XJ starts & runs fine, then rough, then dies

Ok, I'm just about at the limit. My Cherokee will not idle. It runs great for 5 seconds, then stumbles, and then returns to normal, then stumbles, etc. until on a stumble it just plain dies. I've tried to diagnose everything that I could and I'm not sure what the problem is. The only thing that I could find in error is a bad O2 sensor heater (infinite Ohms A>B).

It's a 1987 Cherokee Pioneer 4.0L Auto. Had it in the family since new, I've not owned it until recently.

I have some videos of what's happenening but until I figure out how to recode them to something smaller file wise I can't post them.

1. Vid 1. Everything connected - Starts, runs, stumbles, runs, stumbles, runs, dies...

2. Vid 2. O2 Sensor disconnected - Starts, runs, stumbles, runs, stumbles, runs, dies...

3. Vid 4. I pull the IAC (brand new BTW) connector when Idle is good. It stays alive longer but still stumbles after longer intervals but never to the point of stalling.

4. Vid 5. Still running with the IAC disconnected, I pull the TPS and no change from before


Some of the items I've done/checked are:

These items I did before I knew there was a problem. It was previously running ok, not passing smog.

1. Had the head/manifold off, cleaned out 1/4" layer of oily grime.
2. New CTS
3. New MAT
4. New Knock Sensor
5. CPS not too old
6. Vacuum lines cleaned and joints from plastic to rubber were glued
7. Engine wiring harness reconditioned, crispy flaky connectors and boots replaced.
8. Good ford orange injectors installed
9. Battery cables redone
10. New Cat / Exhaust
11. New Spark plugs and wires

The following was done after everything was put back together and this problem arose...

11. New IAC (I broke the old one by accident)
12. EGR Blocked off with plate.
13. Old TPS has been adjusted, appears to have smooth range checked with DMM (I may buy a new one depending on my mood)
14. Engine compartment grounds checked to be within 0.5 Ohms of Batt -
15. Relays checked good.
16. Most sensor wiring checked good
17. Fuel Pump works / no balast resistor exists ('87)

As I wait for a new O2 sensor to come, is there anything I missed. I haven't been able to test the ECU nor am I sure how to, but seems the outputs that I've checked were fine. I was reading a post here http://www.naxja.org/forum/showthread.php?t=71001 which seems to have the same problems I do. Turned out his ECU was bad? Thought that was rare...

Of course searching gave me all the ideas to do what I've listed already, is great. I've broken down and finally posted... Any and all help appreciated.

First off very nice first post!!!! Pat yourself on the back, twice! A post like this deserves a detailed reply. Obiwan () has taught you well indeed!

Items I do not see on the done list:

New distributor cap and rotor?

Spark coil and Ignition Control Modual (ICM) which is part of the spark coil assy!!!!!! They can be bought and tested seperately.

Try turning the power on, do not start it, and check the TPS idle voltage to see if it is stable, by excercising the throttle from WOT and back to idle (engine off) about a dozen times, checking the idle voltage each time. If the idle position voltage varies (like by 0.1 volts or more) I would replace the TPS as it may be noisy right at the idle spot on the potentiometer, and could cause your problem. It should be repeatable with a variance of no more than 0.01 volts on a new, good TPS. I calibrated my TPS and was convinced it was OK for 2 years while I did as you did replacing everything else only to find out it was bad at WOT, and it had a ground that would read good one day and then the ground would change a day later!!! Some else here recently discovered that he could calibrate his TPS, but the WOT/Idle excersise test failed big time causing his variable idle speed!!!!

Have you run a standard vacuum gauge test on the engine to look for something like a burned valve, or vacuum leak, etc.? (will it run long enough to do one? If not, try this to get it running long enough to run a vacuum gauge test.

I see you tried disconnecting the IAC and O2 sensor one at a time, but did you try testing it with the O2 sensor and the IAC disconnected at the same time, disconnecting the IAC at a good idle speed as you did before? If not that would be worth trying for sure!!!!!!!

There is nice write up somewhere that shows how to run a simple vacuum test to determine if there is a burnt valve, leaky intake vacuum leak, bad rings, bad compression, etc.

You say fuel pump works, but have you tested the fuel pressure and verified that the fuel pressure regulator is also working properly?

Ah, one other possibilty, that is the distributor itself. The shaft bearing could be worn and it could have enough woble in it to cause a miss, or it could be installed off one tooth from the correct position beyond the control limits of the ECU!

