Update. Did some more work on the Jeep today. Couldn’t get rear brake pressure. Pressure out of the master cylinder but nothing to the combination valve. Used a vac bleeder. Nothing. Finally found the issue. PO put a nail head in the line going into the combination valve. Removed that and then replaced the leaking wheel cylinders and shoes that the PO clearly knew were bad. Unbelievable. How cheap can you be.

Anyways concerning my drivability issue: while it was in the air I found that the o2 sensor harness was fused together where PO used electrical tape to fix what I’m sure was a melted harness. Repaired the harness with some good heat shrink tubing and new wires. Reset the pcm. O2 code still came back though however now instead of 0.00 at times or 5.0 volts I see 1.0 volts flat. So the sensor is maxed out due to the misfire on cylinder 4. Gonna do a compression test next after I try switching the injectors around.

 

The flat 1V on your oxygen sensor is not normal. It should not be stuck at 1V like that because of a misfiring engine cylinder. It should always fluctuate and not be going above 0.9V... 1V is out of specification for the sensor

I suspect the oxygen sensor may be faulty with an internal short. This makes sense since you did find its wiring was melted like that.

It could also be in any part of the wiring too though. Would need to trace the wiring...

What does it say the voltage is for this sensor when you disconnect the oxygen sensor? Does it still say 1V for it even with the sensor disconnected?

 

Im sure the sensor is bad. It is my understanding that the o2 sensor voltage rises with decrease in oxygen content of the exhaust gas. With a misfire I would expect this since raw I burnt fuel is present in the exhaust mixture.

I will be grabbing a known good sensor next time I hit the JY

 

Explain to me how to find a known good sensor in the junkyard. I'm a little unclear on this.

 

It requires the item to be fitted to and tested on your engine

its a bit frustrating when a 25yr old OEM part is more likely to be good than an out of the box new replacement

 

Usually I ohm test any sensor from the JY. A lot of times I’ll look for a car that’s been totaled and go from there. For me, it’s a way to save money when a oem sensor costs 10x times more than a used one.

 

I think junkyards are absolute gold mines. I picked up a set of four new tires for my ZJ from the junkyard. Two of them were actually basically brand new. The other two probably had close to 90% of the tread still left on them. The junkyard tire guy even installed them for me. I think it was only something like $15 to change all of them. Grand total I pretty much ended up with a good full set of brand new name brand tires installed for less than the cost of any one of them when new.

I got what looked like a brand new blower motor for my WJ from a junkyard. I think it was $17.. compared to the stores which were wanting close to $100 for a new one.

I got a bunch of back-up sensors for pretty much all of them. They are just sitting in a toolbox for whenever they may be needed.

 

Update:

Tested compression 150 psi all 6 cyl. Replaced cyl 4 injector with new oem. Still misfiring below 2000 rpm. I can watch the misfire stop when I rev above 2k.

Performed cam sync adjustment using our scan tool and watched it move from our of range 12 degrees to perfect zero degrees in range as I turned the cam sensor while running. Seems to have more power on the highway. Still missing at idle though

 

Gotta love JY finds man. I’ve saved so much money in the past.

 

There are only a few things to cause a misfire.

It is one of three... air/fuel ratio, ignition, or compression. That is all it could be.

You have totally and completely ruled out one of those. The compression is normal.

It can only be one of two things now. It is either the air/fuel ratio or spark.

What is more, you know exactly where to go and look for it.. on the problem cylinder.

How do all of the spark plugs look? Are you sure the data from the scanner is accurate? Scanner data is great. It is still no substitute for human intelligence though. I would verify the problem cylinder by comparing its spark plug to the others.

The PCM may get confused if a misfire is happening on another cylinder during its compression stroke, and report the misfire wrongly as happening on the firing stroke of the cylinder adjacent to it.. is a possibility.

 

Noah911,

Spark plugs all appear in good shape. I verified that the cylinder misfiring is #4 by doing a power balance and noted that cylinder 4 doesn’t cause the rpms to drop when the injector is unplugged. However if I hold the rpms above 2500 and do the same test I can hear an audible change in the engine. So it appears that cylinder 4 is only misfiring at low rpm.

 

These are my thoughts on it right now.

You are using a scanner, right? Does your scanner have the ability to measure fuel injector pulse width modulation? There may be an obvious deviation? It is based off of sensor inputs. It could wind up being a cylinder specific issue from a sensor fault which is normally thought of as controlling things in a manner to have more of a global type effect.

The sensor which comes to mind for me is the Crank Position Sensor (CkPS). The reasons why I think the CkPS could be capable of such an effect like this is because the injector pulse width modulation and ignition timing is directly related to engine RPMs. Specifically, one of the things that helps the PCM to determine the injector pulse width modulation and timed ignition is the Engine Speed Sensor - AKA.. the Crank Position Sensor. The CkPS senses the crank angle of each cylinder to know when the piston is at top dead center, and converts it into a pulse signal for the PCM. What if there were pulse signals missing at a certain specific point? The PCM would be blind to know what was happening during that specific position of the crank, and would also skew the percieved engine speed during that part of the cycle. There is a target wheel on the crankshaft with 58 or so longer points on it than the others for producing a higher electromagnetic voltage whenever these long points pass by the CkPS. What if some points along the target wheel are not normal or are broken in some way... A scanner can show the CkPS waveform for you to see the exact signals it is producing; when, where, and how.. to know this. Having a problem like this would adversely affect both the fueling and ignition at a very specific point in the cycle.

