what size header tube?

[spoolinhard]

I am getting ready to make a header for my 8v. Looks like there are many sizes, starting at 1"1/2. Just wandering what other people are using.
thanks
-Kaleb

[Dave]

If you're revving over 6,500 I'd say go with 1 5/8". If you're under that I'd recommend under 1 1/2" primary tubes. The secondaries can be a little larger without hurting velocity and in fact it well help deterring reversion.

[spoolinhard]

I am running a pretty large turbo. It is a T04 housing garrett with .60 on the compressor side, and .81 on the exhuast. Pretty sure I am going to need some RPM;s up to spool it. Primary and secondary tubes?:confused: Explain please with relation to #'s 1-4 runners.

[BadAssPerformance]

What kind of header? Log or equal length?

for equal length, 1-5/8" works well

http://www.badassperformance.com/mstore/bap_parts/header/header_05.jpg

[cordes]

What kind of header? Log or equal length?

for equal length, 1-5/8" works well

http://www.badassperformance.com/mstore/bap_parts/header/header_05.jpg

Show off. ;)

[fishcleaner]

The circumference of the 1-5/8 tubing works out nice with the exhaust port too.

[Dave]

I am running a pretty large turbo. It is a T04 housing garrett with .60 on the compressor side, and .81 on the exhuast. Pretty sure I am going to need some RPM;s up to spool it. Primary and secondary tubes?:confused: Explain please with relation to #'s 1-4 runners.

That is quite a big turbo for our 8valves. You're going to need some big, immediate volume to spool that turbine. I'd suggest a very big head and a large intake plenum to start.

The primary tubes are the tubes you see that BadAssPerformance just posted, if he were to combine those tubes into 2 collectors, the tubes after that collector would be called secondaries, then the turbine housing collector would be at the end.

[BadAssPerformance]

That is quite a big turbo for our 8valves. You're going to need some big, immediate volume to spool that turbine. I'd suggest a very big head and a large intake plenum to start.

The primary tubes are the tubes you see that BadAssPerformance just posted, if he were to combine those tubes into 2 collectors, the tubes after that collector would be called secondaries, then the turbine housing collector would be at the end.

primary and secondary tubes are only used if you go with a '4-2-1' 'tri-Y' header like (pictured below) Aaron Miller (8valves) the '4 to 1' header I made does not have secondaries. ...and for the most part does not need them...

http://i234.photobucket.com/albums/ee189/Eight_Valves/Dsc01242.jpg

[Dave]

primary and secondary tubes are only used if you go with a '4-2-1' 'tri-Y' header like (pictured below) Aaron Miller (8valves) the '4 to 1' header I made does not have secondaries. ...and for the most part does not need them...

http://i234.photobucket.com/albums/ee189/Eight_Valves/Dsc01242.jpg

That was the exact pic I was thinking of yet to lazy to retrieve. That's a perfect example of the 2 into 1 that develops secondary tubes. Most headers will not have them, but if they do they must be larger than the primaries.

[spoolinhard]

Done think that I am going to go secondary. Just primary. and it will be an equal length header. I have already hogged out the head and intake quite a bit, also have oversize valves in the head.

[BadAssPerformance]

Sounds cool :thumb:

[8valves]

That was the exact pic I was thinking of yet to lazy to retrieve. That's a perfect example of the 2 into 1 that develops secondary tubes. Most headers will not have them, but if they do they must be larger than the primaries.

Says who?

http://www.ultimate-racing.com/subaru-gt40r/header4.JPG

[black86glhs]

Hmmmm...that looks like a Subby header.:)

[spoolinhard]

Sounds cool :thumb:

Should be fun to make! Im pretty excited about getting started.

[Dave]

Says who?

http://www.ultimate-racing.com/subaru-gt40r/header4.JPG

LOL books. :D If all the exhaust pulses from 2 cylinders are grouping together would you not want to increase the volume of the secondary pipe to accommodate the additional volume? Sure it might keep the velocity higher, but will in turn develop additional unnecessary backpressure at the merger, especially if both of those 2 cylinders are firing at the same time.

Now in your example, that does appear to be a 2.5 boxer header, if memory serves right they have a firing order of, 1, 3, 2, 4. The way you have it routed, the exhaust pulses will be independent of their sister cylinder, so the merger will go unaffected as the pulses inside the 2 merging secondaries are timed out.

I think... :thumb:

EDIT: Still got the 8V header?

