Subaru

[Mettler]

Having an argument with a coworker here as to why subarus make that pulsing/rumbling exhaust note.

We disagree on the cause of it.

I say it's entirely due to the exhaust manifold design, having uneven length runners, causing pulses to run in together and exiting the tip in that uneven burble sound.

He says it's due to the engine's design, being a flat 4, and the firing order causing a half resonance effect. He sent me this link to back up his argument: http://www.doudna.com/Subaru/

Now riddle me this. Rumbly subarus only run a single exhaust, not a dual exhaust, right?

The exhaust note of a subaru is heard at the exhaust tip, right? You don't hear the sound of it up at the manifolds as it's coming out of the engine.

What this means, is that you're hearing all four exhaust pulses coming out in the sequence that causes the subaru exhaust burble. What makes them come out in that order?

The only POSSIBLE REASON for it, is the exhaust runners being uneven, making the pulses from one side have to travel further than the other side, and them running in together to the turbo, before coming out the back.

Here's a quote from that article:

The key thing to note is that while a boxer-4 does alternate firing fore and aft cylinders, it does not evenly alternate firing between its left and right cylinder banks. It cannot due to the 180 degree orientation of crankshaft pins selected for balance. So instead it must fire twice on one side and then twice on the other. And unlike an inline-4, a boxer-4 must have two separate exhaust manifolds.
One manifold exhausts [fire, fire, wait, wait] while the other side exhausts [wait, wait, fire, fire]

So in addition to the evenly spaced firing of each cylinder (just as from an inline-4) the boxer-4 has exhaust pulses exiting the left and right manifolds at half that frequency. This cadence is perceived as a half-pitch "rumble".

What the author of this article has done, is attempted to separate one cylinder bank from the other, to describe exhaust note. /facepalm

In terms of exhaust note, this is retarded because the vehicles do not run a dual exhaust system. Therefore, it's not [fire, fire, wait, wait] while the other side exhausts [wait, wait, fire, fire].

The engine as a whole, produces exhaust pulses like this: [fire, fire, fire, fire]. It fires one cylinder after another, evenly. It's irrelevant what side is firing, since the exhaust runners all come together to a single point before entering the turbo. It's an even series of combustion events, as even as any i4.

What this means, is the only possible way for the pulses to be uneven like this, is what's between the exhaust ports, and the merge point into the turbo. i.e. what does the exhaust pipe see coming into it?

What we have, ladies & gentlemen, is uneven length exhaust runners.

This is reinforced by the fact that all new subarus as of like 2005 or whenever, no longer have that sound... they have an even, flat exhaust note like an evo or a silvia, due to a new manifold design being in place.

Lastly, this effect is confirmed in my own original exhaust system on the V8... I made some headers for it that didn't have even length runners, and this my exhaust pulses ran together and created an unusual sounding exhaust note, not entirely V8 sounding, but more like 2x angry WRXes.

[Gauge]

Good points, I always thought it was the boxer engine design.

I ****ing love the subaru noise tbh. Sounds awesome.

[Mettler]

So howcome two people now have voted boxer engine design? Are you idiots? WTF is wrong with you, read the OP, clearly you need to learn something.

Can a mod tell me who made those votes so I can laugh at them please?

[Gauge]

I did it just to **** you off.

lol

[Mettler]

I ****ing love the subaru noise tbh. Sounds awesome.

Sweet, for the right price, I can make your silvia/*insert i4 powered car* sound like that too.

[[BT]MikeMan]

I think it is a case of a combination of both.

A VW or Porsche flat-4 also has a distinctive sound and uses equal (Roughly) length manifolds and single or dual exhausts. Not the same as a subbie but not a L4 either.

[Mettler]

Have a good hard think Mikeman, about how it's actually possible for it to be at all caused by the engine's cylinder layout or firing order. (Other than, of course, the cylinder layout forcing manifolds to be a certain design cos exhaust is on one side.)

[rugby_super12]

New sti's still have the boxer rumble u know (even that new hatchback 08 version)

[Lord monkey]

Alfa's boxer 4's have an interesting sound to them aswell.

[Mettler]

Perhaps the new STIs are still running an uneven length manifold. Look it up on google m8, there's plenty of even length runner manifolds up for sale that provide an increase in power, but eliminate the rumble sound.

Just remember that yeah maybe the induction note and the engine's actual vibration is as described, because yes the resonance thing is true about it firing twice per side, but it's specifically and only the exhaust note that I'm talking about.

If I'm wrong, explain then why my uneven length exhaust headers on the V8 made it sound like 2x subarus.

[rugby_super12]

Well yea it's no secret that the most powerful subaru's sound like honda's, but i'd rather sacrifice some performance for a nice aggressive exhaust note. You're right that the new sti's have lost a bit of the rumble, but it's definately still there.

[Mettler]

That's not what the point of this thread is about. It's neither to diss nor praise subarus. It's to establish the cut & dried facts about exhaust system design and how it affect exhaust note, nothing else.

[EZAS]

Have a good hard think Mikeman, about how it's actually possible for it to be at all caused by the engine's cylinder layout or firing order. (Other than, of course, the cylinder layout forcing manifolds to be a certain design cos exhaust is on one side.)

I have no idea, at all, But I thought that it was to do with firing order, like 1 cylinder on either side firing at the same time. Thus giving it that V-twin sound?

[GaR]

TLDR

It's due to the firing order. Each bank fires twice in turn, which is the main reason.

[Mettler]

Oh f'n nonsense, you're wrong GaR.

At any given RPM, a boxer engine is generating as many combustion events as an i4 engine held at the same RPM. The way the sound comes out is entirely dependent on exhaust plumbing.

