Ok, so from the big gas discussion in the General Maxima section, according to LvR and lightonthehill, we can't really measure gas mileage accurately at all from just dividing how many gallons we pumped in at the station divided into the # of miles we drove. Then, how exactly can we estimate our mileage? Then I see all these threads where people say I got 25MPG highway and 18MPG city....then that's just false?...

 

That's the best way to easily get a rough estimate of MPG.

Sure, there are better and harder ways, but eh.

Seriously, delete this thread. We don't need ANOTHER MPG thread here.

 

Sorry I asked, I'm new here lol

 

Thats how I calculate my MPG

 

Generally, yes I feel that the numbers listed contain a lot of error. The variation from one tank to another doing very similar driving (I do 95% of my driving just back/forth from work) can be as much as 4mpg. I often see one tank calculate a low MPG and the next tank calculate high MPG - suggesting that the final tank fill level is a significant variable.

I track fuel mileage and when I quote an MPG I'm doing it based on a number of consecutive tanks. This causes the error at the pump to get very small after, say, 5-10 tanks. I've never been nerd enough to calculate the actual error of the method, but after 5 consecutive tanks you're definitely within 1mpg. Any other method IMO is just a rough estimate.

Dave

 

Fool Proof Method:

Run the tank down to the very instant the low fuel light comes on.
Add 10 gallons.
Reset the tripometer.
Drive it.
Look at the tripometer again the instant the low fuel light comes back on.
The first 2 numbers on the tripometer is your mileage.

For example, after 10 gallons, I drove 240 miles = 24mpg.

 

I would think that would be worse than the ones mentioned above.

 

Here is how to estimate the error. Let say you did the test with N tanks and with similar driving conditions. From each tank you measured Ki mpg ( i from 1...N) . You calculated the average to be:

K0 = SUM(Ki)/N

Now the standard deviation of your measurements is:

STD = SQRT( SUM( (Ki-K0)^2 ) / (N-1) )

This gives you the confidence window. Meaning that the actual mpg lies in the window [from K0-STD, to K0+STD] with probability of 95%.

 

99grn,
Just saying it's "worse" doesn't do much for anyone. How about providing what makes you think that?

My approach is that over the course of several tanks, you'd easly be able to find an average ...and if you're tired of 10 gallon fill ups, go to 15 or more and just divide the same way using the low fuel light as your trigger to get the data.

Shilov, my head hurts now.

 

http://www.omega.com/techref/flowmetertutorial.html

read, purchase, install, measure, calculate, know mpg

 

I would agree with 99grn. Who is to say the light comes on with same amount fuel every time? Or that you get to a gas station the same time after the light comes every time?

If you are ****, fill up the same time of day. Better in the morning when it is coolest. Fill it until it clicks off then fill it very slow after that point until you see the gas. Do it the same time every time and then we are talking decimal points in differences in mileage due to fill up.

 

LOL, now that's nerdy!! I left that stuff in college physics lab, and it can stay there.

Dave

 

Haha, that's just $1500 for the flow meter and line conditioner.

 

the savings in knowing your gas mileage will make it pay for itself