[Mopar318]

LOL books. :D If all the exhaust pulses from 2 cylinders are grouping together would you not want to increase the volume of the secondary pipe to accommodate the additional volume? Sure it might keep the velocity higher, but will in turn develop additional unnecessary backpressure at the merger, especially if both of those 2 cylinders are firing at the same time.

Now in your example, that does appear to be a 2.5 boxer header, if memory serves right they have a firing order of, 1, 3, 2, 4. The way you have it routed, the exhaust pulses will be independent of their sister cylinder, so the merger will go unaffected as the pulses inside the 2 merging secondaries are timed out.

I think... :thumb:

EDIT: Still got the 8V header?

Thats kinda the point of a 4-2-1 header using a twin scroll turbo charger.

[8valves]

I just used that as an example. It's the header from the Big Valley STI... one of the few "fast" ones out there int he WRX world.

With the divide scroll setup, the key is beign equal length so that you don't have two pulses fighting for space.

However, if you want to play the math game, if a 4-1 header has a 2.5" collector that leaves 1.22 in^2 per cylinder if all cylinders flow into the collector area at the same time which is likely to happen at some point, or when pressure is built up in the collector area.

2 cylinders firing into, say my secondary of 1.682" ID runner size, leaves 1.11 in^2 per cylinder. Doesn't seem like that big of a difference anymore huh?

Just some numbers to crunch. I'm in no way saying it's best, optimal, or anything for that matter. Just that some secondary designs are the reason for poor spooling setups more than you might think. It just gets blamed on the turbo right away. I'm guilty of this one :p

And yes, I still have 'ol spider legs (the header) here :nod:

[Dave]

I just used that as an example. It's the header from the Big Valley STI... one of the few "fast" ones out there int he WRX world.

With the divide scroll setup, the key is beign equal length so that you don't have two pulses fighting for space.

However, if you want to play the math game, if a 4-1 header has a 2.5" collector that leaves 1.22 in^2 per cylinder if all cylinders flow into the collector area at the same time which is likely to happen at some point, or when pressure is built up in the collector area.

2 cylinders firing into, say my secondary of 1.682" ID runner size, leaves 1.11 in^2 per cylinder. Doesn't seem like that big of a difference anymore huh?

Just some numbers to crunch. I'm in no way saying it's best, optimal, or anything for that matter. Just that some secondary designs are the reason for poor spooling setups more than you might think. It just gets blamed on the turbo right away. I'm guilty of this one :p

And yes, I still have 'ol spider legs (the header) here :nod:

Wow, never knew the difference was that small. Certainly something to think about... in the Street Turbocharging book I'm currently reading, it states that the rules of cylinder scavenging don't really matter much in turbocharging, for the reason that it is far more beneficial to reduce primary size to spool the turbo than it is to scavenge exhaust gasses out from the cylinders... Now with that said, wouldn't the pressure pushing its way inside the cylinders effectively push the exhaust gas out during any overlap period? What do you think about those two ideas, Aaron?

[8valves]

it states that the rules of cylinder scavenging don't really matter much in turbocharging, for the reason that it is far more beneficial to reduce primary size to spool the turbo than it is to scavenge exhaust gasses out from the cylinders... Now with that said, wouldn't the pressure pushing its way inside the cylinders effectively push the exhaust gas out during any overlap period? What do you think about those two ideas, Aaron?

A) Equal lengths are a form of a better "scavenging" header. They make more power than non-equal's on NA AND turbo cars. It's been proven. The reason people still disagree is because people make LOTS of power on non-equal headers. That doesn't mean they couldn't do better though. Perfect example is outlaw drag radial cars... lots of non-equal turbo headers because of big blocks stuffed into Fox chassis, and a TON of power. Need more? Add more boost. I geuss you would need to weigh your priorities there.

B) I've never personally experimented with different primary sizes, save for once with a car that the 1 3/4" didn't match the port well and was moved up to a 1 7/8". That probably isn't a worthwhile example because of the port mismatch. I certainly wouldn't want to be too small on this though! Seems like the best rule of thumb there is follow what matches your port diameter closely. I'd rather err to the large side personally.

C) Intake pressure would only "force" out exhaust gasses in the chamber if you have a sub 1-1 pressure ratio intake to exhaust. I'd be willing to be there isn't a turbo-mopar out there save for one or two MAYBE that have an at, or better than 1:1 PR. I could be wrong, but I highly doubt it.

D) Remember that heat is more important than volume. And that heat is directly related to internal volume of the header system too.

