Misfires, A Ghost Story.

Jeep:
1996 4.0L Cherokee SE, Manual Transmission, 4WD

Original Symptom, Fall 2005: Rough running. Misfires.
Original Codes thrown: 300, 306

First steps taken:
-visual inspection of cap, rotor, wires, bosch plugs, and coil. Looked OK except cap&rotor. Replaced.
-replaced fuel filter
-had fuel rail and injectors cleaned by a mechanic (some special machine...).
-Fuel pressure checked at idle and rev in driveway. OK at 49PSI. Check valve test seemed OK, too.
-Visual & jiggle inspection of ground points for ignition. Look fine.
-replaced bosch with champion plugs
-wiggle-check all injector wiring connectors, clean w/ spray cleaner.

As luck had it, it actually resolved for a while at this point. I 'closed out' the thread, and it was more than a month before it happened again. It was also not happening much, so I put off dealing with it until last week- 'cause it got REAL BAD. But the issue is SO friggin' intermittent that I don't know whether it was granted by my actions in October or if it was just dumb luck that it went away for a while.

Now I'm getting codes 300, 301, 306, 302, 303. On clearing the codes repeatedly, it seems that 306 is the most common, as it was at the beginning of the problem back in September.

Second round of attempts:
-new distributor housing & CPS (camshaft position sensor)
-new MAP
-new Intake Air Sensor

Frying pan into the fire:

Now throws code 1391 in addition to the same old misfire codes.

I thought I'd botched the distributor install, so I took it to a mechanic. He took it out and reinstalled it. He said that I had it in right, though. Still throwing 1391, he used his Snap-On scanner to read codes and look at it while it ran. Sync on cam and crank sensors was out. He backprobed the Crankshaft Sensor and found it to be dropping out erratically and frequently.

-New CKS.

Still throwing 1391. Still misfiring.
the other things I did sure made it run better BETWEEN the misfire episodes, but none seem to have fixed it.

I am SO stumped.
And now I'm so broke. (all the parts)
Help?



------------------------------------------------------
These threads have the full, long-winded story:
http://www.naxja.org/forum/showthread.php?t=68307
http://www.naxja.org/forum/showthread.php?t=91084
http://www.naxja.org/forum/showthread.php?t=91148

Nixt said:

Jeep:
1996 4.0L Cherokee SE, Manual Transmission, 4WD

Original Symptom, Fall 2005: Rough running. Misfires.
Original Codes thrown: 300, 306

First steps taken:
-visual inspection of cap, rotor, wires, bosch plugs, and coil. Looked OK except cap&rotor. Replaced.
-replaced fuel filter
-had fuel rail and injectors cleaned by a mechanic (some special machine...).
-Fuel pressure checked at idle and rev in driveway. OK at 49PSI. Check valve test seemed OK, too.
-Visual & jiggle inspection of ground points for ignition. Look fine.
-replaced bosch with champion plugs
-wiggle-check all injector wiring connectors, clean w/ spray cleaner.

As luck had it, it actually resolved for a while at this point. I 'closed out' the thread, and it was more than a month before it happened again. It was also not happening much, so I put off dealing with it until last week- 'cause it got REAL BAD. But the issue is SO friggin' intermittent that I don't know whether it was granted by my actions in October or if it was just dumb luck that it went away for a while.

Now I'm getting codes 300, 301, 306, 302, 303. On clearing the codes repeatedly, it seems that 306 is the most common, as it was at the beginning of the problem back in September.

Second round of attempts:
-new distributor housing & CPS (camshaft position sensor)
-new MAP
-new Intake Air Sensor

Frying pan into the fire:

Now throws code 1391 in addition to the same old misfire codes.

I thought I'd botched the distributor install, so I took it to a mechanic. He took it out and reinstalled it. He said that I had it in right, though. Still throwing 1391, he used his Snap-On scanner to read codes and look at it while it ran. Sync on cam and crank sensors was out. He backprobed the Crankshaft Sensor and found it to be dropping out erratically and frequently.

-New CKS.

Still throwing 1391. Still misfiring.
the other things I did sure made it run better BETWEEN the misfire episodes, but none seem to have fixed it.

I am SO stumped.
And now I'm so broke. (all the parts)
Help?

Click to expand...

Grounds, check your grounds, specially the one on the back of the head to the firewall..take em apart and wirebrush them then reassemble..

Check the wiring associated with the crank sensor. Follow the wires from the crank sensor all the way back to the PCM. Use an ohmmeter and jiggle the wires, see if the resistance goes open at any point. If necessary, replace completely the runs from the crank sensor to the PCM with new wire.

RichP-
Spent the evening unbolting, cleaning, and rebolting all grounds.
All looked OK, but they got scrubbed w/ emery cloth and sprayed down with contact cleaner.