I would do the cap and rotor (if not already done) then run the TPS idle repeatabiltiy test, then try and run without the O2 sensor and the IAC connected (disconnect both) at the same time, then run the vacuum test, then the fuel pressure & fuel pressure regulator test (if not already completed), then test the igntion coil (or try a know good one) and the ICM, then lastly take a close look at the distributor, in that order until you solve you problem. One of these has too fix it. If this does not fix it, you are one a million with a bad ECU (or you still have bad wiring harness connection somewhere, which is probably one in a thousand odds, but the tests I listed will isolate the bad part or bad wiring to the seemingly bad part!!!!!).

Hang in there, it sounds like you are nearly done, and like you nearly have a new jeep that should trouble free for a good long while once you find that last gremlin.

Ecomike said:

First off very nice first post!!!! Pat yourself on the back, twice! A post like this deserves a detailed reply. Obiwan () has taught you well indeed!

Click to expand...

Wow, never thunk that I'd get commended on a forum posting LOL. Thanks Ecomike!

To address your questions..

1. During the top end redo I didn't feel the cap / rotor needed a change. They were a good brass terminal cap and I wire brushed off the oxidation.
2. Coil / ICM not changed because for the same reason I didn't think the problem existed in the Cap & Rotor.

Dealing with many people's cars over the years, I just never seen an ignition problem act the way this jeep is reacting. Typically a bad cap/rotor will cause poor running all the time and throughout the RPM range. Nor have I seen ignition work this way either. Bad coils or ignitors seem to cut out completely. My jeep for those 5 seconds before it stumbles sounds like it's running great! I do however realize that cars can be mysterious machines and of course will not totally rule this out. I'll probably save it for the last desperate things to do.

3. I'll try out testing the TPS like you suggested, I believe I did do this test, but I didn't know the tolerance would have to be so tight. 0.1V variance is too much eh? I'll have to do this again. I may just buy a new one anyways because I see them for cheap right now.

4. I can hook up a vacuum gage but I'm not sure what a normal vacuum reading should be at given RPM. I'm pretty sure there is not a burnt valve because I've had the head off and resurfaced / lapped the valves. I'll have to do a search for vacuum testing.

5. Will definitely try disconnecting both O2 and IAC (cross your fingers!)

6. Fuel pressure has not been checked. Me being the cheap guy that I am, I've avoided buying a fuel pressure gage. I guess I don't like spending money on things that will tell me that I have to spend even more money on new parts. I'll look around for a good deal, I guess I could use one having 5 cars an all.

7. During the semi-rebuild I had the distributor out also, didn't seem to have any excessive play. And for the one tooth off thingy, I'd think that would make the motor run poorly all the time. I have those good 5 second intervals! ok, I'll check it...

Thanks for the help, I'll try this stuff out ASAP. And I'm still trying to post the vids, hopefully soon.

These RENIX XJs are a new beast to me. They are quite different than the old carburated and newer self diagnosing vehicles I'm used to..

SuperRA said:

Wow, never thunk that I'd get commended on a forum posting LOL. Thanks Ecomike!

Ya, we may have to make an example of you!
For the nobs that post up a new thread that says "hey my jeep is acting strange, how do I fix it?:doh:
:clap:

To address your questions..

1. During the top end redo I didn't feel the cap / rotor needed a change. They were a good brass terminal cap and I wire brushed off the oxidation.

I have seen perfectly good looking, new even, rotors and caps with a nearly invisible hairline crack that caused running problems.

2. Coil / ICM not changed because for the same reason I didn't think the problem existed in the Cap & Rotor.

Dealing with many people's cars over the years, I just never seen an ignition problem act the way this jeep is reacting. Typically a bad cap/rotor will cause poor running all the time and throughout the RPM range. Nor have I seen ignition work this way either. Bad coils or ignitors seem to cut out completely. My jeep for those 5 seconds before it stumbles sounds like it's running great! I do however realize that cars can be mysterious machines and of course will not totally rule this out. I'll probably save it for the last desperate things to do.

Keep in mind that some parts can move, like wires, and that movement can cause intermitent shorts that come and go. You do the wire and conector wiggle tests with a DMM hooked up to find these lovely little rascals.

