2003 Liberty 3.7L sport, Reman engine from Tri-Star, p0304 code

[dtennes]

I'll share what's going on, and if anyone has anything to contribute I will be most grateful.

Back ground:
1. 2003 Jeep Liberty Sport 4WD 3.7L
2. Blew a ring and trashed the engine
3. Had a Goodyear shop (excellent credentials, and past personal experience with them so I trust them) install a reman according to VIN and everything runs nice and smooth... until the PCM throws a P0304, Cylinder 4 misfire
4. The mileage between tripped codes, always P0304, varies between 12 and 110 miles.
5. The one pattern I noticed is that it seems to happen after the vehicle has been idling warm for 4+ minutes.
6. If I simply erase the stored P0304 code with an Antec diag handheld, the code resurfaces again somewhere between 20 and 40 miles, but only after I idle at a long light or pulled over in park.
7. When I detach the neg battery cable it seems to go 60+ miles before the p0304 code gets thrown.
8. When the code gets thrown I have to reset the PCM or the engine starts to run badly. I've got about $10k into the motor and the exhaust system, so I am not at all willing to risk blowing another motor.

Engine parts status:
1. New engine. 2 conversations with a TriStar engineer after all of the following was done, and I'm 99.7% confident that the motor is good. When the PCM isn't complaining it runs like a sewing machine.
2. The shop checked all vitals and everything is dead on spec (compression, fuel trim, everything)
3. All new coils, injectors, NGK Plat plugs
4. Complete exhaust system replacement, 3 new cats and pipe end2end
4. Eventually ran two bottles of Techron Complete fuel system cleaner through the tank as directed
5. Then ran one bottle of LiquiMoly Ceratec through the tank as directed
6. All original sensors were reused (the reman didn't include that and the shop didn't ask... so the MAP, TPS, IAC, IAT, etc are all in transit as I type this), and will be replaced by an auto electronics expert at another good shop I've used (time for new eyes)
7. The electrical harnesses are all original, and yes there is evidence of age, but no shorts have been detected by the 1st shop. We'll see what shop #2 says.

One last observation: I've got a concern that the existing wiring may be experiencing electromagnetic interference. I am no electronics expert but I had a problem in my man-cave with (a) unshielded signal wires being crossed over each other, and other wires being near electromagnetic coils in DC motors and past strong fluorescent lights... and I wonder if there may be any experiences out there regarding rogue EMI.

 

[klc]

You should replace those platinum plugs with NGK copper. I’m not saying it’s the cause of your issue, but these Jeeps are particular about their electrical parts. Also check the engine grounds for corrosion. A number of problems can be caused by bad ground wires.

I’m an electronics geek myself, and I have no evidence that Libertys are more susceptible to random EMI interference.

 

[LibertyTC]

Yes use the correct spark plugs !
Ensure gap is perfect.

You must be registered for see images attach

 

[dtennes]

You should replace those platinum plugs with NGK copper. I’m not saying it’s the cause of your issue, but these Jeeps are particular about their electrical parts. Also check the engine grounds for corrosion. A number of problems can be caused by bad ground wires.

I’m an electronics geek myself, and I have no evidence that Libertys are more susceptible to random EMI interference.

@klc Thanks for the ideas.

The garage used copper Champion plugs, which I replaced with the NGKs, so the problem originally surfaced when copper plugs were installed.

Regarding the plugs and the grounds, consider that the #4 misfire is specific to #4. A few more forensic inspections that rule out system-wide effects:
1. Cannot be anything in the fuel lines because both rails are shared across 3 cyls each, so logically multiple misfires would occur. So a fuel pump issue and all feeds to the injector is ruled out.
2. In addition to the immediate preceeding observation, the injector for cyl 4 has been changed twice, and in between changes the injectors/coils/plugs were swapped with cyl 1, which has remained problem-free throughout. So the parts are good.
3. I read about, and then discussed the possibility of a bad/collapsed valve on cyl 4 with various mechanics and the engineer from Tristar, and the consistent observation is that a bad valve would be noticeable the moment you start the engine. If the p0304 code is cleared, the motor goes back to smooth operation. So a bad valve seems to be an extremely remote possibility.

 

[dtennes]

@ LibertyTC: Thank you for the observation. Yes, I definitely gapped the plugs dead on at the 2003 3.7L spec. And, I checked the 6 Champion plugs that I swapped out for the NGKs, they were new and were all gapped to spec.

 

[dtennes]

Another system-wide replacement part done (adding to the list of 7 above):
8. All new fuse box relays. Two different kinds, some have nothing to do with ignition, but for $10/ea I replaced all of them... *and* double checked the fuses, which were fine of course otherwise things don't work at all.

 

[dtennes]

I exposed some of the work done to date by various garages and would appreciate some observations. I am going to break these things out into separate sub-posts to keep them each isolated...

 

[dtennes]

"Mysterious heat sink block".
Does anyone know what this is and why it is necessary? The location is the inside wall of the left front wheel well, under the power distro center (fuse box).

 

[LibertyTC]

That block is the voltage regulator.
Live data from a shop that has a Snap On modis ultra shop scanner can view random or particular misfires etc..

 

[dtennes]

"Missing Ignition capacitor"
Apparently there is supposed to be one of these installed (see capacitor_b1 photo) on the left engine compartment firewall (see cpacitor_b2 photo) according to the manual (see capacitor_b3 photo).

The point of a capacitor (I'm no electrician but I can understand definitions/explanations) is to provide a stable charge delivery of _____ amps at the expected level of voltage.

