Manfold/Header testing complete.

XR7 Dave

Registered User
After initiating this test it quickly became apparent that I needed to establish some testing standards and parameters to prevent skewing results and getting innaccurate results so that is why I had to re-do the test. The results have changed.

I realized in the first test that each port wasn't flowing the same in each manifold and I became concerned about exit angles for each tube and possible turbulence in the collectors. This led me to do two things.

1) I added a pipe extension to all collectors. The longtube Kooks already have an 8.5" x 3" collector on them so I extended that slightly to a total of 14" and I used that length on all tests. The other headers all have 2.25" oulets so I used a 2.5" extension on those and the manifolds (and 96 Bird header) all have < 2" outlets so I attached a 2.25" extension on them. I then tested the Kooks longtubes with a 2.5" reducer and tried the 95 Mustang header with both 2.25 and 2.5" pipes. In each case the flow difference at full lift was just .5cfm. Now is also the time to point out that in EVERY case adding the 14" extension improved flow!

2) I tested each pipe in each manifold. I blocked off the remaining 2 pipes when testing each one and the results are noted below. Some manifolds/headers definitely have better balance than others.

The results include flow at all points from .100-.450" lift on a totally stock cylinder head and are compared to a straight diffuser for a baseline. All tests were conducted consecutively on the same day for the sake of consistency. I also included the % variation between all ports and the gain or loss relative to the straight diffuser. This first test was embarked on specifically to show what flow advantage there might or might not be for each manifold/header on a stock engine.

I hope you find these results helpful. I will be performing tests on ported stock manifolds also but that will have to wait as I do not have the porting done at this point. I will update the test when I have that data.

I will also be doing another test on a fully ported set of heads at lifts up to .600" to give a better perspective of what potential these manifolds/headers have.
 
Last edited:
Nice work Dave.

I see our stock manifolds flow slightly more than the 96 style, but look at how unbalanced they are. The 95 one seems to be a good choice. What was different about the 95 manifold?

I see the longtubes flow the most, that was expected. The mid lengths and mac's are very close in average and are pretty equal for each primary.

Going from stock un ported early style manifolds to a set of Kooks longtubes would gain an average of 13.5 cfm per cylinder, but at $700 is it worth it, and does 13.5 cfm make any power increase?

Looking forward to seeing what the ported SC manifolds can do as it is the common theory they are good to around 300 rwhp maybe more.
 
I was pleased and surprised to see that the longtubes were the best flowing of the bunch and actually improved over a straight diffuser. To me that was impressive. The Kooks longtube has a nice evenly configured collector which I feel is a significant factor in the good results. The Kooks that I have contain an 02 sensor in one primary lead and it was this tube that had the lower flow #'s compared to the rest. I did not test it with the 02 out (it's not coming out either) so I can't verify that but that's what it looked like to me.

The mid length Kooks also performed well and I was impressed with the consistency port to port. It may not result in a performance enhancement on an otherwise stock SC but it seems like a good part nonetheless.

The MAC shows what happens when you try to stuff a header where a header doesn't fit. The longer tube had more bends and suffered due partly to it's poor entry angle into the collector. The middle tube was a little better because it dumps at a better angle into the cramped collector and #3 lead the charge because it is extremely short and also dumps straight out the collector. In this case you can clearly see how a very short and direct tube can outflow a better designed but longer tube. The MAC concerns me because of the variation between ports although it is clearly better than a stock manifold.

The 95 Mustang header did better with an extension on it but clearly it doesn't keep up with the rest. Why? Because the tubes are jammed up at the collector. They just don't have enough cross sectional area at the point where they join the collector. It's like 3 little venturies there. It should be noted that it is very consistent though which has value compared to the MAC, Bird Header, and the 89 stock manifold.

The 96 Bird header in this test was ported lightly so the results are most likely somewhat better than a stock untouched header would have been. I expect that this one will look much worse once we get some real flow going through the head but for now the results are not THAT bad. For a stock motor it may be a decent improvement over a stock 89 manifold due to it's improved consistency.

The 89 manifold surprised me. I always wondered why Ford changed the manifold in 94 and now I know. The center port is severely handicapped to the point of possibly causing a real issue. With over 17% difference between it and the front port it is very unbalanced. If you look closely at the design you will see that the center port must make 2 almost 90 degree turns before exiting into the rest of the exhaust. Obviously this is a bad condition.

The 95 SC manifold rocks. I can't believe this thing. It looks ugly enough but obviously someone at Ford was burning the midnight oil to get this one right. Performance was on par with the headers and consistency was better than the MAC shortie. Who would have guessed?

So, in this test the results are as follows.....

