Calculating Fuel used

[Jimster]

I'm trying to work how much fuel I've used up to a point in time on the track. Can anyone help me work this out?

I have the ECU interface so I log injector pulse width. The car has 4 injectors which are batched fired.

I'm using 500cc injectors.

Has anyone done anything lile this before, and can they point me in the right direction on how to calcualte this?

Thanks

Jim

[patje]

I've tried something in the analysis software and made an average fuel use. I'm driving an 4 cyl engine with 475cc/min injectors at 3.5 bar.
There is some explenation on wiki. Don't know the adress, i'll look it up.

formula = (var_2500/3600000)*28.5*2*Var_0010*60

Var_2500 = Fuel inj 1 PW on MS
Var_0010 = RPM

Then i make a summary statistic with average Fuel used.
Then select laps,sectors or runs to see the average fuel use.
I' don't know if its correct, i get about 48L/U on race pace.
My fuel tank is a 100 litres, I know i can run more than 2 hours on a tank.


Found it

http://en.wikipedia.org/wiki/Fuel_injec ... pulsewidth


I would be excellent when able to show on the dash, but i'dont think thats posible.

[osborni]

Assuming you have a fuel tank sensor wired into your dash, you could calibrate it for volume of fuel, not 1/4, 2/4, 4/4, etc of a tank.

[patje]

No, sender. Its a ATL bag tank. And i can't find a cheap level sender

[Jimster]

Thanks for that patje, I'm in the same boat as you, and have no tank sender.

So with your formula it works out an intant, and from there you work out an average?

How would we make it a running total of fuel used so far?

[patje]

So with your formula it works out an intant, and from there you work out an average?

Yes, if you put it in an graph. You can see what you use on that spot. You have to take an average, otherwise it would not add up. And the usage would be out of range. You have to make sure, you're fuelpressure is right on the money.
Otherwise the injector will inject more or less then set fixed in the formula.

Somebody wanting to refine the formula with an Fuelsensor and variable flow rate?

Running total or used so far? Thats up to somebody who can calculate a lot better then me.

I've looked at the motec site and they have this option. Don't know how it works, sombody with experience in this?

[faraday]

Not just pressure, but temperature as well.
The first thing you need to do is characterise the injectors, preferably individually, to voltage.
Some ECUs will not offer user access to apply this information anyway.
I hope all have some built in voltage compensation.

Fuel delivery is not actually proportional to pulse width. The parameters mentioned can heavily affect the delivery at the stop and start of the pulse, so the closer the injector gets to saturation, the smaller their effect.
Apologies if the links given explain all of this; I've not read them.
The saturation point needs to be understood as well.

If you follow this so far, you'll appreciate that pressure and temperature need to be monitored and input to the ECU's compensation table...

To be fair, only the best well financed teams do all this.
Patje seems to be on track suggesting an independent sensor for callibration purposes. The problem is, however, that sensor is likely to be quite inaccurate and dependent on the same parameters itself

The best you can probably do is have a fudge factor that you adjust on a track by track basis. You need to drain the tank before and after a calibration session...