Fuel Consumption varies on full/empty tank?
Fuel Consumption varies on full/empty tank?
Author
Discussion

hutchingsp

Original Poster:

67,893 posts

240 months

Friday 24th November 2006
quotequote all
Got a new 987 and noticed that if I keep the fuel tank pretty much full (i.e. top it up when it's half-empty) I get very slightly better fuel consumption according to the computer (which does seem to be fairly accurate) than if I let the tank run low before topping up.

I wondered if anyone had any idea why this would be?

Not a showstopper, just something I've noticed.

GreenV8S

31,024 posts

314 months

Friday 24th November 2006
quotequote all
Perhaps it's something wierd like the fuel being hotter / less dense when the tank is empty and the fuel flow sensor isn't compensating? If you log how much fuel you're actually putting in and compare this to the mileage, you might be able to see how acurate the computer readout is.

porsche4life

1,164 posts

255 months

Friday 24th November 2006
quotequote all
thats weird, would have thought the other way because of the weight.

hutchingsp

Original Poster:

67,893 posts

240 months

Saturday 25th November 2006
quotequote all
porsche4life said:
thats weird, would have thought the other way because of the weight.



So would I. We're only talking a few tenths of a MPG but it seems fairly consistent.

willibetz

694 posts

252 months

Saturday 25th November 2006
quotequote all
I suspect the range computer is non-linear in its estimations.

While fuel density does vary significantly with temperature, the variation is compensated for by the lambda feedback loop and any problems with this would lead to poor running and warning lights.

Vesuvius 996

35,829 posts

301 months

Saturday 25th November 2006
quotequote all


My 996 fuel gauge drops at a consistent rate until the lat quarter tank, then it seems to drop like a stone, to the extent that the first day I had it I nearly ran out!

GreenV8S

31,024 posts

314 months

Saturday 25th November 2006
quotequote all
willibetz said:
I suspect the range computer is non-linear in its estimations.

While fuel density does vary significantly with temperature, the variation is compensated for by the lambda feedback loop and any problems with this would lead to poor running and warning lights.



Yes the engine management system will compensate for fuel density variations, but the trip computer may not. I don't know how it measures fuel consumption, but it may be from a volume flow sensor, which would probably NOT allow for density changes.

willibetz

694 posts

252 months

Saturday 25th November 2006
quotequote all
Thanks, I misunderstood.


Edited by willibetz on Saturday 25th November 17:33