There's a reason why these things get included in a circuit, and mine seems to be missing from the engine compartment. The hypothesis is that a mechanic ended up with a loose cable/connector, couldn't figure out where it went, and "since the engine seemed to be running fine at the time, we'll just cut it off and forget about it".

If that hypothesis is truth, then the new/replacement capacitor needs to be screwed into the firewall where it belongs... but I have no idea where to splice in the connector whip. The anticipation is that it is a single-wire splice because there is only one connector pin exposed in the connector socket on this part.

 

[LibertyTC]

I believe that capacitor is there to reduce radio static, noise.

 

[dtennes]

That block is the voltage regulator.
Live data from a shop that has a Snap On modis ultra shop scanner can view random or particular misfires etc..

Ok, that's excellent news. Is it an after market gizmo that isn't necessary, and can it be removed safely? If other 2003 Liberty 3.7s have the block, and no misfires, then I'm neutral... but if this is a "nice to have for the benefit of some garage mechanic", I would like to minimize extraneous circuitry in the effort to get to the root cause of this cyl4 misfire problem. Thoughts?

 

[dtennes]

I believe that capacitor is there to reduce radio static, noise.

That's what I got from what I've read too. So is that capacitor connected to the radio or to the ignition circuit?

 

[Ksat]

You might want to do a compression test in #4 just to confirm you have the proper pressure there. Remember to disconnect coil primary and floor the accelerator and crank for about 5 seconds when you do the test.

Another possibility is you have an intermittent bad connection in the wires going to the #4 coil and/or fuel injector from the PCM. Perhaps a connection breaks when things get hot enough under the hood (I had this happen on my car with the fuel injector wires). The problem could also be in the PCM itself.

Does the engine ever feel like it's missing on a cylinder when you drive it? If so, that would help in the diagnosis efforts. It might not hurt to check what the short and long fuel trims are, if you know them, too. Using mode 6 on a scanner would show the number of misfires that cylinder has undergone in comparison to the others. Feel free to post that info, if u can get it.

 

[LibertyTC]

That's what I got from what I've read too. So is that capacitor connected to the radio or to the ignition circuit?

The 2003 Liberty factory manual (8w wiring schematics) should identify the circuit. see link below.
http://www.colorado4wheel.com/manuals/Jeep/KJ/2003JeepKJServiceManual.pdf

 

[dtennes]

You might want to do a compression test in #4 just to confirm you have the proper pressure there. Remember to disconnect coil primary and floor the accelerator and crank for about 5 seconds when you do the test.

Another possibility is you have an intermittent bad connection in the wires going to the #4 coil and/or fuel injector from the PCM. Perhaps a connection breaks when things get hot enough under the hood (I had this happen on my car with the fuel injector wires). The problem could also be in the PCM itself.

Does the engine ever feel like it's missing on a cylinder when you drive it? If so, that would help in the diagnosis efforts. It might not hurt to check what the short and long fuel trims are, if you know them, too. Using mode 6 on a scanner would show the number of misfires that cylinder has undergone in comparison to the others. Feel free to post that info, if u can get it.

Compression tests on cyl 4 have been done at least twice now and every time it comes back normal. That is the reason why the possibility of a bad valve has been rejected.

When the PCM stored codes have been erased (using an Ancel AD 310 handheld plugged into the bus), the engine runs smoothly. No clicking, grinding, pinging, nothing but the hum of the engine.

The scanner only shows that the cyclinder misfire count has reached some threshold. Any recommendation on buying a better scanner without breaking the bank?

Regarding the "...things get too hot..." the mechanic who did the install left all of the wiring harness branches laying all over the valve covers, against the trans bell housing. I've tried to lift everything away from the engine but that mechanic also managed to either break or lose all standoffs that would manage harness contact. Please elaborate on your "things get hot" experience, that may be significant here.

I have a brand new PCM from Flagship 1 installed, to rule out that possibility of bad PCM. Exact same p0304 on both (old and new) PCMs.

 

[dtennes]

The 2003 Liberty factory manual (8w wiring schematics) should identify the circuit. see link below.
http://www.colorado4wheel.com/manuals/Jeep/KJ/2003JeepKJServiceManual.pdf

...that's the same manual I've been decyphering for the last week.

 

[dtennes]

@klc Thanks for the ideas.

The garage used copper Champion plugs, which I replaced with the NGKs, so the problem originally surfaced when copper plugs were installed.

Regarding the plugs and the grounds, consider that the #4 misfire is specific to #4. A few more forensic inspections that rule out system-wide effects:
1. Cannot be anything in the fuel lines because both rails are shared across 3 cyls each, so logically multiple misfires would occur. So a fuel pump issue and all feeds to the injector is ruled out.
2. In addition to the immediate preceeding observation, the injector for cyl 4 has been changed twice, and in between changes the injectors/coils/plugs were swapped with cyl 1, which has remained problem-free throughout. So the parts are good.
3. I read about, and then discussed the possibility of a bad/collapsed valve on cyl 4 with various mechanics and the engineer from Tristar, and the consistent observation is that a bad valve would be noticeable the moment you start the engine. If the p0304 code is cleared, the motor goes back to smooth operation. So a bad valve seems to be an extremely remote possibility.

I ordered the pure copper plugs as recommended by "LibertyTC".

 

[dtennes]

I ordered the pure copper plugs as recommended by "LibertyTC".

..and also recommended by "klc"... thanks everyone.

 

[dtennes]

Has anyone ever created a "bypass circuit" directly from the PCM to a particular coil/injector? The thought is that if a bypass is possible, and I can run wires to the coil/injector on cyl 4, that would either confirm/refute whether there is a short buried in the harness somewhere.

I would rather not speculatively tear apart the ignition harness if it can be avoided.