Swim

if you forced the butterfly closed and it keeps on revving the IAC is the only way to get air passed, or big vac. leak, so IAC prob getting stuck wide open. clean it good or get a good new one. OTHERWISE you can block IAC off to test . Then 0000start the car with your foot while cranking it. then get it to temp. and drive to see if stopped the revving prob...

agreen

The IAC has been replaced recently with a Mopar IAC from a good running XJ. Also, the IAC is definitely being commanded open when this is happening. Momentarily cycling power to the ECU makes the high idle go away. If it were an issue of the IAC pintle being stuck, the high idle wouldn't go away after a power cycle.

I did have a theory about it being related to humidity, but the past couple of days it's been raining and mine didn't even idle high one time.

I also got into the ECU and took a look at the capacitors. No leaks, no cracks, and everything looks mint. Now I have to scrape all that silicone and re-seal it. It's probably going to be a pain to get it to seal right. The ECU doesn't come apart easily. I don't recommend anyone take theirs apart unless it's HIGHLY suspected.

agreen

For 10 days the high idle hasn't happened. It happened again today.

The high idle feature is supposed to idle the engine up to 1000 rpm according to the FSM. And it's also supposed to only do so when in park or neutral. I think that theory is busted too.

At this point, I'm going to try finding a good ECU. I definitely need a manual ECU since I did the swap, so maybe that'll solve all my other related problems too (and possibly this one).

I do have one more test to run. The TPS is still suspect, although I've changed it at least 2x. It should be 0.26 to 0.95v at idle, and <4.49v at WOT. I wonder if I hit WOT today at some point, but not the past 10 days? What's weird is that I've tested the TPS many times and never gotten that high of voltage from WOT, but who's to say it isn't happening intermittently? IDK. It's just another theory. I need to hook up a multimeter to leave on it for constant monitoring purposes.


Potentially Related:
I also have noticed when it's cold outside (in the 40's or so) that it'll rev up to 2k rpm on cold start sometimes. I mean, not many things scare me. But an inline 6 with 350k miles revving to 2k on cold start make me pucker a little. It immediately returns to a normal idle, but it feels like it revs to the moon on cold start. Almost like the ECU thinks it's going to need a lot of IAC steps to get it going, then realizes it messed up a calculation. And no, there are no intake leaks. I've also searched this issue and found a dead end. It's either an older Renix with that problem, or one that ends up having a vacuum leak, easily found with a can of carb cleaner.

agreen

New development:

I guess the dash volt meter isn't too terribly accurate. On a whim I checked voltage and found a wild swing at idle and while driving. It RAPIDLY changes from 12.8 to 13.8. Come to think of it, this only started happening after I changed out my old squeaky alternator. I also replaced all my battery, starter, and alternator cables at the same time.

So, I have a 136a unit about to go in. I guess we shall see...

dmont

OP, interesting you mention the volt meter. I'm chasing the same exact high idle issue on my 1998 XJ 4.0L. I just replaced the throttle body gasket, cleaned the throttle body, and replaced the IAC. I'm still seeing this intermittent issue. I would think that a vacuum leak would create a consistent high idle issue, not intermittent. While messing with my code reader, I did a vehicle charging test that basically tests the generated voltage from the alternator. My reader output stated the charging system wasn't working properly and the voltage from the alternator was 13.67 V RMS. I wonder if an inefficient alternator tells the ECU to increase idle in order to keep charging voltage at a certain level. My next step is to have the alternator tested and possibly replace it...might as well do it in the garage instead of on the side of the road.

I hope we can together figure this out!

BlueRidgeMark

Hmmm. I think that's witihin tolerance. From ther 97 FSM:


Interesting idea! Time to dig into the FSM to see if it's mentioned. Do you have an FSM (Factory Service Manual)? I think I have most of them, but not a 98. Ironic, because that's the model year of my own XJ!

Here's what the 97 FSM says about the PCM and idle speed:


Nothing there about adjusting idle to compensate for low voltage. I doubt the 98 is different, but it's possible.



Where's the fun in that?


BTW, welcome to the forum!

agreen

Yeah I found that in the FSM. I mapped all those as the problem occurs. This is the only thing I've found that looks weird.

BlueRidgeMark

That's seems weird. I can't think of any reason for it to be cycling like that. It's within range, though.
I dunno. Maybe that's normal?

agreen

I also don't know. Like I said, I mapped all that out the same way, I read about the stuff the fsm called out that affects idle, but everything looked good. I wonder if fluctuating voltage is somehow triggering some kind of failure in the ecu circuitry. Maybe at a specific voltage/frequency it's hitting the right time constant.

