"My" ideal header

Hey guys, a little reality check here........

It is kind of a fun enterprise to kick around various header designs, etc. But give the manufacturers some breaks. Don't you think that they dyno tested their headers before getting them to market? Yes, they are compromise designs, but they need to fit as many combinations of the X as possible.

Custom designing and building a header is a time consuming and expensive enterprise. I have designed and built two, both for specialized and racing applications. I do not want to even try to calculate the costs.

I did not dyno tune them, but used some basic (very basic) principles to pull the designs together - Generous radii (because I had room), diameter to suit application (as a practical matter there is a choice of 3 - 1.25, 1.375, & 1.5inch),
length per guidelines set down by Colin Campbell in the 50's (valve head to exhaust tip). Primary length was chosen for a high winding 1300 fuel injection motor with a big valve head & pretty hot cam. A much more challenging aspect is getting the primaries (whatever length) even. Mine were about .50inch from shortest to longest. From my reading this is outstanding. Note too, that I had extra room behind the engine in both cases.

So, are you looking for that last 2% of performance on a street motor? As an esoteric enterprise, great. As a practical matter, why bother? On the street, will you really feel the torque difference between peak torque at 6000 versus peak torque at 6200 RPM?

I would say concentrate your efforts ($$$) on the little things - port matching, combustion chamber volume matching - you know, the basics. You will never win a race with a rice motor, so concentrate on the chassis so that you can outrun him in the curves.
plaiaraceheader001.jpg

1 3/8 pipe diameter, 26" length +/-.25", 3" bend radius, 2.25 collector, pipes and muffler. Note that the X that this goes on (il mela verde) has a heavily modified and angled rear firewall. This header will not fit any standard X.
oregon_10_091_mod.jpg

This is the sort of handling you want. Springs and shocks are not radical at all. 175/275 lb/in coilover springs, front sway bar with center attached to the chassis, 5 way adjustable Tokico shocks on setting 4.

So ends my screed for today.
Ciao,
 
Last edited:
I'm thinking of building a header as well, this has been a good read. I was thinking of modeling the back of the engine and and engine bay so I could model the header.

I would be willing to work on this project, here is a pic of a header i just modeled and built the prototype of. this is a turbo header in to tight of space to get equal length.

Number1.jpg


IMG_5922.jpg


this is 2" primary tube.
 
Getting back on track

I was away from my computer for a couple of days, so just now responding to the latest posts here.

Mark P: Thanks for your input, great to hear from you. While I agree that designing and building a tuned, tubular header is a big task, I do not necessarily feel that it is out of the scope of the Xweb members; I am amazed at the extensive resources available here, and have seen much bigger challenges successfully completed by much less. Granted, the ideal way to come up with a new header would be to develop and test several configurations on a dyno; a costly endeavor. However it is not unreasonable for us to design a decent product based upon calculated parameters and software development tools. It can always be further evolved from there. I think the goal here is not to come up with the 'ultimate' header for a race-level engine, but to build one that works well for street applications. Most if not all of the past and current headers on the market have been either race-tune oriented or otherwise compromised beyond acceptable limitations. To the extent that dyno testing by Xweb'ers has shown many of them to perform below the stock factory cast-iron manifolds [on a street-tuned engine under 'real-world-driving' type conditions]. Even by looking at their basic designs (4:1 configuration, large tube diameters and short primary lengths, short collectors, etc), you can see where the goal here is a completely different approach conceptually; a 4-2-1 design, with tube dimensions (diameters and lengths) to suit our agreed upon 'street-tune' engine, to yield a broad power band within a realistic RPM range (this is a shortened paraphrase of the previously defined goals). Will it prove to work as desired? I guess we will find out. But frankly, even if it is no worse than a stock cast iron 4-2-1 manifold yet weighs less, then we are still ahead. And it is a fun challenge. Mark, I hope I don't sound like I'm discounting your comments...they are greatly appreciated. I'm only trying to clarify our intent here. Please continue to provide your respected experience and assistance with this project. Thanks.

Brian P: Thank you for joining in. The work you did with that turbo header looks excellent. Your skills with computer modeling will be a tremendous (and much needed) help here. I like your idea of generating a model of the engine and it's compartment to be able to work within. The biggest challenge looks like how to fit the rather long primary tubes in/around everything. Kind of like putting 10 pounds of potatoes in a 5 pound bag (as the saying goes). In general it may be best to try and keep things as simple as possible, to avoid unnecessary development and build costs. But get creative and see what you can come up with. Did you see the reference (earlier in this thread) to the discussion on Guy Croft's forum? His design had different goals, and therefore different tube dimensions, but the photos of his final product kind of give an idea how hard it will be to find a good configuration. He lost the rear trunk in that layout, hopefully we can find a way to avoid that. I'm a little behind on the software curve (I grew up before this technology existed), so I don't really know the full extent of your system's capabilities. Please let us know what information we can offer to assist you in creating some ideas. Thanks again for your interest in this project, anything you can offer is highly appreciated. Thanks.
 
