Won't start - electrical woes

[eburrows]

Hi all, I've got a 2000 D2 that only starts every once in a while, and the problem seems to be electrical.

Every so often, it does start, and runs just fine. However, most of the time, especially after having just run successfully for a while, it refuses to start. Just turns over, no sputter, just crank, crank, crank.

A fuel pressure gauge (connected to the test port, annoyingly located on the back of the intake manifold) shows no pressure.

If I manually supply power to the fuel pump, by shorting between the input and output of the fuel pump relay, I get fuel pressure, but still won't start.

In the "on" ignition switch position, the fuel pump relay sees 12v on the input and on the signal in, but a continuity test shows it has not engaged.

The ground side of the relay is supplied by the ECM. I suspect the ECM is not providing proper ground to the relay (and probably others, hence no start), but other than buying a new ECM, I don't know how to test this hypothesis.

Any ideas? Am I missing something obvious? (I hope)

Thanks,
-Erik

 

[eburrows]

The shop manual has this to say about how the fuel pump relay is operated by the ECM:

"When the ignition is switched on and the engine is cranked, the fuel pump relay is activated by the ECM, allowing the fuel system to be pressurised to 3.5 bar (52 lbf.in2). The ECM then deactivates the relay until the engine has started."

When the engine is working, I see the fuel pump relay ground pin work just this way. It kicks on (0 ohms to ground) for just a second after switching the ignition to "on", then turns off. (fuel system is pressurised) When the engine starts up, the fuel pump relay is grounded again to keep it running.

When the engine is not working, I never see the fuel pump relay ground operate.

However, if I short the fuel pump relay to manually turn on the fuel pump, the engine still does not start. I think that for some reason, the ECM is refusing to run. (fuel pump, fuel injectors maybe, spark coils maybe...)


The working/not working behavior does seem to be controlled by under-hood temperature. If I drive it around for a while, shut it off, then start right back up, it works fine. If I let it sit for 10 minutes after getting a workout, it stops working (engine space heats up because fan isn't running) until it cools off.

Maybe it's a bad connection to, or within the fuse box?

 

[eburrows]

Looks like I fixed it, by replacing the crankshaft position sensor, $117 from LR with time for shipping, or $60 at autozone. (Plus about four hours of labor, since I had to drop the exhaust to get at a stuck bolt on the heat shield...)

I'm sure the ECM could have told me the problem, if I had testbook.