My 1998 XJ8L has a rough idle when it reaches normal temp, like it is a misfire.

The MIL is not lit and no codes are pending.

Exhaust gas from bank B is noticeably hotter than bank A. Under heavy load, I noticed it's starting to be a little rough.

The serpentine belt shredded just before this issue started and the idler pulley was destroyed - not sure which one caused which. However, when that happened, by the time i was able to pull over to stop, the coolant got hot and popped the hose feeding the coolant sensor, so sprayed coolant over the bank B side of the engine.

I've replaced 2 ignition coils that gave me mis-fire codes and shuffled all the others around to see if that changes the temperature from bank to bank.

I also switched round the Ignition control modules to see how that affected things. No change in the temperature difference from the exhaust pipes, the "feel" of the rough idle has changed a little (but I may be trying to convince myself of that)

I put in new secondary tensioners a few months ago, but left the primary ones as the chains seemed tight. I heard some noise just before I lost the belt and pulley, but it's all quiet again with the new idler pulley.

Any ideas?

 

You may have caused a meltdown of the catalyst on one bank. Check the cat temp. top and bottom with a infrared heat gun.

 

Would a problem with the catalytic convertor not be indicated by the Oxygen sensors upstream and downstream of the cat? Why would that not generate a fault code?

 

I would have thought so, but I had it happen on one of my cars a few years ago (not Jag) and there have been a few instances on the forum. When coils fail while driving you send unburnt fuel to the cat which then overheats while burning the excess fuel.

 

So, I searched on-line to see if a failed or compromised catalyst would cause the rough idle. i found this pretty informative article:

Catalytic Converter

It seems, that is plausible, by creating a back pressure to stop the exhaust gas escaping. Going through the quick checks - it suggests that there should be a significant difference in fuel trims:

RPM Short Term Long Term
Bank 1 Bank 2 Bank 1 Bank 2
2500 5.468 4.688 -3.906 -5.468
648 1.563 0.000 -3.125 -1.563

Are these different enough? It does not seem a large difference to me?

If I look at the voltages from the O2 Sensors

RPM Upstream DownStream
Bank 1 Bank 2 Bank 1 Bank 2
2500 0.21 0.41 0.16 0.14
648 0.77 0.78 0.90 0.92

So there does seem to be a small difference, upstream on the voltage from Bank 2, but again, is that a big enough difference.......

Recapping what I see/feel - when the engine is cold, idle is good and engine is silky smooth

Once she is warm (Coolant temp > 180°F), then I experience the issue

The roughness is "erratic", so not constant or rhythmic, worse when in Drive than Park , or Neutral. It will be smooth, then "bump", then a series of "bumps" - which are not typically noticeable again unless the engine is truly under load (like climbing a hill)

Any more insights would be very helpful

 

I don't think I saw the cold/hot comment initially. Try disconnecting the MAFS plug and cleaning it with electronic cleaner. Do the same with the temp. sensor plug on the crossover pipe.

 

Following the advice:

I had a spare MAFS unit, so I switched them over, cleaning the sensor with MAF Cleaner, and the connectors with Electronic Cleaner.

The temperature sensor, I gave a good clean of the contacts.

Took it for a run, when it warmed up, there was some improvement. Larger gaps between bumps on idle. Still there when the engine is under load - mid range revs. If i press a lot harder on the throttle it seems a little smoother.

So, tonight, I replaced the the coolant temperature sensor. I have a part load breather pipe on order also, which i'll switch out once it arrives.

 

well, the coolant temperature sensor did not make a difference to the rough idle, but, there was a change in the measured temperature. The coolant temp peaked at 205°F, when previously it's maxed out at 195°F

 

So in trying to replace my part-load breather (and the vacuum hose from the throttle body to the Purge Valve) I discovered that a little rubber hose that's been attached to a plastic fitting for 18 years grips the plastic tighter than the strength of the plastic - the fitting on the purge valve snapped, so now I wait for a new purge valve.

Thank the Lord for eBay!

 

While you're in there you should replace the thermostat tower if it hasn't been done, preferably with aluminum, and the thermostat. They have a preference to stick in the closed position when they fail. The water pump is also a weak point.

The crossover pipe can also crack, and the hoses should be checked.

 

So I got the Purge Valve all changed out, I replaced the part load breather and vacuum hose to the purge valve.

I left the thermostat alone for now, but I have the aluminum tower in there already.

Now the exhaust gases are much closer in temperature, but, the idle is still a little bumpy after warm up - but a massive improvement.

Any more ideas?

 

So, i'd been thinking about RJ237's comments on the thermostat. I've not seen any "high" temperatures, except when the drive belt broke. However, with the benefit of 20/20 hindsight, I was thinking maybe the temperature in the engine bay was a little warmer than I would expect - so could it be that even though the measure temp was OK, i am running too hot around the actual combustion chamber?

I figured i'd give it a shot and changed out the thermostat last night.

Sadly, no improvement

 

I never did come back and update this thread for what I found. It ended up being a spark plug - the one where the Ignition Coil failed after it's coolant spray. Never occurred to me that the coolant would go down the thread and trash the spark end of the plug!

 

Always good to find the cause and solve the problem, thanks for finishing up a thread with the final results.
Happy riding.