HEADER DESIGN COMMENTS:       A Header Design Advice Letter

Super Coupe Club of America (SCCOA)
 
Dear Member(s):

Thank You for taking advantage of our Header Design Advice Service, as it should help you  better Design and Construct a well-performing and longer last set of headers.

A brief history of our Advice Service:  After literally writing all advice by hand for years, several years ago I switched over to using the computer to, hopefully, speed things up (and make everything more legible as well).  As time went by, I combined many "good" parts of different advice letters into what I called a Master Advice Letter.  I then edited this "master" letter at later dates for other Advice Requests (much advice is actually given more in the for of explanations or examples and, therefore, repeats quite often).  Unfortunately, while the advice given was fine, the time involved often took hours per letter and got worse as the master kept increasing in size.  Spending hours per request is simply way too much time as the Advice Service payment doesn't cover that much time (plus not everybody that uses the Service buys parts or headers anyway because they don't like what I tell them or I don't confirm what they want to believe is correct).  To make matters even worse, excessive time spent on any one advice request also causes all other following requests to be delayed even further - sometimes to the point that some had had to wait weeks, even months to get advice - which is simply way too long.

More recently I have been working on a third and hopefully better way of giving advice.  This letter represents that “better way” – a way that not only retains the accuracy of earlier advice but one that also contains better examples and explanations.

Rather than spending lots of time editing the master letter down to fit a specific person’s situation I am in the process of reformatting many of the design comment areas of the master letter into Advice Sections that can be referenced from the first part of an Advice Letter (where the specific advice is given).  As evidenced in this letter, SPECIFIC design specifications and relevant design comments are confined to one area up front with a second area of other relevant comments at the end.  Basic explanations needed to make that advice more valuable/understandable are incorporated in the enclosed multi-page Advice Sections document that can be referenced in the letter.  (In most cases these sections are many of the aforementioned “good” parts of the old Master Advice Letter just rewritten in a more general manner).  Hopefully, this new method will work out better for everyone concerned plus an added benefit to all users of our Advice Service will be the additional information provided in this new Advice Sections format.

(Presently, only those areas whose section number is followed by a letter have been rewritten - the letter indicates how many times they have been rewritten, i.e. “C” = 3.  In all sections, try to pick up on the IDEAS presented rather than the specifics given.)

I have looked over the two filled out Questionnaires that have been sent here, the letter that accompanied the two Questionnaires and have also considered a few ideas I’ve developed from our earlier conversations.  Because you said that these comments will probably show up on the club’s website, I’ve spent CONSIDERABLE extra time going over some of the design basics about header design and also the correct application of these design basics so that you AND your members will benefit from this advice.  Before I offer my design recommendations I think it best to discuss how unusual a design situation the supercharged Super Coupe presents as well as being asked to make two different header design recommendations relative to two different engine combinations as well.

Looking at the situation overall, you start with an engine that normally aspirated (or non-boosted) apparently produces horsepower in the low 200s (210 HP mentioned in the Questionnaire).  However, by adding a supercharger (and other engine modifications) it is possible to get this engine to produce as much as 550 hp (net) (horsepower estimate in one Questionnaire) and being told that the supercharger can consume 80 HP in is own operation means the engine is producing about 630 HP (gross) in “Maximum Street” trim.  This means that the “Maximum Street” version produces three times (300%) the amount of horsepower as that produced by the standard normal aspirated version.  The “Mild Street” version was said to produce 350-380 HP (estimate in the other Questionnaire) which means it produces 430-460 HP with the supercharger load added in.  This means that the “Mild Street” version offers double the horsepower (200%) and, by comparison, the “Maximum Street” version increases that amount by another 50%.  I wanted to use percentages to illustrate just HOW MUCH different these two versions are to each other as well as how much different they both are relative to the standard “basic” engine.

The “Mild Street” version was also “usage” described as:  “Daily, high performance street use, non-boost 95% of the time with occasional track use and still using 2 1/2” chambered or straight through, essentially ‘quiet’ mufflers”.

The “Maximum Street” version was “usage” described as:  “Multi-use – all-out hi-performance street use” with 2 1/2” duals and straight through (louder) mufflers.

