Flow testing headers and manifolds about to begin

Which header do you think will flow the best?

  • Ported Stock Manifold

    Votes: 6 8.7%
  • 95 Mustang Header

    Votes: 6 8.7%
  • 96 T-Bird Header

    Votes: 1 1.4%
  • Mac Shorty

    Votes: 11 15.9%
  • Kooks Mid Length

    Votes: 25 36.2%
  • Kooks Long Tube

    Votes: 28 40.6%

  • Total voters
    69
  • Poll closed .
Damn...I was going to wait until the Word Doc showed up on a filesharing program...:D

I am really interested to see if the Kooks were worth the money...which I'm sure they weren't...:rolleyes:
 
DamonSlowpokeBaumann said:
I personally think it will be hard to determine which will flow best by these means..Or which would produce more power by these results even if compared with a multitude of variables (intake flow head flow etc)
Damon, you're getting this all mixed up. Dave is performing static flow tests. This is not comparable to dynamic testing, and you're right, it won't necessarily be a true indicator of "real-world" performance. However, Dave never said it would.

I get a chuckle everytime I see someone post "headers have to be better right?" No one is really disputing whether or not headers would provide HP gains... the real question is how much? If you have to spend >$800 for a few HP, or ~$100 to port the OE manifolds for say half that gain, then a decision has to be made, and frankly I would choose to save the extra money for other "go-fast" parts. Others may decide the money is worth it (for a few HP now) and set themselves up for future engine mods.

Due to the nature of our engine package, we suffer from high exhaust backpressure (>15 PSI @ 5,000 RPM according to Charles W. and others). With this kind of backpressure, exhaust tuning becomes ineffective (in other words, you can't "tune" the exhaust system because all the pulses are backed up and crammed together into a giant high pressure slug of gasses). Equal-length headers are designed to "align" the exhaust pulses into evenly-spaced pulses in order to "tune" the rest of the system.

By tuning I mean designing the entire exhaust system so that the individual exhaust pulses are "helped along" by correct sizing of the exhaust components so that pressure nodes (or rather anti-nodes) are present at the exhaust valves when they are opening. Sounds easy, but it's not. The main problem is at one set RPM you have very good performance (negative pressure @ the exhaust valve during opening), but change the engine RPMs and you will find you now have pressure spikes at the valves.

Along with all the exhaust pulses (which travel at sub-sonic speeds), you also have sound pressure waves to consider (which, of course, travel at sonic speeds). Sound after all is waves of pressure in a medium (most guys should get this... there are a lot of audio experts here right?). They travel through the exhaust system and, depending on the design, add together or cancel each other out depending on the phase relationships. They also interact with the slower-moving exhaust pulses. What you end up with is a soup of pressure variations throughtout the exhaust system, and this is highly variable depending on engine RPMs.

Bottom line for us is this: If you have any hope for a true "tuned exhaust", you better get rid of as much general backpressure as you can, and there is lots of room for improvement here. Think "free-flow" rather than "tuning". Now, considering all of this, Dave's testing is actually quite relevant isn't it!

PS. I've got a feeling you know all this stuff already, but you're just wearing your "~~~~" hat aren't you? :D
 
Last edited:
Think "free-flow" rather than "tuning". Now, considering all of this, Dave's testing is actually quite relevant isn't it!

If you're talking free-flowing, then generally speaking, bigger = better. Why else would guys have dump-pipes? So according to the "free-flow" theory, short tube headers with 1-7/8" primaries and 3"+ collectors will outflow any of the current options out there. But don't forget that depending on the engine, smaller may = better. You need that back pressure for low end torque.

Anyay, what everyone here seems to be forgetting is that it's not only your headers/manifolds that need to be thought of when setting up an exhaust system. There's a lot of other crap the exhaust has to flow through before exiting out of the tailpipe. Cats, mufflers, the pipe itself. Your whole setup could make or break you, so to speak. Too large of exhaust, you lose low end torque. Too small and you'll lose top end. It's a balancing act that needs to be designed around your individual system.

In other words... the test would be nice for helping users to select the right manifold/header for the job, but I don't think it's going to prove anything or anyone right. You're always going to have that one idiot with 3" headers running 2.25" pipe say his headers hurt his performance. :rolleyes:
 
I guess that depends. Did you say that your headers hurt your performance?

Seriously though, if you have 3" collectors, but the piping right after the collector reduces to 2.25", you've basically cheated yourself of the gains associated with running a large header. Now if your tips were 2.25" and you staged the pipe reduction based on exhaust temperatures to keep the exhaust velocity up, then no, you did a damn good job.
 
my question is, what is the flow rate, and would it conceivably end up with a collision of gas pulses in the manifold?

