I have always heard that you can get better mileage out of your car by keeping the tank under half full (kinda makes sense… the car is lighter that way, right?!). Since there really is no good way of calculating actual mileage by doing that, I never have. Recently, I have topped off before a long trip and was surprised to see about an 8% increase in mileage above my average. This has happened consistently now and I have seen it two different vehicles.
Why would keeping the tank full (and therefore heavier) INCREASE my fuel economy?
If you have a 15 gallon tank and your car weights 3200 pounds, a full tank adds about 94 pounds, or 3%. If you run it from full to half full it’s 1.5% heavier than if you run it from half full to empty. If you run it from full to empty, the average weight difference is only 0.75% from the above scenarios. That’s the most you could save, and I’d guess that it’s really a bit less than that. Does it seem like much to you?
If you were previously keeping the tank less than 1/2 full, how can you now know that your mileage has increased? How were you checking mileage before?
On a long trip you will ALWAYS get better mileage, unless you drive 90 mph. As mountainbike suggests, by what you are doing you can’t possibly get a statistically accurate figure of what your mileage really is.
Generall?, Tom and Ray would say that getting excess weight out of the trunk, pumping your tires up to the proper pressure, don’t take your mother-in-law, and other things will improve your mileage.
Sanders adequately demonstrates that the weight difference would at most increase your mileage by .75%, not 8%. So, if what you say is true and your measurements accurate, then 7.25% of the increase is due to driving at a steady speed and 0.75% due to the lower weight. In stop and go driving, everytime you touch the brakes you are converting kinetic engergy of the car into HEAT. That’s where the mileage goes. Don’t use the brakes and your mileage goes up.
I wasn’t ever keeping it under 1/2 tank… I would just let it get down most of the way and fill it up to the top. That way, I can just divide my miles driven since the last full fill-up by the gallons I just put in! I have been averaging about 48 mpg doing that. But the last couple of times I filled it before hitting the 1/2-way mark and magically got to 52 mpg which is an 8.3% increase in mileage! I was just very surprised that (although the overall weight of an 11-gallon tank is not that much), I managed to do better with it fuller. Just wondering why that would be? Is it how the car is balanced? It is consistent that I do better with more fuel in the car. I just can’t figure the physics of it and it goes against all of the common fuel economy lore I know.
Any ideas?
Well… I guess I wasn’t all that clear in my initial post. I luck out in that my normal commute is almost exclusively highway and I do drive very conservatively (I manage to get 48 mpg average out of a car with an EPA rating of 34 mpg-highway). I drive 60 on the highway and I never accelerate really quickly unless I need to.
So why is it that if (when driving the exact same way as normal) I fill it up before hitting 1/2 tank, I get better mileage than if I let it get down to near empty before filling? THAT is what I want to know. 
Individual gas pumps are very different in the way they shut off. The only way to get an accurate reading is to go to the same pump and stop at the first click; don’t try to top up the tank. I have had a similar situation where I filled up in a distant city, and got fabulous mileage, since the pump there shut off too early.
If you live in a very hot part of the country, a half full tank will incur more evaporation, but should not affect your mileage too much.
All the stories of wonderfull gas mileage I’ve heard over the years nearly all dealt with the pump shutting off too early compared with the pump where the original; fillup was made.
As posted elswhere, it must be the pump you use, and the way it shuts off. Whatever it is, you need several fillups from the same pump to get a resonably accurate figure.
Commuting still takes more gas than highiway driving on a long trip.
Hmmm… yup, I can understand that. I will sometimes use different stations and, I admit, the first time I noticed this phenomenon was at a station that I had not been to before. Normally, I do use the same pump at the same station (just the one on the way, pointing toward home!) though and am still noticing more than just a fraction of an mpg increase. And with only 10.3ish usable gallons in my 11 gallon tank, I would think that a 4 mpg increase would be really noticeable on the gauge (particularly since we are talking about only putting in about 5 gallons at a time). I do also remember this happening fairly regularly in an older car that I used back in college… (I’m a geek - I’ve been calculating my mpg since getting my license) but I never really paid much attention to it then… I always just thought, “Cool! I did a lot better on that tank!”
Maybe it is just me… there is a mysterious, unexplainable fuel consumption phenomenon localized around my personal vehicles! Too bad I don’t live near the Bermuda Triangle! 