I hope that you see (and appreciate) the difficulty of offering header design advice here due to the incredible RANGE of power output relative to what may be considered “normal” - partial throttle usage on the street with no boost (one extreme) and then seeing another usage combination as a supercharged race vehicle but limited to the degree that the car must still be capable of being driven on the street (the other extreme)!

For a normally aspirated V6 (peak horsepower in the low 200s), I would recommend using Equal Length 1 1/2” OD x 38” long header tubes with a collector outlet diameter of 2 1/8”.  This design would produce more bottom end and mid-range power then larger tubes would provide yet would NOT restrict top end power (remember, just over 200 horsepower!).  Compared to much shorter headers and/or headers with large amounts of tubing length variation this design would ALSO increase power down low and into the mid-range yet NOT restrict top end power either.  This design would be a good choice for non-boosted operation.  However, since the supercharger’s positive effect must be included in the recommendation I would then offer the following design advice:

“MILD STREET” HEADER  DESIGN  RECOMMENDATIONS:

Please keep in mind that we don’t do much testing with mufflers or exhaust systems so these recommendations should be accepted with some reservation as to the fact that there might be a better way of designing the exhaust system.

1 1/2” would be the best choice if the engine did not have the supercharger.  By going to the next larger tube size (1 5/8”), the header design is “tilted” so it would be better with the supercharger but, at the same time is only a “one tube” size increase, so that it does not unduly hurt bottom end and mid-range power at part throttle.  For what its worth:  I’ve listened to guys complain about headers from big companies NOT working at all or even hurting bottom end power dramatically to the extent that they felt that adding headers seemed to be a total waste of time and money (which seems to be the club’s experience).  In the majority of these situations, I’ve felt that the headers being complained about were two tube sizes larger than what my experience deemed correct.  Because of this, using a 1 3/4” tube on this engine simply doesn’t make sense for normal partial throttle “street” usage (read also as having good gas mileage potential) as it will primarily do damage over the engine’s rpm range unless the engine is getting boosted.  Keep in mind that the engine’s displacement is 232 cubic inches (equivalent to a 309 cubic inch V8 with equivalent cylinder volume) and, without the supercharger, 1 3/4” has to be considered quite large – actually the word HUGE would be more appropriate.  For those who might think that 1 1/2” OD tubing (with an 18 gauge wall thickness) seems ridiculously small, consider that the internal cross-sectional area of three of these tubes is 4.63 square inches and the internal cross-sectional area of a single 18 gauge wall 2 1/2” tube is 4.53 square inches.  In other words, three 1 1/2” tubes are actually internally larger than one 2 1/2” tube (one side of a dual 2 1/2” exhaust system).

Looking at the exhaust system after the headers, the current described configuration seems OK.  A larger system than 2 1/2” would reduce bottom end and mid-range power due to lower gas speed in the system and would only be of advantage at higher rpms and full throttle/boosted operation.  Since one of the design requirements given me was maintaining good performance in non-boost driving it makes no sense to make the exhaust system any larger than 2 1/2”.  Another thing to consider is that the "best average" collector size for a three tube 1 5/8” header is 2 1/4” so making the collector 2 1/2” (to make it match the 2 1/2” exhaust system) also “tilts” the header design again slightly in favor of the supercharger without going overboard in sizing.

Please keep in mind that we don’t do much testing with mufflers or exhaust systems so these recommendations should be accepted with some reservation as to the fact that there might be a better way of designing the exhaust system.

If this were an all out race vehicle (with more radical valve timing, no concern for power produced below 4,000 rpm, etc.) I would definitely recommend a larger than 1 7/8” tube size – 2 for sure but would be curious to see if 2 1/8” might even work better. 

