The great "octane" experiment

Oh, I know, my point was that compression ratio by itself is only a good indicator if the timing cannot change or adapt itself via an ECU/sensors.

BTW, knock sensors were not "mandated" by OBDII. I had an 02 Dakota with a 4.7, and it definitely did not have knock sensors. Later versions of the 4.7 did, but not my 02.

 

Ah, well, sorry about that. It's what you get when you rely on Wikipedia for your info:

"Higher compression ratios will however make gasoline engines subject to engine knocking if lower octane rated fuel is used, also known as detonation. This can reduce efficiency or damage the engine if knock sensors are not present to retard the timing. However, knock sensors have been a requirement of the OBD-II specification used in 1996 Model Year Vehicles and newer."

My question now is concerning this whole "resetting the ECU" craziness. Should I do that at the start of each phase? Should I not do it at all?

 

I can't help you any more with that. I do know that there are adaptives in the ECU that need to change when fuel type is changed. What I don't know is how long it actually takes the ECU to adapt and if hitting it hard, or resetting the ECU, is going to speed up or slow down the "natural" process.

 

I would recommend you run one tank to allow for computer adjustment and then record the tank after that. If you see a large drop from higher grades to 87 it most likely means that only the 87 is composed of ethanol.

All in all I think there are a lot more variables to account for. I can notice a large behavior difference in this motor on the same tank of fuel from day to day as the temperature alone changes. Your daily driving habits won't always be the same either. From my observations when at cruise (not accelerating) the timing tables are fairly static and only change based on RPM and Throttle percentage. If this is true octane would make zero difference at cruise speeds.

Once you are under acceleration there are definitely at least 2 different fuel cells which use different timing numbers. I can definitely notice a difference in how much timing gets pulled between the octanes however like I said, this is greatly affected by temperatures. On really hot days it seems to pull a decent amount of timing no matter what the octane fuel I'm running is. On moderate days higher octane fuels let the motor run at max timing most of the time, which would benifit your power under acceleration.

Regardless, if your motor is knocking and it's stock..... somethings not right. This motor definitely does not REQUIRE high octane.

 

you are not going to notice a mileage difference- any mileage changes that you do see will be because you got stuck in traffic for 20 minutes on one tank or other variables.

As far as fuel octane requirements, listen to the vehicle mfg. Honda has 10.5:1 CR engines that they say to run on regular fuel, which is 87 octane at "lower" elevations, and they run fine on 87 w/o any pinging. Note that this is a high CR to run on 87, but the engine design allows it. A 1970 designed engine with the same CR would need at least 93 octane.

 

Just fueled up w/87 for second time. My MPG on 87 was 0.9 HIGHER than that on 91. Previous tank's mileage was 21.0, and my first 87 octane tank was 21.9.

Driving style didn't change, however I did use the a/c for about 50 miles.

 

MaineYota,
Thanks for doing this long term experiment. I would not be surprised to see some small differences in MPG per tank.

 

0.9 is probably within the normal variability. I doubt you'd see a mileage decrease due to higher octane. A lot of factors you can't control can contribute to a mileage loss such as wind or traffic. You can even have subtle subconcious impacts. I've seen up to 2mpg swings with just small changes in driving style (e.g. going 70 vs 65, using ac, not shifting at the correct time). The only way to eliminate your own subconcious bias is to have someone else fill your truck and record it when you're not there so you don't know what octane is in the tank.
http://fuelsaving.info/testimonial.htm

What engine do you have in your truck (v4 or v6)?

 

Yet there is an implied hypothesis here, which is this:

Different grades of fuel may affect fuel economy.

I haven't been worried enough about fuel economy to check for myself, but I will say this:

It gets very hot in Fresno 5:00 PM rush hour traffic in the Summer. Temps frequently hit 100~105F at the airport, but on the asphalt in traffic, they can be far worse.

In such weather I can make my truck ping when using low octane fuel. With 91 octane I've never made it ping. The ECU attempts to prevent pinging, but ultimately it can only do so much in extreme temperatures.

Now, with the A/C and the radio on, you may well not hear the pinging, and it is very harmful, so my suggestion would be this:

In very hot weather always use high octane. Even in moderately hot weather, if towing or hauling a big load.

In very cold weather it won't ping, but I'd also use high octane then as it has more additives to prevent freezing & accumulation of condensation.

