I recently experienced failed fuel pump in a '96, and have always read that year has a unique fuel pump module, but that is misleading

While the entire assembly is unique for a '96, the working parts appear to be the same '91-'96

I will further detail what I did later, with photos and links, this is an introduction to the topic

There is 5 issues

1) the fuel pump (same 94-96, I believe 91-93 are also the same DYOR look at pics on Ebay or Rock auto

2) fuel level circuit. Seems to be the same part number 91-96 *edit, the '96 circuit board IS unique, see additional details

3) Regulator. Appears to have the same part number 91-96, I believe pre-96 are externally mounted and have a vacuum hose attached.

4) Assembly "skeleton" . This part differs, '96 is unique, 94-95 is apparently different to 91-93.

The above 3 parts are clipped or bolted to this frame, and can be replaced individually, or removed to allow a clean up

5) The wiring

So in my case, the fuel pump failed, the sender was not working, and when I withdrew the module from the tank, it was corroded

I took all the parts out, cleaned up the skeleton and sender circuit board, and replaced the fuel pump itself with a quality Bosch unit $A60

Everything now works

If you have any year '91-96, you should be able to do this, no need to oder a whole new module, which are not only expensive, but can be hard to get, or NLA

'97 onwards has different parts

Pictures and more details, links etc coming soon...

 

The '96 FSM refers to 2 tracks on the fuel level sender. One is for the gauge, the other is for the PCM for OBDII emissions requirements.

Track 1-— fuel gauge operation: As the fuel level increases, the float and arm move up. This decreases the sending unit resistance, causing the fuel gauge on the instrument panel to read full. As the fuel level decreases, the float and arm move down. This increases the sending unit resistance, causing the fuel gauge on the instrument panel to read empty.

Track 2—OBD II emission requirements: A variable voltage signal is sent to the PCM to Indicate fuel level. The purpose of this feature is to prevent a false setting of misfire and fuel system monitor trouble codes if the fuel level in the tank is less than 15 percent of it's rated capacity."

These are pics from a thread on another forum (not mine).

 

Excellent post @Saudade

I have investigated this further and have some additional info and possible work arounds which I will edit in later.

The wiring diagram in the '96 FSM shows different wire colors than what appears to be on my '96, so when I suss that out I will update

So here is a very detailed thread from NAXJA that covers this info, and I checked it is ok to link, I am trying to get as much info in this thread as I can, which helps me to

1996 XJ fuel pump and pressure regulator repair details - NAXJA Forums -::- North American XJ Association

On the matter of the '96 fuel pump sender circuit;

Further checking revealed the two copper tracks on the circuit board show 5-110 OHMS Full/empty, and 100-1100 Ohms Full/empty

The 5-110 Ohms corresponds to the figure recorded as being correct for a 91-95 XJ and ZJ, so its a fair bet, that is the fuel level sender

That would mean the 100-1100 Ohms would be the fuel level low sense circuit (to be clarified when I work out the wiring colors)

It would appear likely that by far the most likely cause of the sender to fail is actually varnish from fuel on the copper plates causing high resistance

Both the circuit boards I checked were heavily covered in varnish, which I very carefully cleaned up with fine wet & dry

I also carefully dragged the "point" across the wet & dry

One way to test the sender and level sense circuit is operational after repair is to plug the unit in externally, and turn it up side down

be careful passion-fingers, dont break nothin, its all plastic and fragile

more soon...

 

I wonder if a fixed resistor couldn't be used to trick the PCM...

 

In short, I believe a '95 and earlier circuit/float would almost certainly work, and the '96 low sense light & PCM could indeed be "fooled" by a fixed resistance

You would lose the function of the light coming on when your fuel level is low, but you would still have an accurate fuel level gauge

As best as I can tell '97 and up dual circuit boards have a different resistance, so no help there

You need to examine the item under strong light and magnification, but it does appear that wires or tracks could be re-soldered if needed, hopefully negating the need for a "new" part. I do not know whether the pre-96 boards are still available

I will also discuss the possibility of using an external adjustable fuel pressure regulator, if the Crown unit becomes NLA. I cant see why it would not work if needed

 

