5.7 MPI injector question

In my last post I was advised to check the wiring on the injectors. Looking at the wiring diagrams in the Diagnostic Manual I realize I don't really understand how these things work.

I appears that all injectors have a terminal connected to a hot source when the ignition is on (same one that drives the ignition coil and fuel pump relay). Then the other terminal on the injector coil is connected to a "Fuel Injector Driver" in the Engine Control Module (ECM). The Driver will fire the injector by connecting its side of the injector coil to ground for the appropriate number of milliseconds, at the appropriate time.

My problem is that on the 5.7 the ECM only has two Drivers, each with a single wire "output". Driver "A" connects to injectors #2,#3,#5 and #8. Driver "B" connects to injectors #1,#4,#6 and #7. So it appears that the only control the ECM has over the injectors is to fire an entire injector bank of four every time one of its cylinders goes on the intake stroke. This makes little sense to me and I must be looking at something wrong. Does anyone know how this thing controls which injector gets opened?

Perhaps my assumption that there is one pulse on each intake stroke is wrong. Perhaps the injectors pulse 4 times for each cylinder power stroke (spread out evenly over the whole cycle)??

You are correct, they are "batch fired"

CaboJohn said:

Perhaps my assumption that there is one pulse on each intake stroke is wrong. Perhaps the injectors pulse 4 times for each cylinder power stroke (spread out evenly over the whole cycle)??

Click to expand...

I believe it is.....when I went thru the factory training material, the take away was each 'bank' fires every other revolution, so it takes two revolutions to fire each bank (all injectors).....that was the theory with the original TBI setups thru MEFI4.....the 4B enhancements may permit individual injector operations - didn't look it up.

Which bank is triggered is dictated by the MEFI unit (version) and the sensors supported, similar to the manner that the ECU drives the distributor timing .....

Thanks for the information. It restores my faith in my ability to read a schematic and I should now be able to trouble shoot the wiring.

It doesn't really matter to me or my problem how often each bank of injectors fires. I had assumed that if it did fire as a bank, it would do so twice each revolution in order to keep each cylinder looking at the same timing of the injector flow (vis a vis the intake valve opening). But we are dealing with milliseconds so maybe that does not really matter. Mark thinks he remembers each bank firing only every other revolution.

I did the math to see if my flow numbers from the scanner make sense. The scanner showed that the engine using the most fuel had a pulse width of 6.4 milliseconds and a flow of 15.2 gallons per hour, all at 4000 rpm. With gas at 6.3 pounds per gallon, that works out to each injector having a constant flow (if left on, not pulsing) of 14 lb. per hour if they fire twice per revolution, 28 lb. per hour if they fire once per revolution, and 56 lb. per hour if the fire only every other revolution.

Now, having to once find a substitute for a failed injector, I learned that the Crusader ones in my engine are rated to flow at 24 lb. per hour. (My substitute was out of a Chrysler, rated at 22 lb./hr. - worked fine.) The rating is done at a standard pressure of 40 psi (or close to it, I may not remember the exact number). My engine shows 55 lbs of fuel pressure at 4000 rpm so it should flow some more than that but the 28 lb number for one firing for each revolution sounds like a close match to the standard rating of 24 lb.. So I believe each injector bank is firing once per revolution in my setup. Can't speak for other years or engines.

Anyway, none of that is very useful, but it is interesting (particularly since I assumed until now that each cylinder had a singe, individual injector firing).

CaboJohn said:

...

I did the math to see if my flow numbers from the scanner make sense. The scanner showed that the engine using the most fuel had a pulse width of 6.4 milliseconds and a flow of 15.2 gallons per hour, all at 4000 rpm. With gas at 6.3 pounds per gallon, that works out to each injector having a constant flow (if left on, not pulsing) of 14 lb. per hour if they fire twice per revolution, 28 lb. per hour if they fire once per revolution, and 56 lb. per hour if the fire only every other revolution.

Click to expand...

John: the math above appears 'inverted' or the typing took the data 'backwards'....

If rating is done @ 40 PSI, a 24 lb/hr injector should flow just over 28 lbs/hr at 55 PSI....the adjustment is SQRT (Pnew/Pstandard) - just like an oil burner nozzle....you should be able to get the firing frequency for each bank with the data....just gotta watch all of those factors of two...

Here is how I did it:

4000rpm = .015 sec per revolution
All injectors fire for 6.4 milliseconds (.0064 sec), assume once per revolution
Duty cycle is .0064/.015 or 42.7%
Fuel flow is 15.2 gal/hr or 15.2x6.3 lb/gal = 95.8 lb./hr
95.8/8 injectors/.426 = 28.1 lbs/hr (single injector flow at 100% duty cycle)

So the assumption of firing once per rev agrees with the rated flow very well (in fact perfectly with your pressure adjustment). If they only fired for 6.4 ms every other revolution then flow per injector would have to double and that seems unlikely.

Of course the MEFI4 that delivered the data to the scanner only knows what is the rpm and what pulse width it sent to the injectors. It did the same calculation of fuel flow I did and they should agree perfectly (assuming the MEFI4 believes it triggers once per rpm). Actual fuel flow will depend on actual pressure and injector performance to rating, which on this system it has no way of knowing.

My observation was focused on the firing rate increasing and the fuel delivered decreasing.....

As far as the terminology goes, remember you have BOTH BANKs firing to get fuel out of all eight injector....so what I remembered may be applicable for the TBI setups but doesn't hold for the MPI's - I doubt there was any philosophical change at Delphi back then so I'd bet my recollection was flawed...