Fuel pump stuck on prime

1993 cherokee auto 4.0

So the fuel pump will not shut off after the usual 3 sec with key on.....
I already changed the FP regulator at the rail and the relay. No dice.

Try pulling the ballast resistor relay (or sometimes called the bypass relay) and see what happens. Then try pulling the fuel pump relay and see what happens. The relay contacts could be welded closed. The contacts sometimes shed the coating and then can arch and weld themselves together.

If the contacts are welded closed I'd look for a short or very low resistance in the circuit someplace. Have you blown any fuses lately, like the number 2 fuse?

Sometimes the return spring inside the relay wears through and the contacts just kind of flop around in there.

The O2 sensor heater shares a power source with the bypass relay (3 second prime) may be a short in the O2 heater overheated the relay power wires.

I unplugged the wires from the ballast and no change.
I take the relay out and it stops. Out it back on and it turns on the pump....

Did you try swapping out the relay to see what happens?

Yes.
Swapped it with the Aux Fan relay and still the same issue

Maybe a short to ground in the relay control wire from the ecu

How can I diagnose this?

Anyone?

Isolate the circuit, disconnect the relay and the PCM connector. Then ohm test to chassis ground. Might as well check for volts also, there shouldn't be any volts, resistance should be infinite.

You'll have to double check this (my eyes are healing (from an operation). The red/dark blue wire is from the relay coil (activator) to ground through the PCM. The PCM grounds the bypass relay through a timer and)or some sort of simple logic circuit.

It could be the wire is grounding between the relay and the PCM (pin 37) or maybe that particular part of the PCM has failed.

I guess it is also possible the power in for the actuator coil, a dark blue and white wire from fuse 6 has some sort of voltage issue (or high resistance) that is fooling the timer. Hard to tell sometimes just exactly how a circuit is going to act when it gets marginal voltage. I'd change the fuse and have a close look at the fuse socket. Fuses can heat up and act like a bi metal switch, rare but it happens.

And like I said, do your homework before you trust my eyes, I may be wrong.

If I were in the same situation and I was sure the fault was internal at the PCM, Id install a momentary hold toggle switch and prime it manually from the drivers seat.

Thanks. I'll do this in the weekend.

Ok, so took the relay box appart.
Found tht there is a ground wire soldered to the yellow/blue wire that goes from the 'aito shutoff' relay to the fuel pump relay.
I took it apart. When I touch the ground wire to the end going to the 'auto shutoff', the pump starts to run.....

This is drivin me nuts. Now what.....

It's not that complicated. Basic relay has a coil (electromagnet) inside, when the relay coil is under power and grounded it pulls a little spring loaded metal tab towards it, closing two contacts together. The contact are usually numbered 30 (power in), 87 (power out), 85 and 86 being the coil wires.

On non Renix XJ's most of the relay coils are constant power or power with the key in run or start, sometimes power from another relay (contact side, in series) or a shared power source to the relay coil between multiple relays.

The relay coils often (most always in motor management circuits, after 1990 XJ's) switch to ground inside the PCM. The PCM completing the ground, closing the contacts.

Two circuits can run the pump, the fuel pump relay or the bypass relay (usually in combination with the starter), the bypass relay bypasses the ballast resistor during cranking. Pull both relays and plug them in one at a time to see which is powering up the pump.

It's most likely to be the fuel pump relay, barring some sort of weird short, power crossover or circuit feedback.

That yellow/blue wire is most likely the ground side of the relay coil (it sounds like it is for both the ASD relay and the fuel pump relay). It goes to the PCM (pin 51 I think). Either the wire insulation is cut (or melted) and sorting to the chassis or the PCM is grounding it when it shouldn't and it is likely to be a control circuit fault.

You said before you unplugged the ballast resistor and the pump kept running. This isolated the fuel pump relay out of the circuit (I think) and left only the bypass relay to power the pump or some sort of wire short or crossover power fault. Where you wrong? It seems now that the fuel pump relay is powering the pump? That Is why I said to remove both the fuel pump relay and the bypass relay, divide and conquer, if you remove and plug in only one at a time, you only have to troubleshoot half of the circuit.

It's always possible that two wires are touching someplace, have melted together or rubbed bare in the same spot and are touching, but I always troubleshoot from easy to complicated and isolate the circuits and then test them from either end or the middle out.

To the best of my knowledge the ASD relay and the fuel pump relay close for a few seconds when the key is turned to run. Then the relays open and won't close again until the PCM gets a pulse from the CPS to the PCM. The basic function for the prime event (fuel pump runs for a few seconds when the key is turned from off to the run position on the way to the start position), seems the same for all XJ's, exactly how it is wired is sometimes different year to year. On your 93 it sounds like the fuel pump relay and the ASD relay work in tandem.

What you describe now sounds like the PCM is not opening the ASD and Fuel pump relays, pulse signal or no pulse signal, engine running or engine stopped.