Test drive ~1 mile, all RPMs and gears, one minor rough run still happened, but no codes for misfire and largely a good running motor. However, it again threw code 1391. Seems like I may have gotten closer to finding the misfire issue, but now this sensor... I'll post to confirm if the grounds were the misfire cause if no more misfires show up over the next week.

anany91xj-
Feel like giving a novice instructions on how to do that with an ohmmeter? I have a multimeter that I have used for years to do electrical work on houses and at the shop, but have NEVER used the ohm scale on it. Never touched any electronics, either... I can't continuity test without a tester that is self-powered, right? Is it the same for Ohms? Sorry I'm so ignorant.

Nixt said:

anany91xj-
Feel like giving a novice instructions on how to do that with an ohmmeter? I have a multimeter that I have used for years to do electrical work on houses and at the shop, but have NEVER used the ohm scale on it. Never touched any electronics, either... I can't continuity test without a tester that is self-powered, right? Is it the same for Ohms? Sorry I'm so ignorant.

Click to expand...

Yes, the tester should have a battery of some sort in it. If your meter is set up for mainly 110v AC work, it might not be the best thing for testing vehicle electronics though. However, if it has an ohm setting, and is sensative at lower voltages, it should work.

Disconnect the harness at the crank sensor and the PCM. Find the pin on the PCM end of the harness that corresponds with each wire on the crank sensor connector (this is where a FSM with wiring diagram comes in VERY helpful). Connect one probe from the meter to the pin on the PCM side of the harness, the other probe to the corresponding pin on the sensor. The meter should read VERY low ohms (0.01 - 0.03 or so), meaning there's little resistance and a good connection. Jiggle the wires around (start with where the wires meet the connectors, then go from there), see if the resistance jumps or goes open altogether. If it does, you've found the break in the wire. Repeat this with all the wires on the crank sensor until you find a break.

If you find a break in the middle somewhere, you may want to just run a whole new wire from the sensor connector to the PCM connector, bypassing the wire in the factory harness.

Basically, you have a multiple random misfire (P0300), as well as several individual cylinder misfires. What we do know is you now have a code for crank/cam sync signal (P1391). Both sensors are new, that rules out a sensor problem. Your mechanic noticed, with both a scanner and DVOM, that the crank signal was erratic and dropping out. So, we're left with two possibilities. First, and most likely, there's a wiring problem causing a poor crank sensor connection. The other possibility, however unlikely, is that some teeth on the flywheel are broken, chipped, missing, whatever...feeding the sensor an erratic signal.

Geeze, before you stick that ohmmeter on there DISCONNECT THE BATTERY, I hope you read this before you start...
Also just because the part is new does not mean it's good...

Both of you, thanks...

anony91xj-
I've got to go out and buy a meter like that, mine's not self-powered. Once I have, I'll test as you described. When/if I replace the wire, do I splice right near the connector or do I have to open up the connector and go pin to pin? If so; the PCM connecter disassembles, but how the heck do you do that on the sensor connector? It's a sealed unit.

About that flywheel: I recently had the clutch done. First they resurfaced the old flywheel and new cylinders as well as clutch and kit. It was a botched job, had to bring it back next day for massive jiggles when engaging clutch. They had that set of stuff in & out twice, no dice. So they replaced the flywheel (on them, very nice), and it finally smoothed out. It is still not perfect, as there is still a slight warble at high RPMs and really slow engagement, but I don't drive like that so I wasn't going to bring it back in.
Think that's related to this sensor issue?

RichP-
Sadly you are right - new doesn't neccesarily mean working. The mechanic that 'scoped the heep also said as much, and is swapping in a Chrysler crank sensor tomorrow to eliminate that possibility.

You can't resurface a JEEP flywheel, its tapered not flat...
Oh, one other thing, the ground cables do degrade and corrode inside the shielding/cover and the woven one does corrode.
One interesting note, if you use tap water with a high mineral content the cooling system becomes part of the electrical system and conducts, it also plates the inside of the cooling system with whatever minerals are in the water, a good reason to use only distilled water when changing coolant... had a friend of mine flush his system with a 7 hour flush, it started running rough and died, the cooling system was acting as part of the ground ckt, the flush and new clean coolant w/distilled water removed that 'ground'. I had read about that but had never actually seen it happen and I gott tell you it took us a whole day to figure it out, we replaced the ground between the head/block and the firewall and it fired right up but then that woven ground cable practically fell apart when we removed it..
You can pick up a cheap vom at radio shack for $15 or $20, they come in handy...

Hmm. Can't resurface a jeep flywheel?
== Can anyone else verify this? ==
I'm calling the dealer tomorrow. If true, it means I've been bullshitted by my transmission guy. Calls the clutch work & price into even more question... Great.

