No power to the fuel pump, 89 MJ

[xjtrailrider]

I went out to the garage and the pressure had drained back to 0. It would not start. I pounded on the tank and it fired right up. I think I have the problem narrowed down now, just have to find it and fix it. Its ether a wire short/open or the pump.

 

[cruiser54]

I went out to the garage and the pressure had drained back to 0. It would not start. I pounded on the tank and it fired right up. I think I have the problem narrowed down now, just have to find it and fix it. Its ether a wire short/open or the pump.

Yep. Good diagnosis. BFH and determination!! OL.

 

[joe_peters]

Slow down for a minute.

Earlier you said you tested for power at the fuel pump, and that when it had power it would run.

Ok, then I doubt the hammering on the tank did anything.

Check again for power at the pump--you need to determine WHAT IS WRONG when there is NO POWER at the pump. Otherwise you are chasing the wrong issue.

 

[xjtrailrider]

Slow down for a minute.

Earlier you said you tested for power at the fuel pump, and that when it had power it would run.

Ok, then I doubt the hammering on the tank did anything.

Check again for power at the pump--you need to determine WHAT IS WRONG when there is NO POWER at the pump. Otherwise you are chasing the wrong issue.

I'm not sure why I wasn't getting power at the pump for the last few days. Suddenly today when I opened up the wire harness and used the splice to check for power I started getting power all the way back. I don't see anything wrong with the splice or the harness for that matter. So after I started seeing power at the pump connector I still did not have a running pump. I'm going to check the wire harness at the pump where it was hacked up by a PO. that might have become a issue and now pounding on the tank is making it close the circuit or not short.

 

[8Mud]

Just throwing something out there that may or may not be relevant, but don't count on the copper wire being good under the insulation. Twice now I've found spots where the copper wire was broken under the insulation. Usually at a spot where the harness flexes, like the bend at the firewall where the injector harness makes the bend and the fusible link bundle near the starter relay.

Jeep actually put out a TSB for this, *my TSB* after I wrote a letter and sent some pictures.

Only two ways I know to find a spot like this, one is to shake the harness and hope your ohm meter shows a fault or feel the individual wires with your fingertips to find the bad spot. The two broken ends of the wire rub together under the insulation and may or not make contact, a real pain to find.

Also don't count on those splices no matter how good they look. I've found them where somebody forgot to crimp the splice and the wires where just laying in the splice, sometimes they would make contact sometimes they wouldn't.. And I've found them where the wire was pushed too far into the splice and then crimped onto the insulation, maybe they'd make contact and maybe they wouldn't.

You mentioned the poles on the old starter relay looking cooked. Those spade connectors and the pins in the connectors get hot and loose the clamping spring type force to make a good contact. They usually turn blue or black from heat. something like that screams electrical short to ground someplace. A loose connector can act just like a resistor and change voltage and amps (watts) to heat. That is why the plastic connector housings melt melt. I always clean the connectors out and eyeball them good, a close visual inspection may turn up a problem, flashlights are your friend.

Another spot you may want to look at is the relay block, the bottom is open and packed with die electric grease, mine was packed full of mud. It looked really nasty under my relay block, green oxidation, mud and the die electric grease had turned to a tar like substance.

Pinch the wires going into the fusible links, move them around some and see what happens. the copper wire under the insulation can fatigue and break. Also look at the fusible links for discoloration, they can get iffy if they overheat, sometimes they don't blow and can act like sort of a circuit breaker and make iffy contact internally. Fuses or switches can do this also, a crack or a break in the fuse can make contact, heat up and flex (like a bi metal) and break contact cool back down and make contact again.

Just some oddball ideas, if the normal culprits aren't the problem. If your systematic search proves fruitless, you have to move on the bazaar and unusual and hope for a little luck. Been there done that.

One of the best things I ever did was pull the complete harness out of a junkyard XJ. I look at my schematic or wiring diagram and then find the wires in the harness on the garage floor, then I go find them inside the XJ.

