The great "octane" experiment

At the outset, I know I'm going to get flamed for putting crap gas in my engine.

That said, I want to know how much my truck's fuel economy is affected by different grades of gasoline.

Here are the experiment parameters:

- I live in Maine, and there's NO 93 octane in the state (which was a real PITA when I was driving my WRX)
- Maine uses 10% ethanol
- Fuel grades are fairly uniform: 87, 89, and 91 octane
- I'll be tracking my mileage at fuelly.com
- This phase of the experiment will last 2 months or 500 miles, whichever comes first
- Next phase will be 89 octane
- Current fuel economy is 19.7, and that's over 28 fuel-ups covering 7,305 miles
- Truck in question is a 2005 4x4 SR5 Doublecab (short bed) with 38,229 miles at the start of the experiment, and the only mods that would affect aerodynamics are rain guard things on the windows, tube steps, and a tonneau cover (all installed by the previous owner.)
- Engine is bone stock, save for new plugs installed at 30k, and a Goodyear Gator serpentine belt installed at 38k

I ran the last tank of 91 octane (what I normally use) down as far as I dared, and this morning I deposited 16.904 gallons of 87 octane on top of it. I know that the residual 91 will *slightly* increase the overall octane rating of the fuel in the tank, but seriously - this "experiment" is quasi-scientific at best already. I'm not worried about confounds from a gallon of the good stuff.

I'll update the thread with issues, noises, fuel-ups, and the like. My hope is that nothing explodes and I see only a minor drop in fuel economy. Worst-case scenario would involve my engine blowing up and me needing a tow from one of you guys.

 

Good luck man. I did a similar comparison with my truck. About 5 tanks of each grade. I found that my overall $/mi were slightly better with 91. But, the over all benefit was smaller than my driving style. If I were to accelerate a little more gently and was more cognizent of my speed, I could increase my 87 enough to make it better.

Same could be done with the 91. I also noticed that with 91, the truck seemed a little more responsive.

 

What is your hypothesis for this experiment?

 

My hypothesis is that our engines are more octane-tolerant than most believe, and that the loss of fuel economy that people are worried about is negligent at best.

I'm definitely not an engineer, mechanic, or anything of the sort...so the results I come up with will easily be dismissed, and I'm okay with that

I do think, however, that unless you're running a crazy high-compression motor (which we aren't), there's no MPG or engine longevity benefit to running fuel above 87 octane.

 

You won't see much diff in mileage running diff grades of fuel. The energy in 93 oct and 87 oct is about the same. 93 has anti knock aditives that keep high performance engines that have the timing more advanced from knocking but the energy in the fuel is about the same.

 

Hey keep in mind your engine runs on an adaptive fuel map. Everytime you disconnect your battery you basically go back to the factory base map. Your engine will have a maximum allowed timing advance for each "quadrant" of your fuel/load map. 500 miles might not be enough for each grade if you decide to reset your ECU each fuel grade change. And with out reseting your computer your engine has kind of already set it's maximum allowable parameters so your just burning expensive fuel. It will begin to "test" itself and creap back up the advancement with higher grade octane but that also takes time.

 

Cool idea.

The only reason I use mid grade 89 octane is in my state, it has no ethanol. At least I think it has no ethanol.

 

I don't know about NM, but here in all the new england states I'm pretty sure every grade is 10% ethanol, you can't get away from it.

 

Given this then, his best method would be to run premium for 500-1000 miles, get his premium data, then start going down the octanes; 91 -> 89 -> 87. This would force the ECU to set the fuel/timing adaptives at each change since he's moving down. Other thought is to run a tank or two and be pretty hard on the accelerator, that way the ECU again is forced to setting the max timing given the knock sensor data for the fuel being used.

 

I like where you're going with this. Keep us posted.

 

Same in Texas. The ethanol is the anti Knock replacement that proved to be toxic and leaking into the soil around gas stations

 

I'm willing to change up the experiment if it makes good sense to do so. I know nothing about the black magic that takes place in the ECU, so I'll defer to others to tell me what and how to do this the smartest way possible.

 

Contrary to 4th grade science teachers, not all experiments require an hypothesis.

 

I use 87, and get 20.9 mpg consistently with mixed city/hwy driving.
(V6 Access cab)

 

The energy may be the same but if your truck knocks, it'll retard the timing to prevent it. So it's conceivable that it will run better on an octane higher than 87.

I remember reading an NHTSA bulletin about using premium that stated if your engine's compression is greater than 10:1 you should run premium and if it is less than 10:1, you should run regular. Our trucks are right at 10:1.

 

Ethanol or MTBE?

 

Maine's definitely ethanol.

 

MTBE has been outlawed everywhere I believe.

 

Depends on the engine, if it has knock sensors it should adjust to a lower timing level if the fuel isn't sufficiently knock-proof.

 

All our 2nd gens should have knock sensors, as they were mandated as part of the OBDII spec from '96 on.