With increased engine load and RPMs the gap and difference of pulse width modulation (and timing) is effectively narrowed. The fuel density to manifold air mass density goes from being completely non-linear to become fully linear in the upper range. Since this is used in determining the duration of microsecond bursts of fuel spray to match air density.. it helps to explain why the issue would be worse when the engine load is low.

(Edited):
I edited the first paragraph above in case you already read it, as I do not know if it is possible to see the injector pulse width in milliseconds/microseconds individually on each specific cylinder and injector like that or not?

 

You make a good point and this has been on my mind for a while however, one thing that makes me think otherwise is that I inspected the flex plate toner rings. If you could call then that. Little holes outside the perimeter of the flex plate. All appear intact. Flex plate doesn’t appear cracked and bolts are tight. Also the issue comes and goes, so if I’m driving and I come to a stop light, or I get parked somewhere it will run perfectly for a bout 3 minutes before starting to misfire again.

 

Dang.. I had something typed up when my phone died and lost it. Let me try this again;

Let me know if you feel I am leading you astray in any way too. I feel most of the more commonplace simpler resolutions have already been explored. It may be time to think differently for the other less likely reasons. It is still always possible for it to be something super simple and most basic to keep in mind though. With this, I go back to the spark plug. Have you already tried swapping the spark plug out of the problem cylinder with any of the others yet? For cylinder specific problems, ANY part that can be swapped out with the same similar parts coming from the other cylinders should definitely be tried at this point... Looking very closely at the spark plug it could show as having a microscopic hairline crack on the porcelain (even on a brand new spark plug).

The XJ does not have the coil-on-plug designe to try that with swapping those around does it? Is there any other ignition component that is able to be swapped out with the other cylinders? Spark Plug Wires... Have you taken the time with a spark tester somehow to fully scrutinize the sparks over a good length of time and through all of the different sustained RPMs?

For the CkPS. It is.. of course, a Hall Effect Sensor producing a square digital wave with magnetism. You looked extremely closely and surely saw no chips or cracks or anything abnormal on the crankshaft target ring (pulse wheel)? There are few other things to consider with the cause effect relationship where the CkPS signal is being produced.

The first thing I think of is the space or distance between the CkPS magnet and the target wheel pulse ring may not be fully proper. The CkPS should be in a fixed position. However, the flywheel is not. The flywheel rotates with the crankshaft. It is possible the crankshaft is loose or flexing around its rotation. It could be doing this and is intermittently too far out of specification to the distance from the CkPS magnet for signal purposes happening at a specific point during the cycle. The CkPS could be somewhat loose and incorrectly installed itself to be aware of and keep in mind too. It would not take much with these types of issues to cause excessive crank or flywheel movements and vibrations to occur.

The next thing I think of for the CkPS is dealing with Electromagnetic Interference (EMI). The CkPS wires should be well shielded. EMI can occur from the alternator, secondary ignition wires, or even the engine/transmission grounds. I would not want to go too deep into this possibilty though. Mainly since I am not sure EMI would produce an effect intermittently happening on the exact same cylinder like that all of the time. And because this type of issue may be hard to narrow down aside from the obvious reasons of having stripped wiring, overlapping secondary wires, a known failed alternator diode, etc... It would be worth it taking a very cose look for something like this anyway. It would probably require an oscilloscope to see the occurence and effects of EMI.

An oscilloscope would be a great tool right now for comparing and to see synchronization between the CkPS and the Camshaft Position Sensor (CPS). The CPS may be the problem? It could be the wiring for it is stuck or pinched in some way. I would look very closely for everything at the CPS and CkPS for all of the wiring at the connections and through the harness etc... The CPS operates similar to how the CkPS does producing its signals. The two of them need to be in sync with each other in order for the system to work correctly. They both have rings, teeth, and spacing components. Examining the reluctor ring may show faults such as chipped, cracked, bent, vibrating, or out of round mechanical parts possibly interrupting the integrity of signals being produced. Dual side by side oscilloscope readings would be the most ideal solution for verifying the CPS and CkPS synchronization. This would still not be easy, since it would require a good known reference to show the two sensors working together for comparisons. These graphs are out there though, and it could be done. Using an oscilloscope it would be easier to find a problem specific if it were only just happening individually on either one of these sensors alone.

I think doing a cylinder leak-down test is in order for the problem cylinder now, in addition to the compression testing you have already done.

 

I also say to point an unlit propane torch or spray a can of carburetor cleaner all around everywhere at the intake manifold.. just to be certain it is not a vacuum leak as this could certainly produce intermittent misfires.

In addition to a vacuum leak possibilty there is a different type of leak which could occur at the fuel injector. I say to try dowsing the whole area all around where the injector mates up onto the engine at the problem cylinder with ether or carburetor cleaner as well, for checking to see if there is a good seal at this location.

If there was oil in the cylinder it may have been able to thwart what would have otherwise been an abnormal compression reading on the cylinder. Also, you can do a leak-down test with the piston oriented at both, the top dead center as well as when the piston is at bottom dead center positions. Some people will argue a compression test is just as good as doing a leak-down test. Or, at least they will say it is almost as good. This is not true though. Performing a leak-down test is the better test . . Period. There are cases of people, even well trained mechanics, who have performed repeat compression testing which shows normal (or very near normal) readings on all cylinders. A good friend of mine is an independent mechanic. He routinely has people bringing their vehicles in to see him after they have already tried everything, and have taken their vehicles in to see many other multiple mechanics shops who have attempted diagnostics and repairs. He prides himself on fixing problems when the other shops cannot. A compression test can be normal.. with a leak-down test showing the issue. Just another something to be aware of and keep in mind is all.