There's no conceivably about it. There is going to be a joining or colliding of exhaust pulses in the manifold or a header's collector (hence the term). The question should be, is the manifold/header going to be helpful or harmful in exhaust scavenging when used with a specific cam profile.
 
There's no conceivably about it. There is going to be a joining or colliding of exhaust pulses in the manifold or a header's collector (hence the term).
That is where you are wrong. Header collectors refer to the chamber immediately after the termination of a long tube tuned header where sonic and pressure waves are reverted back into the pipes. On street cars it also doubles as the connection of the header into the rest of the exhaust system. In reality only a longtube has a collector. The short tube or manifold versions simple offer a common chamber for connection to the remaining exhaust system. The term "collector" is completely mis-used on manifolds / psuedo headers.

Without actually measuring the exhaust gas velocities you simply cannot make any statement about whether or not the exhaust pulse from two different cylinders will ever share the same space within the manifold. To say otherwise is pure speculation.

:)
 
You guys are cracking me up. I WAS KIDDING ABOUT CHARGING FOR RESULTS. :D I have completed phase 1 of testing and the results are very interesting.

As promised test one is on a stock head measured up to .450" lift only. This was done on purpose to give people an idea of what is happening on a stock motor.

Test 1 - no pipe or diffuser ---------- 106cfm
Test 2 - 6" straight diffuser ---------- 113cfm
Test 3 - Kooks longtubes ------------ 115cfm
Test 4 - Kooks short tubes ---------- 109 cfm
Test 5 - Mac shortie ---------------- 112 cfm
Test 6 - 95 Mustang header --------- 99.5 cfm
Test 7 - 96 T-bird header ----------- 92.5 cfm
Test 8 - Stock 89 manifold ---------- 95 cfm
Test 9 - Stock 95 manifold ---------- 103 cfm

So, for a stock engine, Kooks headers are clearly the winners. I was surprised to see that the longtubes were better than the short tubes. clearly collector design plays an important part in flow! More on this later.

The MAC shortie exemplifies how a short tube header with a good collector can outflow a long tube. The MAC's have a smaller primary tube but still out did the Kooks short tubes. More on that later too.

The 96 Tbird "header" is clearly no good, not even for a stocker. This example had been treated to a nice little clean up porting job but still performed dismally. Bone stock I'm sure it would have been worse. Two things contribute to this IMO. 1) the tubes are too small and 2) the way they join the plenum is not smooth.

The big surprise is how the stock 89 and 95 manifolds perform - completely stock. The 89 was very bad and I didn't expect much more from the 95 manifold but someone was burning the midnight oil on that one. It surprised me so much that I decided to flow all three ports to see what it looked like. To my amazement the front port actually flowed better than the middle one and actually nearly matched the real headers at over 113cfm!

So this started me to testing other ports and I started to realize that there can be great disparity between ports on the same header. The worst was the 89 manifold flowing only 94.5cfm on the middle port but 111cfm on the front port.

I have read things like this before and I know that BBK advertises an improved and symetrical collector design that is supposed to fix these problems. So my first round of testing was in no way comprehensive because I did not test all the ports on all the headers. I did notice though that the MAC has some awkward tube/collector alignments.

So that means more testing before we start to declare any winners.

Anyway, there is your teaser for the evening. Any comments?
 
It's interesting to show folks how control of air flow direction can have a positive impact on total flow. note how left alone, the flow is less than if you have various headers on there. negative turbulence in the air flow. Now imagine how that impacts things in a combuston chamber or the intake ports.
 
Mike that is true. I will be attaching a short section of pipe to the manifolds because some of the primary pipe to collector aproach angles on some of the ports is really suspect. I was concerned that adding pipes to the system would result in restrictions but as the long tube headers demonstrated, longer pipes do not necessarily mean less flow.
 
Yes, but preliminary results have demostrated a need for more fixturing to get more accurate results. Because of this and all the other things I have to do right now it will take a little bit of time to complete testing. The above numbers do not represent the final #'s, only first round testing results.

Expect some things to change.
 
Yes, but preliminary results have demostrated a need for more fixturing to get more accurate results. Because of this and all the other things I have to do right now it will take a little bit of time to complete testing. The above numbers do not represent the final #'s, only first round testing results.

Expect some things to change.

Understood....

BTW thanks for taking the time to perform these tests.:)
 
So David, where do I send the $40.00?
:cool:

I purchased a set of Wynn's ported manifold, how can I determine if they are 89 style or 95 style? How can I determine if I should use them or look for something better for my recently built engine?
 
Back
Top