Can Everybody Do Something Wrong?





HEADER DESIGN COMMENTS - Can Everybody Do Something Wrong?

Under Construction GIF


Over the years in designing and building headers and also dealing with thousands of people as well as thousands of performance related issues and situations, it has become very obvious that too many times we have witnessed too many doing something quite badly. The predominant reasons we see for this to occur is a lack of understanding (fostered by inadequate discussion by magazines), reading magazines and accepting - without thought - whatever their "pushing" as being correct or proper, and watching people copy what others are doing.  (Copying is based on at least two assumptions:  (1) The other person actually knows what he is doing and actually did it correctly; and (2) What the other person did is also correct for the person doing the copying (which often proves to NOT work out well because, in most situations, the person doing the copying:  (1) does NOT have the same identical vehicle and: (2) does NOT have has the same intentions or goals as person being copied).  Copying is one of the WORST things we see people do because of the assumptions that are made prior to copying. The following discussion - which will be expanded as time permits - will hit a number of areas we've witnessed for years - where literally tens of thousands of guys do something rather badly (in other words, there are better ways of doing these same things) - yet the "bad" way of doing something has actually become the "popular" way of doing it! Some will be offended by what is said as the discussion will "hit a little too close to home" but others will benefit by seeing these mistakes for what they really are (design selection errors) and, hopefully, will not repeat these same mistakes. By not repeating (copying) the mistakes that others have made, they will end up with better performing vehicles.

NOT important to have "good" headers for the street.

Good headers - in this case - are headers designed to maximize the performance of the engine/car through a muffled exhaust system. Over the years we've heard guys say something like "Well, this is a street car - not a race car - so it doesn't have to have good headers. I can see having a good set of headers if I was racing the car but, since, I'm not, the car doesn't need to have anything that good. Etc., etc., etc."

These are comments/arguments we've heard for years. Unfortunately, for the performance of the car concerned, nothing could be much further from the truth. These comments/arguments are based on wishful thinking, lack of fact, a rationale to make one think that having something cheap is OK, etc. Considering how many times we've heard comments of this nature, it's obvious that this opinion seems to be held by the majority. Looking at thousands of cars over the years to see what headers are actually on them certainly reinforces that opinion as one doesn't have to look at too many cars to see that most have not done the exhaust side of the engine very good. It's interesting to see fellows spend thousands of dollars on their engines and cars (a lot of which is related to improving the car's performance) yet see the exhaust system done rather poorly.

Unfortunately for the majority, testing as well as day in and day out experiences with headers, contradicts this popular opinion. One thing we've noticed in testing is that the mid-range power of an engine (at or below it's maximum torque peak) is extremely affected by header design. It's interesting to see just tube changes raise/lower mid-range power by 10% or more. It's also interesting to see mid-range power increase up to 10% by just adding 6-10" of tube length to a header. We've seen improving a collector's shape as well as the shape of the tubes where they enter the collector increased mid-range 1-3%. When one makes all of these changes as part of a switch to a "good" or actually better header, the accumulative total gain in the mid-range power can be up to 20% or more. This is a fantastic gain - certainly one that many will argue can't be achieved but we've seen it and we've done it. These gains are not based on magic but achieved by simply making improvements relative to another's inferior design.  

If one doesn't believe that it can be done, we would reference the disbeliever to the dyno test article in Chevy High Performance magazine (a reprint of that test is available from us) that used our headers as well as a number of headers from "brand name" header manufacturers. In a less than ideal test (they used an engine DIFFERENT than the engine they told us was going to be used in the test, did not do the engine tuning for each individual header as promised, etc.), they MEASURED an 80 lb-ft increase in torque over really lousy exhaust manifolds which was a 25% gain, a 74 lb-ft increase in torque at 4100 rpm over a block hugger style header (a 23% gain), 51 lb-ft increase at 4100 rpm over a long 1 3/4" header (15% gain), etc. - all increases below the maximum peak torque rpm of the test engine. To say that "good" headers are not important on the street is to ignore how much better a car can perform when the exhaust is optimized. Keep in mind that this test was not run by us but by a magazine. The article only confirmed what we've seen and known for years - that headers can have a profound effect on the mid-range power of an engine - good or bad. Looking at the same test in the context of comparing maximum HP, the worst loss of power at the highest rpms was 6% with our headers and 9% with the header that hurt the highest rpm power the most. Let's see now - based on THEIR test, a 15-25% gain in the mid-range with a loss of 6% at the highest rpm on an engine that is too radical for most to drive on the street. Also keep in mind that the test engine put out over 100 HP more than the engine they said that they were going to use in the test (the engine that we designed our headers to match).

