[Sparky5501]

For a long time I've always figured my gas milage the old fasioned way. Divide the total miles driven by the amount of gas you fill the tank up with.
After I bought my Insight I just let the car do this for me. But, I started doing this again and noticed that the Insight MPG has variences of +/- 5 MPG from my calculations.

Everytime I fill up I zero out the trip odometer. Next time I fill up I divide the total miles driven by gallons of gas poured into the tank. Rarely does it ever match the readout from the odometer display. Sometimes it's low sometimes it's high.
Is it because the computer is taking the average of fuel the engine is using instead of the amount of fuel used per mile?

 

[Hambone]

This link, to one of the info related sections of this site might shed some light on what you're talking about, if you haven't read it already:

http://www.insightcentral.net/KB/faq-dr ... l.html#fcd

I'd imagine that from tank to tank, there would be some discrepancy between a calculated figure and the figure shown on the FCD, but I guess I'm a little surprised at the amount of variance you report (not that I'm doubting your figures at all, just that I would not have expected the variance to be that large). Given the total miles driven in the car's lifetime versus the total fuel consumed though, my assumption would be that the variance would only be fractions of a mpg since you report both positive and negative variance these should balance out to a large degree. Of course, you'd have to have some great records of exactly how many total gallons of gas have been pumped into the car in order to make that sort of comparison!

 

[Sparky5501]

Thanks for the link Hambone. I forgot to look there...

It looks like the 5 MPG varience is pretty close to the 1% variation mentioned in the article. Mine were more like 2.3% but at least it sheds some light on 1% of the varience.