93 Octane??

When you get 5-10% better mileage on one tank than you have gotten on
ANY other tankful in the past, and all you have done is changed fuel,
it is a pretty good bet the better fuel made the difference.

Also, on the Pontiac, OBD1, the scanner shows a significant reduction
in knock signatures by switching from 87 to 89 octane, not much better
going to 91.

 

But my point is that it is impossible to eliminate the non-fuel changes from the
experiment because of the many variations of daily real world driving. If the engine
was on a dyno in a controlled test facility, it would be a different story.

 

But - people pay money for fuel for trips made that have all these
variables, they don't
drive in the lab. So you got to make your testing real-world, not lab.

An accurate enough real world test would be to run about 3-4 tanks of
regular gas
through the vehicle, followed by 3-4 tanks of premium. You tell the driver
what is being
run. Then you follow this with another 3-4 tanks of regular and tell the
driver it's
premium, and then follow with 3-4 tanks of premium and tell the driver it's
regular.

That would do several things, first it would eliminate a lot of the trip
differences because they
would average out over the 3-4 tanks, second it would show if the driver
was subconsciously interfering with the test, third it would demonstrate
that the benefit of
the better fuel would go away when you switched back to the regular.

The big problem with Greg's method is that he as a driver is conscious that
he's running
premium. Thus if he expects better gas mileage out of premium, he's may
unconsciously
drive in a more fuel-efficent manner.

And of course, understand that once all that's done, the data is only good
for the
tested vehicle. Unless you did an experiment like this with dozens of
same-model
vehicles, you can't draw conclusions that are applicable to other vehicles.

Ted

 

The spread would have to be tiny for premuium to deliver enough better
mileage to cover the additional cost. And if your vehicle does not have
a knock sensor, and a PCM programmed to take advantage of higher octane
gas, there is no reason to expect any increase in mileage with premium
fuel, unless the car was knocking on regular. Out of curiosity, I tried
premium in my '97 Expedition for many thousands of miles. The Expedition
does have a knock sensor, and supposedly can take advantage of premium
fuel. There was no significant change in fuel mileage when running
premium. I also tried the same thing with my Mustang (which did not have
a knock sensor) and, again there was not a detectable difference in fuel
economy.

Years ago, when I was in college, I was taught that premium fuel often
had a slightly lower energy content than regular becasue of the blend of
componets used to achieve the higher octane. As far as I can tell, this
is no longer the case and in fact there are numerous references that
claim premim has a slightly higher energy contnet than regular.
Persoanally, I don't think there is a greater difference in the energy
content between regular and premium than the normal variation in
different tankfuls of regular.

From
http://www.chevron.com/prodserv/fuels/bulletin/motorgas/questions.shtml
:

"Gasolines with higher heating values give better fuel economy.
Differences can exist, but they will be small compared to the benefits
to be derived from the maintenance and driving tips in the above answer.

"Traditionally, premium has had a slightly higher heating value than
regular, and, thus, provided slightly better fuel economy. The
difference — less than 1% better — is not large enough to offset
premium's higher cost. The difference is likely to be less or
nonexistent between grades of reformulated gasoline.

"There can be differences in heating value among batches of gasoline
from the same refinery or among brands of gasoline from different
refineries because of compositional differences. The differences are
small and there is no practical way for the consumer to identify the
gasoline with a higher heating value."

Other interesting references:

http://www.chevron.com/prodserv/fuels/bulletin/motorgas/sidebars/heating.shtml
http://www.f6rider.com/Valk2/every_thing_you_ever_wanted_to_k.htm
http://search4cars.auto123.com/en/info/tips/view.spy?artid=590
http://www.petro-canada.ca/eng/prodserv/fuels/6823.htm

Ed

 

I have kept a mileage book on all my cars for years. I can show you
examples of tankfuls that appeared to get almost 20% better mileage with
no changes in the vehicle or typical driving pattern. You need to check
mileage over several tankfuls (I recommend at least 5 full tankfuls) to
get anything approaching a meaningful result. In my case I ran premium
for 6,299 miles in a 1997 Expedition. The average fuel economy while
running on premium was 14.1 mpg. The fuel economy for the 6,230 mile
before I tried premium (running on regular) was 13.9 mpg. The average
fuel economy for the 3,316 miles after I switched back to regular was
14.3 mpg. While using premium the best tankful average was 16 mpg and
the worst was 11.3. With a five tank average the maximum was 14.6, the
minimum 13.5. Overall, my best one tank average was 17.4 (running
regular) and the worst was 9.6 (also running regular). Using a five tank
average the best was 17.2, the worst 10.5. The Expedition had a knock
sensor, and it was claimed that using premium would improve both power
(3 to 5 hp) and fuel economy. I could never tell any difference in
either, although I know other Expedition owner's that claimed their
personal butt-o-meter could detect the extra power.

Ed

 

How did you know the mileage was higher? Often people compare one tank
of regular to one tank of premium and think the difference they see is
because of the gas.


Matt

 

Not at all. I always fill my tank when I get gas and I check the MPG on
every tank and keep it in a log in each vehicle I won. 10% variation
tank to tank isn't unusual at all. Unless you've checked at least five
consecutive tanks on regular and then five on premium and averaged each
of the five, you don't know anything about whether the gas made a
difference.

That makes sense as higher octane is designed to knock less. That is
what octane is a measure of. If your engine is knocking heavily on 87,
then it is likely that the timing is being retared quite a bit causing a
loss of power and possible mileage as well. In this case, the higher
octane might well deliver slightly better mileage, but you'd have to
check several tanks to have a statistically valid comparison.


Matt

 

But the problem with your method is that every trip is different, even when the
gas stays the same. I've kept fuel records since my car came off the dealer's
lot and the mileage varies considerably, even among trips when the car is just
used for commuting on the same route and when fuel is purchased from the same
station.