 

[xjtrailrider]

All great advice, thank you!

I do think the issue is in or near the tank. I now can repeat the process and get it to start when it wont by whacking on the bottom of the tank. My first plan of attack is to inspect the wire harness between the last connector behind the cab and the pump bung. It has been spliced by a previous owner. If that doesn't work I will drop the tank and pull the pump(not necessary to drop the tank but on the MJ its easier to access the pump and inspect it with the tank out).

One note of weirdness; After it starts now(after whacking on the tank) it will run for 2-3 seconds and then fuel pressure will drop and it will die but it always starts right back up and runs fine. It will restart all I need it to unless I let it sit until the fuel pressure drops to zero then I have to hit the tank again to get it to start. That tells me it may be a fuel pump issue again.

I wonder if Bosch pumps have a lifetime warranty. I bought it from Advance Auto

 

[8Mud]

Just for giggles check your battery voltage, then jump the wire at the starter relay (orange with a black stripe) straight to the battery (better with a fused jumper wire just in case). Then check the voltage at the pump connector with the connector disconnected.

It may be the pump, it may be you are losing voltage (line loss). You will lose some, but anything significant translates into lost amperage. You might typically loose half a volt, much more than that and it is likely you have an issue.

You can get 12 volts through a single strand of braided copper wire, but you aren't going to get enough amperage to do much. Voltage roughly translates into pressure, amperage roughly translates into volume (how much electricity) volts X amps equals watts and watts roughly translates into horse power. You can test good for volts and still have an issue with the actual amount (volume/amps) of electricity flowing through the system.. The loss between point A (say the battery) and point B where the current needs to be, is often lost through heat (lost watts). I've found problem spots with my fingers, they often get hot, blister your finger hot or maybe just warm..

When my connector screwed up on the path to the fuel pump, sometimes I'd get fuel, sometimes I wouldn't and sometimes the pump would make kind of a groaning sound (feed me).

 

[xjtrailrider]

Just for giggles check your battery voltage, then jump the wire at the starter relay (orange with a black stripe) straight to the battery (better with a fused jumper wire just in case). Then check the voltage at the pump connector with the connector disconnected.

It may be the pump, it may be you are losing voltage (line loss). You will lose some, but anything significant translates into lost amperage. You might typically loose half a volt, much more than that and it is likely you have an issue.

You can get 12 volts through a single strand of braided copper wire, but you aren't going to get enough amperage to do much. Voltage roughly translates into pressure, amperage roughly translates into volume (how much electricity) volts X amps equals watts and watts roughly translates into horse power. You can test good for volts and still have an issue with the actual amount (volume/amps) of electricity flowing through the system.. The loss between point A (say the battery) and point B where the current needs to be, is often lost through heat (lost watts). I've found problem spots with my fingers, they often get hot, blister your finger hot or maybe just warm..

When my connector screwed up on the path to the fuel pump, sometimes I'd get fuel, sometimes I wouldn't and sometimes the pump would make kind of a groaning sound (feed me).

That is why I would like to run a new line from the A splice to the pump. Is this a good idea or not? I can cut out a whole bunch of connectors and possible future problems. My plan would be to cut the orange wire at the A splice then route down and along the fuel lines inside a wire loom to the pump with no connectors except the last one before the pump.

 

[8Mud]

That is why I would like to run a new line from the A splice to the pump. Is this a good idea or not? I can cut out a whole bunch of connectors and possible future problems. My plan would be to cut the orange wire at the A splice then route down and along the fuel lines inside a wire loom to the pump with no connectors except the last one before the pump.

Likely to be a good idea, but may not be the solution to your problem. That is why I'd test first, process of elimination.

You may have problems adding a wire to the existing loom. On my Chev pickup I added a plastic flex pipe front to rear and cable tied it to any handy spots. And ran a snake through it and pulled extra wire as I needed them for my trailer, extra back up lights, rear fog etc..

