Where does fuel go?
#1
Where does fuel go?
Hi,
Ever since I replaced the Walbro GSL392 fuel pump with a Bosch 044, fuel consumption has increased by 20% in exactly the same low load running conditions.
I have gone extensively through AFR logs and made sure that AFR has remained the same for the same load conditions as before the pump replacement. I replaced all fuel hoses before and after the fuel filter/pump as I replaced the pump and I have not been able to spot any fuel leak anywhere along the fuel circuit.
Any clue on where the additional amount of fuel disappears?
Ever since I replaced the Walbro GSL392 fuel pump with a Bosch 044, fuel consumption has increased by 20% in exactly the same low load running conditions.
I have gone extensively through AFR logs and made sure that AFR has remained the same for the same load conditions as before the pump replacement. I replaced all fuel hoses before and after the fuel filter/pump as I replaced the pump and I have not been able to spot any fuel leak anywhere along the fuel circuit.
Any clue on where the additional amount of fuel disappears?
#4
Converted to US gallons, I used to get 23.5 mpg with the Walbro against 19.6 mpg now with the Bosch, and that is driving like a grandma with zero boost.
If we consider that I replaced the pump about a year ago and drive mostly all year long, I would say that the climatic conditions have had no influence when averaged out.
A local friend noticed a comparable increase in fuel consumption when he "upgraded" from the stock 951 fuel pump to the 964 turbo fuel pump, although not by quite as much as 20%.
I obviously had to reduce enrichment during the warm-up phase (open loop lambda), but even when warmed up and running in closed loop lambda the engine drinks more than it did with the previous fuel pump, although this does not reflect in the stoechiometric AFR at constant low loads observed in real time and logged.
If we consider that I replaced the pump about a year ago and drive mostly all year long, I would say that the climatic conditions have had no influence when averaged out.
A local friend noticed a comparable increase in fuel consumption when he "upgraded" from the stock 951 fuel pump to the 964 turbo fuel pump, although not by quite as much as 20%.
I obviously had to reduce enrichment during the warm-up phase (open loop lambda), but even when warmed up and running in closed loop lambda the engine drinks more than it did with the previous fuel pump, although this does not reflect in the stoechiometric AFR at constant low loads observed in real time and logged.
#5
Nordschleife Master
Any chance you have a pin hole leak somewhere? Try jumping the pump so it runs constantly and inspecting everything if you haven't already. I had a Mercedes E320 that leaked at the fuel filter after i replaced it, a very fine stream from a pinhole (i neglected to replace copper sealing ring). Only showed up when engine was running.
#6
Rennlist Member
Hmmm.... I'd maybe put the old pump back in and check the mileage just to be absolutely sure the pump is the culprit (and not a change in local fuel formulation, something else that happened around that time, an aging O2 sensor, etc.). And I'd probably check the fuel pressure with both pumps (and flow through the injectors if you can rig something up). The higher flow of the new pump might have induced a leak somewhere, but unless you park it in a windstorm everyday, you probably would have smelled that by now, I'd think. From hours of smog tinkering, I can tell you that all closed loop, 14.7 AFR tunes are not equal. If the new pump added pressure, your closed loop AFR may look the same thanks to the closed loop system -- always returning to 14.7 eventually -- but the amount of time above or below that target will depend on the base tune (and therefore, in part, the fuel pressure). Is the car richer at WOT with the new pump? Two cars might both dither around the 14.7 target, but the one with the lean base tune will burn less gas because it will spend more time on the lean side of 14.7, whereas a car with a rich base tune will spend more time below 14.7. The results on the smog emissions can be significant, though I've never tried to see if that translates into noticeable MPG variations. Just a thought.
#7
Addict
Rennlist Member
Rennlist Member
Try running the tune in open loop all the time. This will add to Tom's point of the correction to Stoich being on the richer side of things.
I like to establish the complete fuel table in open loop anyway so that once closed loop is turned on, the amount of correction is minimal and fast.
I like to establish the complete fuel table in open loop anyway so that once closed loop is turned on, the amount of correction is minimal and fast.
Trending Topics
#8
Rennlist Member
You'd think, though, that even if a more powerful pump is capable of more flow and pressure, the fuel pressure regulator would keep the system at the pressure intended. More fuel will just go back the return line with the stronger pump.
#10
Rennlist Member
You'd think. It was just a theory and why I was suggesting testing the pressure with both pumps. It's possible the big pump is outflowing the ability of the FPR or, worse, outflowing the capacity of the relatively tiny return lines back to the tank (think clogged drain). I use mostly stock fuel lines with an 044 pump just fine, so doubt that's an issue, but of course I tuned my car this way, and didn't compare pressure before and after changing the pump, so it's possible I tuned around a slight bump in pressure. Just brainstorming things to consider though...
#11
Thanks for the replies.
Monitoring fuel pressure in real-time is not possible at the moment and it's perhaps too early to clearly expand on what I have experienced altogether by dialing-in the fuel map, but your inputs lead me to think that I may need to disable acceleration enrichment for a while as even at its lowest sensivity it seems to affect the AFR more than it should in the most modest [load ; rpm] windows, and possibly, or at least partly, explain why the engine is burning that much of an excess of fuel over long slow-pace drives, even if it is not that obvious when looking at the variations in AFR when Closed Loop is enabled. Tuning in Open Loop in my experience does not work that well when loading the engine with different gears (!).
I have a long drive planned over the weekend and will get back with further details.
Could a higher-flowing pump create enough additional pressure that the FPR may not be able to bypass quick enough, thus need some compensation by lowering acceleration enrichment?
Monitoring fuel pressure in real-time is not possible at the moment and it's perhaps too early to clearly expand on what I have experienced altogether by dialing-in the fuel map, but your inputs lead me to think that I may need to disable acceleration enrichment for a while as even at its lowest sensivity it seems to affect the AFR more than it should in the most modest [load ; rpm] windows, and possibly, or at least partly, explain why the engine is burning that much of an excess of fuel over long slow-pace drives, even if it is not that obvious when looking at the variations in AFR when Closed Loop is enabled. Tuning in Open Loop in my experience does not work that well when loading the engine with different gears (!).
I have a long drive planned over the weekend and will get back with further details.
Could a higher-flowing pump create enough additional pressure that the FPR may not be able to bypass quick enough, thus need some compensation by lowering acceleration enrichment?
#12
Addict
Rennlist Member
Rennlist Member
What ecu are you using and how does it work in closed loop? 0-5v or can?
#14
Rennlist Member
How is your acceleration enrichment configured? If you wanted to tune around it, I would think you would need to look at RPMdot (first time derivative of RPM). On my ECU, acceleration enrichment looks at MAPdot and/or TPSdot. Although MAPdot might actually work ok as well since the FPR is MAP referenced.
#15
It looks at MAP, TPS, RPM, ECT and IAT, with MAP being the main source.
It has always worked well but I have reached the point where I am splitting hair on low load efficiency on top of this issue with excessive fuel consumption.
It has always worked well but I have reached the point where I am splitting hair on low load efficiency on top of this issue with excessive fuel consumption.