HEADER DESIGN COMMENTS - Cheap Performance Gains.
While the following discussion primarily involves performance gained at the drag strip, it really deals with improving open exhaust performance so it must be considered completely relevant when applied to other forms of motorsport competition - circle track racing, closed course racing (SCCA, gymkhana, autocross, etc.), pulling, mud racing, sand dragging, etc. - as well as where open exhaust systems are the rule rather than the exception - experimental aircraft, etc.
For those that have drag raced for a long time, reflecting back on those early years often results in thinking about how much cheaper it was to race at the beginning. Something as simple as trying different ignition settings to determine the best overall ignition timing to lower Elapsed Times (ETs) (and/or increase MPHs) didn't even cost money - yet, thru experimentation, might drop a car's ET a tenth or two! Going to the local speed shop to get some different jets and/or metering rods to optimize air/fuel mixtures usually resulted in further lowered ETs (and/or increased MPHs) yet only cost a few bucks. Pulling out the distributor and having it recurved by somebody competent where some of the ignition advance curve was removed so that additional timing could be run at idle had a remarkable effect on bottom end power so the automatic-equipped car would launch harder off the line (and yet made the car perform even better on the street - if the car was street driven). This change also didn't cost much money yet was often impressive when appropriately done. ETC., Etc., etc.
A couple of decades later (and several cars later as well), those cheap and easy tenths (reductions off the ETs) don't seem to be around any more. Getting a tenth off a (much faster) car now not only represents expenditures of hundreds (or thousands?) of dollars (and often many hours of work) but also seems to entail a certain amount of risk (anybody that has raced for a long time can usually relate numerous tales of woe where the "sure thing" they were told about (or read about in a magazine) didn't work at all or, even worse, actually hurt the performance of their car!)
With the above comments in mind, I'll now talk about an exhaust related "subject" we've been involved with for over 40 years - a subject that, to us, is absolutely true (because we've been involved with it so long) - yet a subject that hasn't been discussed to any extent elsewhere. We have continually mentioned this subject in our literature (our Header Design InfoPak) for years and have continually told many of our customers (Header and Header Parts) about it for years as well yet, most disappointingly, find minimal acceptance because OTHERS don't talk about it (or don't do it)!
(Does something have NO importance simply because others do NOT talk about it?) (OR, are there reasons - reasons NOT discussed - WHY they don't talk about it?)
Pictured above are several examples of tapered and straight collector extensions. For over 40 years we have watched the addition of these extensions improve Elapsed Times at the drag strip by as much as a half a second. These impressive gains are achieved by improving the basic design of the header that the collector extensions are bolted on to. Generally speaking: Lengthening the header produces gain when the added collector length establishes that the header was initially too short; Reducing the size of the header's collector produces gain by showing that the header was initially too big. There are many other reasons as well as to why the use of these bolt-on tapered and straight collector extensions can give you a gain. Read further to find out more about buying some of the cheapest performance gains you might ever see at the drag strip! ! ! ! ! ! ! !
Many guys.... (racers included) seem to think that collector sizing (and length) is a non-issue - that the header manufacturer involved has done his homework - so that nothing can be gained by changing either the size or length of the collector.
Unfortunately, for far too many, nothing could be much further from the truth.
Over the years we have determined that collector sizing and collector length can have a major effect on performance - as much as over a half a second change (good or bad) in ET. We know that this many seem unbelievable to many racers who have struggled just to find another elusive tenth of a second reduction off their ET, BUT it is true. The amount of gain possible just depends on the specifics of each situation.
Probably the best way of getting across what can be achieved is to present a few Header Design Experiences:
The very first experience - that showed how important collector sizing could be - involved a 335 HP 1965 401" Buick GS Skylark I built headers for forty five years ago - in 1966. The car was sponsored by a local Buick dealer and ran in a stock class at the local drag strip. The engine had been blueprinted but, otherwise, was stock (as defined by NHRA rules at that time) and the car itself was setup (converter stall speed, rear end gearing, etc.) purely to drag race. The car came to me with the intention of making the car faster by improving performance over the headers that had already been on the car for about a year and a half. The fellow that was responsible for the car gave me header specifications and I built the headers as requested EVEN THOUGH I WARNED him that the collectors, based on my experiences so far (only about four years at that time), were too big. I had no problems with the other header specs but felt the collector size was inappropriate (nice way of saying it would slow the car down relative to what it would run with a more appropriate collector size).
