Mark P

First off, leaving Queens headed for CT, I achieved 37.9 mpg (per computer, which I'll verify for accuracy after ~5 tanks) for 45 miles. This included starting the car from cold overnight, moving it a few blocks, loading it with luggage, and then continuing on. That was followed by some stop and go traffic getting out of Queens, then highway travelling <70 mph, which is likely the ideal range for the Q5 TDI as 50-60 corresponds to about 1200-1400 rpms in 8th, and keeps aerodynamic drag down.




However while driving, I happened to look down and noticed the Average MPG jump from 36.7 to 37.3 and back to 36.7. At first I thought this was some screw up in the computing of average perhaps averaging in some large number (200 mpg seems to be the highest the instantaneous will show). But then after keeping an eye on it going forward, I noticed that the increments for which it would change were a minimum of about 0.6 MPG. This seems really strange, and unlike anything I've seen on other Audi/VW/Porsche's I've owned. I'm almost positive all others have shown 0.1 MPG increments for Average MPG.

Indeed, these were the only values I saw within this time of driving - it just jumped from one to the next:

35.6
36.1
36.7
37.3
37.9

I'll watch each time I drive to capture others as well.

Has this been noticed/discussed before? At first I thought it strange that if this were to happen that it wasn't being done with 0.5 increments. Then when I saw that most of those ranges were 0.6, but one was 0.5, I realized it's likely due to conversion from metric. Sure enough 0.6 mpg is 0.9656 kilometers per gallon. So then I thought it was only incremented every 1.0 km/gal and that it would get rounded up or down every so often and that would account for the 0.5 and 0.6 values. However 1.0 km/gal is 0.62 mpg, so it would round between 0.6 and 0.7 (not 0.5) so that doesn't make sense either. I also tried comparing to L/100km, but that would come to be 400, which seems like an odd value as well. So I'm at a bit of a loss here to the rationale behind this or the increment value/metric used. It seems like a step backwards. Of all the things this car is capable of, why not 0.1 mpg average increments? It feels like choosing between a poor and good bathroom scale or tire pressure meter that have 0.5 or 0.1 increments for their relative measures.

Here are those values:

Parsec

The difference between your last 2 pictures is 17 miles in 11 minutes. That's a pretty brisk pace if you're trying to save gas!

amstrad

The quantization error appears to be happening because the computer is calculating tenths of a L/100km (or whole numbers of liters per 1000km).

Here is a conversion of L/100km to mpg. The display is just truncating (not rounding) to tenths of mpg.

soonerfan

I cannot wait to hear the results of this! And I didn't have to wait long!

idale

So I guess those of us **** enough to want to have the computer be correct should be basing our adjustments on L/100km or waiting until the significance of the conversion and truncation gets small enough to not matter. Good to keep in mind for when mine comes in.

Mark P

Ahh thanks! I forgot to account for them being inverse calculations. One can't merely convert 0.6 mpg to L/100km because that represents a horrible efficiency, and not an increment. It's not like converting inches to cm for instance.

Still disappointing this is the accuracy they've reduced the trip computer down to. I'll be keeping track of it manually anyway so no real issue here, just annoying with all the other technology going on

Mark P

17.5 minutes. That's a "1:08"

uberwgn

Mark, get that fuel consumption displayed calibrated! They're typically 10% optimistic.

Mark P

I will. As noted, I need data first to do the calibration. I should complete 3 tanks by the time I'm home next week. I'll likely wait for 5 to have more datapoints. Oh and I need to get a newer VCDS interface too.

uberwgn

I've tracked my displayed vs. manual calculation over the life of my A4 and found that I can get an accuracy of +/- .3mpg around the city, but more variance on long trips. Most of my driving is in the city so I accept the fact that the highway displayed value is still a bit higher than it should be.