Now in possession of a 2004 Toyota Avalon

[Ecky]

Originally Posted by sarguy01

Did you create a new profile for this car with Torque? Are all the settings correct? It seems very odd that she is 5mpg above the EPA highway number with the AC on in stop and go traffic without any hypermiling.

But, if that car is getting that gas mileage, good for you!

I hadn't yet set up a Torque profile, just plugged it in, drove across town, and looked at the numbers. I'll update the thread once I have it all set up and have some miles on it.

[sarguy01]

Originally Posted by Ecky

I hadn't yet set up a Torque profile, just plugged it in, drove across town, and looked at the numbers. I'll update the thread once I have it all set up and have some miles on it.

Hmmm...on my Torque app I need to switch the profile when I go from the Civic to the Mazda. I know the Torue Wiki says to input the car's data into a profile for the most accurate information.

So if the app was set up to use your other car's smaller 4 cyl engine, it could really throw things off.

[Ecky]

Originally Posted by whichy01

Hmmm...on my Torque app I need to switch the profile when I go from the Civic to the Mazda. I want to say it uses the engine size to calculate air flow, to estimate gas mileage. Don't quote me on that though...

So if the app was set up to use your other car's smaller 4 cyl engine, it could really throw things off.

That'd make sense, looks like the default engine size was 1.6L. I made sure is was in US MPG at least.

(My Honda is too old to have an OBD 2 port)

1.6 / 3.0 = 0.533, 0.533 x 32 = 17MPG, which sounds more realistic. Dunno if the calculation works that way.

[sarguy01]

Originally Posted by Ecky

That'd make sense, looks like the default engine size was 1.6L. I made sure is was in US MPG at least.

(My Honda is too old to have an OBD 2 port)

1.6 / 3.0 = 0.533, 0.533 x 32 = 17MPG, which sounds more realistic. Dunno if the calculation works that way.

Oops, I edited my post while you were posting...

I am not sure how the math is done. I know it also is based on the air/fuel ratio. Maybe the airflow*AF ratio?? Of course the units would have to be converted.

[Ecky]

Update:

Decided to take my Del Sol across the state as I was pretty certain it'd net me at least slightly better mileage. I haven't logged fuel in it in a while but I put ~500 miles on it going south with an average of about 37MPG (~62-65, behind large trucks), and ~500 miles coming back up averaging ~32 doing 72-78 without any drafting. It's incredibly difficult to get below 29MPG and incredibly hard to get more than 36, regardless of driving style.

Back home now though, and had a chance to inspect what she left with me. It's an XLS trim, factory upgraded speakers (JBL) and pretty much every other option that was available in 2004. Currently 47,000 on the clock.

Some pics of the car:







I set up Torque with the correct displacement, tank size, weight, and whatever else it asked for and did a bit of city driving today.

Trip 1
~6 miles with a cold start in town, a few redlights and stop signs but low traffic, some hills but destination was approximately on-level with starting location.






Trip 2
~9 miles, car was mostly warmed up, some more redlights and a bit more traffic.






Torque's average read about 40% higher than the car's "average economy" reading, though the instantaneous consumption as displayed by the car's LCD vs Torque was very similar (with the car's reading actually being 5-10% optimistic). Two things I noticed were that the average economy as displayed by the car didn't really rise while coasting down hills, and that the readings were very similar between the two until I took my foot off the gas, specifically in the 2nd trip when I was pulling out of Whole Foods and going up a long hill to a traffic light.

I think the proof of which is correct will have to wait until I have a fillup, which could be a while. I'm really curious about how well it'll do on the highway now.

[LeanBurn]

I usually manually calculate fuel economy for a few tanks to get a baseline, then see how it measures up to the apps and in dash read outs for error diversity.