Not to mention that fuel guages have a tendancey to stay longer at the top because they are only made to go so high. Lets say the gauge is set to stop on a peg at a certain spot, you might be able to put 2 more gallons of gas in after it shows you its full so basically you could burn 2 gallons of fuel before you even see it budge.

 

Agree, adding wind; my commute to work includes 300 feet of climb, mostly rolling hills, with some very gradual inclines. MPG is typically 3-4 mpg better going back, "downhill," but a 10 mile wind can raise or lower mpg on either half by 2-5.
Lewis

 

There's always the driver psychology, too. Little bit of a placebo effect at the start of a new tank never hurts FE. I won't say I give up 1/4 or 1/2 way into a tank, but be assured I am pulling every trick humanly possible when there's only 75 or 100 miles on a tank b/c it's very pleasing to see that instant gratification of a rise in just minutes. As opposed of course to 650 miles in, when you kinda realize nothing you do - not a late-night empty-road drive of P&G nor a drag race - is really gonna alter the tank's average being as how it's almost empty. My MPG generally does start out hotter than it ends, but part of that is my doing.

 

Not really related, but... I can go 100 miles on my bike and the fuel gauge will read 3/4 tank. At 150 miles it reads 1/2 tank. And at 180 miles, it reads empty. It's as if my gas tank was shaped like a martini glass!

 

I wouldn't think you could actually measure the difference in fuel economy between a full tank and a 1/2 tank,but if you could ,I would bet on the 1/2 tank, since it would be maybe 30 lbs lighter(bit more than 1%).
But,that isn't what you are noticing, so maybe there is a bit of water in the bottom of your tank(if the fuel intake is in the bottom-probably is-them this is BS).If the intake "floats" and sucks fuel from the top of the tank,then there will be less water in the top fuel.
If the above is BS because the fuel intake sits on the bottom,then maybe when your tank is 1/2 full more water condenses from the bigger space above the fuel-more significant in cold weather.Now the tanks probably aren't supposed to allow much outside air in,but they have to allow some in.
More room for water to condense is my bet-no energy released by water-so a gasoline+ water mixture will have less energy per gallon.
Still-I would be surprised if it could be measured.Thanks.Charlie

 

I believe the computer calculates the mpg from the mileage and the fuel mass flowing through the fuel injectors (not the fuel level gauge and its sensor). Therefore, I think you have to look at what the difference is in the properties of gasoline when the tank is full and when it is empty.

One difference is the amount of air above the gasoline. When full there is little air above the gasoline and the fuel injectors get “pure” gasoline. When the tank is near empty, there is more air above the gasoline, and when driving, the sloshing gasoline can entrain and dissolve air into the gasoline. Thus, near empty, the fuel infectors get a mixture of dissolved air and gasoline, which reduces the gasoline to the injector. To satisfy the fuel demand for a given power level, the injectors will thus have to inject more “air and gasoline” near the end of the tank than at the beginning. A quick search of the internet did not find the solubility of air in gasoline, but if it is close to air in water, it would be around 3 to 6 percent (depending on temperature and pressure). The auto makers don’t worry about this since a closed-loop control is used after the car warms up, and thus the “correct” amount of fuel is always supplied for the power level demanded.


Water could also impact mpg if you are getting condensate in your tank. If you have more water (condensate) near the end of the tank, water would also contribute to reducing the mass of gasoline through the fuel injectors. Gasoline has an additive to emulsify water into the gasoline so you do not continue to accumulate water in the bottom of your gas tank over the life of your car.


I don’t believe you can measure what was discussed above by only running on the top half or the bottom of the tank since you really don’t have an accurate measure of the actual fuel consumption (i.e., the gas station’s fuel pump), which can easily “fill” the tank +/- a quarter gallon, an error which would swamp your measurement. That’s in part why you can only “accurately” measure mpg by hand calculations and over several tanks.


As an aside, “Big Oil” has fairly sophisticated equipment for measuring the amount of water and dissolved gases in petroleum when it flows through large pipelines for transfer of ownership; after all, they want to pay for petroleum, not water or gasses.