I never saw the other header - because the car was delivered minus an exhaust system - but was told it was a Tri-Y header design from a "big name" header company.
After the headers were done - and I did a really GOOD job as they were TRUE Equal Length - the car ran no better at the drag strip. The car dealer would not pay me for the work I did (one of many negative experiences I've had with car dealers over the years and why I intensely dislike doing business with them) even though I continually told them that I did NOT design the headers, that I had only built them and that the fellow responsible for the car's performance should be held accountable for what had happened - but definitely not me. To make matters worse, I was continually hollered at for what seemed to be every other problem in the car - "the headers are over-scavenging the engine and I can't get the carb rich enough" was one complaint that I heard a lot.
So, one night I grabbed my Milwaukee Sawzall, torches, and a set of welding tanks and went down to the Buick dealer and cut off the 3" collectors that I had already told them were too big and welded on the 2 1/2" outlet collectors I wanted on the headers in the first place. Guess what? The next Sunday the car was down .4 second!!! AND that was the only change made!
Keep in mind that if it took only 5 horsepower to cut a tenth off this car's ET, switching to the smaller collectors was good for at least 20-30 horsepower (factoring in the retuning of the engine too).
The car dealer eventually paid me but NEVER lettered HEADERS BY "ED" on the car which was part of the deal - so I got cheated anyway plus I never got paid for all the EXTRA work I went through due to THEIR mistake (actually their complete lack of understanding of good header design)!
However, entirely by accident in this instance, I did learn something very important about header design and the importance of collector sizing. I also learned that while the saying "the customer is always right" might apply in many situations (like ordering how one wants a steak prepared) it isn't always appropriate when design considerations must be observed - particularly when the customer is inexperienced. I also learned later on that their tuning of the carburetor was WAY off (probably OK with the other headers but NOT OK with a much better header design) and, by the end of summer, the car was turning its best ETs and MPHs with .003" SMALLER than stock jets! (Which I had already told them - that the carb might already be too rich with the new headers - at the start of all their "hollering" but, like too many others, they wouldn't listen to me.
To me, with optimized tuning factored in (at least another .1 second off the ET), the smaller collectors were worth somewhere between .5-.6 seconds off the ET compared to the larger collectors! Talk about a CHEAP performance gain: If done with bolt-on tapered collector extensions - a .4 second drop in ETs for about $90! HDE001.
A later Header Design Experience involved a 1966 Pontiac GTO. The engine was basically a rebuilt stock 320 HP 389 except for a "hotter" cam. The car had a 3 speed T-85 manual transmission - not a 4 speed - with 4.56 gears in back and was basically a "street" car (with Tigris Regalis lettered on the side!) that was drag raced regularly on weekends. (This was also the car I used to design the HEADERS BY "ED" now offered for the 1964-1967 Pontiac Lemans and GTO.)
The headers on this particular car were a TRUE Equal Length design 1 3/4" OD x 34" long with a 3" outlet collector with a flange located at the end of the collector taper. 3" OD x 12" long straight collector extensions were bolted on to the end of the collectors when the car was raced.
This car was the first car I used to test the idea of using Tapered Collector Extensions - smaller diameter collector extensions with the inlet tapered up to match the outlet of the collector of the header.
The idea of using collector extensions made out of smaller tubing came out of doing some simple math based on numerous earlier header design experiences involving small block Chev headers I had been building:
At that time, I had already built a number of 1 5/8" OD small block Chev headers for certain customers but with two different collector outlet diameters - 2 1/2" and 3". By going out to the drag strip and talking to the various customers with these different headers it became very obvious that the headers with the 2 1/2" collectors were overall improving performance to a much higher degree - usually .2 seconds or so quicker - compared to what the headers with the 3" collectors were accomplishing. (Also - keep in mind that the 1 5/8" headers I was building were TRUE Equal Length designs as well as up to 20" longer on some header tubes when compared to the UNequal length headers that other companies were making.) In fact, in several cases - on TOTALLY stock engines - the headers with the 2 1/2" collectors had improved ETs by as much as a full second over times run with closed factory exhaust systems yet the headers I made with 3" collectors NEVER improved ETs near that amount.
