Hey guys just wanted to fill you in on a little fuel economy study I've been working on. Right now I just completed the 'preliminary investigation' to determine whether there was any merit to scientifically analyzing whether or not premium fuel results in better economy than regular 87 octane.

So far my results have concluded that YES 93 octane results in better economy than 87 octane around town and on the expressway at 70mph (2000 RPM) for long periods of time.

My methodology this far has been relatively simple relying upon the onboard economy meter (which I don't 100% trust), similar routes, and lastly an eye-ball approximation of my fuel gauge to verify what the econometer has said.

In my past 1000 miles of testing I have found e-way mpg with 87 yields about 13-14.5 mpg both into and with the wind. With 93 octane I have consistantly been achieving 14- 16.7 mpg into and with the wind. On the east/ west routes, and north and south routes.

In my latest trip I traveled M-57 east and west at 60mph with an economy of 17.1 east, and 15.4 west on 93.

Furthermore, just to kind of give this approximation some meaning... I have consistently been driving about 140-150 miles on a quarter tank on premium. Over the 4th of July holiday, I drove from Flint to Sanford Lake, to the beer store 3 times, and home all via I-75 and averaged 16.1 from door step to door step on $20 worth of gas.

Now I am aware there are some discrepancies between econometer, and hand calculations... especially on my Sanford runs. But what I interpret this to mean is there is some merit in actually filling the tank up making a trip, refilling and using the actual fuel consumption against the mileage for a real study, also noting route and temperature.

The reason I'm posting this here is twofold:

1. If this turns out to be a true economy gain with fuel type it may benefit all us gassers

2. I need yall's help!

If anyone out there consistently drives particular routes round trip that are 10+ miles each way I'd really appreciate your help in collecting data for me, both using 87 octane and 93 octane specifically from Shell or BP. To actually perform an ANOVA analysis on this experiment I actually need quite a few data points and without a job it could take me a year or 2 to gather enough data.

If any of yall are willing to contribute please let me know so I can give you the details of what I need! Thanks yall!

 

I wish I could do that but not after my insurance went up when in my name

That interested test. If it improve like 5 mpg with 93 Octane then check how much it worth to use 93 instead 87.

 

I get the idea that you're making measurements on small trips and topping off the tank frequently. If so, then this is introducing large error in your mileage calculations. You'd be much better off scrapping all your data and starting over and running the fuel tank low before filling up again. The more fuel you add the less measurement error you will have due to inconsistency in final fill level.

 

This I have to disagree with you on, and maybe after a little explanation you'll see where i'm coming from:

To accurately test fuel economy improvements we need to keep as many variables as we can the same, thus the particular route. If you test the exact same route, multiple times, multiple directions, and the only variable changed is the fuel type we are focusing on one parameter.

When using an entire tank, we cannot control as many parameters. For example, if you have a full tank you are seldom dictated to a similar route due to the nature of life. What is known for a fact is, the more you accelerate the more fuel you use. Thus if a person gets a tank of 93 octane, drives thru the city for all 30 something gallons encountering a stoplight every 1/4 to 1/2 mile, then gets a tank of 87 and drives 400 miles down the eway with a strong tail wind it will drastically skew the results. THis is obviously an exageration, however I think you should get the principle.

If we (being multiple data gather-ers) test a single route in 2 directions we eliminate the effect of elevation. The biggest discrepancy is going to be the amount of fuel used according to gas pump which should be 1 decimal place variation, but for safty's sake we'll say 1 gallon.

An ANOVA test is "ANalysis Of VAriance" which statistically analyses our dedicated route results and attributes the amount of gain (within a confidence level) to the different specified factors. In the study I'm pursuing will be Temperature, Fuel Type, miles traveled, and Economy. ANything that cannot be statistically assigneed to any of those 4 parameters will go into the noise/ error region of the test. Based upon what I have seen I am VERY confident that this method will yield results, and thats why I believe 10+ mile trips each way are critical.