But it's weird though, how I could just be sitting at idle at a red light and it just decides it's time to idle (for me) at 1500 rpm. Or sometimes I could just be driving along and I'll slow down to go to a stop and it's already doing the thing. It just picks and chooses when it wants to happen. Unfortunately, I can't take this thing apart to modify the mounting bracket because my "backup car" is apart for other upgrades/modifications. But once that 136a is on, I'll report back with what I figure out. (Side note, I did check the battery temp sensor to make sure it's not the cause)

dmont

@BlueRidgeMark thanks for your reply! I think I was a part of this forum back in the mid 2000's. I had a 98 from 1998 through about 2010 bud didn't do much wrenching on it. When my kids reached driving age we searched them out and now we've owned 4 more during the last 4 years. Two 2000's, an 87, and now a '98 that my youngest son drives. We love the XJ's, and I can probably contribute to some threads after all the tinkering we've done.

I do have a service manual for a 2000 that I downloaded at some point. You are absolutely correct, I don't think the engine uses charging voltage in the control loop for idle speed.

I think I may have fixed the problem! There was an air leak between the intake manifold and engine block due to a loosely mounted intake manifold. I went through all of the vacuum lines and didn't find a leak anywhere. I removed and cleaned the throttle body, replaced the IAC and it still persisted. However, when there as a high idle I was hearing some serious whistling that I couldn't trace to a vacuum hose. I recall seeing a receipt for an intake manifold in the records of the previous owner. So, I looked at all of the bolts securing the manifold to the block and found a few that were very loose. One was actually barely finger tight! I tightened them all down with a 9/16" wrench...the underside was a PITA to reach. After 20 miles of test driving it seems to be behaving normal with idle at 750rpm in neutral and park. So far so good!

So, the problem in my case, I think, was a loose intake manifold. If the problem comes back I'll probably replace the intake manifold gasket...hopefully this isn't the case.

Thanks for your help guys! I'll watch this thread to see how @agreen fixes the problem.

BlueRidgeMark

Yep. That will do it. It's a common problem. Good sleuthing!


Okay, Green, You're up! Show us whatcha got!

agreen

Well, a new challenger appeared. My backup car needs a new engine. I guess civic engines just can't hang with the 4.0 for longevity.

So that means I really don't want to be stuck without a vehicle. But fortunately, I don't have to go to work until Friday, so that alternator swap will be happening tomorrow probably. The civic engine seems kind of hard to find somehow.

While I was driving home from work yesterday I did another data log. The throttle position has me asking questions now. The attached picture is over the course of a 30 minute drive, so unfortunately it's compressed on the x axis. But you can EASILY make out the data. I drove in town for a bit, then as you see I hit the interstate for a while (cruising around 2300 rpm) and then drove in town. At some point, the TPS stopped going to 0. It got to a point where it would only go down to 15% at idle. But it's weird because it'll still show 100%, and it just suddenly started doing that while I was driving around.

If you go back to the beginning, I said I have already replaced the TPS with a known good unit. So I'm wondering if the system voltage being the random number generator that it is has somehow caused the ECU to have a hard time interpreting the tps voltage? So I'll be doing one thing at a time. First and foremost, I want the higher output alternator to be able to run my winch. So I'll start there. Then I'll go back over the connections and all grounds... AGAIN... and if nothing changes, then the TPS... again.

BlueRidgeMark

Hmmm. Maybe try a decent sized capacitor on the alternator to PCM connection to see if it smooths that out?

agreen

I put on that 136a today. No change. The high idle still happens. Voltage still pulses too, so I'm wondering if that's because of the software I'm using. I might hook up an oscilloscope to see what it's actually doing.

The grounds and connections got refreshed, and I even made a weight reduction upgrade by snapping an AC bracket bolt. So that's gotta get fixed later ...

Next up: another Mopar TPS

agreen

So for 3 days and 75 miles of driving I have not had a high idle once. It persisted after I swapped the alternator, but has seemingly stopped after I swapped out yet ANOTHER TPS. I really don't understand how this thing can be so finicky. As I have said before, I swapped the TPS once and solved a shifting issue. I swapped it again with another XJ for testing, no change. The one from the running vehicle was not a Mopar TPS, but the one I replaced before was. That one is about a year old. I did swap it back after it didn't fix it. So a year old Mopar TPS has already gone bad. I put a used Mopar TPS from the junkyard on this time because I was so skeptical (didn't feel like wasting money on the somethingI already bought once). I guess I'll get another (new) Mopar TPS and keep it as a spare since this one seems to work. Also, my data log is showing that it's returning to 0% every time I let it idle, so it looks like there's some concrete proof that the problem is solved.

So there you have it. If you have idle issues and you've already replaced your TPS, do it again and again until you've spent over $200 on the same sensor.