Looking at this particular 1 3/4" Pontiac header and thinking about what I was seeing with the 1 5/8" small block Chev headers, I began to wonder if the 3" collector diameter was necessarily the best size to use on that particular tube size. From those thoughts I constructed a basic proportion based on header tube and collector tube cross-sectional areas:
Area of 1 3/4" collector Area of 2 1/2" collector X 4.53 in2
------------------------- = ------------------------- OR -------- = --------
Area of four 1 3/4" tubes Area of four 1 5/8" tubes 8.57 in2 7.32 in2
When you do the math, X equals 5.30 in2 which ends up slightly smaller than the inside area of a 2 3/4" tube. Well, isn't that interesting?
According to the math and these other previous experiences with Small Block Chev headers, it appeared that a 2 3/4" diameter extension might prove to be a better collector diameter for this header. SO, I bought some 2 3/4" tubing, tapered it up to 3" OD, welded on a 3" flange and called my customer (who I had already talked to before about this) to come by and pick up some "extra performance".
The following Sunday the customer first made 3 runs with the 3" x 12" extensions - 13.01 seconds, 13.05 seconds, and 13.06 seconds all around 108 MPH. Very consistent and expected as the car typically ran 13.0s-13.20s but had never made a 12 second run. Going back to the pits he swapped the extensions and ran a flat 12.80 on the next run! Better than .2 seconds improvement. The rest of the runs that day were in the 12s and the MPH did not change so the smaller diameter extensions didn't even restrict the engine on top end. Talking to him later on, the car, with engine retuning, ran as quick as mid-12s that summer (I don't know if the extensions were ever cut to match their "burn" marks so it is possible the car could have run even quicker!). Talk about a CHEAP performance gain: a .2 second drop in ETs for about $90!
To put all of this in a more relevant perspective: At the same time this experiment was conducted, in St. Paul, Minnesota there was a 1967 GTO that also ran our headers that was the then current NHRA class record holder at 12.31 seconds and would typically run 107+ MPH. Referencing back to our experiment, this means that our "test" "street" car was running only .2 seconds off a current national record at the same MPH and was doing it handicapped by running a 3 speed transmission. Another thing: If one factors in the improvements made in tires alone in the over 40+ years since this car ran, the same car would now probably be running high 11s on a new set of slicks!
One last comment: Even though the above experience showed that a 2 3/4" diameter collector extension worked a lot better on this particular car, remember that this is what worked BETTER on THIS particular car. This experience can NOT and should NOT be interpreted as proving what the best diameter collector for 1 3/4" headers is as we've seen smaller as well as bigger collectors work better in other situations still using 1 3/4" headers. One has to experiment to find out what is best. HDE002.
Those two collector size-related experiences took place over 4 decades ago. Since then our customers have proven time and time again - on their own cars - and in much the same manner as discussed above how important collector sizing is. This they proved by either buying our headers with two different sizes of bolt-on collector extensions or building their own headers and setting them up to allow collector extension swapping. By running their own comparisons at the drag strip, they have seen collector sizes that big header companies don't even offer work the best and/or they have seen the smaller of two sizes work better (proving again that bigger isn't always better).
NOT YET FINISHED - much, Much, MUCH, MUCH MORE TO BE WRITTEN!
Above: On the dyno, shortening an extension like this - cutting the extension so that its length is about equal to the length of the "burned" area - can create as much as a 5-7 HP gain by further optimizing the length of the collector. (We made these shorter tapered extensions to help you better visualize how to make the cuts on the extensions.) Also - notice how "equal" the burn length is on the two longer extensions - illustrates how "equal" the tube lengths are as well. By the way, these extensions were taken off My Old Dodge after several passes at the drag strip.
Collector Extension Prices and Parts Numbers are as follows:
(3 bolt flanges are standard in design and match all current header manufacturer bolt patterns.
All Collector Extensions are made to order - please allow at least one week for construction.
HEADER DESIGN COMMENTS