Fuel pump issues has everyone stumped

[Clayer]

I've been a long-time watcher, never had to post because I have been able to find the answers I'm looking for. Until now.

1993 XJ 4.0 4wd RHD 285k miles

When I take my tired girl on my mail route she runs great when the outside temperature is less than 90. But when the temp. gets above 90-ish, the fuel pump will get super loud after running her on the route for about 2 hours. She has left me stranded more than once because once the fuel pump starts to get loud, I have about 30 minutes left before she dies completely. Once cooled down, she fires right up like nothing ever happened.

Work that's been done:

3 fuel pumps in 9 months, two I installed (Delphi/Chinese), and one I had a shop do because I was thinking maybe it was me (Bosch).
2 fuel filters
Crank position sensor
Cam sensor
Plugs/wires/cap/rotor
Coil
PCM
replaced all rubber fuel lines in the back and used air to blow out the steel lines
Heat-shield on all lines up top
ECM
Fuel pressure regulator
Transmission
I'm sure there are other things, but I can't remember them all!

EVERY SINGLE mechanic I have brought it to, tells me that I need to replace the fuel pump AGAIN. I have done all of the work on this myself because these mechanics are not mechanics. They all seem to be parts changers.

The only difference I have found, so far, is when I changed out the fuel lines in the back. Before I did that, she would leave me stranded when the temp was about 70.

Please help - oh wise people of this awesome forum - I need a wizard or a witch to chase this ghost from my poor old lady!

 

[Alucard226]

Does the 93 still have the porcelain pump ballast resistor on the driver's fender near the brake booster? These can cause pump issues, can give it a quick test next time it happens by jumping the two green wires feeding the two terminals on it. Also do any of your relays click on/fuel pump cycle on during the stranded times when you go to turn the key on before starting the engine? Crank sensor can cause a hot no start condition if it's failing when it heat up. I had a very similar issue but opposite temps, my ECU wouldn't cycle power to anything (no relay clicks, no pump prime or check engine pre test light up) until the key sat on long enough for it build up enough juice to kick over, was a nightmare to chase in the beginning. Have you flashed the diagnostic code/key sequence to see if it's tripping any?

 

[Alucard226]

Wouldn't hurt to go through all the grounds too if ya haven't already, especially the one on the block at the dipstick, the one on the firewall for the ECU and the one under the dash for the cluster and such. I cleaned up all the factory ones and just searched around here for ideas I liked to add extras everywhere.

 

[75SV1]

I had issues with stalling out on some XJ. Also, a rough running one. I could not pin it down either. Also, I am not definitive on what is causing your issue. Two or three things I'd try. First is the coolant temp sensor. I think yours actually has two of them. One at the thermostat housing and the other at the cylinder hear. I'd also change the thermostat. I might question the O2 sensor if you haven't replaced it in a while. NGK (NTK) or MOPAR. Also, I had a No Start issue. I replaced a lot of stuff. Fuel Pump (Carter), injectors (12 holes), spark plugs, etc. Also, tested the replaced injectors with a NOID light. I replaced the electronic module for the key.

 

[roller4v]

install a fuel pressure gauge so you can see what is actually happening with the pump. If anything.

 

[Unclewolverine]

None of these allow for the loud pump operation though which seems to me the defining characteristic especially across several pumps. My mail jeep is a 95 so probably different, but the ballast thing someone else mentioned would be where I started. Have you replaced the tank? I would also try it one day with the fuel cap off or loose and see if that changes anything. Everyone does need to remember that he has a rhd so electrical troubleshooting is going to be hit and miss because no one can seem to find an accurate wiring diagram. The fuse box is completely different from a lhd and any one labeled rhd that I've seen is not correct.

 

[lawsoncl]

Just curious what happens if you leak the gas cap loose? I'm wondering if you developing pressure or a vacuum due to plugged evap lines. I could see pressure maybe pushing fuel into the charcoal canister if something was amiss in that system.

 

[cruiser54]

bypass the resistor and see what happens. also drive with the gas cap loose.

 

[SouSy]

What is your volt gauge doing when the pump starts to get loud?

Load test the power and ground at the fuel pump connector. Grab a turn signal bulb (preferably with a socket and pigtail so you don't have to get too creative with wires,) hook it up to the power and ground at the fuel pump connector and turn on the ignition. The light should be nice and bright like you have it hooked up to the battery, if it's dimmer than that then you've got a bad/corroded wire or connection somewhere.

Just because you have voltage doesn't mean the circuit is in good enough condition to supply the amperage. As an electrical components get warm it takes more effort for the circuit to run the electrical component. It could be that once your fuel pump gets warm enough the circuit can't supply enough voltage for the pump.

Also, what rubber lines did you replace? Did you make sure and use fuel safe lines? Just regular old hose will break down and plug up the fuel system.

 

[8Mud]

I have no defining answer but maybe a few possibilities. The only time I've heard the pump get loud is when the tank is empty. One thing that comes to mind is the rubber end cap for the pump pickup. It fits in a detent in the bottom of the tank and helps reduce the leverage on the top piping when the pump wants to swing, it in effect anchors the bottom of the pump assembly.
As far as the electrical goes a few things come to mind. One is the fuel pump harness goes through multiple connectors any of which can get corroded or overheat and melt down, seen it before. The heat of the meltdown makes the connecting iffy, the contacts move with the heat.
Another is multiple sensors can have a partial internal short and the PCM will shut down the bus as a protection. The orange wire going to the crank position sensor, the cam position sensor and the speed sensor should be 7 or 8 volts. Not sure if the engine needs to be cranking and getting a good pulse signal from the crank position sensor in the 93, it doesn't on my 96 it has power with the key in run. Low or no voltage on the orange wire will cause issues until the problem sensor finally craps out completely. Look at the cam position sensor, often overlooked and prone to failure.
Keep an eye on your voltage gauge anything over 14.5 can cause issues and it is possible for overcharging to go a lot higher than that. The Denso is not as likely to overcharge as the Delco is, but it is still possible.
Just swapping out relays may not be the answer, I'd replace the fuel pump relay with a new name brand relay like Bosch. The fuel pump takes a relatively high amperage to function properly and poor contact is going to lose some of that amperage to resistance. Double-check all the connectors, a pain but at this point something that needs to be done. At the very least do an Ohm test on the fuel pump harness, ideally, it should be 0 anything over an ohm or two is going to cause issues.

 

[old_man]

Search for heat soak. The worst thing you can do for a fuel pump is to constantly run low on fuel, since the pump runs in the fuel to cool it.