Subaru Turbo Engines

[Ken Gilbert]

no miles per gallon increase, but i do get more Smiles per gallon as
the higher octane allows me to safely run a bit more boost (16.5psi
vs. 14psi stock).

i have noticed that winter gas formulations gave me a 10% hit in
economy.

hth
ken

 

[Niels]

Interesting! Do you have a turbo boost guage? (If so, was it
factory-installed, or after market.) If not, how did you get the
measurements. Being a numbers freak, I'm starting to regret not getting the
turbo gauge.

Niels

 

[The Masked Marvel]

Here in the Northeast US it's 87, 89, 93 (or occasionally 94 -- mostly
Sunoco) R+M/2 (average of "Research" and "Motor" methods) IIRC R+M/2 is
generally equiv to 91 RON (Research Octane Number) I believe "100 octane
racing gas" is 100 RON, not R+M/2 so about 97 or 98 R+M/2 perhaps?

 

[Greg Reed]

The point of higher octane isn't to improve mileage so much as to improve
power output (though a slight mileage increase is theoretically possible).
Higher octane means increased resistance to detonation -- which is the gas /
air mixture exploding before the spark plug tells it to. Such explosions,
which are also called "knock" because of the sound made when they occur, are
*very* hard on the engine because they occur while the piston is still
making its upstroke. The force of this early explosion tries very
forcefully to push the piston back down from whence it came.

Of course, the entire purpose of putting an explosive mixture into the
combustion chamber is to push the piston down, but the timing and valvetrain
are designed so that this happens later in the piston's stroke. All of the
stress-bearing components inside the engine -- the connecting rods and pins,
the crank, and the crank bearings -- are designed to accommodate this
explosion at just the right point in the piston's stroke. When the
explosion occurs early, these components all have to continue pushing the
piston up through its stroke, but against a tremendous resistance. An
occasional detonation won't destroy an engine. But over time, it will
definitely take a toll. (In addition to the use of higher octane, cooling
the intake charge also helps to reduce detonation and therefore increase
power output. This is the reason for the water spray on the STi's
intercooler.)

While any car with a knock sensor should theoretically see a benefit from
higher octane, it generally grants the most benefit to those engines with
very high internal cylinder pressures. Those would be cars with high
compression ratios or some sort of compressor on the intake (be it a turbo
or a directly-driven supercharger).

Those with turbocharged engines don't need to panic if they've been running
low octane gas. All modern engines employ knock sensors that can do any
manner of things to stop knock once its detected. The only affect of
running low octane is that your engine's computer might not be allowing
maximum power output.

I run 87 octane unleaded in all of my cars except the Audi, which gets the
highest octane available at whatever pump I happen to be using to fill up.
My new turbo-charged Subaru will also get the highest available octane.

- Greg Reed

--
1976 Cadillac Fleetwood 75 9-Pass sedan
(FS: http://www.dataspire.com/caddy)
1989 Audi 200 Turbo Quattro 5-Speed sedan
2000 Oldsmobile Intrigue
2001 Chevy Astro AWD (wife's)
2005 Subaru Legacy GT Wagon (when available in U.S.)

 

[y_p_w]

The point of higher octane isn't to improve mileage so much as to improve
power output (though a slight mileage increase is theoretically possible).
Higher octane means increased resistance to detonation -- which is the gas /
air mixture exploding before the spark plug tells it to. Such explosions,
which are also called "knock" because of the sound made when they occur, are
*very* hard on the engine because they occur while the piston is still
making its upstroke. The force of this early explosion tries very
forcefully to push the piston back down from whence it came.

Just adding a few things.

If I recall, it's a combination of increased pressure in the fuel/air
mixture combined with residual heat that can cause the early explosion.
Forced induction or higher compression ratios will cause an increase
in pressure. I think carbon deposits cause a slight increase in
compression ratio, and a larger increase in residual heat - i.e. they
get hot (and may prevent heat transfer?).

The other thing is that some people think that higher octane rating
means the fuel burns slower. I've heard that high octane racing
gasoline actually burns faster, but resists spontaneous explosion
better than ordinary gasoline.

And it's been a while, but there was something about a "second flame
front". The explosion from the spark causes the an increase in
pressure before the flame spreads, and the fuel/air mixture may
explode in a separate part of the combustion chamber. Thus two
explosions crash into each other - making the knock that we hear.
The ideal is a single explosion gradually and evenly spreading from
the spark.

I found some good articles, but this one from Car & Driver is easy
enough to read (+ another decent article):

<http://www.caranddriver.com/article.asp?section_id=4&article_id=3604>
<www.sizes.com/units/octane_number.htm>

 

[CompUser]

Those with turbocharged engines don't need to panic if they've been running
low octane gas. All modern engines employ knock sensors that can do any
manner of things to stop knock once its detected. The only affect of
running low octane is that your engine's computer might not be allowing
maximum power output.

I run 87 octane unleaded in all of my cars except the Audi, which gets the
highest octane available at whatever pump I happen to be using to fill up.
My new turbo-charged Subaru will also get the highest available octane.

The knock sensor(s) doesn't per se do anything
except alert the ECU that detonation (or
something that sounded like detonation to the
sensor) has occurred. The ECU will respond to
its programming by retarding the engine timing to
prevent recurrence of detonation.

This will normally be all fine and dandy, but if
the engine is designed to run on 91 or 93, and
it's got a tankful of 87 (which is more prone to
preignition than 91 or 93, given the same
conditions), the ECU *may* run out of adjustment
range on the timing to prevent recurrence of
detonation, depending on the conditions.

