Show and Tell: Proper header diameters

[na svt]

I wanted to post this to show how large primary tubes change the shape of the powerband. Some assume that the larger the tubes the greater the power but that is only true if the engine is asking for them. Primary tubes like 1 3/4" stepped to 1 7/8" on a stock engines will have an adverse affect on power; killing power below 6k and while adding little to nothing up top.

Below is graph comparing two similar, but not exactly the same combinations. The 1 5/8" combo has 99 cobra cams while the larger tube one has 96-98 intakes, both sets of headers have 3" collectors. As you can see, the larger tubes give up a great amount of average power. In the first graph the larger tube combo is STD factored while the other is SAE (STD produces higher numbers). Despite this the 1 5/8" header combo still makes far more average power. The second graph shows the larger tube combos graph where it would be (very close) if SAE factored and this shows how bad the big headers hurt power on a stock, low RPM engine.

This graph shows the larger tube combo (red lines)adjusted for SAE factoring. This is a much better comparison; apples to apples:

 

[shurur]

NASVT:

I posted this some time ago and go zero response...
http://www.svtperformance.com/forums/new-edge-cobras-51/862324-1-5-8-vs-1-3-4-headers.html

 

[na svt]

Header diameters have much less of an affect the hp output of FI combos.

On n/a combos, horsepower doesn't dictate the need for a certain diameter primary tube. Some 400rw combos would benefit from a 1 5/8" while others may need a 1 7/8" primary. The entire combo and the type of driving/racing it's going to encounter must be considered.

 

[shurur]

The RPM range for the FI dyno shows pretty much what you are showing though...

I'd like to see where and what NA combo for a 400ish RWHP motor that the bigger headers would show some advantage....average HP specifically..even over a small high (racing) rpm range.

Even someone with say an BB-stoker 5.3L/ 5L stoker with say...106400 cams and a 5.5 SRI running in the 9K rpm range...I'd have to see it.

I'm guessing larger straws as you've called them, are only going to be useful in the upper track/race RPMs..making them a flop for the DD.

These relatively small engines have to be spinning pretty fast to move enough air to make the larger header shine at all over the 1-5/8..??

 

[black4vcobra]

Good post as usual NA SVT. It's crazy how much misinformation there is out there. I got a good deal on the 1-5/8" Macs I have and just figured I would have to be content that I would be leaving a few ponies out there but I guess I'm not.

You probably know this but even in Sean Hyland's book How to Build Max-Performance 4.6-Liter Ford Engines, he states "Our dyno tests have proven that even a dead-stock 4-valve Cobra engine can make more power in all RPM ranges with a 1-3/4-in primary-tube header; gains of 7 hp and 8 ft-lbs of torque over 1-5/8-in headers are not uncommon" He does not specify an NA vs SC Cobra motor.

Do you have anything to repute that statement?

EDIT - link with article- http://www.mre-books.com/sa82/sa82_9.html

 

[98_Undertaker]

Who even makes 1-3/4 besides kooks and ARH?

 

[OW99]

Primary pipe area in inches x 2,511 will equal where peak torque wants to be for a 4.6. It's not incredibly accurate but it should give a general rule of thumb. A=pie x inside radius squared. Considering we don't have any selection of tuned headers for our cars we can't really play with primary length but a good way to help with horsepower under the curve is collector length. Usually the longer the primary tubes/collectors will help with torque under peak. Collector diameter should have an increase of 120-150% of the primaries for na cars and boosted cars like them a little bigger. That means that 15/8" header will benefit from 2.25-2.5" primaries about. 13/4" will like 2.5" too. 3" is too big of an increase and will hurt torque under peak. I know some headers for our cars have this. I don't remember where I got the equation from but it was in my notes on my phone. I'm no expert with exhaust tuning but I figured I'll add that in there for anyone who wonders this crap like I do lol.

 

[na svt]

I suggested to an writer for MM&FF that they do a header test, showing the results of different primary a diameters on the powerband. He thought it would be a good idea but his editor would have to approve it.

 

[na svt]

Considering we don't have any selection of tuned headers for our cars we can't really play with primary length but a good way to help with horsepower under the curve is collector length. Usually the longer the primary tubes/collectors will help with torque under peak. Collector diameter should have an increase of 120-150% of the primaries for na cars and boosted cars like them a little bigger. That means that 1 5/8" header will benefit from 2.25-2.5" primaries about. 13/4" will like 2.5" too. 3" is too big of an increase and will hurt torque under peak.

Collector diameter matters but their length doesn't when an exhaust system attached to the header. Merge collectors can add a good amount of hp to an n/a combo and cost little when compared to other common mods that cost a lot and add no power.

The primary tubes of most off the shelf headers are too short by about 4" and of course none of them have equal lengths.

