Subaru Outback Forums banner

Is There A Reason They Used Equal-Length Manifold on The H6?

9146 Views 16 Replies 6 Participants Last post by  Numbchux
I discovered a local hackerspace/makerspace that has a metal shop and all sorts of fun stuff, which means I can fix my exhaust, among other issues. I'm going to go the cardoc route and make my own manifolds/headers since mine could currently best be described as laughable. I'm debating some sort of unequal length, "classic Subaru" sound (as much as could be had with 2 more cylinders), since there appears to be plenty of room for it. Back to my question: there's probably a reason they made the H6's equal length versus the H4's unequal. I'm just trying to figure out why and if it would negatively impact anything if I went that route.

Y'all know me, I wanna be as blatantly unique as possible.
1 - 17 of 17 Posts
The 01 - 04 EZ30, aka EZ30D only had a single exhaust port from the heads.

The 05-09 EZ30R and '10+ 36R have mutiple exhaust ports from the head and with it different length headers. Not sure if they are tuned in any particular way.
Ah, so the 05+ have unequal length as well?
Maybe that's why this thing sounds so wicked badass, and mine sounds, well, kinda boring.

Different firing order, different breathing requirements. The H6 can never have the same sound as any of the H4s. Some would argue it's better :)

The strength of the H6 is in smoothness as far up the tach as possible. I guess you could corrupt that a bit with unequal pipes.

Suggestion: Make #6 longer to merge its pulse with #3. Make #5 longer to merge with #4. Leave 1 & 2 alone to cut down on spaghetti- it'll probably be good enough there.

EDIT: oh rats, you've got the single port heads. Yeah, kinda boned there. Best you could do is put a delay loop on one side to merge it with the other. Could make a good throb note.
Different firing order, different breathing requirements. The H6 can never have the same sound as any of the H4s. Some would argue it's better :)

The strength of the H6 is in smoothness as far up the tach as possible. I guess you could corrupt that a bit with unequal pipes.

Suggestion: Make #6 longer to merge its pulse with #3. Make #5 longer to merge with #4. Leave 1 & 2 alone to cut down on spaghetti- it'll probably be good enough there.

EDIT: oh rats, you've got the single port heads. Yeah, kinda boned there. Best you could do is put a delay loop on one side to merge it with the other. Could make a good throb note.
Yeah, my car loves making it tough. Can I please has pretty picture of what you're talking about with this loop? Maybe a video of what this might sound like, if it's been done?

Also, I'll never take the H4 noise over my H6 noise, too common/popular and the H6 can sound like one mean muddafukka if you want it to...
Yeah, my car loves making it tough. Can I please has pretty picture of what you're talking about with this loop? Maybe a video of what this might sound like, if it's been done?

Also, I'll never take the H4 noise over my H6 noise, too common/popular and the H6 can sound like one mean muddafukka if you want it to...
Not in a good position to produce a video on the spot, close to bedtime and flying out to a gig tomorrow.

But let me see if I can get you on the concept...

Resolve the firing order of the engines into stereo for your listening pleasure.

A Subaru H4 fires in a cadence of ...LLRRLLRR...

The H6 fires as ...LRLRLRLR....

Now, the turbo subarus with the awesome sounds have that extra feathering/smoothing effect from the turbo.

You can put the exhaust pulses exactly spaced and get an overall quiet system. You can lengthen a given exhaust tube to make that pulse a bit late. If you make it long enough, you can make it show up at the same time as another one at a given point in the system. That's how you tune the sound.

Given that you have two single port heads, you only get one layer to shift phase on. You can lengthen one pipe to make its pulse meet the other one squarely, or a bit less long to give it some off-beat rumble.

Hopefully that helps.
See less See more
  • Like
Reactions: 1
This sounds like I have to math things. Speed of sound vs. length of pipes vs. boom noise beginning time.

but I think I get where you're going with that.
  • Like
Reactions: 1
@rasterman you happen to have the firing order off the top of your head, and if there's any simultaneous firing going on in these old H6's? I'm trying to visualize the order in which the pressure waves make their way through the pipe. Since mine's strictly one bank and the other bank, pipe wise, is it simply just (bank) 1, 2, 1, 2, 1, 2, etc. ?

edit: I might take two speakers and turn them into fake exhaust ports, generate a single high-energy pulse every [insert number here] milliseconds and emulate with plastic pipe or something.
in the latest iterations of the WRX (edit FA20F) it is equal length exhaust, why...faster.

and yes the little wrx whiner girls through a fit,...spending all these UEL years 2 stepping lighting up the night with farts behind the mcdonalds,.
...now being forced to do it with less rumble IF they dump their wallets on a the latest.



firing order diagram, ...and the firing order is in the back / inside the owners manual.