At that point, the ECU programming says, "we
gotta stop this preignition!! cool the intake
charge, with more gas!", and the injectors spray
more and more fuel into the cylinders. See where
this is going?

Excess fuel begins to wash the cylinder bores
clean of oil...a really bad thing. Google
(detonation timing bore wash) and you'll see
pages of stuff.

The concept of "It says 93, but it'll run fine on
87" is risky with today's engines.

If you have to fill up somewhere on 87 in your
WRX or STi, do what you have to do but don't push
the engine...stay off the boost as much as
possible and get good gas in there, ASAP.

Steve

 

[Niels]

Steve,

Thanks for the explanation. My dealer had mentioned that in a pinch, use
the cheap stuff if you ABSOLUTELY have to. He never told me 'why'.

Has anyone ever been to a gas station where premium was unavailable? NEVER
have I seen such thing.

Niels

 

[CompUser]

Steve,

Thanks for the explanation. My dealer had mentioned that in a pinch, use
the cheap stuff if you ABSOLUTELY have to. He never told me 'why'.

Has anyone ever been to a gas station where premium was unavailable? NEVER
have I seen such thing.

I've seen it occasionally...usually a confluence
of events, like finding an open station in a
rural area off the interstate late at nite, on a
big holiday weekend.

Steve

 

[Greg Reed]

Just adding a few things.

If I recall, it's a combination of increased pressure in the fuel/air
mixture combined with residual heat that can cause the early
explosion. Forced induction or higher compression ratios will cause
an increase in pressure. I think carbon deposits cause a slight
increase in compression ratio, and a larger increase in residual heat
- i.e. they get hot (and may prevent heat transfer?).

The other thing is that some people think that higher octane rating
means the fuel burns slower. I've heard that high octane racing
gasoline actually burns faster, but resists spontaneous explosion
better than ordinary gasoline.

And it's been a while, but there was something about a "second flame
front". The explosion from the spark causes the an increase in
pressure before the flame spreads, and the fuel/air mixture may
explode in a separate part of the combustion chamber. Thus two
explosions crash into each other - making the knock that we hear.
The ideal is a single explosion gradually and evenly spreading from
the spark.

I found some good articles, but this one from Car & Driver is easy
enough to read (+ another decent article):

<http://www.caranddriver.com/article.asp?section_id=4&article_id=3604>
<www.sizes.com/units/octane_number.htm>

Wow!! I didn't realize just how much of this stuff I'd forgotten. Thanks
for the additional details.

- Greg Reed

--
1976 Cadillac Fleetwood 75 9-Pass sedan
(FS: http://www.dataspire.com/caddy)
1989 Audi 200 Turbo Quattro 5-Speed sedan
2000 Oldsmobile Intrigue
2001 Chevy Astro AWD (wife's)

 

[Ken Gilbert]

At that point, the ECU programming says, "we

gotta stop this preignition!! cool the intake
charge, with more gas!", and the injectors spray
more and more fuel into the cylinders. See where
this is going?

Excess fuel begins to wash the cylinder bores
clean of oil...a really bad thing. Google
(detonation timing bore wash) and you'll see
pages of stuff.

to add to steve's comments,

the subie rex ecu has a "high det mode" (aka mode B, mode 2) in which
if ignition advance cannot be retarded enough to prevent knock, more
serious measures are taken.. dumping a LOT of fuel, and restricting
boost to wastegate pressure. hopefully your ecu will never have to do
this.

ken

 

[y_p_w]

I've seen it occasionally...usually a confluence
of events, like finding an open station in a
rural area off the interstate late at nite, on a
big holiday weekend.

You mean they're sold out or never had it in the first place?

 

[LCB]

Good thread guys, excellent information - Thanks. I think this also
applies to those with EJ25 DOHC engines that may be wondering what to
do to keep from blowing their head gaskets. The lesson being that
better gas would keep engines with weaker gaskets from breaking
sooner.

 

[CompUser]

You mean they're sold out or never had it in the first place?

Sold out.

 

[Newsy]

Steve,

Thanks for the explanation. My dealer had mentioned that in a pinch, use
the cheap stuff if you ABSOLUTELY have to. He never told me 'why'.

Has anyone ever been to a gas station where premium was unavailable? NEVER
have I seen such thing.

Niels

Australia evidently as some of the worst fuel in the world - baring third
world countries of course.
We have 98 RON in the big cities, but there is none between Adelaide (where
I live) and Melbourne - a trip of over 1000 Kilometres. There is also none
between Adelaide and Darwin - 3000 Klm's.

Rod

 

[Bruce Hoult]

Australia evidently as some of the worst fuel in the world - baring third
world countries of course.
We have 98 RON in the big cities, but there is none between Adelaide (where
I live) and Melbourne - a trip of over 1000 Kilometres. There is also none
between Adelaide and Darwin - 3000 Klm's.

Eek.

According to the following web page (which doesn't claim to be
exhaustive), there is 98 available between Wellington and Auckland in at
least Levin, Palmerston North, Taupo, and Hamilton.

http://www.palmy.net.nz/mr2dog/guides/petrol.html

As long as you can manage at least 200 km (250?) on a tankful there
should be no problem with polluting your engine with that inferior 96
octane stuff.

-- Bruce

 

[Toby Ponsenby]

We have 98 RON in the big cities, but there is none between Adelaide (where
I live) and Melbourne - a trip of over 1000 Kilometres. There is also none
between Adelaide and Darwin - 3000 Klm's.

Rod

Rod, you should really tell the oil companies that. They really do give a
shit - honest!

--
Toby
quidquid latine dictum
sit, altum viditur
Warning: All posts to be treated with a grain of salt.
By Order
K.K.