[GaR]

Oh f'n nonsense, you're wrong GaR.

At any given RPM, a boxer engine is generating as many combustion events as an i4 engine held at the same RPM. The way the sound comes out is entirely dependent on exhaust plumbing.

If that were true, then half the inline 4s around would sound all weird as well - there are that many different manifold setups for them.

(According to an article I read many years ago,) On a Subaru, one bank fires twice, then other bank fires twice, rather than alternating like pretty much all other boxers and V engines. The same article referred to the sound as a "3-cylinder beat", whatever that's supposed to mean.

[Mettler]

Maybe if you stopped with the TLDR (and who are you kidding, you love to read), you'd realise that YOUR FACE.

If that were true, then half the inline 4s around would sound all weird as well - there are that many different manifold setups for them.

Ever consider that many 2.0 I4 engines do sound way different to each other? i.e. 4G63 vs SR20, etc?

It's just that on an I4, runner length doesn't vary to the same extreme as on a boxer engine, because they're effectively all coming out of the cylinder head on the same side of the engine and even if they vary in length, it's only a small variation and not enough to cause pulses to overlap, merely get a fraction closer to each other thus giving a slightly rougher note, but not a full on dual pulse burble.

[GaR]

Ever consider that many 2.0 I4 engines do sound way different to each other? i.e. 4G63 vs SR20, etc?

It's just that on an I4, runner length doesn't vary to the same extreme as on a boxer engine, because they're effectively all coming out of the cylinder head on the same side of the engine and even if they vary in length, it's only a small variation and not enough to cause pulses to overlap, merely get a fraction closer to each other thus giving a slightly rougher note, but not a full on dual pulse burble.

There are some pretty weird exhaust manifold designs out there.

Okay, so I read it. And precedent would seem to support the guy who wrote that article.

The Gen 3 V8's only real architectural difference to earlier small blocks is the firing order. That's also what's been blamed by both journalists and engine builders for the flatter, less staunch exhaust note. Therefore, a single system can sound significantly different due to firing order.

On the Subarus, I imagine some weird interactions going on in the 2-1 collector, as one side pulses twice in a row each time. The idea that that would result in a distinctive exhaust note doesn't seem so far-fetched to me.

If I really cared, I'd suss out the various exhaust systems of VW boxer 4s, Ford V4s, etc (neither of which have the SUbaru "3-cylinder beat", seemingly regardless of the exhaust system fitted).

[Worst]

An I4 would have a hard time replicating the sound without some funky-ass manifold which would be pretty much superfluous. In the same, as you said, differing manifolds can reduce or remove the distinctive noise.

http://auto.howstuffworks.com/engine-flat.htm

So the firing order is front to back, alternating sides. I imagine that by the time the pulse from each cylinder are all in a single exhaust, the net result would be something like,

X........X................X........X........


X being pulse, dot being gap. Crude, but you see where I'm getting at. I surmise the engine is the important factor in the distinctive burble.

[Jewjew]

I originally thought (and voted) for the engine design.

But afer doing some quick googling, it certainly does appear to be the exhaust design, not the engine design.
I would have thought that an extra meter or whatever of mainfold wouldn't effect the sound so much, considering its travelling at the speed of sound, but I guess the engine pulse and airflow would affect it greatly.

[Darkness]

Exhaust manifold design FTW and when i took off the ex manifold of my mates wrx ra it sounded like a Honda .If you know some one with or seen a Subaru which has tuned length headers/ex manifold then you will know it's just sounds like a normal modified car.Thats the main reason why i haven't changed my manifold yet even thou it will give me better top end i love that rumble.Might just get a after market set just non tuned ones.

[Mettler]

Worst: I've seen evo rally cars with exhaust systems that sounded just like street WRXes I've heard around (i.e. burbling).

I don't think it's a combination of being a boxer engine AND exhaust manifold to make that sound, it's solely the exhaust manifolds. However, the exhaust manifolds being this way is a result of the boxer engine design.

GaR: consider this, exhaust pulses have to actually travel... the sound of the combustion event isn't an instant delivery to the rear of the car, the sound & exhaust wave has to fly at *whatever* speed, through the piping, into the collector, through the turbo, and down the exhaust & out the back.

Now think of this. If a cylinder fires, then another cylinder fires, and the first cylinder's exhaust runner is longer than the second one's exhaust runner, then both pulses will arrive at the collector at the same time (roughly), or within very close proximity to each other, merging together to create a single larger pulse (and sound). That's what was occuring in my old headers.

Lastly, I went and discussed this with Stefan out in the workshop just before, and he says for you lot to go and listen to the exhaust note of an Alfa 33.

[Mettler]

An I4 would have a hard time replicating the sound without some funky-ass manifold which would be pretty much superfluous. In the same, as you said, differing manifolds can reduce or remove the distinctive noise.

http://auto.howstuffworks.com/engine-flat.htm

So the firing order is front to back, alternating sides. I imagine that by the time the pulse from each cylinder are all in a single exhaust, the net result would be something like,

X........X................X........X........


X being pulse, dot being gap. Crude, but you see where I'm getting at. I surmise the engine is the important factor in the distinctive burble.

WTF, are you seriously actually attempting to say that a boxer engine doesn't have a combustion event every 180 degrees, but are spaced like... 200 degrees, and then 160 degrees apart?

[GaR]

^ I get what you're saying, and it makes sense. I was merely relating what I'd read, and why it made sense to me. I'm gonna go with calling it a combination of things - especially since boxer and V4s sound completely unlike inline 4s, even if most of them don't sound like Subarus.

[Mettler]

BTW I'm fairly certain that V4s (boat engines at least) have two cylinders firing at the same time.