• Protip: Profile posts are public! Use Conversations to message other members privately. Everyone can see the content of a profile post.

Header Thread * Which Makes Most HP per $$? Best Bang

What is the advantage of 6-2-1 over 6-2-2?
To be honest I'd need to research how a Honda C engine reacts to the balancing effect of single pipe versus the splint bank design,
Joining the exhaust banks makes for a more balanced flow between banks although position, piping and diameter are critical.

V6 engines like ours generally run 6-2-2 or 6-2-1 in OEM, 6-2-1 being technically cheaper and would in basis be more high revving that a 6-2-2 that would make more torque but weigh and cost more.
Now how that translates in to real world figures on a NSX is something that would need some further calculating, (see how the exhaust pulsations and firing order align between banks)

I'd need to dig up my notes on Acura Legend exhaust design as i recall 6-2-2 with a X pipe improving torque down low versus there standard 6-2-1 layout.

The odd firing character of the V6 and especially our '90 degree ones make it a tad more complicated.
A Straight 4 engine running a conventional flat plane crank would require 4-2-1 for optimum balanced performance & torque, although a long primary 4-1 setup improves high RPM capability and max power.
Cylinder numbers 1 & 4 are joined down line and 2 & 3 are joined as the first merger of the primary runners, secondary would combine 14 and 23 together (generally near the oil pan of a transverse engine)

Straight 6 engines must run 6-2 manifold setups to give satisfactory performance due cylinders cross breathing badly in 6-1 manifolds, then you'd traditionally run a 6-2-2 without H or X pipes.
Although there are examples by Toyota and I recall on classic British cars running 6-2-1.
A V12 can be seen as a dual straight six although often running 12-4-2 exhaust lines, Older V12 Ferrari's have very heavy 12-4-4 layouts.

V8's in traditional American/German cross-plane layout run differently to straight 4 engines there heavily off set cranks & firing orders negate the need for 8-4-2 manifold and thus nearly all would run a 8-2 manifold in performance applications,
Most V8 cars have 8-2-2 exhaust systems, often with H or X pipes in the middle to balance the left and right bank this improves smooth running and torque curve.
Side pipe's on for example a old Big Block Corvette run a 8-2-2 setup with out any gas equalization, you can hear the much rougher idle as well.

V8's in Italian/Racing flat-plan configuration are comparable to two straight 4 engines driving a common crankshaft, as you'd see on a Ferrari's and certain GP engines they run a 8-4-2 layout with each cylinder head having a 4-2-1 crossed layout.
These engines have very high RPM potential and very low rotating mass yet do suffer from low torque out put in low revs and higher idle speeds, this makes these engines undesirable in anything other than a race or sports car.

Boxer engines run different yet again on a H4 you'd run 4-2-1, there isn't any cross breathing as the engine runs in a crisscross pattern canceling it out.

I'm not going to bore you guys further with inverted induction engines (like new BMW V8's and 1.5 V12 Honda engines) these have the exhaust on the inside and the intake placed on the outside.
 
Last edited:
Thanks Adnan for the explanation!

Don't know if anyone here can translate japanese, here is the information about KSP exhaust (From translator it seems the exhaust is tuned properly not to sacrifice torque and to keep high speed for exhaust gases...):

"NSX専用に製作されたマフラーで、前後バンクの排気を集合し、適切な長さを維持した後に分離して大型サイレンサーに振り分けるというレイアウトが特徴で6気筒らしい揃った排気音と中間トルクを落とさずに高回転出力を得ています。
アイドリングでの排気音量は非常に静かで、アクセルを踏んだ際には快音を響かせる理想的なマフラーとなっています。"


Google translate:

"The without sacrificing intermediate torque and exhaust sound with uniform layout that distributes large silencer separately after Scarf that is fabricated NSX dedicated to set the exhaust front bank , and maintain adequate length like six -cylinder Features. I'm getting a high speed output .
Exhaust volume at idling very quiet , making it the ideal muffler that sounded a Kaion When I stepped on the accelerator."
 
The KSP setup is a high velocity flow setup with very smooth mergers (which aids flow)
Final result will also depend on the muffler chosen for example a TaiTec JGTC or a typical RFY muffler with long pipe lengths will tend to make more Torque than a short runner Center exit style system which aids top end power.

Then there is the cross breathing factor for the X or H pipe design (if used).
That being said I've seen a few quite radical setups in Japan that don't really take in to account flow reaction but more focused on sound (running a apposing flow x pipe, which literally means Forward and Rear cylinder bank gas crash into each other head on)
 
In all these years, has anybody got a hand built header w/merge collectors for the NSX made? Hytech, or other customer builders? Stepped headers seems to work well on the B series so I think C30A should respond well to it.

- - - Updated - - -

I have 2 NSX.
One turbo, and one other 100% stock.
I have the stock header, a new OBX header in a box, and a old TopSpeed header in a box.
I have a dynometer, so you will have my result in some weeks with my stock NSX.

did you ever post your dyno results?
 
In all these years, has anybody got a hand built header w/merge collectors for the NSX made? Hytech, or other customer builders? Stepped headers seems to work well on the B series so I think C30A should respond well to it.

- - - Updated - - -

did you ever post your dyno results?
Lovefab has one of the nicest headers out there for the NSX.
 
The V12 is different,
Basically exhaust design is straight 6 x 2

And since straight six generally works best with 6-2-2 (some times there are the odd 6-2-1 with the 2>1 being ~ five feet down stream)
As 1-3 and 4-6 would fight each other with exhaust pulsations slowing down extraction from cylinders dramatically.

