Intermittent pump engagement

[Keller]

I was at the track this past Tuesday. Was ~10th in line to race in my lane. They started moving the line, so I cranked the car to move up. *crank* No start. Checked, and everything was plugged in. Eventually all the cranking started dragging the battery down (damned Optima) so I had some people push me back to the pits.

So I started checking the car out. The pump wasn't even doing key-on priming. One of the wires at the pump harness seemed to have battery voltage during key-on, but I couldn't recall which one. And the pump itself shouldn't have voltage all the time at simple key-on, anyway. Removed the connections to the hot-wire harness and went back to the stock connections; no change. Started tapping on both passenger's side relays, since I couldn't recall which as the fuel pump relay. Still no change. I then pulled off both relays. Was sort of hoping they'd interchange. However, the plugs are different. Reinstalled them.

I'd resigned myself to packing up and calling for a rollback. I had my AAA card in one hand and a phone in the other. I sat down in the driver's seat to make the call....when I gave it one more crank. It started! So I went to make a few runs.

After the runs, I packed things up and got ready to go home. Got in the car, cranked and...no start. Removed the hot wire...still no pump prime, and no start. Pit lights are already out, and everyone is already gone. Still trying to get this thing started. I'm beating on the relays, while Jim is beating on my tank hoping to awaken some myserious loose connection. More cranking, and it catches. We head for home.

Just over the bridge into Missouri, I lose the pump again. This doesn't make sense. Always thought that the oil pressure switch would take over while the car was running if the fuel pressure relay failed during operation. Didn't give it much thought then, as I was a slowing roadblock on a major east-west thoroughfare of the nation's interstate system at near-midnight. Coasted to a stop in the slow lane (no medians at this point) and we did the gentle beating procedure and got it running again. Decided moving as fast as possible without attracting too much attention was the best plan. Managed to get home without another stall. The hot-wire was not connected at all during this time.

I'm trying to determine the true source of the problem here:
- Not likely a fuse. I checked the FP/INJ fuse, as well as the hot-wire system fuse. Neither was blown. And the hot-wire system was not used for the ride home.
- Not likely the hot-wire system relay, as it wasn't being used during the ride home.
- Not likely the oil pressure switch, but I'm suspicious of it at this time. It should not be a factor at startup. But it should be a factor at speed, regardless if the fuel pump relay works or not.
- The fuel pump relay. Everything seems to point toward this. I suppose it may need changing every 19 years or so. Temperature and vibration could aid in its failure.
- Shorting the underhood jack behind the alternator to the positive behind the alternator did not help, either. That should still kick the relay. It didn't when the problem was there. That seems to point to the relay.

Any thoughts/comments?

[socalgn]

Oil pressure was good (no bad signal (less that 4 psi) to the ECM)? I think you're right in checking the oil pressure switch (#1) grounds (#2) then the relays (#3). You might want to drop the tank, and check the connections at the pump itself, as well as the usual suspects (fuses, etc).

[Keller]

Oil pressure is 0 when the motor is off, of course. But when running, its decent. That doesn't mean the switch on the pump sees it that way, however.

Have new relays, a new pump, and a new battery. They'll all be going in this weekend.

[Keller]

I have the relay out, and the hot wire circumvented, and was still able to use a jumper to get the pump going to drain the tank. Checked the schematic, and that relay is never used in that part of the circuit.

So...since I could not get it to work when I was having the problem, it must have been either a wiring problem or a pump problem.

At least one of those will be cured shortly.

[IL KIM]

Sounds peculiar. I have had an issue before where the male terminal pulled out of the connector; the wiring connector in the trunk by the rear left wheel well feeding the fuel tank sender. You've probably checked it?

[Keller]

No, but I will. Does the stock wiring come through there? If so, that harness may get beat around by my tires and tools in the trunk at times.