• Welcome to the new NAXJA Forum! If your password does not work, please use "Forgot your password?" link on the log-in page. Please feel free to reach out to [email protected] if we can provide any assistance.

Strange misfire/stalling especially under load issue

Milford Cubicle II

NAXJA Forum User
Location
Out there.
2000 4.0 AW4

So last night I was driving along and noticed what at the time I thought felt like my tranny slipping. So I dove into the gas to investigate and felt what turned out to be a very harsh misfire. While accelerating, the engine had a tremendous misfire and eventually just shut off just like I had turned the ignition off. Upon attempting to restart it, the engine wouldn't even act like it was going to start. So out of desperation, I disconnected my battery, then reconnected it then it started up and idled fine. So I took off down the road and it would run like a top but as soon as I got to about 30% open throttle it would start missing badly again, and eventually die then I'd have to go through the whole disconnecting of the battery thing again to get it running. Except this time the idle was pretty rough too and it was no longer so smooth below 30% throttle. I decided to just get home before I stranded myself then I dove into it today some more and now it pretty much just runs like crap all the time but is still significantly better below 30% throttle.

What really freaked me out was as I was just idling listening and watching the gauges, my speedo jumped up to 100mph then back down to 0! Then just a couple seconds later it did it again except it went to about 55mph.

Then that made me realize, lately I've noticed a couple times when I've turned off my ignition with my headlights still on and opened my door and it's not beeped at me like it normally does. And the beeper is in the instrument panel so that makes me thing maybe there's a loose connection in there somewhere. Now I could care less about the stupid beeper but it made me wonder if my ECM was f'd. But I'm pretty sure that beeper is just a mechanical AND circuit but the speedo, that's controlled by the ECM iirc. Btw, I held down my odometer button while turning on the ignition to put it through what is apparently supposedly some kind of test and apparently it passed cause all the gauges danced for me I dunno, just thought I'd throw that out there.

It's not RPM dependent at all, I can go easy on the pedal and bring the tach up to 5k smoothly all the way. In fact, if anything it even seems more load dependent than TPS dependent because I can squash the gas in neutral and peg the tach... roughly albeit but I can do it without it dying. However, if I tried that in even first gear there's a good chance it would just shut off like it did.

I've unplugged every sensor and injector individually to see if they make any difference whatsoever and the only one that even kinda makes a difference is the MAP.

Also, almost every engine management sensor is less than 2 years old including the MAP, TPS, CPS, CMP, IAC, and both O2's. Also, I checked the TPS for linearity and dead spots with both my scan gauge and an analog DVOM and it's good. The MAP readings on the scan gauge show that it's working fine too.

Sorry for the huge post but I really try to figure things out or at least dig as deep as I can before I have to bother anyone else about it. I'm baffled, can anyone give any insight?
 
Try plugging/un-plugging the crank sensor a few times and see how it is.
 
Same thing! Blowing in an NES cartridge might perhaps clean some dust out.

Also check your grounds - specifically, the one from the rear driver side of the head to the firewall.
 
That one happens to be 4awg welding cable. But I'll certainly check it anyways. Thanks! Keep it comin guys. The part that has me baffled is the speedo thing, that's seems just plain weird...
 
One thing that has been reported multiple times is a "no bus" signal on the IP after a crank sensor has shorted to ground, and an inability to make a connection with a scan tool. You might be on your way there.

Also, 97+ common issue is the connector at the back of the instrument panel making poor contact. There was a TSB for changing the connector. Most pull the instrument cluster, clean the contacts, apply dielectric grease, and I like to tug on the harness a bit to try and get some slack on the connector.
 
So I tested my CPS and it "passed". Meaning it wasn't shorted, the resistance was infinite across pins B and C. So my question is could it still be bad or is that test wholly conclusive? Because the more I poke around with it the more I think it's a bad CPS. I took it for a spin around the block and put my foot in the gas and got it to shut off. Then I crank to no avail. Then I turned the ignition off, unplugged the CPS and cranked also to no avail, of course. Then I turned the ignition back off, plugged in the CPS, cranked and it started right up with no hesitation whatsoever. The frustrating part about this is that the CPS is MAYBE a year old!
 
Put it into fault mode again and then test the CPS/CKP--see what it reads at that time.
 
I just dealt with a similar issue and long story short, i threw no code which made the misfire hunting fun:doh:.
After i replaced the same stuff as you did 2 years ago it wasn't any of them:wow:.
For a lark I swapped in a new coil and voila:party:.
If its not the coil no harm as a spare on the road or trail is a good idea.

What you are experiencing sure sounds like a crank sensor problem which is what i thought i had AGAIN. Keep us posted i'm curious to know what your solution turns out to be. Good luck! -B
 
Put it into fault mode again and then test the CPS/CKP--see what it reads at that time.

Well when I tested it the connector was unplugged. And I don't think watching the pulses go by on my dvom would tell me much if I tapped into it to watch it live. I guess I could check for 5 volts with the ignition on at least. :dunno:
 
Well when I tested it the connector was unplugged. And I don't think watching the pulses go by on my dvom would tell me much if I tapped into it to watch it live. I guess I could check for 5 volts with the ignition on at least. :dunno:

When it dies, pull the connector apart and measure the sensor's resistance AT THAT TIME, not when its cold or has been running Ok.
 
I just dealt with a similar issue and long story short, i threw no code which made the misfire hunting fun:doh:.
After i replaced the same stuff as you did 2 years ago it wasn't any of them:wow:.
For a lark I swapped in a new coil and voila:party:.
If its not the coil no harm as a spare on the road or trail is a good idea.

What you are experiencing sure sounds like a crank sensor problem which is what i thought i had AGAIN. Keep us posted i'm curious to know what your solution turns out to be. Good luck! -B

Geez i'm am idiot, I didn't pay attention to the year of your XJ, it still may be a coil starting to go.
 
Geez i'm am idiot, I didn't pay attention to the year of your XJ, it still may be a coil starting to go.

Coil or coil pack, you aren't an idiot! LOL!
 
For 5-Speed XJ:

Ok, you have spent a lot of money and time and not gotten anywhere.

First, have you put a scanner/code reader on it, and if so, what is the code number(s) you got?

Second, the crank sensor--CPS/CKP--is a signal generator that has to have a correct air gap and be in good condition to do its job. It actually generates both a low and high power signal to the PCM, and that data is interpreted so the PCM knows the physical position of the mechanical components. If the CPS/CKP bolts are loose that can screw up the data, if the resistance values are wrong--all the time or intermittently due to thermal failure that will also screw up the data--yes, the engine can still run with a faulty/failing CPS/CKP.

So, please post up the codes.
 
Back
Top