Jeep XJ - Will not cycle check Engine Light, Fuel Pump Runs Continuously

96 Jeep Cherokee Limited 4X4, 4.0, AW4

WIth the key on run, the pump wont shut off after 3 seconds like it should.
Because this doesnt happen it will not cycle the check engine light (and gen light) on for 3 seconds.
If it is cranked without having cycled the cel on it will not fire (no spark)

Hi guys, I am almost at the end of my rope with this problem! I have watched a number of Pauls vidoes on YT (there is one with a jeep with a similar no start, pump running all the time which i thought was exactly the problem I had, but that was resolved by finding a shorted 02 which my vehicle does not hav) I saw Paul say a few times to post problems on his forum here if you cant work them out so here I am!

- I have a 5v reference circuit works (I do not have a probe on my MM so i tested this by unlugging the MAP and probing the plug, then doing the same thing on the MAF just incase it was the MAP that was bad) I have 5.15v on 2 of the 3 wires with the key on.

- Also believe that sensor ground is present by the same process as above

- I have no visually burnt or damaged wiring and have unplugged everything I can find trying to get the cel to come on

- On this model pump priming is done by a the pcm on a timer, not based on pressure (i know this is good anyway) so this is not the reason for pump not turning off.

- hotwired ignition switch to the run position with no change

- Have swapped Pump & ASD with known good items

- I also have the cruise control light on the stering wheel on solid which should not be there if cel light is working properly

--- BONUS ROUND --

During testing things I discovered I could sometimes, but not reliably, get it to turn the pump off and cycle the GEN & CEL lights on if the voltage is very low, i.e stuffed battery hooked up to it, or a big drain on system getting it under 9v or so. (this will click the asd and pump relay slowely several times but after a few seconds GEN and CEL will go soild and the relays stop clicking and the cruise control light will go out.

I can then remove the drain (again not reliably) and it will fire immediately and it runs/revs/drives so the pump and pressure is good as is all the other sensors and the pcm are all capable of running it perfectly.

If then shut off, it will happily cycle the CEL on on numerious key turns and will fire again normally if done within 5mins or so. I got it to do this once but have not been able to reproduce it since.

I'm not sure how the computer is turning the pump on is it always hot and adding ground or is it adding power? Could it be a bad ground somehwere not part of 5v ref or sensor ground circuits and, when the voltage was low, it wasnt quite enought to ground out and break the system, but full 12.4 volts was enough for it to "leak" it into a busted ground wire, is that even possible or is that theory shot?

This is about as much info as I can give at the moment, i am a DIY guy in Australia, I dont have access to a nice snap-on scanner like dans only a multimeter. I am mechanically minded and practial but I am no expert when it comes to wiring diagrams or these types of systems. Any direction on what to do next or if you have a paid service for more attention would be so much appreciated I really dont want to have to junk this car.

I think that you are dealing with a bad ground. When a ground is bad, circuits will find "alternative" paths to ground. Sometimes, this path is flowing, backwards, through other components. This can cause a relay and/or a lightbulb to remain powered, when it shouldn't be.

I would do a Ground-to-Ground voltage drop test on all Grounds, while the Fuel Pump is not shutting off. Start with the Fuel Pump and ASD Relay control.

ozbarra wrote: - On this model pump priming is done by a the pcm on a timer, not based on pressure (i know this is good anyway) so this is not the reason for pump not turning off.

There isn't any feedback from the fuel pump circuit but the PCM does monitor the switched side of the ASD. IIRC both relays turn on for a couple seconds when you first turn the key to run. It kinda sounds like the PCM software is waiting for the ASD signal to go high. So what I'd do first is see if the ASD is also stuck on when the fuel pump is. Voltage check at the ignition coil, or one of the injectors, looks like the easiest places to check the ASD is switched on.

If the ASD is on then check for ASD signal at the PCM. You'll have to backprobe the connector, or stab the wire a couple inches from the connector, to test for this. If that's OK unplug the PCM and check its powers and grounds. Best to use a test light made from a turn signal or headlight bulb for that.