I’m not a bigger is better” kind of guy as I’ve seen too many times over the years the damage that idea has created when applied to a vehicle’s engine but, here – because of the supercharger’s increased boost to 20 pounds per square inch (from 15 pounds per square inch in the “Mild Street” combination) as well as other engine changes, there is a lot more air going into and through the engine – and that means a lot more air also has to get out.  Looking at the earlier estimated power output of 630 horsepower coming out of this V6 and converting that output per cylinder to a V8, we would be talking about a 840 HP V8 and a 2” header would be considered small on that engine so 2” doesn’t seem at all too big when put in that context.  The counterbalance to recommending that size here is having to drive the car on the street.  With a 2” header and little or no boost plus the more radical valve timing, I think the engine would perform rather badly down low and in the mid-range until the boost really started to climb.  Up until that point is reached I believe the overall “feel” of the car (throttle response, etc.) might even approach the point of being disgusting or at least disappointing.  So I have to recommend no larger than 1 7/8” due to the necessity of maintaining some semblance of bottom end and mid-range performance (especially at part throttle).  Keep in mind here that the 1 7/8” tube size does represent a two tube size increase over the 1 5/8” header and this means that this much larger header design will cause the engine to produce less power at the lower portion of its rpm band.  The smaller header/milder engine combination would definitely be quicker on the street up to a certain rpm but the larger header/wilder engine combination would overall be faster.  This would definitely show up when tested/compared at the drag strip. 

Now, looking at these two designs and comparing their advantages AND disadvantages one could also consider a third (or “middle”) header design – 1 3/4” OD x 36” long but I don’t want to confuse you with more information (or have I already done that?)  Well, guess what, I AM going to make things more difficult (even more confusing?).  If one were going to run the “Mild Street” version more at the drag strip and desire the car to perform better there and be willing to lose some street performance (trade reduced performance at lower rpms for better performance at higher rpms) then a 1 3/4” OD x 36” long header along with a 2 3/4” exhaust system would be a better matched design to that engine and use combination!

Now back to the “Maximum Street” recommendation.  The "best average" choice of collector size for a three tube 1 7/8” OD header is 2 3/4” but with the supercharger I would go up to 3”.  This isn’t a huge exhaust tube size when referenced to 630 horsepower but it does point out that using 2 1/2” on an engine like this doesn’t make any sense.  (Note that a 3” tube is 46% bigger in inside area than the inside area of a 2 1/2” tube.)  I read a comment with the Questionnaires that headers have been a disappointment on this engine – not working out well.  However, and particularly with the “MAXIMUM STREET” engine, a 2 1/2” exhaust system AFTER the headers has to be considered THE primary restrictor at full throttle and full boost so it wouldn’t be exactly fair to criticize the previous headers as being ineffective (even though the ones I saw in earlier provided pictures didn’t look very good) when they may not have been entirely responsible for the disappointing minimal (or zero) performance gains.

You may have noticed that the second design header is 4” shorter than the first design header.  This was done to make the second header a better match to the higher rpm capability of the second engine.

Keep in mind that the longer tube length of either header as well as making the headers truly equal length will improve performance over the engine’s entire rpm range.  Shortening the lengths by 4-6” will basically reduce bottom end power and even up into the mid-range.  Shortening the tubes even more will just further reduce bottom end and mid-range so these lengths are important and this is particularly important to maximize street performance.

While I was asked to provide two header designs, I was also asked to provide “flexible” designs.  I guess that means designs that will work equally well on many different engine and usage combinations.  Unfortunately really good header design doesn’t work that way.  Header manufacturers WANT guys to believe that they can make one header that will work well on everything but that is wishful thinking on their part because it really doesn’t work that way.  If one looks elsewhere in the performance industry, the companies that offer camshafts haven’t figured out a “flexible” camshaft design – in fact, they might offer 20, 30, 40 or more camshaft designs plus offer custom cam grinding services as well.  Carburetor and intake manifold manufacturers offer multiple choices of size and design variations.  Cylinder head manufacturers offer choices of head designs, port volumes and valve sizes.  The list goes on and on except where headers are concerned because in the header industry having a choice of even two designs from any one manufacturer is definitely an exception – certainly NOT the rule.  And, to make things even worse, the header industry doesn’t even provide an accurate product in most situations.  Would you buy a camshaft that had different lobe lifts and/or durations for each of your engine’s cylinders?  I doubt you would yet that’s what the header industry essentially offers when it makes headers with different tube lengths!