[Aries_Turbo]

aaron, any recommendations for primary/secondary size/length for a 2.5L 16v head feeding a split scroll borg warner S200?

Brian

[Mopar318]

A) Equal lengths are a form of a better "scavenging" header. They make more power than non-equal's on NA AND turbo cars. It's been proven. The reason people still disagree is because people make LOTS of power on non-equal headers. That doesn't mean they couldn't do better though. Perfect example is outlaw drag radial cars... lots of non-equal turbo headers because of big blocks stuffed into Fox chassis, and a TON of power. Need more? Add more boost. I geuss you would need to weigh your priorities there.

B) I've never personally experimented with different primary sizes, save for once with a car that the 1 3/4" didn't match the port well and was moved up to a 1 7/8". That probably isn't a worthwhile example because of the port mismatch. I certainly wouldn't want to be too small on this though! Seems like the best rule of thumb there is follow what matches your port diameter closely. I'd rather err to the large side personally.

C) Intake pressure would only "force" out exhaust gasses in the chamber if you have a sub 1-1 pressure ratio intake to exhaust. I'd be willing to be there isn't a turbo-mopar out there save for one or two MAYBE that have an at, or better than 1:1 PR. I could be wrong, but I highly doubt it.

D) Remember that heat is more important than volume. And that heat is directly related to internal volume of the header system too.

Very true. The fox body at work is running enequal headers. With a 94mm turbo, 30psi of boost, and 1600whp, its more than enough needed before it starts spinning the tires. 7.80's is pretty darn fast too.

[8valves]

aaron, any recommendations for primary/secondary size/length for a 2.5L 16v head feeding a split scroll borg warner S200?

Brian

This might sound over simplified, but to be honest build whatever fits and get as close to equal as you can with the divided. Seperate wastegates is more of a preference thing I think.

This has no math/science basis, but I would put the merge of the two pairs right on the turbine flange and say yahoo. I only merged my secondaries early for fitment purposes. Some of the fastest DSM/Honda's on the planet are on a simplified design such as this.

1 1/2" Schedule 10 SS cast pipe works fantastic. Keeps a lot of heat in and has a 1.682" ID. It will also match to an 8V port well, and should oval to a 16V port without issue. If not get a steup up in size for the head flange then reduce to the 1 1/2" S10. Note that Schedule 10 1 1/2" is FAR different than standard tube 1 1/2".

I'm sure Frank could give you some sort of math breakdown, but that's what I would do. A simple topmount design would work great.

Sound like it would be a rather mean street combo too :)

[Aries_Turbo]

i dont trust franks math ;)

thats what i was planning on doing but i was gonna just use mild steel as i can weld it at home and its cheap. i just gotta get some flanges and start the layout. i just wondered if there was any specific benefit to doing a tri Y.

Brian

[Shadow]

i just wondered if there was any specific benefit to doing a tri Y.

Brian

Faster spool

[Aries_Turbo]

hmmmm i want faster spool. maybe ill try a tri Y and see if gazer frank can work his BS magic formulas to determine the lengths. (though i dont trust the results totally hehe.)

Brian

[BadAssPerformance]

I think the last time the BS calcs came out it was like 48" primaries? ;)

[Shadow]

hmmmm i want faster spool. maybe ill try a tri Y and see if gazer frank can work his BS magic formulas to determine the lengths. (though i dont trust the results totally hehe.)

Brian

No offence, but IF I listened to paper calculation and NOT real world proof, I'd be less than 1/2 way to where I've already gone! Just use some common sense along with a practical approch to fitment and ease of instalation. 1 1/2' to 1 5/8' diam. with at least 6-8' primary (more if you want), and at least 6-8' secondary feeding a dual scroll or merging into a final 3-5' before turbo. I dyno'd 480whp on a stock ported turbo exhaust mani, a header like the one I described would gain me 20-40whp and faster spool for sure.

[Aries_Turbo]

dont worry, i follow real world results far before i listen to calculations.

i take it you meant Inches when you typed ' ......

Brian

[Shadow]

ya, ' = inches. I'm not saying this would be THE most optimal header you could build either. All I'm saying is even something this simple would work well.

[Aries_Turbo]

packaging will be an issue so i will be trying to keep it simple and equal length and divided. ill be combining the secondaries for an external wastegate though at some point on the header before the turbo.

the reason i asked about the lengths of the different stages is because i was curious if any pulse tuning applies given the pressures in the exhaust manifold. its a debated topic and i wondered if anyone had some experience with it.