5.15 volts on two wires ? Which 2 wires had those readings? using the diagrams chad provided tell us the colors and pin assignments.

Mate incredible thank you so much for the coloured and absolutely 100% correct wiring diagrams for this vehicle. I have searched high and low for this, they were not even accurate in the supposedly factory FSM I purchased!

Does your source for this also have labelled diagrams for the PDC and two seperate Fuse & Relay Junction Box's? I cannot find these either. 97 and 95 are totally different and thats all you get anywhere you look.

To give an update, I have done a bit more testing now as my understanding of the system impoves. With the key on (pump running ofcourse) I have discovered that the ASD relay is not sending 12v to the coil/injector/02 heaters. (Assume ASD not getting ground from the PCM to energise the relay)

In turn, the PCM would then not be getting the ASD sense response into C12 (which I suspect is the reason the FP wont stop and the CEL & GEN lights are not cycled on - not sure if you can confirm or deny? )

Power to the coil is one of the first things I tested by that was a week ago now in which time I've learned a lot more so I probably performed that incorrectly. Either way its not working now.

I do not have probe on my MM so I have tested this by removing battery neg, removing the grey PCM plug, clamping C3 pin of the PCM with neg side of my test light and clamping pos side to the battery. Reconnecting battery neg and turning the key, test light does not illuminate = no ground.

(I also did this with C19 the FP relay ground, it did light up and not go off so that lines up with the pump running continuously, so I assume no harness short for this)

So thats where I am at now, not sure if the way I have tested is correct but dont know how to move past the fact the PCM is not supplying a ground to the ASD relay.

At this point I'm thinking A) PCM faulty, or b) PCM is getting (or not getting) some other input that is telling it NOT to give gound to the ASD, if thats possible.

According to the FSM I have (for a 97) the only other things implicated in the Ignition Switch Key-On open-loop mode are IAT, MAP, TPS and IAC, CTS. The first 3 being on the 5v ref which is working (I know there is more on the 5v supply but they are not listed by the FSM as implicated in this key-on mode). This seems like the last resort to assume that there was some signal that these are giving to the PCM that it doesnt like and is stopping it from giving a ground signal to the ASD. Although this doesnt seem logical that if it was that clever at finding problems that it would chose to run FP. Not sure how to proceed at this stage.

- UPDATE, Matt and hardtopdr2 your replies were'nt there when I started replying to chad, had been typing this up for a while haha

Matt I think thats what I came to the conclusion of and did earlier today! problem with asd but not sure if how i tested (not backprobing) is acceptable have a read and let me know!

hardtopdr2 good question, now that I have this proper diagram I will retest and get back to you

Yeah ok so retested & confirming 5v ref and signal ground is good. 5v where is should be and no voltage on the signal ground circuit. I also tested IAC - as per chads diagram labels ABCD is 10v, 9.8v, 0.02v, 0.02v respecitvely not sure what is correct for this one.

ozbarra wrote: Matt I think thats what I came to the conclusion of and did earlier today! problem with asd but not sure if how i tested (not backprobing) is acceptable have a read and let me know!

Testing with the connector unplugged is iffy because it might disconnect something the PCM needs to see before it'll switch the ASD. Though I'm not sure what could inhibit ASD but allow the fuel pump to run.......

Better place to check would be the ASD relay socket control terminals. Relay should have a little diagram on it that'll show which are the control terminals. Put a test light across them and see if it lights up when you first turn the key to run. If it doesn't check one of them has power and if it does then confirm the wire to PCM is good.

Matt T wrote: Testing with the connector unplugged is iffy because it might disconnect something the PCM needs to see before it'll switch the ASD. Though I'm not sure what could inhibit ASD but allow the fuel pump to run

Thats exactly what I had some doubts about, I did it this way because I was trying to elimate a wiring issues between the pcm and the pdc. I should have done it your way first though or atleast aswell as. Done now, still no ground coming in to B20. B18 has power from ignition when in run/on position.

Impossible to visually check B18 to C3 is good hence the other test. Would have to cut harness and run a jumper between the two points I guess, not too keen if I can avoid it.

ozbarra wrote: Impossible to visually check B18 to C3 is good hence the other test. Would have to cut harness and run a jumper between the two points I guess, not too keen if I can avoid it.