V12's generally run
12-4-4 (Lamborghini, Older Ferrari, McLaren, early Jaguar)
12-4-2 (Later Ferrari, Certain Lamborghini, BMW, Late Jaguar & Aston Martin) Again the 4>2 is at the very end of the exhaust.


Last year i had very nice example of a simplistic V12 layout in the shop. (i'd need to check if have pictures)
A full exhaust line for a 412I Ferrari, running 6 mufflers in full 12-4-4 layout.

12-4-2 do appear often,
The secondary collectors where the 4>2 happens are down stream and at enough angle so that cross breathing can't happen.

In 456,550,575 Ferrari's this happens prior to the primary cats due to emission requirements.
The headers them self are 4x 3 runner equal length running in to curved secondaries that feed straight in to the catalytic chambers.

In a old-school Lamborghini setup, ('60s early 70's)
The secondary merger is placed in front of the central silencers (let's say 6-7 feet down stream)

BMW V12's ran simple 3-1 log headers (4x) in to four long primaries that then merged Ferrari style in the cat's (12>4>2)
 
Last edited:
I just fit a set of OBX headers and decat pipes, no issue, perfect fit. Only had to extended my O2 sensor cables. Like all things hand made, there is margin for error and so many variables that nothing can be a 1:1 match hence why most brands have fitment issues including the overpriced ones.

Cheers
 
Recently fitted Lovefab long tube headers with test pipes. Perfect fit. Maybe more expensive than some others, however my seat of the pants Dyno tells me they are "great bang for my bucks".

About to make a few more improvements then will post Dyno results. I do feel confident I made the right choice and the headers will help optimise the planned mods.
 
RM Is definitely far superior compared to DC sports headers with FI or not. In search for a CA carb legal system I have to sacrifice Hp and Tq the difference is very noticeable. Comparable Hp/Tq; RM headers with Cats installed = DC Headers with test pipe.
 
RM Is definitely far superior compared to DC sports headers with FI or not. In search for a CA carb legal system I have to sacrifice Hp and Tq the difference is very noticeable. Comparable Hp/Tq; RM headers with Cats installed = DC Headers with test pipe.

The RM/B&B headers remain the best FI option available, except for Cody's custom headers. They use a large 2.5" collector, which is the biggest on the market. Even better than the 2.36" on the Fujitsubo/Pride offerings. If I had a SC, I would go B&B. Just make sure they put the O2 bung on the collector and not the primary.
 
Dropping cats on an NA application yield negligible amounts of power, if you were referencing a turbo car then I would agree more. Deleting the cat on an NA car is going to make it louder, smell, lighter and add 1-2 hp tops and you are not going to feel or notice the power of that what so ever. Louder does not mean more power.

RM Is definitely far superior compared to DC sports headers with FI or not. In search for a CA carb legal system I have to sacrifice Hp and Tq the difference is very noticeable. Comparable Hp/Tq; RM headers with Cats installed = DC Headers with test pipe.
 
Great thread!
For everyone that got their headers replaced, whats the ballpark price? lets say.. no mods NA2 or DC header swap?
For those of you who installed the NA2 headers, how much did you buy them for?
thanks in advance!
 
Great thread!
For everyone that got their headers replaced, whats the ballpark price? lets say.. no mods NA2 or DC header swap?
For those of you who installed the NA2 headers, how much did you buy them for?
thanks in advance!

https://www.lovefab.com/Store/Product-Detail/ProductID/13/nsx-fi-spec-headers

Select your test pipe, and rock out. We hand-build our headers, TIG welding, backpurging, with proper merge collectors. All built in-house, not outsourced as all others' are.
 
I know this thread is old, but I found it very informative when reading about the different types of headers available. Given I am in CA, my choices were DC or CT Engineering. CT Engineering was difficult to source used and cost $1500 new. DC is about $750 for 95-05, and $950 for 91-94. I recently passed smog with DC and I am happy with them. Installer said they fit perfectly so maybe the newer design fixed the prior fitment issues. Also, someone stated on their NA2 that DC produced 2-3 HP. If we do the math, 15 NA2 headers + 2-3 DC = 17-18 HP. A step below CT possibly in quality, but half the price and still Carb legal. DC may not be ideal for NA2, but I think they are a great deal for NA1.
 
I know this thread is old, but I found it very informative when reading about the different types of headers available. Given I am in CA, my choices were DC or CT Engineering. CT Engineering was difficult to source used and cost $1500 new. DC is about $750 for 95-05, and $950 for 91-94. I recently passed smog with DC and I am happy with them. Installer said they fit perfectly so maybe the newer design fixed the prior fitment issues. Also, someone stated on their NA2 that DC produced 2-3 HP. If we do the math, 15 NA2 headers + 2-3 DC = 17-18 HP. A step below CT possibly in quality, but half the price and still Carb legal. DC may not be ideal for NA2, but I think they are a great deal for NA1.

My understanding is that the DC actually use proper collectors, so they may be better than you think. Neither the CT or DC are equal length, so I think you made a good decision and saved 700 bucks!
 
as a data point I have had the same DC headers on 2 different blocks for 20 years.....many hot track miles and still expelling exhaust gasses....:wink:
 
Dropping cats on an NA application yield negligible amounts of power, if you were referencing a turbo car then I would agree more. Deleting the cat on an NA car is going to make it louder, smell, lighter and add 1-2 hp tops and you are not going to feel or notice the power of that what so ever. Louder does not mean more power.

And cleaning soot off the bumper all time gets old even if you don't care about the air you breathe
 
Back
Top