On my XJ those connectors contained a lot more wires than just my pump wire. I just crimp spliced them around (bypassing) the connector, soldered and used shrink tubing to protect the splice. Running a whole new wire may be the easier solution. All I know for sure is the fewer connectors and splices, the less chance a connector or splice is going to cause issues.

Many of the connectors in the XJ seem to have inferior plating (coatings) on the pins in the connectors and seem to be prone to corrosion and oxidation. My guess is the lowest bidder won the contract.

 

[cruiser54]

This is never a bad idea on an old Renix Jeep:

I suggest unplugging EVERY electrical connection in the engine bay you can find, whether engine related or not, and spraying it out with a good electronics cleaner, visually inspecting the terminals making sure they haven’t retracted into the plastic holder, and then plugging it back together. There’s a critical 10-pin connector for the front lighting system located in front of the air cleaner and behind the left headlight assembly. Don’t miss that one. Also be sure that the connectors to the ballast resistor mounted near the air cleaner housing are clean and tight.
ALL of the relays should be removed, the terminals wire-brushed until shiny, and the receptacles sprayed out with contact cleaner. Then plug them back in. I do this on every Renix Jeep I purchase or work on for someone else.
Revised 07/23/2012

 

[cruiser54]

Likely to be a good idea, but may not be the solution to your problem. That is why I'd test first, process of elimination.

You may have problems adding a wire to the existing loom. On my Chev pickup I added a plastic flex pipe front to rear and cable tied it to any handy spots. And ran a snake through it and pulled extra wire as I needed them for my trailer, extra back up lights, rear fog etc..

On my XJ those connectors contained a lot more wires than just my pump wire. I just crimp spliced them around (bypassing) the connector, soldered and used shrink tubing to protect the splice. Running a whole new wire may be the easier solution. All I know for sure is the fewer connectors and splices, the less chance a connector or splice is going to cause issues.

Many of the connectors in the XJ seem to have inferior plating (coatings) on the pins in the connectors and seem to be prone to corrosion and oxidation. My guess is the lowest bidder won the contract.

Harnesses hecho en Mexico FWIW. I always picture old Mexican ladies putting them together for some reason.

 

[xjtrailrider]

Likely to be a good idea, but may not be the solution to your problem. That is why I'd test first, process of elimination.

You may have problems adding a wire to the existing loom. On my Chev pickup I added a plastic flex pipe front to rear and cable tied it to any handy spots. And ran a snake through it and pulled extra wire as I needed them for my trailer, extra back up lights, rear fog etc..

On my XJ those connectors contained a lot more wires than just my pump wire. I just crimp spliced them around (bypassing) the connector, soldered and used shrink tubing to protect the splice. Running a whole new wire may be the easier solution. All I know for sure is the fewer connectors and splices, the less chance a connector or splice is going to cause issues.

Many of the connectors in the XJ seem to have inferior plating (coatings) on the pins in the connectors and seem to be prone to corrosion and oxidation. My guess is the lowest bidder won the contract.

I plan on just cutting the orange wire at splice A and at the pump connector and bypassing all the other connectors to the pump, essentially making the orange wire "dead" between the bulk head connector and the rear of the cab. I don't see where anything else is spliced into it anywhere else on the schematic posted.

I talked earlier of the starter relay giving my issues and I had to replace it. It had a bent post INSIDE the box and was no longer making contact for the pump to prime.

EDIT; Those are some terrible pics!, the one post that you see bent is the one for the fuel pump and the contact would sometimes hit it and sometimes not. I can't see how it got bent, its not discolored anywhere, just bent! I had a hard time finding this issue.

 

[xjtrailrider]

Okay, might be on to something!

Here is a pic of the last connector before the pump, it shows orange to orange(pump?), tan to purple(fuel level sender?) and black to black(ground?)

At the pump, which was re-wired before I bought it shows purple to red?, orange to black? and then the black wire is connected to ground on the bung as it should be.

