Not sure if this is supposed to happen, but when I drive hard and the average mpgs go down and I stop and turn the car off and restart it the fuel gauge drops significantly. The opposite happens when I go from city to highway and my average mpg goes up the gauge goes up. I thought the fuel gauge was a function of actual volume not an estimate based on MPG? Does anyone else have this problem or is it something I should have the dealer look at? Thanks

 

What gauge are you talking about? The gradations on the bar that show whether you have a 1/2 or 1/4 tank left? Or the trip computer setting for "miles to empty"?

If it's the former then get your dealer to check as you may have a sender problem.

If it's the latter then the computer calculates, based on how you have been driving, how many more miles you have until the tank is empty. All trip computers work this way so if it's the latter then you don't have a problem.

 

its the bar, the actual fuel gauge

 

The average fuel consumption number is calculated from the fuel tank sender and the ECU.

If the sender is inaccurate then so will the average fuel consumption number be.

The average fuel consumption number reacts to the same input as the fuel gauge readout in the base of the instrument panel.

How are you so sure this is actually happening?

The fuel gauge sender will react to changes in vehicle attitude when parked, e.g. on hills and so on.

 

Confirming that you are just talking about the bar graph itself and not a number then there's something wrong. Assuming that the car does not move between switched off then switched back on, then you may have a sticking sensor.

 

I have noticed that when fuel is getting below the last one third of the tank that the fuel gauge reading when I leave the car in the evening and the reading in the morning tend to be noticeably different (I even started to note the readings to make sure I was not seeing things!!). It is usually lower in the morning (it is never higher!!) and I put this down to temperature changes (fuel changes its volume by about 0.5 % per 5 deg C change and is why gas/petrol pumps in Canada, for example, use temperature compensation, but not in the US for some reason....), although the observed discrepancies seemed somewhat higher than this would account for. Note that I live in Texas where temperature swings are quite high.