IMO the off the shelf headers with the best specs for a bolt on n/a Cobra or Mach are the MAC's and BBK's. However, I'm not a fan of BBK's collector design and their fit.

 

[na svt]

The RPM range for the FI dyno shows pretty much what you are showing though...

I'd like to see where and what NA combo for a 400ish RWHP motor that the bigger headers would show some advantage....average HP specifically..even over a small high (racing) rpm range.

Average hp/tq over what rpm range? idle to 7k, 2k-7k, 3k-7k? Each range has different header requirements.

 

[OW99]

Collector diameter matters but their length doesn't when an exhaust system attached to the header. Merge collectors can add a good amount of hp to an n/a combo and cost little when compared to other common mods that cost a lot and add no power.



The primary tubes of most off the shelf headers are too short by about 4" and of course none of them have equal lengths.



IMO the off the shelf headers with the best specs for a bolt on n/a Cobra or Mach are the MAC's and BBK's. However, I'm not a fan of BBK's collector design and their fit.

Yeah I was thinking about old school collector extensions for open header cars. I think the location of the cross over can help with tuning for full exhaust cars but I don't think any gains or losses will be that noticeable/worth it.

I always liked hooker headers but their collectors are sized to big. They would be great if they had 2.25-2.5" collectors but I couldn't buy a $700 set of headers and modify them lol. Might as well build custom.

 

[fangs99]

The only combo that I can see really needing a bigger primary tube header is a nitrous setup on 200+ shot.

 

[na svt]

The only combo that I can see really needing a bigger primary tube header is a nitrous setup on 200+ shot.

...but that is if you want the combo optimized for the nitrous without any regard to "off nitrous" power.

 

[na svt]

I always liked hooker headers but their collectors are sized to big. They would be great if they had 2.25-2.5" collectors but I couldn't buy a $700 set of headers and modify them lol. Might as well build custom.

I would buy a set of inexpensive 1 5/8" longtubes and install onto them a properly sized merge collector.

 

[shurur]

Average hp/tq over what rpm range? idle to 7k, 2k-7k, 3k-7k? Each range has different header requirements.

Idle to 7k at least...with most interest in 3k-7k+...just interested in what combo and rpm makes the 1-3/4 shine over the smaller headers...

 

[fangs99]

Yes nasvt, i would only do that to a deadicated track car.... I'd happily give up 30 or so rwhp on a nitrous setup to get it back na in a street car.

 

[RPM4DAZ]

Not trying to dispute the findings/post, because my studies and experience say the same. But, one thing that has not been stated is the running or valve duration. Headers in general only really begin to show their worth when overlap ( hence the fifth cycle ) begins to come into play. Today's engine breath better for sure, but, they also have very little overlap area. In most cases it is null and void to the negative side. I know it is all about emissions and a smoother running engine... But, still a valid and probably the most important point. Tube/runner volume demand goes up as volume goes up. Hence, rpm levels, displacement, overlap, lower compression. Alot of times a higher compression or higher V.E. engine likes a smaller tube/runner than some people think initially. I have seen it and watched it on the dyno more than once on all sorts of different engine styles. Hell, even Kaase and the now famous 409 had two 1.5" tubes. That equals to the area of a single 1.9" I.D. tube. On a 409 cid V8 that made a restricted 720HP and an astounding 665 lb/ft of torque. Todd should be able to attest to it from hands on experience of that build. Again big cubes ( in relation to modulars ) and a high amount of overlap increasing the 5th cycle in both time and volume... I myself have chosen to run a set of JBA 1 3/4" mid length headers in hopes of sliding through smog without too much attention and trying to reduce the lower/mid range ( 2,000 - 3,500 rpm ) torque some. My car is more of road racer set up and traction limited big time on street tires in the lower gears. Header tubes can be used to tune the power band just like the intake manifold runner/plenum and cam timing/lobe locations! This is a great thing to discuss for sure!!!

 

[na svt]

Not trying to dispute the findings/post, because my studies and experience say the same. But, one thing that has not been stated is the running or valve duration.

Post #1 states the cams used in the those two combos and there were 202 and 204 degrees. Neither set was degreed so the LCs are somewhere near 119/110 which provide about -29 degrees of overlap. This is why larger primarys are worthless on a stock cammed engine.

I'm trying to get MM&FF to a header comparison to show the masses what is best for a given combo.

 

[RPM4DAZ]

Yes I saw them. I was trying to add to your post not debunk or debate. I was just trying to state to the masses some of the reasons why headers don't "seem" to be that big mod they thought or think it should be. That's all... The "volume" that a lot of people think they need isn't really supplied by a set of headers, it is about the timing and separation along with the fifth cycle!

 

[na svt]

In response to the "fifth" cycle, engine's like our 4Vs having a large intake valve dia/bore volume ratio don't need a lot of overlap and in most cases need none.