Dcoy

Ok so i have been having a cylinder 4 misfire here for about 2 months now and just can not figure it out. I have a few threads on here about it but just want to start over by asking ---- How does my 2000 XJ even know that a single cylinder is misfiring, its not like there is a sensor on each cylinder thats saying " hey im misfiring " Could there be something telling my ECU that cyl 4 is misfiring but it really isnt? Also if the engine light comes on for whatever reason how long should it take for it to shut off by its self ( given the problem is fixed )?

car5car

You have to do 2 tests to make sure mechanical condition of engine is OK:
compression and intake vacuum test.
then check ignition.

rideordie66

CEL may not go away if you don't remove the code after repair. Since its obd2 you can use a simple scanner/reader and clear the code once you fix the problem.

If you have a misfire then your engin should jump and hop around kinda while running and run poorly. Is that the case? If not then you may not have a misfire and some other issue.

Radi

By detecting the slight increase in RPM every time a cylinder fires. The PCM knows what cylinder it just lit up, if there is no corresponding "kick", or the rpm increase does not match that of the other cylinders, it considers that a misfire.
IMHO the 2000-2001 XJ's seem to have a problem sensing "misfires" that aren't. Some folks tear their hair out trying to fix those misfires and never find the cause. I think it's a PCM programming snafu, personally. (provided there isn't an actual misfire, that is)

Kalali

Maybe I give Jeep engineers too much credit but I would have thought the algorithm the PCM uses for detecting a misfire is more sophisticated than just the RPM differential across the cylinders. The PCM is continuously monitoring the fuel and spark delivery to each cylinder. If/when it detects a fault in spark/fuel delivery it could raise a flag. It is also monitoring combustion efficiency via the O2 sensors when in closed loop and that may also trigger non-cylinder specific misfire code. Of course this would all be relevant if there is a real misfire not a PCM software bug.

salad

Eh close. The PCM can detect if an injector is not functioning electrically, but thats about it. There is no flow meter in the intake valve or manifold. The XJs with distributors have no way of verifying that spark is actually delivered, and the coil packs on the later models are paired between cylinders. As far as the O2s go, there is only one in 88-99 that see the exhaust flow. 2000+ isn't much better with two sensors. Even if each cyl had dedicated sensors it would be difficult to detect anything but a constant problem due to lag in the exhaust stream (expanding gasses). The sensors would need to be so close that they'd burn up. If something isn't behaving right the ECU relies on the signal from the crankshaft position sensor like Radi said.

Kalali

So that's what I don't get: how does the PCM trigger a "single" cylinder misfire based on the signal from the crankshaft position sensor? It must either be getting some sort of a signal directly from individual cylinders which is unlikely or its comparing the signals from the CKP and the CPS (or other sensors) to derive an out of phase condition.
Glad someone asked the question. I've also been wondering about this for a while.

salad

I found this from some Mazda information, I suspect it's similar for the Jeep ECU:

"The ECU determines which cylinder is misfiring by using the crankshaft position sensor to detect variations in the rotational speed of the crankshaft. While the engine is running, the ECU monitors the number of misfires that occur at 200 crankshaft revolution intervals and 1,000 crankshaft revolution intervals. Calculating the misfire ratio for each crankshaft revolution, if the ratio exceeds the preprogrammed threshold, the ECU determines that misfire significant enough to damage the catalyst(s) and/or negatively affect emissions performance has occurred."

The fly wheel has position data into it, as it passes under the CPS the ECU can determine exactly the state that the engine's in. In other words, if the CPS doesn't pick up a "push" after it reads TDC for a given cylinder, the ECU knows that it's not working right.

CKP I believe is only used for fuel and spark (on 2000+) timing

Dcoy

[QUOTE=salad;1995186]I found this from some Mazda information, I suspect it's similar for the Jeep ECU:

"The ECU determines which cylinder is misfiring by using the crankshaft position sensor to detect variations in the rotational speed of the crankshaft. While the engine is running, the ECU monitors the number of misfires that occur at 200 crankshaft revolution intervals and 1,000 crankshaft revolution intervals. Calculating the misfire ratio for each crankshaft revolution, if the ratio exceeds the preprogrammed threshold, the ECU determines that misfire significant enough to damage the catalyst(s) and/or negatively affect emissions performance has occurred."

The fly wheel has position data into it, as it passes under the CPS the ECU can determine exactly the state that the engine's in. In other words, if the CPS doesn't pick up a "push" after it reads TDC for a given cylinder, the ECU knows that it's not working right.

CKP I believe is only used for fuel and spark (on 2000+) timing


Makes sense, just need to figure out whats screwing it up here and there. Drove it 200 miles today and it missed and needed restarted the last 15 miles of my trip.

Radi

That is the long and short of it.

The inline 6 has a relatively long crankshaft with some inherent torsional flex. The CKP is mounted at the back, necessitating some correction tables to calculate how much the crank is flexing when the forward cylinders fire.
Remember what the PCM is looking for is an extremely small rotational change and that event could easily be altered by mechanical factors.
Those correction tables are selected based on engine load, speed sensor and TPS input, sometimes by ABS sensor data...probably some other stuff too.
It wouldn't surprise me if some of those sensors could be doing their primary job just fine but be at the far end of "acceptable tolerance", with the cumulative errors screwing with the misfire detection.

One of the reasons Chrysler abandoned the 4.0 (and that Mercedes dropped it's venerable inline 6 in 1999) was difficulty meeting tightening emission standards without a top-to-bottom redesign. This is one of those difficulties.
When these engines were designed, "misfire detection" didn't exist and crank flex was only an issue when it was bad enough to mess with valve timing.

MtnHermit

Not likely to be the problem with your 2000, but I chased a misfire for 2-years before finding it was the wiring harness.
See this thread: https://www.cherokeeforum.com/f2/new...re-help-32066/

Dcoy

[QUOTE=Radi;1995645]That is the long and short of it.

The inline 6 has a relatively long crankshaft with some inherent torsional flex. The CKP is mounted at the back, necessitating some correction tables to calculate how much the crank is flexing when the forward cylinders fire.
Remember what the PCM is looking for is an extremely small rotational change and that event could easily be altered by mechanical factors.
Those correction tables are selected based on engine load, speed sensor and TPS input, sometimes by ABS sensor data...probably some other stuff too.
It wouldn't surprise me if some of those sensors could be doing their primary job just fine but be at the far end of "acceptable tolerance", with the cumulative errors screwing with the misfire detection.


Just wish I could figure out which one. I replaced just about all the sensors under the hood ( never heard of a speed sensor thou ) and crazy thing is that if I run my ac it almost completely prevents the misfire. I think it has to do with the idle because if I put it in neutral and rev the engine abit it wont do it either and the ac increases the idle a bit also. I also turned a set screw up on the throttle body that the linkage rests on and made it rev higher and it helped to but I didnt like the higher reving engine all the time.

salad

That's strange. When the AC is engaged the load on the engine increases. Do you have any clicking/tapping engine sounds, like a bad lifter? What kind of spark plugs are on there?

Dcoy

First tried champions then put in the recomended NGK's. The only sounds I hear is if i rev it up at the throttle body then let of it I can hear a light knock as it idles down, then once its down to a normaly idle it goes away

Dcoy

any new ideas? I plan on changing the valve springs on # 4 to see if they are weak.