Refill a empty tank or half full

[ST2008]

Hello

Does someone compare the mpg with refilling the gas tank half full and almost empty ?

I ask this question because I have way better mpg if I refill the gas tank when it is almost empty.

half full: 35mpg
almost empty: 44mpg

The difference between the two mpg is too high that I doubt this is because of gas weight.

Btw, my car is a Honda Civic 2000, automatic with 45L gas tank.

[Xist]

45L of gas only weigh 33.7KG; I do not know what could make such a large difference, but the only time I refill at half a tank is when I am about to have a road trip.

[cowmeat]

I tried filling at half a tank before a couple of times in my Insight, but the mpg was all over the board versus the readout of my FCD so I went back to filling at empty.

Normally I try to fill up after I break 700 miles for the tank, which puts me somewhere between 9.25 and 9.75 gallons for the fill up with one bar left on the gas gauge. I fill at the same gas station, the same pump, at the same time of morning and fill to the second click, which takes away a lot of the variables you get using different gas stations, pumps, and filling at different times of day

[JRMichler]

I get exactly zero difference in mileage if I fill at 3/4 tank or 1/4 tank. I've never gone below 1/4 tank. I also fill at a station where I can glide down to 15-20 MPH with minimal interference to other people. And that station is 39 miles from home, so I can top off the tank without concerns about heat expansion causing overflow.

[redpoint5]

The weight of the fuel can't make a perceptible difference in MPG. Your method of calculating MPGs is either not accurate, or other variables are at play. I can have 1,000 pounds in the back of my truck and not discern a difference in MPG.

The 3 biggest things that affect my numbers are speed, amount of brake use, and temperature.

[ST2008]

I suspect this has to do with fuel pressure change but I'm not sure how to diagnose this.

Edit: I also want to add this:

I always compare my mpg with the same paths and the same way of driving.

[niky]

The only way to really figure out if there is a difference is by doing an A-B-A-B-A-B-A-B test.

There should be no difference between the two.

[ST2008]

Not sure if you're serious or sarcastic. If you're really serious, could you explain more about your ABABAB test ?

For myself, I already verified several times with half empty and almost empty tank. The almost empty tank has always much better mpg.

[JRMichler]

Different people measure their gas mileage differently. Only one method is accurate. It is as follows:

1) Fill the tank to a specific point Every.Single.Time. Some use the point where the pump shuts off, others use the second pump shutoff, others dribble in until the gas level is up in the filler pipe. Any method is correct, but the exact same method must be used on all fillups. It's best to fill at the same pump in the same gas station so the vehicle is always in the same position while filling.

2) Record amount of gas and trip odometer reading. Reset the trip odometer.

3) Repeat.

There will still be variation in how full you fill the tank, so you need to average over several fills to get a reliable mileage.

Many vehicles have a gas mileage readout. Some are more accurate than others.

[ST2008]

It's what I did every time. Same gas station, same pump, two clicks, same path, same speeds and etc. I always compare only with freeway trips because the speed is almost always constant.