[Trevor]

Also, I need to ask exactly what gets unplugged and what gets plugged when the engine is able to start. Does the engine start when the control module is unplugged and the resistor is plugged? Or, does the engine start when the resistor is unplugged and the control module is plugged? Or, does the engine start in either case? If the engine starts with either one unplugged is it true that when both are plugged then it does not start?

I am going to assume, for the time being, that the engine starts when the control module is unplugged and the resistor is plugged. I am assuming that there was a mistake at some point when it was said that the engine will start when the control module is plugged and the resistor is unplugged. Given that assumption, I would say that the resistor is probably fine and that the measurement of one ohm is probably correct. In fact, one ohm makes a lot of sense. That resistor forms a voltage divider with the pump when the pump is running. Electric motors tend to have a low resistance so the series resistor would have to be low as well. Therefore, I would conclude that the problem is the control module and not the resistor. Most likely, the final transistor in the circuit is blown (base and collector are shorted) and the control module is supplying 12 volts-ish, through the internal resistor, to the pump when it is plugged in. If that is true, the external resistor is probably getting very hot and the internal resistor in the control module is probably getting very hot.

Fly, can you either measure the voltage across the external resistor or, with a clamp-style current meter, measure the current through it? I would be curious to know how the voltage and/or current changes when the control module is plugged versus unplugged.

1. When transistors “blow” they normally go open circuit and in any event if the power transistor did short base to collector, it would have to simultaneously remain open emitter to collector, and this is in all very unlikely.

2. Furthermore should this unlikely event occur, a positive current would have to be simultaneously supplied by the small driver transistor which can pass only a negative current unless it also “blows” such that it opens to ground and also shorts to positive and in this doubly unlikely event the high current would pass, via a bias resistor capable of very low current, and it would certainly burn out.

3. There is no way the suggested substantial supply could be delivered to the negative pump input so as to form a voltage divider between it and the resistor in the event of the several unlikely possibilities taking place, without a high current becoming available. This would have to be delivered via the TCU which is able to supply only a small control signal, hence the need for a driver transistor. What is more the signal is likely to be delivered at only 5 volts and internal protection would interrupt an excessive load, otherwise the internal TCU switching transistor would be in danger in the event of a fault codition.

4. I agree that if, and only if, the problem has not been correctly described, several other scenarios are possible, including an obscure fault affecting connections or wiring. However a logical sequence of fault location must prevail and unfortunately here the mugwump is now in full flight. LOL

P.S. The word "module" as become introduced, whereas all Subaru technical literature specifically describes the control unit as a "modulator" for good reason. The pump speed is variably controlled in order to adjust fuel pressure in accordance with engine requirements using pulls width modulation. The system does not operate on the basis of simple on/off switching providing only two speeds and therefore only two fixed pressures.