It really depends on the car. If you drive a light weight economy car, the weight of a half a tank of fuel is a larger percentage of the total weight than if you drive a large truck or SUV. A half a tank of gas (even with a larger tank), makes up a smaller percentage of the total weight of an SUV. So your method of only filling up half a tank at a time (which I use too on my car), will save a small amount of fuel. The problem is that your savings get wiped out if the price of gas goes up. Let’s say you have a choice of buying a whole tank of fuel at $4/gallon, or filling it up halfway at $4 today and at $4.10 next week. In that case, any money you saved in conserving fuel is now spent on the more expensive fuel. So your and my method only work when the price of gas is either stable or declining. If it is going up, you should just fill the whole tank. I only fill my tank halfway because it is all about conserving fuel, not necessarily conserving money.
I suspect your method of calculating fuel economy isn’t precise. The methods we use are seldom precise.
Dont you just go by the amount despensed,regardless of when the pump shut off? Isnt it simple,mileage traveled divided by the amount of gas used. You know I posted this how to figure mileage question and got back some less than helpful questions in regards to “don’t you know how to do basic math”.
48 mpg? 52 mpg?
I’m jealous. What kind of car is this? That can be a huge factor in the difference.
Well, yeah… it IS about conserving fuel. But the question is, why does it work?
I start out with a full tank, run 250 miles to when the gauge reads 1/2, then put 4.8 gallons in… so 250/4.8=52 mpg. Now, if I run 460 miles, I’m almost on empty and put in 9.6 gallons. 460/9.6=47.9 mpg. This is considering that I am driving under nearly identical conditions every day and have a tendency to use the same pump at the same station at every fill up.
I drive a sub-compact that is fairly light weight. 11 gallons full-tank is really not that much weight at just under 100 lbs so… again, why does it work?
Oh… I love my car! It is a little Honda Fit. Aptly named as you cannot even imagine the amount of stuff I have crammed in it and still managed to get at least 45 mpg!
With the fluctuations in the price of gas these days whenever I see low price I fill up. Why worry about the little stuff when there is money to be saved. I probably save more this way then trying to calculated the weight of the car/volume of the tank. When you average out your mileage for a month or two I’ll bet there is no net savings.
This is more of a curiosity for me than anything else. It goes against the common theory that the lighter the car is, the better the mileage.
I am a scientist but I look through telescopes so I don’t have more than a running knowledge of cars… I just want to know why it works, that’s all!
It doesn’t work. You and I are actually doing the opposite tactics. I only fill my fuel tank halfway. You are claiming you get better fuel economy filling the tank full each time but using less gas between fill-ups. First, if you are getting better fuel economy that way, there is no way for you to eliminate all the other variables. You have no way of knowing whether your method is affecting fuel economy at all. Second, in order for your method of calculating fuel economy to be accurate, you would have to fill at the same pump using the same nozzle each and every time, at the same time of day, with consistent temperatures. Lastly, the fuel pump gives you the amount of fuel pumped to three decimal places, but your odometer is only precise to one decimal place. Therefore, you can only calculate your miles per gallon in whole numbers. In other words, you only really know you got about 52 MPG one day and about 48 MPH the other day. You don’t know why and you don’t really seem to understand that is a very small difference. Let’s look at gallons per 10,000 miles instead of miles per gallon. When you got a measurement of 52 MPG, that means you used that fuel at a rate of 192 gallons per 10,000 miles. At 48 MPG you averaged 208 gallons per 10,000 miles. 16 gallons of fuel over 10,000 miles isn’t that much, especially when you have not eliminated all the other variables that affect fuel economy. In order to know the cause, you would need to eliminate all those other variables.
The idea that a lighter car gets better fuel economy depends on all other variables being equal. In the real world, you have no way to eliminate all the variables that affect fuel economy.
Not enough difference to worry about. You will save a little if you reduce the fuel in the tank as it will reduce total weight. However then there are down sides. You may run low with no station around that has a good price or if you run lower you might be stressing the fuel pump and they are not cheap to replace. Overall, Don’t worry Be happy!
16 gallons is still an 8.3% increase. But you have sufficiently cowed me into knowing that I have no idea what I am doing when it comes to estimating fuel economy. Cheers!