#1 - Kooks Longtubes
#2 - MAC Shortie
#3 - Kooks Mid Length
#4 - 95 SC Manifold (unported)
#5 - 89 SC Manifold (unported)
#6 - 96 Tbird Header (ported)
#7 - 95 Mustang Header

More to come.....

:)
 
I'm willing to bet a ported 95 manifold would outflow the Kooks mid length and the MAC at stock lift levels. With more lift, I'm not so sure as of right now.

Are you going to test the ported manifolds on the stock and modified head with both stock and higher lift?
 
Thanks for sharing

Thanks Dave,

Very cool of you to

a. do the testing

b.share the results

I'm also impressed with your attn to detail
 
Thanks for going through all the trouble of testing and sharing the results! :) Looking forward to seeing how a set of ported stockers stack up...
 
Me too......

I have a set of stock 89 manifolds that I just got finished porting and are ready for installation.....

It looks like I might have to go find a set of 95s and port them.....:rolleyes:
 
I couldn't help myself and started on phase II before spending any time porting any manifolds. For one thing I need to get the loaner headers back to their owners and second I have to complete some other work before I start porting exhaust manifolds. Some of you know what I'm talking about. ;) I'll do the manifolds later but you are not going to BELIEVE what I saw when moving up to a better flowing head and higher lift. :eek:
 
What, 65 views and only three people have any comments about the test results?

Maybe I'll just send out an email with the final results. :rolleyes:
 
I see our stock manifolds flow slightly more than the 96 style, but look at how unbalanced they are. The 95 one seems to be a good choice. What was different about the 95 manifold?
The center port on the 89 manifold on the passenger side in particular has some very poor exit angles and then the front port has no restriction at all and badly outflows it's neighbor. The 95 manifold adds some direction to the middle port and smooths the transition into the collector as well as changes the outlet angle of the collector. Then the 95 manifold also crimps the front port slightly. Whether this is done to limit front port flow or if it is done to minimize backflow between the two I can't say but it seems to work.

I see the longtubes flow the most, that was expected. The mid lengths and mac's are very close in average and are pretty equal for each primary.
I did not expect the longtubes to outflow the rest. I thought that the longer tubes would form a restriction of sorts but I was clearly wrong. The length of the tubes does not seem to have any bearing on flow. The key is collector design. On the MAC headers one tube has an awkward approach angle to the collector and that port is significantly down in flow compared to the other two. Where it not for that issue the MAC would outperform the Kooks mid length. I don't see how you can fix that issue so if you use those you'll just have to accept that.
Going from stock un ported early style manifolds to a set of Kooks longtubes would gain an average of 13.5 cfm per cylinder, but at $700 is it worth it, and does 13.5 cfm make any power increase?
Therein lies the basis for all the claims of stock manifolds making as much power as headers. You have to look closely at the individual port flow numbers to see what is happening. With a stock manifold some ports are flowing darn near as much as a header where as other ports are lagging far behind. The average may be close, but each cylinder does not look at averages. So you have some cylinders pulling well and other cylinders having issues. Could this lead to detonation in some cylinders more than others? Many questions, not many answers.
Looking forward to seeing what the ported SC manifolds can do as it is the common theory they are good to around 300 rwhp maybe more.
Ported manifolds have supported over 400rwhp, not 300rwhp. Was it efficient? I don't think so. But the truth remains that Jim Oneil made 390rwhp through stock unported manifolds, Bill McNeil made 380rwhp through ported manifolds, I made 425rwhp with ported manifolds.... And the list goes on. How much power would longtube headers provided over those manifolds? The answer is inconclusive.

One thing is very clear. The use of Mustang headers is a big mistake if you want to make power. Anyone who has them on their car might as well start making other plans. :cool:
 
Hey Dave,
Thanks for doing this test. I always wonder what would work the best and for the money, especially the cost of the kooks headers. Looks like I'm going to have to bite the bullet and buy those over price headers.

David

(p.s If you decide to email the results, I be interested)
 
...
Some of you know what I'm talking about. ;)
...
What, 65 views and only three people have any comments about the test results?

Just trying not to distract you too much is all.;)

Let's start with some reality checks. Perhaps avg hp found on cars with each exhaust manifold. I.e. do we see a block occuring. Or if we work back from exhaust flow, to hit that cfm of exhaust flow, how much fuel and air is combusted in the chamber for a standard 3.8liter bore and stroke? Then is this what we are even capable of creating with our current intake capabilities? Say we have a 8.5:1 compression ratio with 12.5psi of boost at 4000rpm. Could we figure a needed flow from that input?