To put it another way:  In this discussion, based on differences in an engine’s parts combination as well as a Super Coupe’s intended use or purpose, one of these FIVE different header designs and exhaust system design combinations has to be considered:

(1)1 1/2” OD x 38” long, 2 1/8” collectors, factory exhaust – no boost

(2)1 5/8” OD x 38” long, 2 1/2” collectors, 2 1/2” exhaust – boost, broad power band

(3)1 3/4” OD x 36” long, 2 3/4” collectors, 2 3/4” exhaust – boost, higher rpm power

(4)1 7/8” OD x 34” long, 3” collectors, 3” exhaust – higher boost, much higher output

(5)2” OD x 34” long, 3” or larger collectors & exhaust – higher boost, race only

Note than on the 2 3/4" system one would use 3" inlet and outlet mufflers with 2 3/4" to 3" cones (which we can supply) to adapt the 2 3/4" exhaust diameter to the 3" mufflers.  One could also use 2 1/2" tailpipes with 2 1/2" to 3" cones adapting the muffler outlet to the tailpipes.

When I read in the Questionnaires and accompanying letter the disappointment with headers previously used on Super Coupes, is it any wonder that their power gains were minimal or non-existent when the appropriate header designs weren’t even available?  Keep in mind that I’ve also been told that these other headers that have been tried did not have tube lengths anywhere as long as the above recommendations, were NOT equal length, and, I doubt, because it wasn’t mentioned, that these other headers had an ideal collector transition shape either. 

Having made the above design recommendations, where are these designs?  No one makes them so you can’t buy them.  That leaves only one alternative – build them!

For those who might want to build one of these designs, keep in mind that we’ve been helping others build their own headers since 1964 so we’re definitely not novices in this game.  Our experiences with others building their own headers establish that most guys can build a header within +/- 2” of target length.  While this is not perfect, it is better than what most big header manufacturers are making (because they are mostly concerned with costs and that dictates simplicity of design and that means equal length designs are minimally offered due to their increased complexity) so if someone wants to argue about the merit of building headers they must consider that they are indirectly arguing that performance isn’t important.  Plus, based on what we’ve observed for many years, if a guy can’t build a header with less than a +/- 2” error on tubing length, he isn’t trying very hard or he just doesn’t care enough to do a really good job.  Another thing to consider is that we are talking about building a V6 header here – a three tube header on each side of the engine rather than four tubes (as on a V8) so building an equal length header on a V6 Super Coupe isn’t going to be near the challenge of building a four tube equal length header.  Another construction advantage of this particular engine is that the cylinder heads have round exhaust ports and it is a lot easier to build headers by being able to put round tubes into round ports than round tubes into any other port shapes.  One construction tip to keep in mind when building these headers is that if you can’t hit the recommended target length, build the tubes longer rather than shorter because going longer isn’t as bad an error as is going shorter (shorter tubes will reduce bottom end and mid-range power).

Relative to one being “intimidated” by the thought of building a header:  It’s unfortunate that magazines do not present the discussion I offer relative to one being able to build a header BETTER than what one might be able to buy.  Obviously, that’s not going to happen because that type of discussion basically pisses off everybody in the header business but it’s the truth.  Guys simply don’t know about the design mistakes and shortcuts that are made by many header companies.  From that they obviously don’t realize that by NOT repeating the mistakes made by these same companies and also by NOT cutting corners, they CAN build a BETTER header.   I look at what I do here as giving guys a good push in the right direction.  By taking advantage of the Header Design Advice Service I offer (as illustrated here), that basically eliminates the possibility of a guy building a header that has many “performance” design errors.  What is left is the basic construction process.   I never tell anyone that building a header is an easy thing to do.  It isn’t and it really can be a good test of patience – particularly when trying to figure out an equal length header design is part of the project.   At the same time, building that BETTER header is what insures the best outcome of the overall project.   I’ve been building headers since 1962 and have been helping guys design and build their own headers since 1964.  I’ve watched a lot of guys put a lot of well performing vehicles together in the process.   The irony of this, however, is that MOST OTHER guys don’t see header design as important (which is the basic message coming out of magazines due to their unwillingness to say things that might offend advertisers) which often gives considerable performance advantage to the guys who ARE paying attention to good header design.   Note that the letter received with the two Questionnaires pointed out several issues related to having less than desirable header design experiences and, when I read over the provided information, I saw situations that required header designs much different than those currently or previously available.

Please read the advice/discussion sections that follow this letter as they pertain closely to your situation (others are enclosed for their informational value).  Some sections haven’t been rewritten to go with this new advice format so ignore the specifics of those sections - just try to understand the “ideas” being presented. The sections are:

Didn’t like what I saw in the pictures sent here.  Too short, not equal length, didn’t like the way the collectors were shaped.  3” outlet collector on a three tube 1 3/4” header doesn’t make any sense as 3” is too large for THREE 1 3/4” tubes.  If it was a FOUR 1 3/4” tube header that would make more sense (but that still may not be the best choice – as that still depends on other variables).

