Hello everyone. I have a stock '99 4.0 Cherokee with automatic transmission. In the morning while I am driving east the idle stays elevated after I stop for the light. I know heading east hs nothing to do with it but that is the direction I drive in the morning. When I am headed home, west, it doesn't seem to do it.
All kidding aside,, the jeep idles around 1800 rpm when I stop for the one and only traffic light on my 30 mile commute. Shutting the engine off and restarting allows the jeep to idle at 700-900 rpm. I have to shut it off and start it again after I reach the parking lot because the idle goes high again.

It seems we should be able to rule out the direction and the time of day but what else sets the idle? Jeep has 125K

 

You know, I am currently diagnosing the same issue. Most people will say it's a vacuum leak. It's absolutely not a vacuum leak. I have gone through mine, swapped the intake manifold, replaced the gasket, capped off all vacuum ports, replaced the throttle body gasket and throttle body shaft seals. I have checked a hundred times for a vacuum leak. I've changed the TPS a couple times with another Mopar TPS. I'm about to swap the ECU with a known good unit just to see if it changes anything. But yeah, I'm in the same boat.

 

The idle RPM is managed with the throttle body. It might ben your idle air control valve that is stuck of gets stuck after driving a while. You want to clean that, and the hole it's in, out.

It's throttle body related, I think.

 

See, it doesn't make sense because we can both shut the engine down and turn it back on and it goes back to idling normal. I've also swapped the throttle body and the IAC. I've been swapping these parts with known good parts from another running XJ. I'm fairly certain it's something to do with the ECU.

 

When you are driving before your only traffic light...is your cruise control on? If so next time it revs..stomp on the gas and then rapidly let off...see if your idle goes to normal

 

Hmmm. A connection to the cruise control system is worth exploring. In theory, stepping on the brake should disengage it, but maybe it's hanging up somehow.

 

I've researched this extensively on several Jeep forums. It seems like every one of these threads ends without an answer.

This problem has been intermittently happening to me since around January of this year. I swapped intake manifolds because I figured if I was going to change the intake gasket I might as well get the 99+ manifold. The intermittent high idle persisted.

I changed the TPS after having shifting issues, which completely solved my late 2-3 shift. The intermittent high idle persisted.

I put another IAC and TPS on from another XJ. Still happens.

I swapped to an AX15 in late Feb, and it continued happening. I'm only adding this info because I was wondering if the auto ECU and manual transmission was causing some idle confusion in the ECU. And I'll actually acknowledge that it does. If I downshift and engine brake, it'll idle very low when I come to a stop. Like 200 rpm. I have talked to others who have swapped theirs and it's pretty common for that. I can blip the throttle once and it goes right back to idling normally.

I haven't quite nailed down what exact conditions it needs for a high idle though. It doesn't seem to do it until closed loop. I can also make it stop (while driving) by leaving it in gear and cycling the key from run-acc-run. It's only a quick, momentary loss of power, but it seems to reset the IAC. I can clutch in and the idle returns to normal immediately. This tells me it's an IAC issue. No codes are being stored other than the ones for the TCM. So the ECU is commanding the IAC open, but the question is why? What's making it want to get stuck at a high idle like that?

Something I've dealt with before on other cars about this same age is electrolytic capacitors blowing out in the ECU. It's usually an easy fix, so I might need to crack the ECU open and take a look.

 

I state this because occasionally after driving home on cc, when i hit the sidestreets, i idle at 1200 plus until i stomp on gas then goes back to 600/700. I dont know if its the cc cable sticking or main throttle cable sticking because i cant make the minute throttle adjustments with my foot that the cc can do but it usually doesnt happen when i drive by foot only

 

Yeah I could definitely see that happening. That's not my situation, though. It could very well be what's happening to OP though. They might want to give that a try.

I have pulled over when it's idling high and tried to shut the throttle body further, but it's already fully shut. I've done harness wiggle tests, whacked on the MAP, IAC, TPS, etc. It continues to idle high until I shut it off and restart it.

 

I don't use the cruise control, but since I have to apply the brake to stop at the light, especially when the idle is high it may not be cruise control related.

 

Where is Augusta (fish). I stay in Millen so maybe its a Georgia related issue.

 

For years, I was chasing a similar issue with my idle being elevated at different times. Then my dad gifted me a code reader that actually connects to my computer and the idle that shows in the cluster is not the same as what the reader shows when viewing live data. In live data mode, the code reader shows the idle is fine, around 800-1000. But the cluster needle hovers around 1300-1500 at times. I think it's an issue with the cluster.

So I would recommend you check your idle with a code reader that displays real time info.

 

Thanks for the input but it doesn't feel or sound like the cluster is reading wrong. There a physical difference in an engine that is idling at 800 rpm vs 1800 rpm.

 

I say Augusta (ish) because I'm on the SC side of the Savannah River. So near Augusta, but not in GA.

 

Still haven't had the chance to dig into my XJ's computer yet. The last thing I want to do is go from an annoying high idle to not running at all. Here shortly when I get off work, I'll have a week off. Perfect time to start the fiddling.

In the meantime, I did some digging and found this:

https://wranglertjforum.com/threads/...bad-pcm.13876/

The TJ has a VERY similar ECU as I recall. My theory is that an electrolytic cap might have leaked a ground path to the high idle feature. (It's not a problem, it's a feature!) So when I kill it, it takes a while to break through the dielectric and create juuust enough of a ground path to engage high idle mode. Maybe. I don't know. It's just a theory. I'll have to dig in and find out.