 

One of the comments from the Ford Hybrid event in the fall was that the change in FE due solely to fuel weight was on the order of a tenth of a mpg or so. 10 gallons of fuel weighs 65lbs- about 0.2mpg (@1mpg hit for each 300#, from here:https://www.greenhybrid.com/learn/ar...s-passion.html)
It would be difficult at best to detect the change in FE from that weight, given that everyday variability in vehicle weight (from passengers, groceries, etc.) is on the same scale as a fuel weight change.

Water content could affect FE...however maximum water content in gasoline before phase separation at room temp is ~0.5% by volume- and that's in E10 which can hold 25x more water than 'regular' gas. Losing that volume in gas would result in a FE reduction of ~0.15mpg at 30mpg average. Given that most of us use non-ethanol gas with a much lower possible water content, the magnitude of FE reduction due to water in the gasoline is even lower. Pure modern gasoline can only 'hold' 0.02% water by volume at room temp, less when its cold. MTBE-laced gas can hold 0.06% Even an Insight driver isn't going to be able to detect that kind of change in FE.

http://www.epa.gov/otaq/regs/fuels/rfg/waterphs.pdf

For all practical purposes the fuel we pump into our tanks nowadays is never contaminated with enough water to cause phase separation. If it were, the slug of water delivered to your engine would make the engine sputter out. Any excess water in the gas is sitting on the bottom of the station's storage tank, and their suction points are not from the bottom. If you refuel very infrequently or go through a large temp swing, you may get some condensation on a mostly empty tank (lots of air) that could make a tiny difference, but reaching phase separation is still pretty unlikely.

 

If you do a lot of solo driving up and down hills that don't have a lot of traffic lights at the bottom (especially if they have a high "RED" duty cycle), then driving with a full tank will actually save a measurable amount of gas. The proof I had was a trip I made back in the '70's with a V8 Buick Sportwagon, getting a 30mpg average. Another thing to remember is that even with the most aerodynamic new cars wind resistance results in peak efficiency at between 40 and 45 mph, and THIS IS WINDSPEED, NOT LANDSPEED, so add the speed of any head wind to estimate your windspeed. Also remember that this does not mean that you should accelerate as fast as possible to between 40 and 45 mph windspeed, but to be gentle with your gas pedal and with your brake - even when driving the latest hybrid!

 

Thats hooy. A modern electirc fuel pump is going to make x psi. They make the psi thats required no matter what the static suction pressure is. Vapor expansion plays No role nor does static psi in FE. Would have helped a mech. old style pump.. maybe. But even then carb jets are the final metering device as are injectors in modern fuel metering devices. Then you are looking at psi drop across said metering device as the final point of flow control.

A often over looked item is the low fuel warning light. You think thats there for us as drivers? Well maybe... But over all its there to protect the pump motor. Running the last drop of fuel out of a car plays hell with the pump. A modern auto fuel pump is a very complex sump pump. Sump pumps require that there cooling be from the fluid there pumping. Most low fuel warnings come on at about the time the pump motor is exposed to air. Air is not a good cooling conductor. Not to mention the cavation and near dry states of pumping the fuel pump has to deal with as one slushes fuel around in the tank. Ever ran your tank down to below E on the gas gauge just to open the filler cap and get hit with a large gas vapor rush. thats because the pump motor is injecting heat and causeing great vapor expansion. At a half a tank it might well be a whisper. Empty its a rush.

Im not a real greenie. Heck I didnt buy a Civic hybrid because of CVT and no rear disk brakes. But I dont over fill my tanks and I dont run gas gauge to near bone dry empty.

Bottom line is tank fill state has nothing to do with mpg/fe. The pump and metering devices dont care whats in the tank. And with a tank bone dry empty or spends alot of time at empty its life will be short. Full it will cool itself and be ready to pump more fuel for you......

 

There, their, they're. Loose, lose.

(Sorry -- pet peeves.... no disagreement with intent, tho)

Don