The Central Hub for DSM Community and Information

For 1990-1999 Mitsubishi Eclipse, Eagle Talon, Plymouth Laser, and Galant VR-4 Owners. This is where the DSM platform history is documented and archived. Log in to help us in our mission, and to remove most ads from the browsing experience.

Unequal length headers

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

Breezio69

10+ Year Contributor
1,109
20
Mar 5, 2012
SLC, Utah
Ok Ok, I already know I'm opening a can of worms here. BUT, just for sake of conversation and since I found nothing through searching I just have to ask.

I know the benefits of having equal length headers, and the spooling drawbacks of unequal length. Which brings me to my question.

(Every thread Ive searched on here and Google does not have an answer)

Say you were to build a header and you WANTED the rumble of a WRX/STI. How uneven do the pipes have to be? Would two be same length (longer) and two be the same length (shorter) or all 4 descending by 1"?

Example, 1&3 @ 12" pipe and 2&4 @ 10"
-or- 1 @ 10" 2 @ 11" 3 @ 12" 4 @ 13"

You can flame me if you want, but I was just curious. :D
 
Wow...

On a serious note I do engine design and development for a living. If you'd like we could setup a project where I design an exhaust system with the acoustics you want for your vehicle. Probably could get it done for around $20k.

I'm not trying to be a jackass either, but unless you hire an engineer to do it, any numbers you hear would be a shot in the dark.
 
Well, like I said, this was a curiosity question. I figured if Joe Shmoe could build an equal length header with a welder and some tricky bending/ cutting then this could be plausible.

I would figure the 2 @ one length and 2 @ another would be the more likely for success? Since a WRX has two short side and two long side... and they'd fire: short, long, short, long.

But the question would be how long/ short... The WRX is a matter of feet in difference, where this is inches
 
EDIT,

And I fully know this question is not "traditional" by DSM "standards", but who always wants to live by what everybody else does?

Ill probably never do this, but it doesnt hurt to ask and at least have something to talk about with some tech... instead of another "My car idles funny" thread
 
And no, im not gonna buy a Subaru. I understand this is against the DSM gods, and yada yada... But if an STI can get great power running it then it must be a minimal loss. Or they woulda come from the factory all equal.
 
I'm not saying it can't be done. It's just that you want a prediction of the acoustics that would result. The options are either trial and error, or have it proffesionally designed (not by a welder or parts supplier, but by an engineer).

Also like you mentioned, anyone can build a manifold, but why is there so much trust in those parts being that great? I mean after all, not too many people put the design time in it. Just use shiny pipe and good welds and it will sell.

I do like your thinking though. Like you said, its not a thread about how to make my bov louder.
 
And no, im not gonna buy a Subaru. I understand this is against the DSM gods, and yada yada... But if an STI can get great power running it then it must be a minimal loss. Or they woulda come from the factory all equal.

A lot of the big subaru guys swap to an equal length setup for the power gains, just fyi. Anything pass 400+hp is normally equal length headers.
 
Tkelly- I could see how going 1-2/3-4 would change the sound... But in thinking how the subie design is, It fires on opposite sides which results in the exhaust traveling in pipes like so; long short long short.

Which leads me to believe that having 1-3 and 2-4 (our firing order) together and at different lengths would get you closer

example; 1 short 2 long 3 short 4 long

96ett

I get what you are saying, there could also be a mathematical equation to figure this out as well? And thats kind of what Im getting at. From the sound of the WRX it sounds like the pulses from the short and long side almost meet at the same time with a short pause between the next set.

If you could figure out an equation on flow rate. X travels down 12" of pipe @ Y, then on the next firing cylinder you would shorten it up to flow "faster" down the pipe IE, shorter pipe.

Is this correct or am I just way off in my thinking ?

A lot of the big subaru guys swap to an equal length setup for the power gains, just fyi. Anything pass 400+hp is normally equal length headers.



I know of a few STI's and wrx's over here making close to 500, still have unequal length.. BUT, they are not hard core into drag strip or anything like that.

Same thing as myself. If I make 350 ill be happy, and probably never see a tree or 1/4 mile of black top in my day.

It would just be nice to have something different, like a subie guy going, why the hell does that Eclipse sound like my car? haha
 
Last edited:
I'm halfway tempted to try them. I think it is 6 of one, half dozen of the other. Pairing 1 and 2 and not putting any thought into how the lengths are would probably be really easy. You'd get all sorts of weird harmonics going on in there then.

On the other hand, I like power and spool up much better than I like a cars sound. I know pairing 1&4 properly with a twin scroll turbo is absolutely worth it and there is no sound that can make up for an extra 500rpm of spool. I didn't believe it until I did it myself.
 
I'm halfway tempted to try them. I think it is 6 of one, half dozen of the other. Pairing 1 and 2 and not putting any thought into how the lengths are would probably be really easy. You'd get all sorts of weird harmonics going on in there then.

On the other hand, I like power and spool up much better than I like a cars sound. I know pairing 1&4 properly with a twin scroll turbo is absolutely worth it and there is no sound that can make up for an extra 500rpm of spool. I didn't believe it until I did it myself.

I realize the drawbacks of this, and im not running a twin scroll turbo and probably never will. Id assume with how short our manifolds are in comparison it would be a minimal loss anyhow.

And I will correct myself here... the exhaust is short long long short. Same principle as short long short long.... since it fires in an x pattern. short meets long, long meets short
 
Last edited:
Oh, I wasn't saying why you shouldn't try, I was saying why I won't be trying. Just talking for the sake of doing it LOL.