Already commented on.

Don’t look at what others are doing as most DON’T KNOW what they are doing.   I have been building headers and selling header kits and header parts for over 48 years now and have dealt with thousand of fellows over all of those years and I have met very few that have investigated their exhaust system design to any extent.  Very few test - most copy.  To beat these fellows or just do better overall, you CAN’T copy them as that almost GUARANTEES you will, at the very best, either just keep up with them or end up with a vehicle that will NOT perform up to its full potential.  One of the many reasons why we design and build better headers here is that we DON’T copy what others are doing.

While most manufactured headers are made in one piece because it is cheaper to do so, many of our customers find it easier to make a header (Equal Length or not) if they design it so that it can be taken apart somewhere.  We’ve done this in many of our headers by using two bolt tube flanges in a header tube (or tubes) so that the upper section(s) of the tube(s) can be removed from the main header body.  By being able to remove some portion or portions of a header, this reduces the overall bulk of the header so that it installs/removes easier plus it allows you to route header tubes in and through areas that would be otherwise impossible.  (Note that the header flange is also cut apart.)  Because these tube flanges are sealed by gaskets, the connection is leak-free plus it is capable of being easily disassembled at a later date and also makes the header stronger (as compared to slip-type connections usually used by large header manufacturers).

We have header flanges for the 3.8 liter V6 engine used in this car.  Available in 1/4”, 5/16” and 3/8” thicknesses as well as with 1 1/2”, 1 5/8”, 1 3/4”, and 1 7/8” ID round ports.  With tubes welded in, a finished 1 3/4” ID flange port (a 1 3/4” OD tube has an 1 5/8” ID) matches the 1 5/8” ID exhaust port just about perfect.  The two smaller diameters will obstruct flow slightly due to some port blockage.  For an extra charge, we also offer FINISHED Header Flanges where the tubes are formed into the header flange ports, TIG welded, ground flat, and internally deburred.  On this particular engine - if we make FINISHED Header Flanges for a customer – we can use tapered starter tubes rather than straight tubes out of the port.  In other words, we would start with a full size 1 3/4” ID port in the header flange and then use a starter tube tapered up from 1 5/8” or 1 1/2” to match the full size header flange port.  This costs a little more money as we have to use a hydraulic press to taper the tubes but then there is no blockage of the exhaust port.

I often find myself not agreeing with header design recommendations from the automobile companies.  In some cases, I think that the header dimensions given represent a minimal design effort on their part or, in some other cases, come from “all out” racing situations that do not, in any way, resemble their normal, assembly line produced products.  I also don’t think they test correctly and must use headers that happen to be lying around in or near a dyno cell when an engine comes in to be tested.  Then, dyno tests are run quickly just to get some numbers for the marketing department (catalogs, brochures, press releases, etc.).  I have even read in factory provided sales literature the comments that they used a certain header design (and carburetor size) when they tested the engine to determine its power output capabilities.  The literature does NOT say that the header design (or carburetor size) is recommended, it only states what was used in the test to get the numbers.  This has proven to be very misleading as, after years of this, I have heard way too many guys (who totally misinterpreted the information provided) talk about these “factory recommended header designs” when, in fact, the factory literature did NOT say that at all!!!  This doesn’t say anything nice about their comprehensive reading skills.

