So the only way to be sure of getting the same level of fuel, without using a jack, is to only fill at the same station and same pump.
Yeah. I've been trying this even if the station/pump is not at the ideal point of my commute loop. Now most of the MPG variable is based on if I break out of the commute loop and drive time temperatures.
Just filling to the brim isn't a guarantee of getting the same amount of fuel in. Wayne Gerdes of CleanMPG uses a jack to tilt the cars in order to make sure he lets all the air out of the tank. He got 10 gallons over the F150 rated tank size in.
I've been with Wayne on drives. Being 6'4" and 240 lbs I always seem to be bouncing on the back bumpers to get more fuel in.
The computer MPG on my car has been 7.5% optimistic for my entire ownership (5500 miles so far) but I have seen some 'blips' as high as 12% out. I put this down to gremlins in the recording system. My MPG via fuelly is pretty consistent, so I'm happy its not more serious than a gremlin. I'm keeping tabs on the average discrepancy as it is possible to enter a correction factor into my computer, to make it read closer to real measured figures.
Well I ran the tank down as far as I dare go, I squeezed the most fuel in I've ever got and the dash MPG was about 2-3 Mpg higher than the fuelly calculations. Safe to say then that it's working fine, and the discrepancy was probably due to the fact that I used a backstreet filling station with a pump that maybe hadn't been calibrated in a while? Who knows, but at least I know the cars calculations are working ok, the 2-3 mpg optimism being perfectly normal of course.
Draigflag: I don't expect the in-dash fuel-economy meter to be accurate, and mine isn't. My Audi Q5 3.0 TDI's display is very optimistic. According to a close friend of mine who is a retired Naval Aviator, test pilot, and rocket scientist (no joke), it's costly to build and implement an accurate fuel flow meter, so car manufacturers don't. They just give you a "pert near" reading as guide. Even though it displays precision to 1/10 of a liter per 100 km or 1/10 MPG, it is certainly far, far less accurate.
Although I have no way of knowing what's really happening on your end for sure, I believe the similarity between your computer read-out's MPG and your actual MPG was likely a pleasant coincidence while it coincided.
Scangauge only has tools at its disposal which are already present on the vehicle. There are no fuel flow meters on cars, and all MPG meters only use the pulse width from the car's ECU to the fuel injectors as a measure of fuel flow. So the scangauge and all other MPG meters (in production vehicles) are limited by the accuracy of this signal.
luckypants: Can you tell me where I can get VAGCOM cable and required software? I'd like to get into these VAGCOM mods.
The cables are available on Ebay, but not all are fully connected, so I think it is buyer beware. The Ross-Tech software is best bought as an official copy otherwise you might find all the menus in German.Ross-Tech: Home. This software is often referred to as VCDS, so another good term to use when searching.
I have found the SEAT forum very helpful in this regard, particularly for the MK3 Leon which has the same infotainment system as current Audis. You might like to search on SEAT Cupra.net - SEAT Forum for some help. I have come across some good Audi and Golf forums when researching the same sort of issue.
I do not have the VAG-Com software, I paid one one for the SEAT forum members who has the software and cable to make my mods.
Here is a video demonstrating making the change to the MPG display of the trip computer. https://www.youtube.com/watch?v=RY8nfw6CTgI to show how it can be done. I'll be getting my Leon corrected to be closer to my measured fuelly average soon. (my car shows a long term average of 61.7 UK mpg, fuelly says 56.7, needs an 8% correction or thereabouts!)