Alright, get as technical as you can. Looking for the best design for a short runner header. Primary sizes, stepped runners vs. straight through, downpipe size 2.25"/2.5", obviously equal legnth runners is best.

What would you like to see in a header? Lets here it.



15.9 @ 86.7mph
97 Eclipse RS

Injen CAI, 55mm TB, P&P Intake, Crower Stage 2 Cams, Hooker Header, GS-T Muffler, AFX UDP, Crane Ign Amp, S-AFC

Hmmm, I'll look over some info I have when I get home, do some calculations, then give my reccomendations. You want technical Mike? I can give ya technical... }(

But, before I figure anything out, you said "shorty" so I'm assuming it would be for drag racing?

http://www.cafefoundation.org/aprs/EPG%20PART%20IV.pdf

Just a little info for you

There are a number of items to consider when designing a header. This will be my description for a header designed for our cars, for drag racing purposes. This will be a LONG post, so don’t say I didn’t warn you. You wanted technical Mike, here it is... }(

There are eight (8) different design aspects I can think of that have to be considered: runner length, runner diameter, design (tri-y or 4 into 2 into 1 design, 4 into 1 design, or stepped design), down pipe diameter, collector, material of construction, coatings, and secret tricks.

1st Runner length. Runner length is probably the most critical aspect of the header design as it pertains to where in the RPM range you want to tune for maximum torque. (It is better to consider the torque curve rather than the horsepower curve when it comes to designing parts to tune for maximum power.) We are trying to get a runner length where the exhaust pulses from one cylinder meet the collector at JUST the right time to help ”pull” the pulse from the next cylinder. Since the pulses are really waves of pressure followed by a lower pressure area, the next pulse tries to move into this vacuum, basic physics says a high pressure area will try to fill a low pressure area. This tuning effect was initially called “supercharging” (we’re talking back in the 1920’s), where you matched your header, intake and cams on a naturally aspirated car to actually get above 100% volumetric efficiency.

Ok, now for a little more practical stuff. The general rule of thumb for runner length is: the longer the runner, the lower in the rpm band you will get the maximum tuning effect. So, since this header is to be designed for high rpm power, we want shorter runners. What runner length exactly? Well, I do have a formula, but it assumes several things, I’ve written some incomplete notes down with it which help a little, and to be honest I don’t remember where I got it from (maybe from the book “The Scientific Design of Exhaust and Intake Manifolds” if that is the correct title).

The formula is: L=V*T/RPM

Where:
L - runner length in inches
V - speed of sound in feet per second
T – time in crankshaft degrees, from the time the exhaust valve opens sufficiently to start an effective wave
RPM – the RPM at which you will see the maximum tuning effect

Simply speaking (I’ll get into the problems with these assumptions later), going by the variables I got from the book (or wherever I got the formula from) it was assumed V=1700 and T=120. I’ll use these in the calculation since I have nothing better right now.

I would want a drag racing header to tune at 6000 RPMS. Why 6000? Because if you are drag racing and have the stock rev limiter your RPMS will generally run your engine from 5000-7000 RPMS. So the maximum tuning effect will occur right in the middle of that RPM range. If you are running an AFX computer, this will increase everything by 1000 RPMS, so when you shift from one gear to the next you will catch the next gear at the point of maximum tuning effect. So, 6000 RPMS will work for both applications.

Therefore using the formula, at 6000 RPMS your runner length will be 34 inches.

Now, here are the problems. The V value is speed of sound in feet per second, but why use 1700? At sea level is would be somewhere around 1100. But, remember that the exhaust gas is at a different pressure and temperature than air at sea level, so this probably explains the 1700 value, but who knows how close this is to reality.

Then there is the T value. I think the 120 crankshaft degrees is the total time the exhaust valve is open enough to actually get enough exhaust gasses to have the tuning effect, but I could be mistaken, and who says 120 degrees is the right value? That would depend on the CAMS you are using. (see how everything in an engine all has to work together?) An aftermarket cam will need a different runner length than a stock camshaft. And, how far open does the valve have to be to open “sufficiently to start an effective wave”? Is 0.050 lift enough? (it probably is, that is probably one of the reasons why camshaft manufacturers use that 0.050 lift specification on their cams, besides the fact it makes for a more consistent comparison between cams)

So, with the problems with V and T, what does that mean for the calculation of runner length? Well, I’d leave V alone, and maybe check on T to make sure it is accurate. But, if people have different cams, this could be a problem. So keeping it at 34 inches seems to be a good compromise. The way to REALLY test this is to make several headers identically except for different runner lengths and see what they do on a dyno. So, until then, I’ll stick with 34 inch runners.

2nd Runner diameter. I have no formulas for runner diameter. Again, the rule of thumb is: The larger the diameter, the higher in the rpm band you will get the maximum tuning effect. For a drag racing header you will want slightly wider runners. You can have a runner which is too narrow (too restrictive) or too wide. I don’t have a number off hand for runner diameter, but I’d say the runner diameters used for aftermarket headers now are probably about right.