3. I'll try out testing the TPS like you suggested, I believe I did do this test, but I didn't know the tolerance would have to be so tight. 0.1V variance is too much eh? I'll have to do this again. I may just buy a new one anyways because I see them for cheap right now.

I did some very thourough Renix tests (beyond the manuals) the last 9 months and documented that a 0.1 volt change in the TPS sensor at idle resulted in a 700 rpm increase, yes that sensitive. The TPS is a simple variable resistor and the internal contacts wear and at some point the resistance starts to get noisy, jumpy causing all sorts of odd behavior. Most of it is in real long thread called the "ReniX Files" (clever name I thought!).

4. I can hook up a vacuum gage but I'm not sure what a normal vacuum reading should be at given RPM. I'm pretty sure there is not a burnt valve because I've had the head off and resurfaced / lapped the valves. I'll have to do a search for vacuum testing.

I found some excelent instructions at:

http://www.ncf.ca/~ag384/vacuum.txt
and
http://www.obd-codes.com/howto/diagnoseengine.php

5. Will definitely try disconnecting both O2 and IAC (cross your fingers!)

Could be interesting!

6. Fuel pressure has not been checked. Me being the cheap guy that I am, I've avoided buying a fuel pressure gage. I guess I don't like spending money on things that will tell me that I have to spend even more money on new parts. I'll look around for a good deal, I guess I could use one having 5 cars an all.

I only spent about $35 on mine. It gave me peace mind to know that I did not have to spend the agonizing time and dollars on a fuel pump replacement to find out it wasn't the problem after all.

7. During the semi-rebuild I had the distributor out also, didn't seem to have any excessive play. And for the one tooth off thingy, I'd think that would make the motor run poorly all the time. I have those good 5 second intervals! ok, I'll check it...

Hmm, let's definately keep that one on the maybe burner. I forget the details, but I had mine off one tooth for months early on, but then again, it ran like Chit, but it did run. Thanks for the help, I'll try this stuff out ASAP. And I'm still trying to post the vids, hopefully soon.

These RENIX XJs are a new beast to me. They are quite different than the old carburated and newer self diagnosing vehicles I'm used to..

Click to expand...

In some ways the newer ones with OBDI and II can be even more of pain, been there done that. Once you get it fixed and figure out its quirks, you may find yourself hooked, addicted to Renix Xj's.

Hello. I'm Ecomike, I'm addicted to ReniX-XJs. :clap::clap::clap:

Ecomike said:

In some ways the newer ones with OBDI and II can be even more of pain, been there done that. Once you get it fixed and figure out its quirks, you may find yourself hooked, addicted to Renix Xj's.

Hello. I'm Ecomike, I'm addicted to ReniX-XJs. :clap::clap::clap:

Click to expand...

Yes, I've experienced what you are talking about too. (about OBD)

Haha, since I only have one right now I won't consider myself addicted. Should I happen to ever get another Jeep XJ or MJ, it'll probably be a RENIX.

I know, I'm always piping in about the CPS.
Try a new one.
Mine would barely, just barely idle, and would go thru spurts of trying to speed up.
When it did that, I couldn't get any rpms or power from it.
If I removed vacuum from the MAP sensor it would idle up ever so slightly.
The exhaust also smelled rich.
After letting it try for 10-20 minutes the Cat was becoming red.
It was the sensor.

I had the same problem last week with my 96. It would start up, then a minute later it would almost cut out, catch itself, repeat and then die. If I kept on the throttle (2-3Krpm)when it tried to cut out it would run very rough like the timing was way off. It turned out the culprit was the Accel coil that I've had for a year. I removed it and reinstalled the factory one and have had no problems since.

this is a really good post and it sounds like yall know a lot more about this in me. (and im not even sure if this is referenced above)

My car behaved like similar to this when my car computer was broken.

... continue on please.

Update:
Ok, so today I started the jeep with no IAC (Same as #3 in first post) and same thing happens as it did before. At least it is repeatable...
I then disconnected the O2 during a good idle. This resulted in an idle that would cycle the following:

1. 20 seconds of good smooth idle.
2. A drop in RPMs (unsure how much, I don't have a tach, I'm guessing 100-200 rpm drop) but still smooth, for a few seconds.
3. Loop back to #1.

This 20 second period is the same period as when I disconnected the IAC alone with the O2 connected, only that when the O2 was connected the RPMs would drop more resulting in a stumble. So something is still causing a fluctuation in a cyclical fashion. Maybe its the open loop program running with the last data it had from the O2 before it was disconnected?

