Accurate MPG Readings

you need to try to fill up more consistantly.

anc what kind of accuracy do you want??

there is less than 1% difference bet the gas pump and the computer...

if i were you, i would discard the gas pumps as they are less likely to be accurate.

 

what most of us do is enter all the data in a spread sheet and look at the long term.
53.7 from your 3 average tanks
total mileage, fuel used 53.42
MFD say's 5.4, pretty close.
normal for the car. Three tanks tells you little about the gas mileage of the car. 15 tanks will reveal more, but I'd suspect the pump kick off on the second tank.

 

I try to use the same filling station as much as possible, but of course this won't always work. If I have to fill my car up at an 'unknown' filling station, I choose a spot that's level (in most cases this is not that hard in the Netherlands ) and be careful to slowly fill up the tank until the level of petrol almost reaches the top of the 'filler tube' (if that's the correct term).

The biggest difference between calculated and MFD was caused by the MFD switching from 4.7 liter/100 km to 4.8 liter/100 km just when I arrived at the filling station.

 

I suggest that you stop calculating mpgs the old-fashioned way. The Prius calculates mpgs with great accuracy, so what's the point of doing it manually? :roll:

 

Agreed with all the other posters. I am currently recording both displayed and calculated mpg figures, and over the last three months, I've found the monthly figures (generally 4 or 5 fills) to be within 0.5mpg, whilst individual tanks may differ by as much as 5mpg. Which leads me to believe that the onboard computer does an accurate job and that consistent filling is next to impossible.

My lifetime figure is calculated from input fuel only (not also from display). Perhaps I should use both for that, too?

 

It's pretty simple in terms of mathematics, as everyone has mentioned here - it makes no sense to track tank by tank data from hand calculations. However, by tracking that you can come up with an average, which over time will closely figure to your actual travelled mileage.

Each gas pump is unique & always a variable. IOW, the pumps shut off differently at different stations (or pumps even, or with different pressures even), the flow rate can be different, and so on. No pump knows exactly how much gas was consumed on your last tank, therefore it doesn't know to fill to that many gallons - it's arbitrary. On your first tank, your car could've easily consumed 7 gallons, or maybe only 5, but really only the car and computer know that.

On my last tank, I went 100 miles over the warning to add fuel. The only reason I did this is because I knew I filled 11+ gallons on the previous stop to the gas station, so I knew for a fact there were at least 11 gallons in the tank. If I'm getting say 54 MPG and the light starts blinking at 440miles, I know I can easily make it to 540 miles with still having 1 gallon of leeway, in case the computer is off.

Of course, many people aren't looking to push the car and gas tank like I am, and that's fine! But, since the gas gauge is a little quirky in this car, especially because of the bladder, I do think it is important to learn what you can expect from the computer mileage, your simple calculation mileage, and so on.

As I said, for compairing one tank to the next, the simple calculation can be greatly off and shouldn't be used for that sort of thing - however, in the long run average, you'll see it will quite closely mirror your car's data.

-m.

 

the only reason i track mileage by pump data is to keep track of which pumps give me favorable fills.

 

You have discovered the major blunder in the design of the Prius. It doesn't matter which pump you use, the "bladder" prevents consistent and repeatable fillups. My solution defeats the whole purpose of the bladder, but that's a design flaw beyond my control. It is just silly to not top off when I KNOW that the tank is 2 gallons shy of being full.

Since the design of the tank prevents repeatable results, and since the displayed MPG is also inacurate, the only reliable way I have found is by mileage modified by the indicated MPG, and topping off every time until the tank overflows.

I drive an 03 prius which I have had for 20 months now and there are around 32000 miles on the odometer. The fuel guage is useless. I still don't know how the designers intended it to work, but it certainly isn't linear. There is no excuse for that, since the consumption is monitored constantly and the indicated MPG is consistently around 5% high as opposed to computed MPG, consistently topping off. If the cumulative MPG were used to drive the guage, it could be a) linear and B) a lot more accurate.

I have been searching for an alternative, but I fear it's a fatal design flaw. It is also dangerous, as I posted on another thread.

 

In thirty seven years of driving (gee, I just dated myself) and eight cars, American, European and Japanese, I have never owned one that had a linear reading gas guage.

 

neither have i.

i have also never owned a vehicle that i didnt learn the eccentricities of the gas gauge after a few weeks (or months in the case of the Prius and its extended range)

 

In other words, it is ACCURATE. Being consistent means you can trust it to be RELIABLE.

Simply subtract that MPG percentage from what you see.

The variance over the 171 tanks (1,318 gallons) with my 2001 Prius was an overall 2.1 MPG difference.

The variance over the 56 tanks (454 gallons) with my 2004 Prius has been an overall 1.4 MPG difference.

 

the more statistics you have the more accurate average you get.






[hr:2b3247a4ec]
auto.1sek.com