Hope this helps you see where I'm coming from... if you still disagree please let me know why because I dont want to dedicate lots of time to a test that is fundamentally flawed.

 

It seems like you need someome like me who tops off every 430 miles going basically the same route everyday all week in every kind of traffic from stop and go to highway? The cost of 93 would have to save me at least 2 gallons on a 30 gal fillup? Good luck with the survey! 02 5.4 reg cab 8'bed 6sp 4.10's with 3/4 ton in the bed all the time

 

Up here premium is only $0.20 more per gallon than 87, so if you fill 30 gallons its only $6 more (if my mental math is good). ANy solid distance will work, as long as both directions are the same!

 

Not sure this will help you or not......all 2009 i used 93 with my 5-star tuner and avg. 12.8 during the week pulling a trailer, and on weekends it went up to 13.2 w/ no trailer. In 2010 i changed tunes to 87 and my avg remains the exact same, pulling the same trailer traveling the same routes. The difference in 87 to 93 around here is .30 cents so it adds up quickly when filling w/93. I also noticed less power w/87 than last year running 93.

 

So you run it down 1/4 tank and alternate between 87 and 93 octane fuel? Or did you already conduct a 1K mile test with 87 octane and have moved onto 93 octane?

 

I officially started me test at 159,600 at my last oil change. I've almost always ran 87, and premium on and off prior to that. I ran 87 all the way to 161K making note of my long trips, routes, etc. and at 161K to 162K and change i've been running premium.

Aside from what appear to be economy gains, i do notice all around driveability has improved. Engine runs smoother, the seat of my pants dyno shows power gains, and the noticeable eway economy increase.

I figured if there was some indication of premium benefits I'd pursue a scientific test, now that there is some signs of awesomeness i want to get official testing underway

 

its good to know these figures but personally I could give a rats *** about gas mileage lol I didnt buy my truck for mpg. Im happy with my consistant 12-13 town mpg outta my truck (in sig)

 

All I'm saying is run the tank to near empty before you fill to improve accuracy of your measurements. Maybe you're already doing that. If not, then do so. The rationale is that you don't want 1/2 gallon error on a 10 gallon fill when you could have 1/2 gallon of error on a 30 gallon fill. Your analysis is only as good as the measurements and statistics don't fix poor data. You're trying to measure small changes in fuel economy. You need to do whatever you can to tighten the distribution of data so you can see differences and verify correlation of the "fuel type factor" in your ANOVA.

 

We'll just have to see this is the only method I can find to eliminate the other 1000 factor that effect mileage, however if you want to drive the same route on a full tank for this study I wont complain

 

Sorry, I have a diesel.

You're better off repeating the same route many times on one tank then filling after every single round trip. It's still the same route regardless, but a more accurate measure of fuel used since it is over a longer period. Would you measure a hundred feet with 100 separate 1 foot measurements and accumulate the error in each of the measurements, or break out a 100 feet tape measure and just induce error once? I don't know how else to convince you.

 

A couple of points:

Fuel quality has a large play in this kind of analysis.

Buying cheaper no-name fuel (commonly found at grocery stores) will sabotage your results before you even get off the ground. Kroger fuel always resulted in poor running and a consistently verified 2-3 MPG drop when I was burning gasoline.

Additionally, given the expansion ratio of gasoline, fuel temperature will also skew your results significantly.

One way of eliminating the "fill up variance" factor is to fuel ONLY at one particular station, at the same time of day, at the same pump. Fill only till the nozzle shuts off -- don't pump in any more. You'll have to do this for 3-4 MONTHS to get enough data to be meaningful. This will also eliminate any skewing from frequent switching between oxygenated fuel and non-oxygenated fuel.

-blaine

 

We can ask our friend who own 97 E250 with 5.4L that have over 1,202,000 miles.


He say he average 19 mpg in summer and 15-17 in winter.

Home


I am sure he would be glad to try your test and see what happened.