I even put it in gear a few times and let the tires spin (its on jackstands) and applied brakes. It died when I broke more suddenly.

I would bet a 6 pack that your fuel pump ballast resistor is open. There is a relay across it when you crank, thus giving you fuel pressure. As soon as you quit cranking, it opens, and without the resistor, it removes all power to the fuel pump. The engine then runs on the residual fuel pressure in the lines and rail until that is depleted and dies.

The ballast resistor is mounted on the drivers side fenderwell, forward up by the hood latch. You can simply jumper across it to test. Heck, many years had no ballast resistor and ran the full 12v to the pump all the time.

I had a cyclic rpm idle pattern when I mistakenly connected my O2 sensor and my TPS to the wiring harness connectors backwards. In other words make sure pin A on the harness is attached to pin A on the TPS sensor (not pin C!), and likewise on the O2 sensor. My connectors are old and damaged enough that it is possible to connect them backwards, this is not normally possible, but I did it with an 87 XJ, so others are likely to make the same mistake sooner or later. I made the mistake working in the dark one night on the TPS sensor.

SuperRA said:

Update:
Ok, so today I started the jeep with no IAC (Same as #3 in first post) and same thing happens as it did before. At least it is repeatable...
I then disconnected the O2 during a good idle. This resulted in an idle that would cycle the following:

1. 20 seconds of good smooth idle.
2. A drop in RPMs (unsure how much, I don't have a tach, I'm guessing 100-200 rpm drop) but still smooth, for a few seconds.
3. Loop back to #1.

This 20 second period is the same period as when I disconnected the IAC alone with the O2 connected, only that when the O2 was connected the RPMs would drop more resulting in a stumble. So something is still causing a fluctuation in a cyclical fashion. Maybe its the open loop program running with the last data it had from the O2 before it was disconnected?

I even put it in gear a few times and let the tires spin (its on jackstands) and applied brakes. It died when I broke more suddenly.

Click to expand...

old_man said:

I would bet a 6 pack that your fuel pump ballast resistor is open.

Click to expand...

Sorry old_man, I'd say you're out a 6-pack. Listed in the first post, it's an "87 which doesn't have the resistor. Hmm, lets see, I don't want beer so I'll take a Dodge six pack. But really, I did see many posts about the resistor and looked long and hard wondering "where is this stupid resistor everyone is talking about" Thanks tho.

Ecomike said:

I had a cyclic rpm idle pattern when I mistakenly connected my O2 sensor and my TPS to the wiring harness connectors backwards. In other words make sure pin A on the harness is attached to pin A on the TPS sensor (not pin C!), and likewise on the O2 sensor. My connectors are old and damaged enough that it is possible to connect them backwards, this is not normally possible, but I did it with an 87 XJ, so others are likely to make the same mistake sooner or later. I made the mistake working in the dark one night on the TPS sensor.

Click to expand...

I'll check that too, I did take apart the harness to recondition it. It's definitely possible I screwed up.

i dont know if any one told you to try this but try your crank postion sensor

SuperRA said:

Sorry old_man, I'd say you're out a 6-pack. Listed in the first post, it's an "87 which doesn't have the resistor. Hmm, lets see, I don't want beer so I'll take a Dodge six pack. But really, I did see many posts about the resistor and looked long and hard wondering "where is this stupid resistor everyone is talking about" Thanks tho.

Click to expand...

What year did the resistor appear on the XJ's? Im having a similiar problem, however mine is an 88... :dunno:

My '88 has one.

anyone have a pic of this resistor or know where i could find one, it'd be a great help

It's on the driver side, on the wall by the airbox, just forward of the EGR relay.

I'd post the pic but imageshack seems to be acting up at the moment

Well anything you can do would be appreciated. The Jeep that I recently purchased is modified, so its hard to know locate and identify where certain oem items are in relation to other/stock jeeps. Being an unfamiliar jeep newbie doesnt help. If need be I can give you an email addy to send the pic if that would be easier. Thanks.

OK, it seems to be back..... the ballast resistor is circled in red

Thanks! Now realistically, if I were to jumper the ballast as the previous poster had mentioned before, and was sending a full 12 V to the fuel pump constantly, should i expect any great issues. I do have a spare tank with a new(er?) pump in the tank in the event that I burn out the current one. Is this the only issue I risk doing this?