Brian

[Shadow]

packaging will be an issue so i will be trying to keep it simple and equal length and divided. ill be combining the secondaries for an external wastegate though at some point on the header before the turbo.

the reason i asked about the lengths of the different stages is because i was curious if any pulse tuning applies given the pressures in the exhaust manifold. its a debated topic and i wondered if anyone had some experience with it.

Brian

Any equil length header IS pulse tuned IMO as long as the firing order is split properly. Wether the primaries are 8' long or 60' long, as long as they merge equily into the secondaries and so on, the header should PULSE exhaust strokes evenly. (BTW I just realized why you asked me if ' = inches! lol guess I should be using " instead)

[Aries_Turbo]

yeah i know that the pulses will reach the collector equally and make a difference in spool but i was curious if different lengths would favor certain rpm ranges.

yeah, i was gonna say... i am not going to be able to fit 15-21 feet of pipe in the engine bay. :) maybe a rear mount turbo hehehe.

Brian

[Shadow]

yeah i know that the pulses will reach the collector equally and make a difference in spool but i was curious if different lengths would favor certain rpm ranges.

yeah, i was gonna say... i am not going to be able to fit 15-21 feet of pipe in the engine bay. :) maybe a rear mount turbo hehehe.

Brian

I don't believe that's AS crutial in a turbo application. What kind of RPM are you wanting to turn?

[Aries_Turbo]

i didnt think so either but there is some debate on the issue as far as max efficiency is concerned. probably 7-7.5k max. its a 2.5L. ill probably build a short primary and as short a secondary as i can keeping them equal length to the divided inlet on the turbo.

ill probably start mocking up an engine sometime this winter. ill get a few small pieces of pipe to see which one ovals out the nicest to match the exhaust ports on the head.

Brian

[Shadow]

i didnt think so either but there is some debate on the issue as far as max efficiency is concerned. probably 7-7.5k max. its a 2.5L. ill probably build a short primary and as short a secondary as i can keeping them equal length to the divided inlet on the turbo.

ill probably start mocking up an engine sometime this winter. ill get a few small pieces of pipe to see which one ovals out the nicest to match the exhaust ports on the head.

Brian

For higher RPM I would run larger diam pipe before I would go to extreams on length. Matching diam. to the head should give you a good idea what to start with. Don't forget, Stephan turned 7600rpm in an 8v with a very short 2" diam header. Wish I would have had more time this year to go back to the dyno and see how much my powerband will change when I dial in the stock roller cam. (retard it) Depending on how that goes we'll see if it isn't time for something a little bigger.

[Aries_Turbo]

yeah, im not going to go super short, getting it all in there and equal length will determine some of the length constraints. i dont want to go too big on the piping, this turbo is pretty big already and i dont want too much lag.

Brian

[Shadow]

What size turbo?

[Aries_Turbo]

borg warner s200. they used to be schweitzer s200's. id have to measure the wheels. i dont know the exact measurements.

Brian

[turbovanman²]

Look at what Warren built, I think that would do the trick. Also the stock TIII manifold works quite well for alot of turbo's so something else to think of there, design wise of course. :nod:

[Aries_Turbo]

yeah the one he built is nice. im not going to make one that fancy. :)

Brian

[Shadow]

yeah the one he built is nice. im not going to make one that fancy. :)

Brian

Warrens header could support 1000whp, that should give you an idea of something shorter and smaller diam and what it could support.

[Shadow]

borg warner s200. they used to be schweitzer s200's. id have to measure the wheels. i dont know the exact measurements.

Brian

If I compare the s200 info to the different turbos I've had on the Charger I would think I would see full spool 30+psi by 3500rpm. That's on a stock cam'd stock ported exaust mani'd 2.2 8v. You put that on a 16v lotus exhaust mani and I would think that you'd reduce spool time by at least 300rpm. Build a header like I described and as insane as it sounds, 2800-3000rpm full spool should be doable! (depending on other mods of course)

[Aries_Turbo]

If I compare the s200 info to the different turbos I've had on the Charger I would think I would see full spool 30+psi by 3500rpm. That's on a stock cam'd stock ported exaust mani'd 2.2 8v. You put that on a 16v lotus exhaust mani and I would think that you'd reduce spool time by at least 300rpm. Build a header like I described and as insane as it sounds, 2800-3000rpm full spool should be doable! (depending on other mods of course)

we'll see. :)

Brian