Let’s say that you are in a place with a high temperature of 85 degrees F and a low of 63 degrees F on the same day.
That afternoon, you realize that you need a full tank of gas(let’s say 20 gallons), but decide to wait until later. Later on, it’s cooler outside and the gas temperature is lower. Is the temperature difference enough to change the density of the gas so that you get more of it by waiting until it’s cooler out?
Well, let’s see. Gasoline has a thermal volume expansion coefficient of 950 x 10[sup]-6[/sup]/[sup]o[/sup]C. So, if you buy a gallon of gas at 63 [sup]o[/sup]F, ( 17.2 [sup]o[/sup]C), when it reaches 85 [sup]o[/sup]F ( 29.4 [sup]o[/sup]C) it will have expanded to 1.0116 gallons, a difference of about 1%. So, yes, you could get a bit of extra gas, but it probably wouldn’t be worth it.
The density change of liquid gasoline under the relatively small pressure rise of a petrol station pump is over that small of a temperature difference is likely smaller than the measurement error on the pump itself.
The problem is, in practical terms the petrol is most likely at a relatively constant temperature, being underground in the big tank and all. So my point is it would only change temperature very slightly during filling, and thus you would end up with much less than a 1% difference.
Very true. But I like to work with the best-case scenario. 
If I have free access to ginormous storage facilities for gasoline, I should buy en masse in the winter and use it all year long. I could use all the extra money to buy a new tinfoil hat.
Thanks.
Gas station owner here! 
The gas temperature inside the tank is supposedly stable at 15C regardless of what the ambient temperature might be.
When the big truck comes to refill the tanks, the invoice I receive has two values:
(1) Liters in ambient temperature and
(2) Liters in 15C.
While I make sure that the amount of gasoline is the same as denoted under Liters in ambient temperature, I pay for the liters in 15C, because that is the temperature that the fuel will eventually reach.
When each refill is in the order of 15000 liters, you can imagine that even the slightest temperature difference has a measurable effect.
how far off are the two numbers on your invoice, if you can recall?
It depends on the ambient temperature. If it is 15C, then the two numbers are going to be the same.
I am curently 2 time zones away from my gas station, so I’ll have to quote some numbers off the top of my head, i.e. highly inaccurate 
For a typical ambient temperature of 25C and say, 7000 liters, I think the difference in 15C would be in the order of 500 liters.
That is, the fuel will eventually reduce in volume from 7000 to 6500 liters
Ooops! Make the above 50 liters, not 500 :smack:
So if it’s not because of thermal expansion, why do they tell you not to buy gas in the middle of the day when it’s hot & humid out?
Aren’t gas pumps corrected to 15C anyway? So on a cold day, if you buy 10L of gas, the amount you get will not quite fill a 10L container?
Who tells that? Someone official, or some guy at the pub? 
I don’t know…
A local airport that I took flying lessons out kept their planes out of doors. They had their airplanes fueled up every night. This was done mostly to reduce condensation in the tanks. The planes are capable of holding 50 gallons. I would usually show up at the airport at about 10:00am. On a sunny day every plane had fuel coming out of its overflow tube due to fuel expansion. I would estimate about 3 to 4 ounces per wing of wasted fuel. That equates to approximately .125% of spillage.
FWIW
Jim Mac Millan
Somewhere in between… on the news, in places where it gets really hot & humid, like Baltimore or Washington, D.C., they tell you to limit the amount of time you spend outdoors, and driving and filling up mid-day.
Does pumping gas have something to do with ground-level ozone?
Yes, gas fumes are a source of ground-level ozone. Sunlight breaks the fumes down to create ozone. Warm air in cities creates stagnant conditions that don’t allow pollutants to disperse, so that’s why you get bad ozone days when its hottest.
In addition, what little temperature variation you get underground will be out of phase with the temperature changes at the surface. It’s a rather standard intro-to-heat-transfer problem to calculate the temperature due to conduction at various depths caused by a fluctuating surface temp. It’s been awhile since I did HT so I won’t even try to scale it, but for the depth of the average storage tank, it’s quite likely the max temp at depth is delayed enough from the max temp at the surface that you’re better off buying gas in the hot part of the day when the cooling of the night before has finally affected it.
The reason that I heard for filling up your gas tank in the middle of the afternoon was a better time to do it, was because of the thermal expansion of the gasoline after it was put into your car. Many people top off their tank, and the expansion forces gas out of the vapor lock, they said.
I heard this explanation on the news, though I’m not sure if it was from a local TV news broadcast or from the radio. It sounded reasonable, though there are plenty of things that sound reasonable that are completely wrong.