Considering that - in this less than accurate test - the 6% loss in top end power was at full throttle at almost 7000 rpm, how often would this loss - on the street - ever even be felt? At the same time, how many times would one hit 4100 rpm at full throttle on the street? Lots of times and think what a 15-25% gain at that rpm would feel like! When someone tells us that "good" headers are not needed on the street, isn't that like saying having a 15-25% gain of usable power is also not needed? Hmmmmmm!

Also - this test was run without mufflers - all of the headers were tested open to atmosphere. Depending on the exhaust system and mufflers actually used on a car - and the restriction they might represent - it is entirely possible that the actual difference is high rpm power between the headers tested might diminish considerably. So, in a real world environment, the engine's actual peak HP differences (based on header changes) might be much less than the test measured yet the mid-range power - which would be most likely unaffected by the restriction of the exhaust system and mufflers - would probably remain about the same. It would be entirely possible to have an engine (based on changing headers yet retaining the same exhaust system) produce little or no more additional power at highest rpms with larger headers yet produce amazingly different amounts of power (note that the magazine measured 15-25%) in the mid-range simply based on the design of the headers selected. When somebody tells us that "good" headers are not important on the street - that is simply not true.

One last thing. This test was run with a 355" small block Chev engine. Using simple proportions, we calculated that, if the test was run with a big block 454" Chev engine - which is 28% larger in displacement, and comparing to the difference between the output of the block hugger style header used in the test to the output of our headers at 4100 rpm, the same test results with a big block Chev engine would be almost 95 lb-ft of torque. Wow!!! To put a "real world spin" on this, I mentioned this to a customer in our shop and he said that he saw more than that comparing a set of headers of ours he bought used at a swap meet and, just for the hell of it, made a comparison on the dyno against the headers he had been using on his big block Chevelle. He sent me the dyno test comparison he had run and at 4500 rpm the difference was 126 lb-ft of torque!!! Keep in mind that this test just compared two headers and doesn't even represent what might actually be achieved if one attempted to optimize the design of the headers. Good headers not important on the street? Facts do not support that opinion. 

Magazine Content.

I was reading a magazine the other day and, by looking at the various pictures of headers in it, I started to realize that there didn't seem to be many headers in the magazine that one would consider to look very good - good as related to performance improving capability. After noticing this, just out of curiosity, I reread the magazine - from cover to cover - looking very thoroughly through all of the ads and all of the articles, examining closely all of the headers pictured. When I got done I had NOT seen one single header that I considered to be well designed as none appeared to be Equal Length nor long enough to function well in the low to mid-rpm range of an engine (which is quite important since most people who buy headers are on the street where bottom end and mid-range power gains are the most noticed and beneficial performance improvements).

I had not seen ONE decently designed header in an entire magazine!

If anyone read this particular magazine with the intent of buying a well performing header, it would have been virtually impossible for him to do so because there were none to be seen. How could any one do anything correctly, when there were no examples of this to be seen? Simply by looking through this magazine, most people would develop the conclusion that what they were seeing was correct, and, therefore, correct to buy and would then end up buying a header that was not very good. (By being good, remember that I am assuming that the header is being purchased to maximize performance - NOT to just decorate the side of the engine). Due to the assumption that all of the headers being pictured must be good (NOT!) - and, because they all look so much alike, that assures that EVERYBODY would then do it wrong!  

(Some have asked us - if these headers aren't any good, then why are they made and sold? The reason is quite simple. These headers exist because people buy them. Sales justify production. When people buy these headers and then wonder why the vehicle doesn't perform to their expectations, they can't get their money back so they are stuck with them. Others find that they have spent so much time and money in the process of getting a poor header on their vehicle that, even with bitter disappointment, they don't take the headers off - or change to a better header - due to the additional expense and time involved.)


Subjects to be discussed in the future (notes to us).

Flathead Ford header design errors.

Big Block Chevrolet header design errors.

5.0 Mustangs header design errors.

Collectored header designs becoming "acceptable" far later than they should have been.

Header parts wrongly designed in the early '60s still messing up headers designed in 2005.

1/8th mile drag racing (with headers rarely optimized to perform best in a 1/4 mile drag race).

Effect of altitude on Performance & Header Design. Nobody seems to pay any attention to the effect of altitude on performance and the necessity to factor this into a header's design so EVERYBODY at higher altitudes (above 2500') loses some portion of the potential performance gain of a header!

Circle Track - owner built header outperforms mass-production headers that everyone else uses.

"STICK" headers on cars with automatic transmissions.