That the long tubes are proving better is very interesting. As we try to make our engines reliable at higher HP levels, cylinder balancing will be important. I.e. are we pushing that much fuel just because one cylinder is leaning out? Or is one cylinder on the ragged edge but it's not showing up in the wide band o2 sensor at the tailpipe?

So I guess I need to become familiar with we need going out for what we put in to be able to better place things on the priority list.
 
XR7 Dave said:
What, 65 views and only three people have any comments about the test results?
Nice WORD doc there Dave... your hidden macro opened up my Mastercard account and sent $40 to "Super Coupes Unlimited" :eek:













Just kidding... :D

-----------------

Obviously, next to true headers, OE manifolds appear to be the next best choice. I look forward to seeing the results after porting :)

With regard to the middle runner, and knowing what we know now, perhaps a little extra porting to that one would help (I know there's not much material to work with, but every little bit could help).
 
The center port on the 89 manifold on the passenger side in particular has some very poor exit angles and then the front port has no restriction at all and badly outflows it's neighbor. The 95 manifold adds some direction to the middle port and smooths the transition into the collector as well as changes the outlet angle of the collector. Then the 95 manifold also crimps the front port slightly. Whether this is done to limit front port flow or if it is done to minimize backflow between the two I can't say but it seems to work.

This brings another thought to my mind. The last two 3.8's with Blown head gaskets Ive seen were between the passanger side last two cylinders, and Ive heard of others blown in the same place. Im wondering if it might have anything to do with the exhaust manifold.

- Dan
 
Let's start with some reality checks. Perhaps avg hp found on cars with each exhaust manifold. I.e. do we see a block occuring. Or if we work back from exhaust flow, to hit that cfm of exhaust flow, how much fuel and air is combusted in the chamber for a standard 3.8liter bore and stroke? Then is this what we are even capable of creating with our current intake capabilities? Say we have a 8.5:1 compression ratio with 12.5psi of boost at 4000rpm. Could we figure a needed flow from that input?
From what I've seen it's not so much a matter of flow volume as it is efficient design. The fact that the 96 bird header flowed as much as it did points out that cross section area isn't the biggest factor. Velocity is probably a more important consideration and things that kill velocity will kill flow.

That the long tubes are proving better is very interesting. As we try to make our engines reliable at higher HP levels, cylinder balancing will be important. I.e. are we pushing that much fuel just because one cylinder is leaning out? Or is one cylinder on the ragged edge but it's not showing up in the wide band o2 sensor at the tailpipe?

I think this is an important consideration. The harder you run the motor, the more important consistency from cylinder to cylinder will become. If you believe that pulse tuning is important for a supercharged engine then longtubes are the answer. If you don't believe that then the next most important thing is a smooth collector and pipes that have similar flow capability. A header like what Wynn was working on might be just the answer for some people. I think the mid length Kooks also represent an excellent header choice.

I won't be flow testing the ported manifolds that were on the XR7 when it made over 400rwhp because - well - they are still on it. So I need to do another set. I spent quite a bit of time on those and I won't be spending that kind of time on an 89 manifold because I know that there is no way to balance out the extreme problem that exists there. You just can't get to the place where the problem is. I do have some 95 manifolds here that will be getting the extreme treatment though and we will see what that brings.

The only thing I'll be doing to the 89 manifold is a collector port and a minor cleanup around the port entry. I'm guessing that it will improve flow but I doubt it will do much if anything for the flow imbalance. I don't see any point in increasing flow if you can't cure the imbalance.
 
Nice WORD doc there Dave... your hidden macro opened up my Mastercard account and sent $40 to "Super Coupes Unlimited" :eek:

Damn. I always mess up some mundane detail like a decimal place or something. It was supposed to take $400. :(


You wait until you see the results of all these parts on a head that flows good numbers. On a professionally ported cylinder head the choice of manifolds/headers can mean a big difference!

Dan, I don't know if there is a correlation or not. Could be a lot of these things adding up though.
 
Last edited:
nice work Dave, I was very interested to see how the Mustang headers did, great info, can't wait for ported numbers.
 
What, 65 views and only three people have any comments about the test results?

Maybe I'll just send out an email with the final results. :rolleyes:

Dave,

I was waiting to see the rest of the results before opening my yap. But since you asked.......I'm not really concerned with what they flow on stockish heads. I want to see what they do with a set of Steig or similar heads and at higher lift.

So far the results are about what I expected, except for the late model factory manifolds....Is that what Bill McNeil had on his car ?

David
 
BTW, here is a pic of a Mustang header with the collector cut off, you can see how they are pinched into a triangular shape. At least they are all fairly evenly pinched which may be why they are fairly consistant. I dunno

119920.jpg
 
Back
Top