A mathematical exercise for me

Because I am a science nerd by occupation and nature. And I am a dreamer. I like reality checks and apply them around here all the time, but I also like to dream and to theorise.

Not manufacturer or vendor knocking at all.

Simply pointing out that our options are limited to 4-1 systems, which is not ideal for street use. Simply pointing out the collectors of the available units could/should be longer as they present a restriction to flow if too short.

Will I go to the trouble of getting one made? Not sure, still lot's of basic work to be performed on the barn find before we even get it back on the ground and running.

Will I be pursuing that last 2%? Nope, because none of the other options mentioned have been applied yet. But there is no doubt that getting exhaust gases out efficiently is key to any plan.
 
Coordination of tasks

While this may be getting a bit ahead of things here, I was wondering:

Kevin B, what does Walt need in order to be able to put a design into actual creation? For example, if Brian P puts together a computer model (like the example of his turbo header), can Walt build directly from that? Or is something additional helpful for him?

Brian P, do you require any particular type of data beyond the dimensions that Rob's mathematical calculations will generate (like in his prior posts)?

I'm certainly not trying to rush anything, I know it will take time to continue developing these ideas. I'm merely trying to think through some of the logistics that might help in pulling our resources together.
Thanks to all.
 
Rob's calculations are all I need to start laying out the header, The modeling of the engine compartment and back of the engine is the time consuming/difficult part. My x is in storage for the winter, I still have snow on the ground, so I don't have it to model. that will come later when the weather is better as it has lost it's garage to the 20B RX7 project which will continue into the summer. I will also need access to a car with A/C if the goals are to include fitment for them.

I did read the discussion on Guy Croft's site, I don't want to give up the rear trunk like he did. The first thought I had was the primary tubes meeting at a flange at the factory pass through hole and the secondary tubes continuing on into the muffler space. It all fits nicely in my head, that's why I have the CAD, my head doesn't always have a firm grasp of reality.:)

I have a head I can pull the flange data off of. To get started we would need some engine compartment measurements.

As for the capabilities of my system... they are far greater than the capabilities of the operator..:)
 
Thanks Kevin and Brian

So it sounds like the idea of combining everyone's efforts here should work well. That is really cool. Again, there is no hurry...any and every effort is greatly appreciated as time allows. I am very grateful to all for sharing your interest in this.

Kevin, as I mentioned earlier I have a possible resource for some/all of the materials to help with the build. We can discuss that further when the time comes.

Brian: Thanks again, I'm amazed at what can be done. I wish I had the knowledge and gear to do this type of modeling. I am always thinking of ideas but seldom able to get them into any format where I can develop them further. I had a similar thought with the primaries coming down through where a stock system goes, before merging into the secondaries, which then carry into the muffler area. But that is just a mental image without any real measurements. All of my play cars are also in storage (in another state) and won't be accessible for a long time (due to my current family situation).

Thanks everyone.
 
Checking my numbers

I'll soon post the equations I used, which all come from A. Graham Bell's book.

Then someone can go over them and check that I have not made any errors!
 
Re: 4-2-1 Headers

Hi rjplenter. Very interesting thread. I happen to agree with your thoughts pretty much 100%.

From everything I have experienced and read, I agree that a 4-2-1 system is best for STREET use, and a 4-1 is best for RACING.

I have in my possession two sets of 4 into 2 headers for an X1/9. I'm 99% sure that they are both the upper portions of an old ANSA system from a '74 X1/9. Like the one in this this thread:

http://xwebforums.com/forum/index.php?threads/10995/

Interestingly, the primary tubes measure 1.25" O.D. about 15-17" long. They taper nicely into collectors of about 1.58" O.D.. Amazingly similar to your "ideal header". !!!

I do not have the rest of the system, but from the picture I would estimate the secondary pipes on the ANSA system to be maybe 18-20" long. And the single tube - including the length of the muffler - would appear to be maybe 1.8-2.0" O.D. and 40" long.

I guess we have no idea how much - or how little - design and testing went into this system. But I'm sure the engineers at ANSA had at least some knowledge of proper exhaust flow and tuning. It actually seems like they read the same book that rjplenter did !!!

I will try to post some pics of my headers installed on my engine.

Cheers, Doug
 
Last edited by a moderator:
More pics.

Somewhat different from rjplenter's plan, my current thought is to add a pipe something like this :





which should lead almost directly into a muffler like the Miata one in this thread :

http://xwebforums.com/forum/index.php?threads/10275/

Tailpipes should exit through the rear grill.

Any thoughts ?

Cheers, Doug
 
Last edited by a moderator:
Very cool Doug

Thanks for posting the pictures of your 4-2-1 header. These ANSA units must be very rare, how did you ever find two of them? Looking at the pic's in the referenced thread, the original lower half looks like it was all one piece with the muffler. I wonder if that lower half was prone to rust/rot, and therefore not retained by a prior owner, leaving only the upper halves that you have? Cool item none the less.

