The Gen 3's avg FCD is a tad optimistic compared to the Gen 2 and even moreso for those of us using L/100km compared to those using mpg (L/100km only displays to 1 decimal place). We also know that the mpg tests are done with straight fuel to ensure that all vehicles are tested in the same manner. Petrol stations dispense fuel with various concentration of ethanol blended into the fuel (from 0-10% and even E10 fuel is sometimes between 8.5-9.5%). Additionally, cold weather and winter blend fuel (more additives, less fuel by content) seem to affect the computer's calculation even further. Plotting the difference between what I calculate by hand and what the Prius tells me reveals a surprisingly consistent trend between summer and winter display differences. In addition, there's a very small difference between E10 and straight fuel. It's about 0.1L/100km difference in calculated vs. aFE in favour of the calculated FE. I don't know if I'm actually getting better mileage since I don't know if the computer display accounts for it. (In other words, do I look at all the other summer numbers with the same aFE? If so, then the calculated is lower than those. If I look at aFE numbers that are 0.1L/100km better, then the calculated is slightly worse, about 0.02L/100km). Anyway, I'm on my 3rd tank of straight fuel (which, around here means running more expensive 91 octane as 87 runs up to E10, 89 up to E5 and 91 has no ethanol). Anyway, if you look at the graph, the winter numbers are 0.6L/100km worse than displayed and the summer ones are 0.2L/100km worse. The calculated has only beaten the aFE 3 times and they're all in the summer.
Seems unusual that the 91 E0 is not better MPG than 87 E10. But MPG relates to energy content which is not a spec. on the pump. If I lived there, I'd be getting a few samples in my gaso can to measure weight to get density which relates to energy content. Where I live, there is no energy content difference nor ethanol difference between grades, so I don't have any fun checking anything. I assume you do not have the EPA reformulated gasoline like we do many places, so its more interesting variations where you are.
I've never seen calculated fuel economy better than dash display. I do see a very regular and consistant offset, with 99 fill ups the variation's at 7.5%. Seem to me Toyota's pushed it to the credibility limit, anymore and it'd be laughable. Here's my last couple of years:
When I changed tires, I selected a set whose revs/mile matched the ~5% error seen before. Since then the MPG indicated and calculated has been less than 1%. Bob Wilson
I think the more important point that it is not an ERROR but a BIAS. Since it is always positive and about 5% (for me is currently 4.7% for the full life of the car)