The beast was at the mechanic today. He put in the Chrysler sensor, and that has evened out the 'scope reading. But it is still throwing the 1391 code, so the cam & crank are not registering as in synch still. He needed to keep it another day to test more stuff. He's going to "reasearch it tonight, and call some freinds". He was dismissive of the ground idea, yet it looks like the ground cleaning may have solved the initial misfiring condition, as it is still not acting up.

Will post again tomorrow with any new news.

And BTW, thanks very much to both of you for the help. I appreciate it.

Well, it's back and worse than ever.
Last Monday and Tueday the Mechanic 'fixed' my dist. intall job, said it was off, and the only way he could get it in was to have it off 4 degrees at the pulley. That's exactly what I had done (though I thought it was only 2 or 3 degrees). Anyways the 1391 code is gone.

Drove it 3 miles and suddenly the same old misfire, but then it got BAD. The jeep actually 'bucked' from it. I have no option but to keep driving it, and no time to wrench (I've lost enough time at work to this already). No engine light yet, but the code is in there,. (P0306)

The week since has been all rough running, with several severe rough-runs (like the truck has a seisure or a fit or something). Scary when I'm in heavy traffic...

I'm making a last sensor attempt, just for the hell of it, tomorrow - CTS and upstream O2, as was recommended in another thread for similar symptoms.

If this fails I will be forced to go buy a new truck of some kind - I can't have a daily driver this bad - more about the lost time fuc**n around with it than the cash for me. This sucks.

BTW -clutch now wobbly. Great. Will have to go back in to the dick that has botched it twice, but not if it doesn't run decent - otherwise what's the point?

Any words of wisdom?

I'm really no expert here but wanted to throw this out. I just had a cracked head that behaved exactly like this. Low oil pressure and low coolant pressure were easily seen symptoms, but the misfires and bucking became much more common before I had the head replaced. Runs fine now except it really hates starting whith the AC still on.

Good luck!

Back at it, with the last-ditch sensor swaps...

Just replaced:
-TPS
-Idle Air Controller
-O2 sensor (pre-cat)
-Sprayed down the intake manifold (mostly just the butterfly) with intake cleaner. (It was wicked dirty.)

On first start, threw Pcodes for sensors being out of range. Ran awful.
Restarted it after battery dis-then-reconnect.
Runs strong on a 'round the block drive.

Only time will tell, since this has been an intermittent thing.

As always, will post w/ any new news soon.

Alrighty then. Been a while.

Still does it. Now in addition to the misfires, it feels like a bucking bronco randomly.

Fuck it. I bought an '04 van for work and DD.

The heep is parked - collecting bird shit and dust.

Anyone near Ft. Lauderdale want a crack at this? I'll drop it at your place and leave it for you to tinker with..

;-)
ah well.

I'm having the same conditions in my 96!

Any updates????!!!!
I've been getting P0306 "cylinder 6 misfire" codes randomly.

So far I've:
cleaned throttle body (not too bad)
new: bosch +4 plugs, wires, cap and roter.
fuel filter
cleaned the K&N air filter


could it be a bad injector?

i am reposting on this tread too.


i was reading you original tread. how where you getting them codes.... where you using odb II or an odb I reader. also do you have a Haynes manual.
there is a way of checking you codes by doing an on off combo. try this on...off...on...off...on.then the check engine light will flash. on and off. then pause and then flash again. count the flashes till the pause and then after.
2 flashes pause then 3 flashes = 23.


also it could very well be the fuel injector. i am in no way an expert and more like a beginner novice. and i replaced all my injectors my self. the only trouble i had was getting them in and out. it would be a good test to switch the injector with injector 1. and see if the code follows if so there you go replace the one or all ( i recommend). also i do believe there recommended replacement every 100k. just my 2 cents

90xj06 said:

i am reposting on this tread too.
i was reading you original tread. how where you getting them codes.... where you using odb II or an odb I reader. also do you have a Haynes manual.
there is a way of checking you codes by doing an on off combo. try this on...off...on...off...on.then the check engine light will flash. on and off. then pause and then flash again. count the flashes till the pause and then after.
2 flashes pause then 3 flashes = 23.

Click to expand...

Those are industry standard codes, which are the numbers you will get if you hook up a code-reader to the OBDII port. The 2 digit MIL/CEL "blink" codes are jeep specific - though most of them have corresponding 4 digit DTC's

http://autorepair.about.com/cs/troubleshooting/l/bl_obd_main.htm

that is true but my book says that 98 and up has odb II and 1996 if that is what this is. has obd I

90xj06 said:

that is true but my book says that 98 and up has odb II and 1996 if that is what this is. has obd I

Click to expand...

96 has both. All vehicles sold in the US in 96 and up have OBDII, as far as I know.

yes thats what i thought but the haynes says 98 and up.