The "two-piece" flange like this was very common on 4-2-1 headers for older British cars (for some reason there were a lot of 4-2-1 systems for them back then). I don't know if it makes any difference with sealing the header to the head (for better or worse), but in some cases it does make it a little easier to get the header on and off.

Thanks for sharing. Let us know how your project goes.
 
Wow!

That looks excellent Doug. And amazing how close Ansa's dimensions were to Bell's equations, which were not of course "invented" by him.

Love seeing all those photos.

Cheers,

Rob
 
Reply

Hi Dr. Jeff. Strangely enough I have absolutely no memory of how 2 of these very rare items fell into my possession. But I somehow think your theory is correct. I have had them at least 25 years. Back then the ANSA mufflers rusted out pretty quickly up here. The header portions seem to be some kind of low grade stainless. Both of mine are still very solid. Much better than they actually look in the pics. The set I installed on the engine were actually in really good shape until ... which leads me to a very bizarre story.

Look carefully at the set installed on the engine. Notice the large dent on the #1 cylinder tube. Well...

A buddy of mine helped me build the engine. He had a shop out in the countryside with living quarters above. Late one night (after the engine was totally completed but before I could pick it up) he was in his usual drunken stupor. He was upstairs and awoke and thought he heard an intruder downstairs in the shop. He reached for his trusty shotgun and while loading it the shotgun discharged while pointing downwards. The shot ripped through the floor and - yes folks this is the truth - scored a direct hit on my precious header on the engine in the shop below. Yeah, drunk people and guns are a great combo. But I guess I should be thankful I wasn't working on the the engine at the time.....
 
Available still... for a price...

http://www.ebay.com.au/itm/Fiat-X1-...=AU_Car_Parts_Accessories&hash=item589296e84f

though the Ansa mufflers tend to fall apart pretty quickly, certainly on the several systems I've seen over the years...and the muffler is really too small, and the tail pipes have tiny pea shooter outlets...but the header pipes work really well.

I would graft the ANSA headers onto a system like the one I posted pictures of in this thread...

http://xwebforums.com/forum/index.php?threads/8908/

and it would be about the best you would get without custom designing and fabricationg to suit a particular camshaft/capacity/compression ratio and other mods...

remember the headers and muffler are all part of the air flowpath, and altering one component in the flowpath has a chain reaction on the other components...so anything off the shelf will always be a design compromise.

SteveC
 
Last edited by a moderator:
G'day Steve

That's a great photo of your muffler set up. I particularly love the looong collector taking 2 pipes into 1. That is key.

We have a 4-1 system on the rally car back in Australia. I was never happy with it and I don't think we gained much if any power over a stock system and I think it is becuase the collector is so short. This is especially troubling in a system trying to merge 4 pipes into 1. We watched as the paint burned off the collector within a couple of minutes of starting the engine and I remember thinking there must be a lot of friction (read: back pressure) in that spot.

Your points about a suitable muffler and not oversizing the tailpipe are also valuable. I certainly don't want be to droned out of enjoying my 4 speed X on the interstates where the speed limit is generally 75mph (125kmh).

Cheers,

Rob
 
If this were a Chevy...

It would be called a "Tri-Y" header... 2-1, 2-1, 2-1 configuration used for each bank of usually a small block V8. Many have applauded the power gains with stable torque numbers as well but they were more complex and expensive to design and build, hence were never as popular.
 
All excellent stuff

Thanks to everyone for your valued input.

Doug: That story about the shotgun damage to your header is as funny as the 'hooker' story from Larry. It is buried in this post:
http://xwebforums.com/forum/index.php?threads/16617/

SteveC: Thanks for your comments, very glad to hear from you. Please review the overall idea of this project for us. Do you think it is reasonable to design a header based on the general assumptions here? Granted, ideally, each engine's unique variables would require a slightly different header design. What we are attempting to do is choose a set of engine parameters and intended usage (as defined earlier in the thread) and come up with a header that will be close enough to work for that configuration and application (basically a hot street set up). While it may not get that last percent of performance out of the engine, hopefully it will be an improvement (with the added benefit of some weight savings) over the stock manifold. Naturally compromises are required, but we hope to find the best overall package. Its a work in progress, the final design yet to be determined. But looking at the currently available (limited) alternatives on the market, all are either designed for a much higher level of engine build (and intended use) or are extremely over priced (like the ANSA in the link referenced). So we hope to be able to come up with a good option at a reasonable price; the combined resources here on Xweb are not only amazing technical assets, but will allow the costs of the design/development to be contained and distributed. Please feel free to add your highly valued knowledge to assist with this. Thanks.
 
Last edited by a moderator:
You must log in or register to reply here.
Back
Top