ECU pulse width calculations...?

[Andrew Brownsword]

> Date: Sun, 12 Dec 1999 12:07:47 -0800
> From: "John Dammeyer" <johnd at autoartisans.com>
> Subject: Re: ECU pulse width calculations...?
>
> You haven't mentioned what the actual pulse widths were with the old injectors
> and then after you scaled them.  The Open/close time of injectors remains
> somewhat fixed and if the pulsewidth is made too short the injectors operate
> erratically.  So if the pulse width with the old injectors was, say, 2.5ms and
> you change it to 1.7ms by your scaling algo. then you just barely get the
>pintle
> off the seat before it starts to close again.  This causes erratic mixture as
> sometimes it opens more than other times.  Been there done that.

I'm fairly certain that this is not the case -- the engine is running
uniformly too rich.  It may become a problem at high RPM, but with the
turbocharger I doubt it.  The injectors are less than 50% higher flow rate
than stock, so I don't think there will be a problem.  There's another Probe
owner who is planning to go with 550cc/min... I think he is going to run
into the issue you're pointing out!

My current theory is that my O2 sensors are carbon fouled and the ECU thinks
it is running lean and thus richens up the mixture.  When I cut the values
down far enough it goes beyond the ability of the ECU to compensate.  We're
going to test the O2 sensors next, and see if we can diagnostic information
out of the car's ECU.

> It's for this very reason that picking an injector for an engine is so touchy.
> Big enough for gobs of power at high end and you dispense, with those
> injectors, too much fuel at idle.
> It's for this reason that the concept of staged
> injection is used by some performance enthusiests.  Small injectors for idle
> and midrange and large injectors for high end.  This is much like a two barrel
> carburetor.

Unfortunately my application has very little space for the extra hardware.

> One alternative,  is to use a throttle body injector for idle and cold start
> enhancement and use the multiple injectors for high end.  The throttle body
> injector has to supply all cylinders and so has a much better chance of never
> being asked to be open less than a minimum time.  The new problem introduced
> with that is then you need carb heat and an ability to make a smooth
> transition between the two injector set.

Considered this as well, but it is a V-configuration engine with a
convoluted pair of intake manifolds, so even fuel distribution would be a
big problem.