Some positive diagnosis information:
With the fuel pump connector disconnected, and the fuel pump being driven directly by a variable voltage power supply, the engine runs for several minutes without stumbling, slowing or stalling.
I varied the voltage from 10.6 V down to just under 6 V to see if there were any issues, and there were none.
The implication (to me at least, please check my logic) is that if it runs with the pump driven directly, the ignition circuit is fine, and the issue is somewhere in the fuel pump circuit, upstream of the pump itself (Fuel Pump Control ECU, wiring, ECM).
The wiring doesn't seem suspect since it works reliably for a minute or so at a time, and every time.
I would like to test the FPC ECU. It has the following wires connected (numbers match terminals in wiring diagram):
- Black/white - power to the fuel pump
- White/black - ground to the fuel pump
- White/black - ground to chassis
- None
- White - signal wire from ECM
- Black - power from battery (via EFI Relay no 2, which is in turn triggered by the A/F relay)
So I can test some stuff while running with a multimeter by back probing them.
- Black/white - power to the fuel pump - does it lose power prior to or at the stall? If it does, why? Problem in FPC ECU or upstream?
- White/black - ground to the fuel pump - is ground solid and not lost prior to or during the stall (I expect this one to be fine)
- White/black - ground to chassis - is ground to chassis solid (if so, this shouldn't change. I expect this one to be fine)
- None
- White - signal wire from ECM - ??? - monitor to see if the pump signal is lost or continues during the stall?? What should I see?
- Black - power from battery (via EFI Relay no 2, which is in turn triggered by the A/F relay) - monitor during the stall to see if power is ever lost (if it is, maybe I have an issue with the relays that lead to it)
The problem with monitoring some of these is that I don't know if I am seeing a symptom or a cause.