Not Nissan or Infiniti question, but I would appreciate an answer.

[szh]

Hi, all.

This is not an Infiniti or Nissan question, but I wonder if you guys could shed some light or comment on this. My wife's car is a 1997 Acura 2.2CL. Essentially a stock Honda Accord EX engine (140HP, 4 cylinder) that the manual says only needs 86 octane fuel.

Well, I have always used 91 for the heck of it, since the price difference was not that dramatic per se for what is a decent mileage car. Usually Chevron. When gas prices went up late last year, my wife told me to try 87 octane - well, after some "discussion", we compromised on 89. Well ... after a few tank fulls, she started complaining that her miles-per-gallon had dropped and she was filling the tank more often.

Hmmm ... I said.

So I tried a more methodical test. Ran about three tank fulls of 89 octane - same gas station, same driving habits. The tanks all lasted about 325 or so miles. Then I repeated with about three tank fulls of 91 octane - again, same gas station, same driving habits. The tanks lasted almost 400 miles! So, I went back and repeated the 89 octane with a few more tank fulls. Sure enough, they were about 325 miles per tank full.

This surprises the heck out of me! I had always assumed that using a lower than recommended octane would not be a good idea, but that using a higher than recommended octance should not have any effect (like all the newspaper like to talk about!)

Since the cost difference between 89 and 91 is only about 10 to 20 cents per gallon (on a base near $2.10 or so for premium here in California), it appears that the cost per mile for the 91 octane is better on this Acura.

My question is: WHY?!?

Z

[DAEDALUS]

The short answer is that it appears the car is simply tuned to run better with the 91, for whatever reason. Given a small difference, I would have suggested differences in gasoline between Japan and your location or slightly advanced timing might be to blame.But 75 miles per tank (20%!) is a big difference. I'll submit that maybe the engine was knocking with the 89, and retarded the timing in a big chunk. Even that wouldn't seem to explain it though.My understanding is that increasing the octane can lower mileage in the same way that dropping it can do the same. It all depends on what the car's tuned for. Ideally you want combustion to end at a very specific point along the crank's rotation. So the burn rate of the gas is important to achieve this, along with the spark timing of the vehicle. The burn time is fairly constant...10 ms IIRC. But the engine speed varies infinitely, yet you still want combustion to stop at the same place every time...multitude of variables the ECU must contend with. If combustion finishes either before or after this optimum point, mileage will drop, since the engine will need more gas to produce the same power.

[szh]

Thanks for the response, Daedalus!

I do not have any answers. One thought I had that I had had the car tuned (at the Acura dealer) last time when I was still using 91 octane. Maybe ... they "tuned" it for the gas that was in there, so it effectively now wants to live on 91 octane.

But it seemed strange to me that the drop would be so large. In fact, I just filled up this morning, and the tank read empty at 289 miles (even below the the 325!) But it has been quite a bit colder recently, so that could account for this change. In any case, I put in 91 to see what happens on the next tank full.

Another thought was that the knock sensor (assuming that this engine has one - I am not knowledgeable about it) might be acting flaky and giving the wrong signals to the engine management. It is, after all, an eight year old car. Nothing major has gone wrong with it, touch wood (as I wrap knuckles on my head), but I sure would like to avoid having it start now.

Z

[97Q45t]

If you've been running the car with 91 gas for a long time, the ECU must have been adjusted to run best in that condition. I think the time to burn 1 tank of gas is not enough time for the ECU to adjust itself. Plus when we tend to drive just as fast when we change to a lower octane gas, hence the gas pedal has to go deeper a little bit. My suggestion is to reset the ECU (disconnect the battery maybe?) when you start to use the 87 octane gas so the ECU will be adjusted right away.

[avatarr]

If you've been running the car with 91 gas for a long time, the ECU must have been adjusted to run best in that condition. I think the time to burn 1 tank of gas is not enough time for the ECU to adjust itself. Plus when we tend to drive just as fast when we change to a lower octane gas, hence the gas pedal has to go deeper a little bit. My suggestion is to reset the ECU (disconnect the battery maybe?) when you start to use the 87 octane gas so the ECU will be adjusted right away.

This is sound advice. Disconnect the battery and press down on the brake pedal. This will reset the ECU. Run the car on 87 octane for a couple tank-fulls. Take your data, reset the ECU, and do the same thing with 91. Come back with your results. My guess is that you'll have better mileage with the 91 but I'm guessing the results won't be as staggering as you found out they were originally. I would also suggest that you get the same type of gas (Chevron) with the 87 and the 91 octane - more consistent detergents.

-Jason

[Q45tech]

ECU are fast learners no more than 30 miles at highway speed teaches it all it needs to know.

Modern ecu advance the timing till the knock counts per second exceed a thresohold then back off [too much] till it stops then creep forward again till it learns what doesn't knock till you turn the car off then starts all over again on a new crank up.

Ignition advance at cruise can dramatically affect MPG at highway speed but probably not 20%.......more like 10%.

The reason premium grades might get less MPG is that premium may contain more MTBE or ethanol and thus average BTU/gallon might be 3-7% lower but the converse could also be true depending on refinery and the day fuel was made.

California is funky as they are playing with aromatic% on daily basis to drop MTBE and avoid ethanol........this weeks premium just might be 100% gasoline without oxygenates -----varies daily .........that would explain it.

Oxygenates screw with the O2 calibration making the mixture 5% richer than desired thus less MPG.

Oxygenated regular/89 blend vs. non oxygenated premium?

[szh]

I think the time to burn 1 tank of gas is not enough time for the ECU to adjust itself.

Like you, I was not sure of this, plus I wanted to make sure that as little of the previous octane was still in the tank, so I measured over three tank fulls each time ...

Z

[szh]

I would also suggest that you get the same type of gas (Chevron) with the 87 and the 91 octane - more consistent detergents.

Agreed! I did my more methodical tests by buying from the same gas station every time (Chevron 89 and then Chevron 91.) I didn't even go to another Chevron station in the hope to get consistency.

[justjuiceit4]

If you've been running the car with 91 gas for a long time, the ECU must have been adjusted to run best in that condition. I think the time to burn 1 tank of gas is not enough time for the ECU to adjust itself. Plus when we tend to drive just as fast when we change to a lower octane gas, hence the gas pedal has to go deeper a little bit. My suggestion is to reset the ECU (disconnect the battery maybe?) when you start to use the 87 octane gas so the ECU will be adjusted right away.

According to an engineer at Saturn, the new Redline supercharged model takes 3 tanks of premium gas to get optimal performance.

[scoobas]

higher octane less petrol needed for each detonation dah

[ARKQX33V6]

Running an ICE, internal combustion engine designed for 87 at 91 is a waste of money, at 91 the engine can carbon up more because of the slow burn 91, with carbon deposits the 87 engine running 87 may ping because of this carbon increasing the compression. But you know how your engine is running