Easy way you can verify that wiring. Unplug the PCM and ground C3 on the plug with a jumper. Then connect a test light from battery positive to B20 and if the light comes on the wiring is good. Which I'm guessing it probably will be since you did get the truck to run at one point.

I'd go after PCM powers and grounds next. If they check out I'd jump B16 to B17 and see what happens.......

I lost my reply when the internet went out but basically using your method with the test light;

ASD Ground
Wire in car from Grey PCM plug C3 to B20 is good

PSM Power/Ground
Power verified at the Black PCM plug A22 & A2 (when key on)
Ground verified at the Black PCM plug A31 & A32

ASD Power
Have jumped B16 & 17 before did not change the situation, tried again and still has no effect on fp or cel/gen lights (verfied that when jumped, it is delivering power to the coil circuit as expected though)

So still left with PCM not switching a ground at C3 with the key on. Is there a way of knowing if the PCM is waiting for something in order to switch this on or if by design it is purely based on the key coming to the run position, like a schematic for the PCM itself?

A bit more learned today, in the PDC there are a bunch of large fuses which I didnt originally pull (I tested them instead with the multimeter over the top of both contacts and found none of them were blown so moved on).

When I pull what I think is Fuse 6, i get:

- CEL/GEN lights back solid (wont go out though)
- ASD relay gets energised by the PCM ground and is correctly sending 12v to coil circuit.
- The cruise control light on the wheel does not come on
- Fuel pump wont run obviously as this is the fuse for it

With the fuse added back in, pcm drops asd ground feed again.

I dont have a labelled diagram for this PDC it is p/n 56008894 if someone has one please post it I cant find one anywhere and my fsm is wrong.

As this fuse is also giving power to the PCM at A22, I ran a 12v supply directly by probing the back of the plug to see if this would produce normal behavior on the dash, still no luck (For this test I removed the FP relay so power didnt feed back and run the pump incase theres a problem here with voltage leaking out of this wire in the loom)

At this point I'm not sure what to do with this information or if it is helpful. Any ideas?

I want to double check with you guys also on the 5v sensors, I have 5.15v on the 5v supply A17 and ground at sensor ground A4. Question is, should there also be 5.15v on all the individual signal wires? (A8, A18, A23, A27)

ozbarra wrote: So still left with PCM not switching a ground at C3 with the key on. Is there a way of knowing if the PCM is waiting for something in order to switch this on or if by design it is purely based on the key coming to the run position.

Found the OEM 1996 powertrain diagnostic procedures "81-699-96103" manual. Apparently the ASD and Fuel Pump relay control circuits are monitored by the PCM. So the PCM seeing the fuel pump relay stuck on could be what's causing it to hang up.

Faults on these circuits can also set codes in the PCM. You mentioned you don't have an expensive scanner but do you have anything you can pull codes with?

ozbarra wrote: I do not have probe on my MM so I have tested this by removing battery neg, removing the grey PCM plug, clamping C3 pin of the PCM with neg side of my test light and clamping pos side to the battery. Reconnecting battery neg and turning the key, test light does not illuminate = no ground.

(I also did this with C19 the FP relay ground, it did light up and not go off so that lines up with the pump running continuously, so I assume no harness short for this)

Did you also test C19 with the key off to check it wasn't grounded then? If you didn't that's the next step.

One other thing I noticed is there's an additional 5V feed to the VSS on B31. Wouldn't hurt to check it isn't being pulled down by a short.

Hi Matt please have a look at my post #42099 I have some updated findings in regards to getting ASD circuit to stay awake.

Dont have anythign to pull codes, I could buy something but im not sure it will have any codes to give (last time I was able to get codes by key on-off method there werent any, also without the cel light working properly I dont think it will communicate codes.

Yep c19 not grounded until key on. no problem with that side.

I think the problem at this stage can be summed up by saying: PCM not adding ground to ASD relay when fuse 6 is hot, which it is all the time by design. It can be seen in reverse by removing fuse 6 and seeing the PCM add ground to the ASD.

