I have personally found, particularly in relation to touring in France, that day-to-day mpg variations can be quite significant and suspect that this must be purely as a result of ambient conditions, if the nature of the roads changes little and it is the same fuel in the tank. I first experienced this strange variation back around 2004 to 2005, when I was touring France in an Audi A2 1.4 TDI 90. Variations of the order of three to four mpg in around 50 to 55mpg were typical. We know that ambient temperature has some ill-effects, if it rises sufficiently that the intercooler can’t do its work properly, and the charge air density drops. Wind level or direction is usually noticeable, and I don’t think this was an effect on the occasions that I am quoting. That really leaves only ambient air pressure, in that lower pressure means lower density and therefore less oxygen available to burn the fuel. But diesels run with excess air and therefore there’s generally no shortage of oxygen. Whether the air density effect operates through the MAF (mass air flow) meter to change any key operating conditions and fuel injection settings I do not know. Then there is humidity, which might just be the answer – and, as we know, there are sometimes gains to be had from water injection in both diesel and petrol engines, although they don’t seem to be exploited much.
So the day-to-day variations of mpg in one car, and on the same fuel, remain very much a mystery to me, and I do think that they could cloud the sort of mpg variations that you are quoting. But I don’t think it would be incorrect to suggest that some engines “like” some fuels, and not others. I ran a Rover 75 2.0 CDT back in 2001 to 2003, and the only fuel that it seemed not to like was Morrison’s diesel, on which it ran so badly that I thought that I had an engine problem. But a double dose of Millers sorted it out, although I have never bought Morrison’s diesel since.
You also suggest that the variations that you have noted might be related to cetane values of the fuel. This could be true, but all EC cars are produced to run on the lowest cetane number allowed by the EC specifications – CN 51 – and most are found on testing to be well above the lower limit. Here again though, in relation to engines liking one fuel and not another, I would suggest that some engines have a different optimum efficiency cetane number from others. You refer in particular to Esso vs BP, and I am presuming that you refer to standard BP diesel, and not the outrageously overpriced BP Ultimate. With regard to Esso, if you dig around you will see that they specifically advertise their super grade diesel as (simply) having double the additive package that the standard stuff has. So in many ways the whole issue is a bit fuzzy, and I wish there were more specific answers and conclusions available. But it seems that one man’s or car’s ideal fuel is not necessarily another’s. With the complexity of multi-pulse injection these days, I guess that’s no surprise. In fact I can even imagine that a Ford 1.6-litre TDCi engine and a Peugeot 1.6-litre HDi unit, which essentially are the same thing, may well not respond identically to different fuels.