3rd Design. A 4 into 1 design would be the best to use. The “stepped” and “tri Y” setups are usually used to spread out the tuning effect, so you don’t get as much peak torque, but the torque increase is over a wider rpm range. For drag racing, you want as much power as you can right in that upper rpms range, so the4 into 1 setup is the best. I’m not as familiar with a “stepped” design, but the “tri Y” (4 into 2 into 1) design is usually used because it is easier to make very long runners with this setup, making it better for low end torque applications. Also, theoretically with a “tri Y” design, you probably get reversion of the exhaust pulse up the first branch the pulse sees and that is probably what causes the torque increase to spread out over a wider RPM range, but also reduces the beneficial tuning effect of the pulses lining up at the proper time. Since we want maximum possible upper RPM torque, this is not ideal.

4th Down pipe diameter. The down pipe should be the same diameter as the rest of your exhaust. If you’re stock then 2 inches, if aftermarket then whatever the aftermarket exhaust is. I’d say 2 or 2 ¼ inches, no more.

5th Collector. You don’t want to go cheap on the collector. The collector for a 4 into 1 header should be well crafted, have a smooth internal design with no obstructions. The collector should not open up too much where the four pipe go into it, then it should taper down gradually (not abruptly) to the 2 or 2 ¼ inch diameter of the downpipe. The center of the collector should have a small “point” in it in between the four pipes to help smooth the exhaust gas transition from runner into collector; it should not just be an empty space if possible.

6th Material of construction. If price is a problem then 14 gauge mild steel is fine IF it is coated. If it isn’t coated it will rust and disappear VERY quickly. I think 16 gauge is a bit too thin, I’d prefer 14 gauge. Of course you could go with stainless steel if you didn’t mind spending more money, but it will eventually rust too (it will just take longer) and I’d use the same gauge thickness.

Of course, if price isn’t an issue…. You could make a titanium header. It would be much lighter, you could use a much thinner gauge, and it will also last FAR longer. Best bet is grade 5, it is overkill for the application but it’s a common grade. Grade 2 is even more expensive and even more overkill, so I wouldn’t bother with it since grade 5 is MORE than enough.

So, I’d go with a mild steel header, 14 gauge to keep price low, with the right coatings of course.

7th Coatings. I’d have the header coated, regardless of the material used (even titanium). If it was titanium I’d get it coated inside and out with ceramic. But, for a mild steel header, I’d coat the inside with ceramic (I think it’s heat retaining properties do help a little bit) and the outside would be aluminized. The aluminized coating seems to stand up better than the ceramic coating to wearing away and rusting. When the FSAE team I help out with used this coating on one of their mild steel headers, the aluminized coating held up to the elements well.

So, ceramic inside, aluminized outside for the mild steel header.
8th Secret tricks. Ok, I would do this, I’d take….. HEY! I’m not giving away any secrets today :nono

OK, so this is it, a 34 inch long runner tuned for maximum torque increase at 6000 rpms, 4 into 1 header, moderate runner diameter, 2 to 2 ¼ inch downpipe, an efficient collector, 14 gauge mild steel, ceramic coated on the inside, aluminized on the outside.

Well, what do you think? :D

34 inches doesn't sound short.

I'm no expert, and generally have no clue....but as per your 1700 feet per second, I don't think that's right. The relative speed of sound (I took this from a website) in standard atmospheric pressure and relatively dry air is 331.45m/s or 1087.43f/s

Just used "Ask Jeeves" and a metric to english conversion chart.

So.....


1087.43 X 120 / 6000 = 21.7486 inches for the runners.


I agree with Dino, 34 inches is very long...that's almost 3 feet. 21.75 (rounded) is less than 2 feet which sounds closer.


Edit:

Using the same formula you can calculate the RPM at 34 inches...

1087.43 * 120 / RPM = 34


Ends up being 3837.98 RPM, that is assuming the 120 is right.

I think you won the prize Todd! I don't expect anyone to beat that post. But, now if only someone could build a few prototypes in various sizes and dyno them...then we'd have something. And then design and test intake manifolds...I'd be interested in that.

Corbin

'95 ESI-T
HRC+FMIC+ETC...



Gimme fuel...Gimme fire...Gimme that which I desire

>And then design and test intake manifolds...I'd be
>interested in that.
>
>Corbin
>

I have formulas for intake manifolds too ;)

I think modifying the calculations as Aaron did seems to make more sense. But, in the exhaust gas (it's hot nd might be pretty dense, sound travels faster in denser air), what could the speed of sound be? Maybe the 1700 was a typo in the book, maybe I copied it wrong, I can't be sure.

>I think modifying the calculations as Aaron did seems to make
>more sense. But, in the exhaust gas (it's hot nd might be
>pretty dense, sound travels faster in denser air), what could
>the speed of sound be? Maybe the 1700 was a typo in the book,
>maybe I copied it wrong, I can't be sure.


hot air is LESS dense..

i'm not gonna mess with the rest of what you guys are talking about though.. don't know any of it off hand.. unless I do some research.