I still have not got around to sort out the wiring and pinouts to allow for external testing with a multimeter...but I will

some pics instead ( this is from my spare '96 and has been in tank ~8 yrs

Here is the removed module, it looks very corroded, the fuel looked like ****




All parts removed, circuit board and regulator cleaned of varnish,

the "skeleton" soaked in CLR, and use a rotary wire brush. Cleaned up well

This is the vital "unique" to '96 component

Do not lose any of the small parts and fasteners




Here is a pic of the re-assembled unit with new Bosch pump, cleaned up and tested working dual track circuit board, and original regulator (left)
Next to it is the spare module I have that is tested working

The overhauled unit is now in and working



On a '96, the module can be withdrawn with the tank in place, but installing it again may be difficult, as it only goes in by twisting turning it at a certain angle. I would recommend you video withdrawal if you do want to try

The (any old) tank definitely should come out and be cleaned of years of built up water, scum and rust

In an ideal world, you would add a drain bung

Dropping down the guard and tank is not much fun at all

The guard bolts are a turgid ***** , very hard to get out, I damaged the threads.
I would be more patient if I did it again, more soaking, heat, and my 3/4" drive gear instead of just an impact gun

the tank hoses are stuck on the neck and access is not easy, pry them a little and soak some wd40 between the hose & neck

I add anti-seize to stop this happening in future, and also grease up the hose clamps

I have done this work on much older cars than XJs, working on ancient fuel systems and repairing rusted out fuel tanks is not much fun, preventative measures are best

 

Been an interesting thread.
I had an early 96 I bought new. After about 8 years /50 k the tank rusted thru from the outside.
... pulling the tank was fun as there factory red loctite on the skid plate bolts. Using heat to get them to move was interesting with a leaking tank.

... while I was in there decided to change the pump, got a new one from the dealer. Fit in the tank fine, put everything back together. When tried to connect the pump to the wiring harness the connectors were different.. ARRRRGHFiled a rib off one of the connectors and they mated ok.have no idea what that was all about. Everything worked fine including the gauge. Still have the original pump.

 

Yes, you want to give your original tank some love tank, and not take it for granted.

If you do not drain them every few years, there can be a buildup of particulate matter that is in fuel, and H2O in the air becomes displaced into the fuel when the tank heats & cools

If left to sit, the water will precipitate to the bottom of the tank, mix with the particulate matter, and begin rusting the bottom seam of your fuel tank

After 30 years, rust will be present.

I occasionally put some E10 in my near empty tank as H2O is soluble in Ethanol

The external tank and guard should be treated with rust retarding compound, as should every part back there, especially the exposed tank guard bolts, which go into nutserts

My guard has took a hard hit which distorted it and made fitment a real mongrel

Steel tanks can be successfully treated inside with POR15

As to the wiring with '96, I know it to be the case that there is at least 2 variants, early has the OBD1 style sensor plugs, later OBD2, some may have mixed !

The wiring colors in mine are not the same as the FSM

I will add more detail on this later for testing purposes, basically, there is a fat wire for 12V pump, black wire is earth, 3 thinner wires which should be power in for sender circuits and 2 wires out, one for sender, one for sense circuit

I will also add the Crown regulator part num (is available) and discuss the use of an adjustable external FP regulator should the Crown ever become NLA

If you own '96 (like me) and live far away from a source of some new parts (Australia), and sometimes travel far into the forest (as I do sometimes), you need some spare parts like Fuel pump, alternator, PCM, CPS etc

 

woopee, I get to do a few updates, as I have 3 x '96 that I (try) to keep running. (My DD, my farm truck, and my "spare", which resides in the backyard), this means more messing around with fuel pumps than one would want.

At any rate, I can confirm that my method of cleaning the circuit board and pointer appears to be successful at restoring the function of "fuel level" and "low fuel light"

This is important, as the circuit board and associated parts are NLA to the best of my knowledge, and it does become a problem with older vehicles

(I swapped a couple of cleaned units around and they worked)

I have personally hand repaired and re-wired senders over 70 years old, where no aftermarket part was ever available, and original style replacement parts are extremely hard to come by


One thing I am not sure of, and that is where is the "non-return valve" in the fuel pump module located? The FSM indicates its is in the regulator?

Extended cranking is a sign that this is an issue, and can be tested by cycle ignition key on/off 3 times, it starts immediately

I am still working on what is causing this issue, I have a suspect, and have not yet tested fuel pressure for any possible injector leaks