I put the last pump in but I do not remember exactly how its wired inside. I remember saying to myself what a mess but I had to put the pump in on the ground at the house where I bought the truck, I did it fast and on the fly.

Is it possible that the pump has been running off of the fuel level sender voltage? My fuel gauge will never read full or empty but it does rise and fall with the fuel level between 3/4 and 1/4 tank?

Would it be a bad thing if I swap the orange wire to the red wire and see what happens or could I fry the sender if its correct the way it is now, once reversed?

Worst case scenario I drop the tank and pull the pump to make sure this thing is wired correctly. It may not be a bad thing to replace the pump at this point as well if its been running at a much reduced voltage.

Edit; The 88FSM that I have on pdf shows only two wires going to the fuel tank unit, purple(violet) and black? I know for a fact on the 89 that the orange wire carries the voltage for the pump. I just verified that with my test light. I get a bright light when cranking. The purple wire gives me constant "dim" light while the key is in run or start. That has to be the wire to the fuel sender. I'm so damn confused :/

 

[8Mud]

The tan wire is the gauge wire. I think maybe you hit the same spot I found, the orange wire changed to a purple wire at one connector and then back to an orange wire at another connector. It sounds simple in retrospect, at the time it totally confused me. But this was an XJ and not an MJ, bound to be some minor differences.

www.bteventures.com//mj1988electricalmanual.pdf

I took one of those starter relays apart once and it was full of mud and rust, but it still worked. My guess is it got submerged in muddy water at some point and then the water eventually dried out and left the mud.

 

[8Mud]

I never have seriously investigated the XJ fuel gauge sender circuit. But on my YJ a dirty connection would mess up my gauge reading. Most times it was the contact strip plugged into the gauge cluster that was oxidized.

Th XJ may be different, but to the best of my knowledge the current at the tan wire at the pump and through the fuel level sender to ground is a micro current (milliamps) and the ground for the tiny winding inside the gauge that moves the needle. Doubtful it would pass enough current to run the pump IMO. This is just an educated guess after troubleshooting my YJ.

 

[xjtrailrider]

I called Advance Auto today and my Bosch pump is under warranty so i will be replacing that tomorrow as well as dropping the tank and looking this wiring over good for issues. This will be the 4th time in two years I have had to drop the tank. Once for the fuel pump bung leaking, once for the check valves that were leaking vapor, once for when we replaced the bed from the roll over and now the pump again.

I'm getting good at this unfortunatly

I can make it start whenever I want to by hitting the skid plate under the tank, as soon as pressure drains off after a few hours it will not start. My son, who owns this truck has run it out of gas many times in the last few years due to the gauge being off so I'm going to address the sender as well. Running the tank empty probably shortened the pumps life as well.

 

[xjtrailrider]

I found the problem at the pump. The ground was loose on the connector to the pump. The new pump will get this connector eliminated.

So what I found was a intermittent positive connection at the bulkhead connector(orange wire) then the burnt/loose ground at the pump.

I think I have this sorted out now!

 

[cruiser54]

Had the same issue with the ground wire.

 

[xjtrailrider]

Had the same issue with the ground wire.

Yep, you mentioned that earlier in the thread so I knew then I was going to have to drop the tank. I found lots of issues that I'm going to fix while I have the tank down this time.

Is there any special wire I need to replace the crap wire that someone put in there? The wire to the pump looks like 12g house wire. It could just be that it has a special coating but it has the clear stuff over the insulation. I plan on wiring it up with 14g to replace the 12g that is in there. It has 14g the rest of the way to the relay. I'm not sure why that heavy gauge is between the connector and the pump.

 

[cruiser54]

I just used a 10 gauge and crimped the terminals on each end. I DID tweak the terminals so they grabbed tighter. I think that was the problem with the original replacement on mine. The small terminal must not have grabbed tightly. It was the most burned.