5.15v on VSS feed, only if fuse 6 is in place though. I assume this means that the computer must see power into A22 to turn on the 5v system. All the other sensors on the other 5v supply dont drop to 0 though with fuse 6 removed, they only drop down to 1.9v. All sesor grounds are good

ozbarra wrote: Dont have anythign to pull codes, I could buy something but im not sure it will have any codes to give (last time I was able to get codes by key on-off method there werent any, also without the cel light working properly I dont think it will communicate codes.

Cycling the ignition key on-off 3 times and then once more to ignition on will display powertrain codes in the odometer on this Jeep product.

Well yes, normally. CEL not cycling is one of the main parts of this problem

ozbarra wrote: Hi Matt please have a look at my post #42099 I have some updated findings in regards to getting ASD circuit to stay awake.

I did see that post but don't really know what, if anything, to make of that test. Re. fusebox layouts see if you can download an owners manual from the OEMs website. They often have fusebox layouts in them.

ozbarra wrote: Dont have anythign to pull codes, I could buy something but im not sure it will have any codes to give (last time I was able to get codes by key on-off method there werent any, also without the cel light working properly I dont think it will communicate codes.

The possibility of a no comm is one of the reasons I asked if you could try to scan it with something. Maybe one of your friends has a code reader or bluetooth dongle you could borrow?

ozbarra wrote: I think the problem at this stage can be summed up by saying: PCM not adding ground to ASD relay when fuse 6 is hot, which it is all the time by design. It can be seen in reverse by removing fuse 6 and seeing the PCM add ground to the ASD.

Summary of the problem is the PCM isn't doing several things correctly. Aside from the ASD not coming on the fuel pump is staying on when it shouldn't. MIL and GEN lights aren't lighting up and you can't retrieve codes by cycling the key.

Other than attempting to communicate with a scan tool I think you've done everything a good shop would do before calling the PCM. And way more than the dealer would do.

I really dont think its the PCM. After more testing I have worked out a way to get it to behave that I can reproduce at will;

I put an old battery in it that only puts out around 10v. the car will cycle the cel lights on and turn the pump off after 3 seconds, asd awake, everthing it needs to start (except it wont because the battery is low) I can however come in at this point and jump start it.

As an experiment to see if the extra voltage was escaping in the pcm internally, I delivered a full charge 12.4v directly to the PCM on A22 off a good battery and it's still happy, as long as the low voltage battery is still hooked into the vehicle putting a lower voltage through everything else. Seemingly the PCM is not the part that is glitching out when normal voltage is running through the system. (Reversing this situation, low voltage to pcm and normal to the car doesnt work)

It is just not happy with 12.4 volts during the inital key-on stage, after its got itself self-tested or the MIL cycled or whatever goes on there its happy. I think I could litterally put a switch in it that only fed the system 9v in the on position and then back to 12 after 3 seconds and it would be happy.

Safe to say I have no idea what I'm doing other than experimenting and trying to learn by trial and error at this stage, I have singed up to mitchell DIY to dig into more wiring diagrams for the systems in the car but I am a novice when it comes to reading them still no idea what the hell could be causing this to happen. A PCM for this jeep is extremely difficult to find as it was unique for the 1996 model year only, which probably makes it an expensive trial even if I could track one day (just for the sake of it even though I'm not entierly convinced I have been trying with no luck yet) I like to at this point pretend I have already changed it and it didnt fix the problem, what would I do next? Thanks for everyones input so far, if there are any more ideas please let me know and If I manage to work it out one day I will update.

I have the exact same problem. I have not tested it with a low battery but really all other symptoms are exactly the same. It is a 1996 XJ RHD. Because the fuel pump won't shut off, i ended up replacing a capacitor in the PCM which made no difference.

If i tap on the white part of ignition switch with steering column covers off, check & Gen light flashes and i can hear the ASD relay at the same time. With the help of wifey and MM, i was able to confirm ground signal at ASD while tapping on the ignition switch. I tried to hot wire but really i cannot reproduce the check engine light through any other means.