2005 TJ 4.0 - rough idle only after warm up

[RMac80]

2005 Wrangler X with 105k miles
4.0 automatic

I have had quite the saga on this thing trying to solve misfires, idle surging, and rough idle. See this thread:
https://wranglertjforum.com/posts/1672902/

After a couple weeks of driving since the last time I worked on it, I think the misfires are solved. I was able to solve the idle surging a while back.

But I still get the lope/rough idle. Unlike before, I get no misfire counts in live data, I get no CEL, and it doesn't try to stall and die, but it is loping like a cammed out 396. Runs smooth, quiet, and fantastic at first, but as soon as it warms up I get the lope.

And it is intermittent. Some drives I take it runs great, other times it does this.

So I had been driving around with live data on my misfire counts. Today I started driving around with upstream o2 voltages and fuel trims for both banks AND I CAUGHT IT IN THE ACT!

Bank 1 fuel trims are negative on the drives when it acts up. Short term is big negative like -12-14% and long term is getting down around -4% negative. THis is through the whole drive, but when the temp gets up aroun 210'F the computer starts listening to that bank 1 trim and starves the thing for fuel -> lopey idle.

Here are my thoughts:
-B1 upstream o2 sensor - but it was replaced a few months ago, and the voltage is in range and fluctuating like it should, and the temp is up where it needs to be. Heater and sensor grounds are good. I can pull/manipulate this wire when it is loping with no change, haven't tried unplugging it, though...

-MAP sensor - I need to check the live data on this, but if this was bad wouldn't it be causing negative trims on both banks?

-I also plan to check live data on air inlet temp sensor and engine coolant temp sensor just to be safe...

-Clogged cat on bank 1 maybe?

-Leaky intake or exhaust at the manifold/gasket somewhere around cylinders 1-3? I don't hear or smell any exhaust leaks, don't hear any hissing at the intake...

-Crank position sensor? I have had a valve cover gasket leak (now fixed) that has been dumping oil all over this sensor and its wiring since before I owned the jeep so this is suspicious to me...

-The dreaded PCM failure since it is intermittent? Please no...

What are your thoughts, my jeep gurus?

 

[macleanflood]

My fuel trims only vary by a few percentages. It was 27 out yesterday and my IATs were 79. Check the MAP sensor reading...mine measured in milibars and it was around 1000. At around 700 ft above sea level. If your O2 is reading correctly then no point in replacing it. Get an IR gun and measure CAT temperature to see if it's any good.

Keep up the good diagnostic work. I thoroughly think folks need to watch and graph all their PIDs when their Jeep is running well so you know if something is off baseline.

-Mac

 

[JKP]

You need to ask Mark @Wranglerfix if that's a PCM issue.

 

[RMac80]

Checked live data on the ECT, IAT, and MAP today. Everything seems fine.

Emailed wranglerfix and got immediate response, seem like nice folks. We will see…

Going to use a laser thermometer and check the exhaust and cats tomorrow…

 

[RMac80]

Temperature of the cats are both the same within 40'F of each other and almost exactly the same as the exhaust manifold temp. So probably not a clogged cat.

I tightened the bank 1 O2 sensors up and downstream, no movement. I unplugged and re-plugged the wiring and made sure wiring was routed away from heat sources. I tightened the intake and exhaust manifold bolts (that I could get to easily), little to no movement. No change in behavior.

Wranglerfix does not seem to think these symptoms are PCM related (thankfully)... But I appreciate their time and attention when I contacted them!

I currently have a couple theories, let me know what you think.

1) Still haven't ruled out crank position sensor. Seems less likely since I found the bank 1 fuel trims going negative correlate to my lope/rough idle.

2) There was a lot of oil crusted on the tops of piston #1 and #6 from the leaky valve stem seals, running 87 octane fuel with that much deposits maybe I am getting some detonation? Will try adding 93 octane today since I am down to half tank again. I don't think this is likely, if it was happening I would probably see a knock sensor CEL/code.

3) It is possible I have a bad injector at 2 or 3 which is dumping fuel causing the trim to go negative for bank 1. I did not replace injectors, I only cleaned them. And in the process I swapped injector 1 and 5 (due to P0301 misfire code). It could be that #2 or #3 injector was always dumping too much fuel causing the whole bank 1 trim to go negative, and the misfire just showed up in #1 due to the oil leaking in that cylinder. To rule this out I will swap 3<>6 and 2<>4 injectors. That would put the original bank 1 injectors all on bank 2 now and all on different coil/firing sequences. Or I may just bite the bullet and get the new/refurbished OEM injectors. I feel like this is the most likely scenario given the data/history.

4) I never really ruled out ignition coil wiring, just changed plugs and the coil rail. If all else fails I could go down that route.

 

[JKP]

The 4.0 doesn't have a knock sensor.

Take a multimeter and measure the resistance of each injector. This will tell you if the coil inside is going bad. They should all be within a few ohms of each other.

 

[macleanflood]

On my 97 I purchased a balanced set of rebuilt Siemens injectors for my 97. Discount code Jeep gets you a little percentage off.

Mine were old, looked rough but we're functioning correctly. One less thing to worry about in my mind.

-Mac

 

[RMac80]

HOPEFULLY FINAL UPDATE from me on this thread!

I did indeed have to replace the injectors to finally fix my issues. I definitely don't regret any of the work performed earlier, as it seemed to be a combination of issues coming together. My final diagnosis is:

1) Injector 3 was dumping too much fuel. It was not leaking down and the resistance was in spec (but lower resistance than my other injectors. I did not detect/fix this earlier in the process because I cleaned and rebuilt my original injectors and put in new o-rings. At the time I swapped injectors 1 and 5 because P0301 had me focusing on cylinder 1.

2) The fuel trims for bank 1 were going negative on bank 1 due to injector 3 dumping too much fuel. It would eventually pull so much fuel from bank 1 that...

3) cylinder 1 would start to misfire probably due to the very excessive oil/carbon build up from a long time valve stem seal leak.

4) Any time I would perform any work to the system I would reset the PCM, which would reset the fuel trims, and it would take a week or two of occasional driving to get the fuel trims low enough to cause the P0301 (rough idle would be present first before the code was thrown)

Been over 3 weeks now with the rebuild set of flow matched OEM injectors (opted for this instead of 4 hole upgrade). Got them from K Suspension, awesome product and great people to work with!

No rough idle, no P0301 code, and live data is showing my fuel trims on both banks to be right around zero at idle.

 

[RMac80]

The 4.0 doesn't have a knock sensor.

Take a multimeter and measure the resistance of each injector. This will tell you if the coil inside is going bad. They should all be within a few ohms of each other.

This is what led me to order the rebuilt flow matched set from K Suspension. I had done this measurement before but was only checking for them to be within spec, not comparing to the other injectors... THANK YOU