gas mileage question

I was just wondering what everyone thought was the BEST way to determine gas mileage. Here's the thread I read in the Impala forum that made me question how I calculated my mileage...

"The gas tank is divided into 2 parts, The top 1/4 is used for highway and the bottom 3/4's is used for city....:smokin: This is not an uncommon thinking on your part as it does seem to be that you get better mileage on the first 1/4 used. However the float on the gauge does not register the fuel above the bottom of the float which can be as much as three gallons in some vehicles. So when your gauge starts to register the drop, you have used that portion not measured which includes that which is in the fill tube as well. This can very with fill up, that is why when measuring mileage it is best to use the same pump and same number of fill clicks for accuracy." (57chevyragtop)

Anyone have any thoughts on this?? I was just curious because I never thought about using the same pump or worrying about the number of clicks. Thanks!

Fuel guage level sensors are not 100% accurate, as tank shape, sloshing, etc... can throw it off a bit. The best way to calculate fuel mileage is to fill the tank until it shuts off automatically at a pump with a level parking lot, record how many gallons it took. Then take the miles driven, and devide by the gallons used. As the previous poster stated, the more consistant the variables (using the same pump at the same station) will keep it the most accurate. I believe that's a little anal, but it is the MOST consistant way. A few percentage points aren't that critical, and most gas pumps these days are inspected regularly and are pretty accurate. The most important part is to fill your tank the same way each time, and don't continue pumping until the total is an even dollar, as that will skew your results.

thanks. I usually top off my tank, but now I'll stop. And I will use the same pump at the same station if possible. I'm trying to get a visual of the float as described. What does it look like? How can there be as much as 3 gallons of fuel above the bottom of the float?? Would a fishing bobber be a good comparison?? The bobber is somewhat submerged in the water so therefore there is going to be some water above it. Is that an accurate conclusion. Can you elaborate and help me make sense of this? not that it's real important though, just more information that could be useful. Thanks!

the float in the tank is on sort of a piece of wire that is connected to a variable resistor that alters resistance for the guage as the level in the tank goes down. What rich is saying is that when you fill it up, the float will only go as high as the wire its attached to will let it. So if you continue filling, the float stops as high as it can, and you continue to dump more gas right on top of it and the whole float is submerged below "3 gallons".

thanks t. Now I get it. One more question. If someone were to let the pump stop filling on its own rather than 'topping it off' would their [calculated] gas mileage improve or get worse?

It doesn't matter. Beyond the fact that there is so much misinformation in that original post, and there are numerous incorrect presumptions about how fuel level sensors operarate and how to properly fill a fuel tank, the mileage doesn't change.

If you want to know fuel mileage, divide the miles travelled by the volume of fuel used, period. It doesn't even matter whether you fill the tank. Eventually, the mileage will average out. If you are checking mileage one tankful at a time, it's going to be inaccurate regardless of what you do. You really need to log mileage and fuel consumption over a period of several refills to get accurate.

Hey GW. It's probably not the most accurate, but the way that I do it is reset the trip od. every time I fill up. I make sure the machine spits out a receipt when I am done and I write down the trip reading on the receipt. Then, do exactly as Blue Bowtie said. I take it a step further and keep an Excel spreadsheet with all the data.

It may not be the most accurate the first few times, but if you continually collect the data over a period of time, statistically, the standard deviation will decrease the more samples you take.

Microsoft has these templates (http://office.microsoft.com/en-us/templates/results.aspx?qu=milage&av=TPL000). I use one that is slightly modified. It's on my computer at home, so I'll edit and attach a copy later today.

thanks highroller. I've been recording my mileage and gallons for quite a while now. I also record the cost per gallon just for shits n giggles....to see how it's changed over time.