See less See more
2
You need to look very carefully at both the pipe length as well as the diameter - changing either will change the timing of the reversion pulses, in turn changing the rpm range that the exhaust scavenging is the most effective. If you don't do it correctly, you can really screw it up big time.

If I were you, I'd confer first with a good Sub race tuner before contemplating anything.
in the latest iterations of the WRX it is equal length exhaust, why...faster.
Then why were they UEL to begin with? Space? Curiosity? "Cuz we can?"

Also, Richard's got a good point, although I don't know much about that side of things, except basic educated guess on this aspect, as an example: if the pulse wave from bank 1 made it to the Y-pipe or whatever joining portion when the pressure wave from bank 2 was the weakest, then the wave from 1 might push backwards up the bank 2 side and screw up the flow? anyone got anything more on that?
Then why were they UEL to begin with? Space? Curiosity? "Cuz we can?"

Also, Richard's got a good point, although I don't know much about that side of things, except basic educated guess on this aspect, as an example: if the pulse wave from bank 1 made it to the Y-pipe or whatever joining portion when the pressure wave from bank 2 was the weakest, then the wave from 1 might push backwards up the bank 2 side and screw up the flow? anyone got anything more on that?
when I look at all the H4 turbo headers and cats over the long period of time as designed by the factory: I see they are dealing with exhaust pressure, temp, and space,


they can go from simple to just plain nuts.
...I guess they need the same pressure on both sides of the engine, and some exact temp for each side of the cat set, all trying to keep exhaust up in the car so as to not let it get crushed by whatever a subaru runs over. (sand dunes, rocks, hard ice in snow piles ).
when I look at all the H4 turbo headers and cats over the long period of time as designed by the factory: I see they are dealing with exhaust pressure, temp, and space
but this is my H6 we're talking about, and we know how bizarre this thing is compared to every Subaru engine ever built... lol seriously though, my pipes, factory style, don't seem to be anything special or fancy, ****, the system's basically symmetrical all the way to the back axle.

I'm just after making simply the most unique and badass sounding (first gen/single port) EZ30 ever.
but this is my H6 we're talking about, and we know how bizarre this thing is compared to every Subaru engine ever built... lol seriously though, my pipes, factory style, don't seem to be anything special or fancy, ****, the system's basically symmetrical all the way to the back axle.

I'm just after making simply the most unique and badass sounding (first gen/single port) EZ30 ever.
replace the ultraquiet muffler with a straight pipe....cheap easy,...

I ran around for 6 months with a rust hole before my muffler,...

here is some video,...not mine:

Yeah, did that. Sounds wicked, but also, under load, like a pisssed off weed whacker. Also got me quite a number of fix-it tickets/warnings from my small hometown police department.

Have built ( or the guys in my shop built) 100's of race headers - ALL were designed using LOTS of dyno time. Seemingly small changes could easily hurt HP 5hp or more if you were not careful. Granted, these were race headers, tuned for maximum peak HP as well as area under the curve ( most HP over a given rpm range), but the same laws of physics apply.

There is lots of info on the net that can at least guide you in the right direction, BUT, you still will want to consult with someone who has experience with these engines. Here's on of many links to get you started:

Performance Exhaust System Design And Theory
AFAIK, all the non-turbo Subaru headers are equal (or very close to) length.

Certainly not unequal length like the borla headers.

The point of a tuned unequal length header is for the exhaust pulses to get to the collector at the same time, therefore doubling up the power of the sound, but halving the number of pulses we hear. But this also creates a pressure spike that is actually a restriction.

Having equal length means the exhaust pulses arrive at the collector perfectly between the 2 pulses from the other side, which improves scavenging and therefore power.


All the H6s (except the ER27) have almost the exact same layout, straight back from the manifolds to a collector in the middle of the car. I think the reason for this, is the H6s have 2 primary cats, whereas the 4s only have one.

As mentioned, the first gen EZ30D has single port exhaust, and due to the firing order, it would not be possible to separate the exhaust pulses and match them up like that. It would theoretically be possible on the EG33, and second gen EZ30D, and EZ36D, but a huge amount of work for a reduction in power...


Because of the considerably higher exhaust pressure before the turbo in a turbo car, they're much less sensitive to these kinds of changes. As a result, all the Turbo Subarus (in the US anyway, I think the twin scroll ones might be different...) with the turbo behind the RH head have an unequal manifold, because it goes straight from the LH head, to the RH head, and up to the turbo.

The FA20F (2015+ WRXs) and EE20 (diesel, for those that are interested), have the turbo under the front of the engine, so the manifold is equal length again. The STi still has an EJ257, with the turbo behind the RH head, and the same old unequal length header.
See less See more
1 - 17 of 17 Posts
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top