Do you know how to make a manifold? If you don't I can give you the web resources I used to make my own.
 
Oh, I wasn't saying why you shouldn't try, I was saying why I won't be trying. Just talking for the sake of doing it LOL.

Do you know how to make a manifold? If you don't I can give you the web resources I used to make my own.

Honestly, I wasnt intending to try this. I was seeing if other people had experience with this or had numbers etc.

But, now just thinking of it, I was pondering getting a $99 manifold off ebay and adding/ subtracting length to two runners. If it doesnt work im out $100 plus some pipe. Not bad for a trial and error cost in my opinion, Ive seen more money wasted in worse ways... like lambo doors LOL

And also, after double checking the firing order of a WRX you would modify runners 2&3, either longer or shorter (most likely shorter for ease of putting it back under the hood), If I could get accurate measurements of a WRX header (which I havent searched yet) you could find out the difference between the two in a percentage and make new header based on that percentage. Sounds simple now...
 
Looking at pictures of some aftermarket Subie headers, it looks like there is no rhyme or reason to a specific length, some look 30-40% shorter and some have a J pipe off one side that looks like its closer to 70-80% shorter...




Alright, now I have been looking at headers on Ebay for the last little bit, finding the easiest one to hack up and try out this theory on. I found one but I have a question on problems that could occur by changing where each pipe enters the collector... Ill post the picture then explain my concern below it.

You must be logged in to view this image or video.


Ok, in the description this claims to be all equal length, one of the only ones that says so in the ad, $80+ $30 to ship.

My question is this, will I run into problems if I cut where runner 3 goes in the collector and have runner 2 replace it, vice versa for runner 3. Based on eye balled measurements it would cut runner 2&3 about 40-50%

Update with measurements:
so I went out and checked out some stuff under the Hood, the factory 2g manifold seems to have a variance of up to 3 inches in the design. But I did measure up a potential design that would be pretty cheap to make considering its a lot less pipe.

Basically runners 1&4 would come up in a ram horn style 12-13" runners and take up the front section of the turbo flange. Then runners 2-3 would come almost straight off and 90*into the back section of the turbo flange. If set up like this I can get away with an exact 50% difference in runner length and may be the ticket to the rumble? Any thoughts?


I found out information on this 4 banger, running a 6" difference in 2 runners... a larger distance would net a larger rumble

http://youtu.be/hy8UOA-qY1M

Definately more distinct than my current 3" Thermal turbo back system




Well I have priced everything out to give this a shot, Im going to do the idea I thought of, two ram horn style runners and two straight in... It will be about $130 to do it all myself and try it. I figure if it doesnt work I can buy the other 6 90* fittings and just make the full manifold a ram horn style.
 

Attachments

You must be registered for see attachments list
Last edited by a moderator:
your not going to change the sound much by changing the header runner length. the turbo deadens all the bangs and smoothes all the pulses a lot. ive seen some tuned turbo headers on v-6 and v-8 engines that were designed to have multiple exhaust waves hit the turbo at the same time. so they were as uneven in length as you could get (sort of like 180 degree headers on a nascar) the end result was a SLIGHT difference in exhaust NOTE. but it did not sound like a differnt engine.

the only way to change the exhaust sound into a different style would be change the crankshaft. like the guy up there said. that causes a difference in firing timing that will cause a different exhaust sound. and a big bang DSM would be foooking schweet.
 
I believe I understand everything in general about the UELH's vs ELH's on here, and how they are the main source of the famous "boxer noise", and I get the impression that something like this would only be doable on a non-turbo 4G63 (or any I4 for that matter) Examples here:

You must be logged in to view this image or video.

Unfortunately for the turbo applications:

...the turbo deadens all the bangs and smooths all the pulses a lot.

You are 100% right about this, but I can't help but wonder if there is still a way to make it at least kinda possible for those with stock turbo applications, and I am sure there is with enough money and time (neither of which would be worth it to many unless you knew how to weld in the first place and had the time to experiment with it) but like its been said, it wouldn't make much of a discernible difference...and since Honda's can now sound less like gigantic bees (THANK GOD) I am sorta jealous...

OH WELL! I guess Ill have to live with the wonderful music my 4G already makes :D but I am going to have to play around with exhaust weights this weekend to try and get rid of my droning issue.
 
Last edited by a moderator:
My $.002 is why not focus on the piping/muffle from the turbo back? I've never made a manifold etc...

However I picked up a 1g AWD 2 years ago that had a 3" back with some muffler (sadly I can't remember the name) that he fabbed and it was the closest i've heard to a WRX or Sti on a non-subie.

I like the idea of doing it yourself and messing around with lengths. Love an update on what you end up doing.

Thoughts?
 
the only way to change the exhaust sound into a different style would be change the crankshaft. like the guy up there said. that causes a difference in firing timing that will cause a different exhaust sound. and a big bang DSM would be foooking schweet.

In theory to do a big bang motor with a 4gxx, wouldn't you just need to switch the cams and get 2 plugs to fire together? Just a thought, I don't know if there would be any power to be made that way, but it sure as hell would sound cool.
 
I've got a better idea; put rubber caps on one or two of your spark plugs. I had spark plug wires go out on me and the car sounded like a Subaru.
 
Add Value - Be Respectful - No Trolling - No Misinformation - Participate Often!
Support Vendors who Support the DSM Community

Build Thread Updates

Latest Classifieds

Back
Top