In moderate weather, use whatever you want.

 

Third 87 octane fill-up this morning. MPG on last tank was 21.7, which is down .2MPG from the tank before.

I used the AC for about 100 miles' worth of travel.

 

Not true. My '97 Honda accord I had before my truck did not have a knock sensor.

 

Another issue to consider in a long-term test is the effects of seasonal climate and driving conditions. I have tracked mileage in my vehicles for over a decade. Every fuel up, I write down the mileage, amount of fuel, and price. Every now and then I put the data in a spreadsheet. When plotted (even from vehicle to vehicle over the years) you see a seasonal trend. I am sure there can be a bazillion reasons (temp, rainy/snowy, gas blend, driving habits, etc.) but a trend nonetheless. My climates have been in Phoenix and then Tulsa, both places exhibiting wide variances in seasonal climates and gas blends. So if you are performing a long-term test (over several months) this may be a factor to consider. This is further complicated if you go in order increasing or decreasing octane because that would match a seasonal trend as well. A true test would have to be randomized. Unless the differences were huge, you couldn't rule out other causes.

Just a thought.

But I do have a question on some of the posts in this thread. I tried an experiment looking at the differences between 10%-EtOH and 'real' gas. I switched back and forth, using a couple tanks of one then a couple of the other, and so on. I did this over a period of several months and got no major difference. I did get an 'answer' but I do not want to hijack this post and go off on an ethanol tangent. But I am wondering based on what I read in this thread if I was giving the computer enough time to adjust to the different formulations and thus did not get good data.

 

I ran 89 for the first 2+ yrs I had my truck and then decided to see how 87 would do, and have been running it ever since, and havent noticed a real difference. gas mileage sucks in the winter and is comparable in the warm months. fuel up and roll baby.

thanks for doing this btw, i look forward to the conclusion

 

I would be shocked if you can not by non ethanol gas in your state. If you really cant what do all the motorcycles run? Boats? the marinas must be running "real Gas" if you put ethanol into a PWC that has a high performance motor that has been running real gas i have seen a ton of the fuel injectors clog and then no more run.

If you really wanted to see a dramatic difference and prove this one way or the other once and for all go down to your local airport and tell them that you want a tank of 100 octane fuel... it's for sure ethanol free and its taking it to 100 octane. All the bass boats run it around here. I know my local airport will fill you up.

 

100 octane = $$$
not to mention it isnt pratical.

 

Aye - Ethanol can't be used if tanks are fiberglass as it can dissolve some of the resins and phthalates that can be deposited as the gas evaps.

 

i did this and with my style and needed duties of my truck the 91 was best, put the motor under a load with REAL world situations and the 91 will win, when i towed, when up a grade, needed power for passing i felt the lose and response i had with 91.. ran 91 since day 1 except for a tank or 2 of other grades for testing.

i did the same thing with my ruckus scooter 50cc... ran way better on 91 then on 87 or 89

 

I think there is a lot of issues that are too difficult to place a value on when it comes to this issue. More power, torque, etc. may be subjective and difficult to prove without an 'industry level' or 'research facility level' type of study. So I say, "if it feels good, DO IT!"

Right now here in Tulsa 87 is $2.59 and 91 is $2.79. [Always an arbitrary price difference of $0.20 which is a clue to me that the cost is not based on the value] Just ten cents more for mid-grade and then ten cents more for premium. At these prices, the difference in cost to drive is one cent per mile if use use 20mpg as an average value for both. If you get less mileage the difference increases, if you get better mileage the difference decreases. For example if your average mileage is 18mpg, the difference between $2.59 and $2.79 is now 10% higher (1.1 cents per mile). At 25mpg the difference is now about 20% less (0.8 cents/mile).

But safe to say about a penny per mile difference in cost to drive (assuming mpg remains the same).

So that means if you get 20mpg, your mileage better increase to at least 21.5mpg to break even. This is strictly an mpg comparison and does not include performance issues or anything else. If you get 18mpg on regular, you better be getting at least 19.4 to break even. If you get 25mpg, then you better get 27mpg to break even.

In a nut shell, an improvement of around 2mpg is needed to justify the cost of premium with a $0.20/gal price difference in the fuel. So it will be quite interesting to see how this trial comes out. Hope my math was right, otherwise I just wasted a bunch of time for everyone