Very good info, both of you, thanks. Keep it coming.

Please tell me that's 34" spread across 4 runner, so 8.5" each? Because 3' on a runner can't be right.

The stepped runner is done to help prevent a vacuum being created and pressure being pulled back toward the combustion chamber and help to aid in the scavanging effect. The increase from smaller to larger diameter is oppse to prevent backflow. The same principal as to why you leave a lip on the head when porting the exhaust ports, and don't exactly match the gasket/header/manifold.

My thoughts:
4-1 design, stepped runner, I think could be beneficial.
As always equal legnth runners.
I wouldn't go smaller than 2.25" on the downpipe, a 2.4L swap and I think 2.5" would be better, but for 2.0L 2.25" should be good..
I say ceramic coating inside and out. keep the heat in to romote exhaust gas velocity and keeop the engine bay temps down. Plus the added bonus of preventing rust.

Thanks again guys, this proves very helpful! }(
Any more? Keep going!



15.9 @ 86.7mph
97 Eclipse RS

Injen CAI, 55mm TB, P&P Intake, Crower Stage 2 Cams, Hooker Header, GS-T Muffler, AFX UDP, Crane Ign Amp, S-AFC

Okay, I can't find a calculation for it off hand, but you are right. And here are some things I found.

Relative air density decreases with temperature.

Speed of sound varies based on temperature and gas saturation (CO2)

Therefore, with the ammount of CO2 created by combustion and the extremely high temperature 1700 could be correct. Sorry for correcting you, I didn't anticipate those facts.

A prgram that caculated out header tube legnth, taking into account more variable, even exhaust cam duration, nubmer of cylinders...
Anybody know what the calcs they are using is?
http://www.rapidline.com/calc/engine/pchdr1.htm
http://www.rapidline.com/calc/engine/pchdr2.htm
http://www.rapidline.com/calc/engine/pccollec.htm

A random Calc I found (with pipe diameter):
(CID * 1900.)/(rpm_max * tube_od^2)
Using this with a
7200 rpm max It comes to 20.6"
8500 rpm max is comes to 17.45"

Thoughts?



15.9 @ 86.7mph
97 Eclipse RS

Injen CAI, 55mm TB, P&P Intake, Crower Stage 2 Cams, Hooker Header, GS-T Muffler, AFX UDP, Crane Ign Amp, S-AFC

damn those are some pretty serious calculations.....But, 2 or 3 feet worth of runners isnt going to fit under the hood....This is why all of the header manufactures dont try and make them that long...They try and make them as long as possible while still running the stock exhaust system. Also, shorty headers use more "round" design snaking the tubing so that it fits in a smaller space....Of course, when i say that, i am refferring to header designs for PushRod V8 motors, but i assume the term is the same regardless of engine type.

1700 ft/s is the speed of sound at approximetly 600 F. The speed of sound is temperature and pressure dependent.

Then 1700 is low, my car at idle doesn't get below 900 degree of exhaust temp... WOT 600 degrees is a distant memory... I thin the 1900 used in one of the other calculations may take into account a higher temp.




15.9 @ 86.7mph
97 Eclipse RS

Injen CAI, 55mm TB, P&P Intake, Crower Stage 2 Cams, Hooker Header, GS-T Muffler, AFX UDP, Crane Ign Amp, S-AFC

Actually I screwed up the equation I used was a low temperature approximation. Not to mention it was for dry air at 1 atm. I'll have to go and look up the right equation.

If you guys get a chance can you measure what diameter your runners are?

AFX's Neon header uses 1.75" I'm curious what others come out to be. I'll measure mine when I'm home.



15.9 @ 86.7mph
97 Eclipse RS

Injen CAI, 55mm TB, P&P Intake, Crower Stage 2 Cams, Hooker Header, GS-T Muffler, AFX UDP, Crane Ign Amp, S-AFC

The Hooker Header:
Stepped runners, from 1-1/2" to 1-5/8"



15.9 @ 86.7mph
97 Eclipse RS

Injen CAI, 55mm TB, P&P Intake, Crower Stage 2 Cams, Hooker Header, GS-T Muffler, AFX UDP, Crane Ign Amp, S-AFC

somone should talk to clay rathburn at Black Dog Motorsports hes spent hundreds of hours designing headers for our 420a motors even though they are neon ones but still they are awesome units produce great gains...


in fact if we got some people together im sure hed do a group buy to geta product started....

his tri y are very nice products...

justa thought... i think hes got the most r&d time on any 420a exhuast system out there =)

Bill

My OBX header's specs I measured.

Runner length - 17.7"
Runner diameter - 1.75"

What about the variable intake manifolds.. that are in the Ferraris or Lamborghinis (either one, or maybe both)...

How do they work?

~Tacoman