Reflecting on the above paragraph relative to a personal “header design” situation of my own:  In December 2006 I purchased a 350 HP RamJet 350 crate engine from Chevrolet for a 1952 Dodge (that my Dad bought new) that I have had in storage for 43 years.  The “recommended” header dimensions given in the brochure that came with the engine would only serve to reduce bottom end and mid-range power and not give a top end gain over what I know to be a much, much better design.  Their header design recommendation of 1 3/4” OD x 32-36” with a 3” collector is too big and too short for the 350 horsepower rating of the engine (which will NOT rev over 5500 rpm as there is a rev limiter in the engine’s computer).  To make matters worse, their recommendation when referenced to use on the street (which is where this engine is used the vast majority of the time) makes even less sense – particularly if one uses an automatic.  The collector size is too big for the same reasons.  (While not being set up to just drag race, I am hoping that this car will run low to mid 13s at 100+ mph at the drag strip and I know that their collector size recommendation of 3” will not work on a car like mine that is that slow!)  The 1 5/8” headers I have built do have a 3” collector outlet but I will be experimenting with smaller/tapered collector extensions at the drag strip and I fully expect a 2 3/4” collector diameter to work much better than 3” in the quarter mile and I won’t even be surprised if the car runs the quickest with a 2 1/2” collector diameter – particularly if I run the car at the eighth mile drag strip near here.  The car weighs 3875# (about the same weight as a Super Coupe), has a 700R4 transmission in it along with a relatively low stall speed converter (2500 rpm) and 3.73:1 gears in back so it will need all of the mid-range power the engine can muster to get the car off the line as quickly as is possible.  I expect the smaller collector diameters to work better because I’ve already seen this happen many, many other times on many other cars!) 

Picking on Chevrolet again:  To make their header design “recommendation” even worse, Chevrolet does NOT, in any way, point out how important having an equal length design is (and the fact that they list the length as 32-36” implies that length or being equal length is NOT all that important yet we’ve seen mid-range power change as much as 6-7% based on a 4” change in tube length) NOR do they talk about the importance of a good collector transition shape – which I regard as two very serious omissions if they are truly concerned about maximizing the engine’s performance.  (Then again, getting more power might even cause more of these engines to self-destruct - increasing warranty claims so, perhaps, there is a method to their "madness" after all!)  It is also interesting to look at the fact that one of the reasons why I bought this particular engine (and spent extra money to do so) was because I was led to believe by Chevrolet that the port fuel injection included will improve bottom end and mid-range performance over what I might obtain with a carburetor – yet their header design “recommendation”, if followed, would actually take away what the fuel injection was supposed to give me as an improvement!!  This is really great – “pay one guy” to get a gain and then “pay another guy” to take it away!  It is quite disappointing to see “advice” of this nature.

Also - collector SIZING is much more important than most realize.  I’ve proven many times that collector sizing has to be made available in no larger than 1/4” increments if one has to even have a chance of getting the collector sizing near correct.  When one gets down to smaller tube and collector sizes, I feel that collector sizes must be controlled in 1/8” increments as 1/4” changes start to represent area (velocity) changes of 20% or more!!!!!!!!!

One of the advantages of “merge” collectors, that so many seem to think are so great, is due to the fact that they are usually custom built to order and, therefore, are often sized more accurately that what most guys obtain in a mass-production header.  Since I’ve been telling guys to experiment with collector sizing since the late 60s, the primary advantage of a merge collector for many is actually obtainable in a much cheaper fashion.  Another advantage of the merge collector is the basic improvement in shape as well as a longer taper when compared to many of the stamped collectors used by the large header manufacturers but, again, I make collectors by hand to get a better/longer shape and also tell guys to form the tubes in V8 headers where they enter the collector into a cross pattern (+) so having an overall improved transition “shape” can be accomplished in another manner.  (ALSO - Keep in mind that merge collectors are NOT new.  I saw them being used in the 60s and there was even a header company around in the late 60s/early 70s that made headers with merging areas IDENTICAL to that found in merge collectors so, again, they definitely aren’t anything new.)

Collector sizing and merge collectors:  I know a fellow that was or maybe still is a current NHRA record holder and, when he switched to merge collectors on a “new and different” set of headers he had built on his “new and different” car, he “thought” he gained 3 HP on a 580 HP engine and this was after a day of fooling around on the dyno.  He wasn’t positive that he actually got 3 HP but “thought” he did even though everything else was new and different.  I don’t know what other collectors he compared to so who knows if the gain was even of merit or just existed in his mind?  One thing this fellow told me was that the smallest part of the merge collector measured 2 5/8” OD and he was amazed that the collectors were that small.  I had told him to experiment with collector sizing to possibly find more power but, like too many others, he did NOT listen to me – instead he looked at what others were doing and copied what they were doing (assuming that they knew what they were doing).  (Yeah, right!)  It’s interesting TO ACTUALLY KNOW that there are guys out there who can set and hold national records yet have no idea if their headers are even near right (yet others copy them!).  Talk about the blind leading the blind!

