to backpurge the welds, or if headers are a little more lenient than manifolds -> i don't have a setup readily available to me, and CBF setting one up

7OXIN, what you want has pretty much described my whole exhaust system. long (over 1000mm) runners, smooth curves, 1 3/4 dia. runners into a 2.5" collector and 2.5" system

i wont lie, you would need a big motor to make this work but so far it certainly has shown there is a lot of potential in it.

[Rice86]

make 3, 4 and 6

and sell them if not needed

[70XIN]

Thanks heaps gunner and xero. If i hadn't already bit the bullet and ordered all of my piping and collector, i would definitely have stepped up to 1 3/4", i guess i should've asked on here as well first.

Ah well, this will be good practice nonetheless

[Celica RA45]

i would make a 4 .2 into 1 set up .what ever the port is, copy that as flange plate and go say 1and 1/2 inch for 800mm long then 1and 3/4 for say 300mm and then up to 2ich dia and then use taper cone up to 2 and 1/2 inch for your exhaust
i have not found a 4 into 1 set up better than my race trd pipes which are 4 into 2 and then 1

[xero]

could it be considered that given the length of your 4-2-1 headers is the biggest factor there?
do you think that if the pipes were much shorter they would make the power they do, or would it be a restriction?

7OXIN, as rice has said have a go at it, see what happens, if its not to your liking sell them off to cover your costs and have another go. i am genuninely interested to see how this turns out :)

[xero]

doppelposten farken.

[Shifty]

I have nothing technical to add to your decision, but the headers marked #3 (with the bike in the background) look absolutely fantastic. I also like 1 and 4.

Honestly with most of these things as long as they are a reasonably sound design then the final direction etc has little impact. Something I don't hear discussed enough is the length of primaries, secondaries and the final exhaust itself. On rotaries for example, massive gains are available by running the primaries (remember there are only two) all the way to the diff and then join into a single pipe over the diff and out the back.

Try and get in touch with some racers / workshops who have experimented with different exhaust designs on a similarly modded 4AGE.

[Gunner]

On that note, there is a video of an NA porsche engine (havent seen it for years, dunno if its still about), on a dyno with 10m of exhaust, made huuuuge power over the standard stuff

[Celica RA45]

i have played around with 5 sets of different lengths even with slip joints on the dyno to do back to back ,hell i even changed ram tubes have run 4into 1 from 31 inches back to 21 and have found 26 seems to work the best and from motor to motor the length of inlet can change over from 10mm up to 120mm in ramtubes ,but at the end of the day the trd 1s make power from idle all the way up till red line .the 4 into 1 make the same power but make a big power drop from 2 to 5and also torque
where there is no drop in power or torque with my pipes that i use

[Shifty]

What are people's thoughts on header design/layout to maximise low-down/midrange power/torque?

[xero]

a simple 4-2-1 design, 1.5" diameter runners around 600-750mm long, with a 2.25" collector and 2.25" system.
a similar set of shorter 4-1's should also do the trick.

[70XIN]

Okay, so i have finally finished making them.

It ended up taking about 30 hours, maybe a little less. Only took about 6 to make the design, then another LOT of hours adjusting them so they fit perfectly and were equal length (honestly probably 20 hours), then another 3 or 4 to weld them.

The only thing i'm SLIGHTLY annoyed about, is i didn't expect the actual pipes themselves to warp much under welding - they did enough so that two of the runners were touching each other - after a tiny bit of persuasion and grinding, i now have about 0.5mm gap - hopefully stainless doesn't expand enough for them to touch during driving conditions?

http://img847.imageshack.us/img847/2905/manifold1.jpg

http://img851.imageshack.us/img851/4374/manifold2.jpg

Anyway. Runner length ended up being 855mm - with less than 10mm of variation between all four pipes (smallest was 852mm, largest 859mm). Just for complete interests' sake, when i just knocked the design up without measuring lengths, there was almost 11cm of variation - i made runner 4 waaay too long - the others were within 4cm of each other.

Overall, for my first ever manifold (hell, first ever anything except cooler piping and exhaust) i'm fucking happy. I guess i'll see how it goes, and then maybe down the road i'll make one with larger 1.75"+ runners and a larger exhaust.

[ke_70]

they look really good man!

im alittle worryed abount the .5mm gap. if they start touching they'll might crack.
mabey you could throw alittle bead of weld between the 2? stop it from vibrating.

[Sam-Q]

looks amazing, I like the quick release clamp idea as well. At what revs is the calculated primary resonance point?

Now I have a sneaky idea for you, as above fix those two pipes together but my idea is to put a stainless tag between the two. Make it look like it's meant to be there, like a build plate. Some 20mm wide and 3mm thick material should be heaps. That said I don't see a probably with two pipes touching

[timbo]

straight to the pool room!

[Vance]

my vagina is wet

[KurtHS]

Well done man...

They look fckn amazing! Engine bay pics!

[lolwat]

my vagina is wet
+1 huge patch

[turn69up]

they look amazing :D just wondering why does it have a slight funnel shape (looks like it goes from 2.15-2.25) at the end is that so it has a bit of back pressure?

[Skylar]

Huge props for going to all that effort. Turned out real nice. I'd get sick of trying to make them all equal length and give up if I were to try something like that.

One thing I stumbled upon the other day. You can make the runners equal length but it doesn't mean they flow the same due to having different bends in each of the pipes at different points distances from the port, right? Would extractor design be better off trying to get each of the pipes to flow the same CFM instead of having the same length?

Not trying to rain on your parade or anything, just something I came across in a book the other day. Am I overthinking it?

[70XIN]

Haha thanks guys. And yeah Skylar, i probably would have given up on trying to make them equal length, but because the mate i had over at the time has as bad OCD as me, he wouldn't let me haha. That worked out well for me i guess

And yeah, you are indeed correct - but because (in my case) i used a similar number of the same bends for each runner, they end up pretty close - as far as my brain told me, the real way to know if the *lengths* are the same, is to CC each tube. Because i welded each of the runners off the the flange/collector, i could do this, and the result were pretty similar (within 20cc or so of each other) - but this wasn't exactly easy or perfect, as you have to refill the burette many times to fill up one runner, which leaves a margin for error.

To be honest though, there are SO many variables that come into making a perfect header, that i'm not educated enough to even go into any more detail ;)

Sorry if any of the above doesn't make sense, i am faaaarked. ;)

[xero]

looks really good mate, have you thought about just chucking them on a 20v and seeing if they dont leak or crack or whatever?
possibly chuck them on a car on the dyno for shits and giggles? (it should only take about 30mins to swap them over on a already converted car)

[Sam-Q]

the speed of the resonance waves should be fairly constant regardless of the bend radii. If the length is the same and the pipe diameters are the same then it should work out exactly right.

[marvis]

http://86fighters.files.wordpress.com/2011/05/dsc_0035_resize.jpg