idea..

[Dave J. Andruczyk]

> 
>   You would need to know more than just the pulse length, as a 1ms pulse at
> 1000 rpm and a 1ms pulse at 20000rpm are quite different in terms of fuel
> consumption. Forgetting about injector opening/close times, 1ms at 1000 rpm
> is keeping the injector closed most of the time, opening once every 60 or
> more ms. This means its delivering fuel at less than 2% of its max. flow
> rate. At 20000 rpm, the injector is opened for 1ms every 3ms, delivering
> fuel at 33% of its max flow rate. Measuring duty cycle would be the way to
> get instantaneous fuel usage, this can be calculated from pulse width and
> frequency. And yes, I'm sure it can be done quite easily, although I've
> never done it. I'm sure there's plenty of averaging circuits or duty cycle
> circuits already designed that could be used.

Thats not true.  In simplistic terms, (i.e. close to ideal...)  A
injector has a nearly constant pressure drop across it, due to fuel
pressure on one side ( assuming its constant), and manifold on the inside.
( I'm neglecting flow differences due to vacuum..)  The injector will flow
a certain amount of fuel per millisecond of open time, minus the varying
flows during opening and closing of the injector,  but once it is open it
will flow a constant rate of fuel.  All the ECU has to do is record the
total "open time" of the injectors into a register, adn do a bit of math
on it with the input from the distance sensor ( pulse wheel on the drive
axle)

Dave