The three tube collectors we make by hand have a much longer taper than those I saw in the header pictures received here.  This longer taper improves flow (reduces restriction depending how you want to look at it).  The collectors also have a deep crease on the outside so that the collectors “hug” the contour of the tubes entering them.  This is also important to reduce gas expansion and turbulence.  They also have a constant taper which means they are tapered over their entire length of 5 1/4”.  Some collectors in manufactured headers have a straight area on their inlet so that the header tubes can end at different locations and still go inside the collector.  This straight area allows the manufacturer to make his headers a little cheaper because he doesn’t have to cut the ends of the tubes square to fit the collector but then the collector’s tapered section is actually shortened which can increase restriction.  (Because of this, in some cases, the welding on the ends of the tubes looks like hell when the tubes aren’t trimmed even with each other and the excessive amount of weld on the ends of some tubes – “goobers” - can’t be considered anything but bad when you see them as obstructions to flow as well as capable of creating additional turbulence.)

Even though our three tube collectors are made by hand and are more expensive then the four tube formed collectors we also sell, they are also much cheaper than the “merge” collectors I’ve seen and considering that this whole “header thing” we’re talking about here involves performance through an exhaust system I can only see the “merge” style collector as a waste of additional money.

Since everybody will undoubtedly want to know the price of the V6 header kits, 1 3/4 and smaller tube header kits cost $592 with the parts needed to hookup to the street exhaust system (collector flanges, gaskets, and tapered cones or short sections of straight tubing to match the collector outlet to the exhaust system diameter).  If exhaust hookup parts aren’t needed, the 1 3/4” and small kits drop to $528.  1 7/8” and 2 V6 kits cost $620 with hookup parts or $560 without those parts.  These prices cover the cost of 1/4” thick header flanges but 5/16” thick flanges can be substituted for $16 more and 3/8” thick flanges for $24 more. Note that these prices will be good through 11/30/10 – after that check our website to see if the prices have changed.  On a 2” kit, three 1 7/8” bends will be substituted for 2” bends so that a stepped header can be built (1 7/8” tube off the head for 4-6, then the rest of the header built in 2”).  Because we don’t have header gaskets for this engine, they have to be sourced elsewhere and we substitute an extra U-Bend for the unavailable gasket (which brings the total number of mandrel bent bends in the kit to 11 U-Bends and 2 J-Bends plus 10’ of straight tubing, header flanges, collectors, header bolts, street exhaust hookup parts – if ordered, etc.).  Because we only have the hand built Custom Collectors available for these V6 kits, they raise the cost of the kits about $116 compared to what they might be if machine made formed collectors were available.  However, the Custom Collectors are much better shaped (more flow efficient) and are also available in a number of inlet and outlet sizes so their availability eliminates design compromise (design compromises made when the CORRECT part isn’t available so the WRONG part is substituted) so they are well worth their extra cost.

In conclusion, (WOW, this IS long isn’t it?  But keep in mind you wanted advice to do this better.) I’ve tried to give experienced design advice based on what I’ve been told so far. 
Exhaust system design to me is quite logical and follows certain defined paths.  I have learned over the years that equal length header designs with tubes long enough to function properly in the rpm range of the engine involved are critical to maximizing performance.  I have also learned that the sizing of the header (tubes as well as collectors) has a profound effect on an engine’s capability to produce power over a wide rpm range (oversizing a header to any degree GUARANTEES diminished bottom end and mid-range power gains).  I have also witnessed how important the tapered transition area (length AND shape) of a collector is relative to its ability to improve performance over a wide RPM band so its value to improve performance is not just a race oriented improvement but actually has MORE importance in a street vehicle – where the engine has to produce good performance over a very wide rpm range.  The presence (or absence) of any of these desirable header design elements definitely affects performance.  By building one’s own headers one can move away from the performance (and construction) compromises so prevalent in many manufactured headers.  Hopefully you will move on to something much better.

Well, I hope that this helps you.  Thanks Again.

      Our Header Kits (prices, contents, and purchase options) are described on:
      http://www.headersbyed.com/headerkits.htm 

      If you would like to look at the Header Questionnaire that was filled out for me to
      analyze and, from that information, offer these Header Design recommendations, look at:
      http://www.headersbyed.com/questionnaire.htm