Author Topic: A question on how the consumed gas is metered  (Read 3657 times)

Offline chenripp

  • Newbie Gearhead
  • *
  • Posts: 1
    • View Profile
A question on how the consumed gas is metered
« on: July 09, 2011, 01:15:58 AM »
One of the most attractive feature of UltraGauge is its real-time monitoring
of the fuel economy by displaying the instant MPG value in every second.

There are two ways to calculate the MPG, or more specifically, the gas that
was used during the trip, since mileage is always much easier to obtain from
ECU with good accuracy. In the simpler or more straightforward approach,
the amount of gas used is measured by reading the MAF signal and multiplying
it by 14.7, assuming an ideal stoichiometric air-fuel-ratio. In the more complicated
approach, the gas consumed was metered by monitoring the fuel injector duration
in each spray, and adding them up to obtain the total amount of gas consumed in the trip.

Could you tell which method UltraGauge used to report the instant or average MPG??
And what does MAF 2 do on this purpose?  Thanks a lot in advance.

Offline tfuk

  • Newbie Gearhead
  • *
  • Posts: 10
    • View Profile
Re: A question on how the consumed gas is metered
« Reply #1 on: July 22, 2011, 12:58:08 PM »
As I understand from the User Manual, UG uses the simple calculation MAF * 14.7 when MAF sensor is present.
Otherwise it uses an approximation - a calculation based on MAP sensor, % of load, RPM, typical torque_vs_RPM diagram for atmosphere-charged engines and some adjustment constant VE (I call it a Bulgarian constant  :) )

I suppose both methods will work very badly with my car. I have a 260hp turbo-charged engine with very flat torque_vs_RPM diagram and use a mix of gas and ethylalcohol (from 3% to 45% of ethylalcohol depending on locally available fuels). As burning 1 liter of ethylalcohol needs 40% less air than burning 1 liter of gas, any calculation based on quantity of air without knowing the percentage of ethylalcohol must be wrong.

The only method which can give accurate results in this situation is the injection-monitoring-based one. My built-in board computer must use this method as it shows reasonable values (103% of real fuel consumption) regardless of fuel mixture composition. Alas, it does not show many other interesting gauges  ;D