TJ Wrangler 4.0L Intermittent Fast Idle

[SomeBuckaroo]

Hey, Moses...I had another high-idle episode a week ago, so there may be an intermittent vacuum leak somewhere. I have removed & reinstalled the intake and exhaust manifolds about a dozen times over the years and am very familiar with that large gasket surface area. For years I have used a Remflex carbon gasket https://www.amazon.com/gp/product/B00IQHL6KC because I read (and observed) that it compresses much more than the usual blue Fel-pro gaskets, allowing it to better handle slight variations in the intake and exhaust surfaces. Do you have any experience with that gasket? Maybe it is unreliable and is the source of my intermittent leaks?

I took my front window-uppers off, and I definitely hear what sounds to be an exhaust leak. I had carefully inspected the exhaust system about 2 years ago for leaks (using the reverse-shop-vac & soapy water spray bottle method). Maybe I missed something. Your mention using a smoke generator, which has me thinking that would be the most reliable method to find leaks. I think I'll look into that.

Thanks!

[Moses Ludel]

SomeBuckaroo…If the high idle is intermittent and not continuous, the intake gasket is an unlikely culprit.  Depending on the leak’s position, however, this could be a source of extra oxygen in the O2 area.  That would throw off the fuel trim.  Follow up on the exhaust leak, regardless of whether it cures the idle speed issue.  CO is not a gas to have in the cabin. 

I looked at the Remflex gasket, it’s similar to the gasket Borla furnished with the header on my 4.0L.  The thick flanged Borla header has pluses and minuses.  It’s very heavy and stiff.  I’ve retorqued the manifold nuts and bolts at least a half dozen times since installing the header seventeen years ago.

As for actual causes of the idle flare-up, a vacuum leak is a possibility.  I had a vacuum leak on the XJ Cherokee that may be worthy of your attention.  The vacuum hose to the bulb reservoir ran past the battery.  Over time, the hose corroded and began leaking.  The engine management system compensated under most driving conditions.  My only clue was that the cruise control would kick out on a grade.  The manifold vacuum signal was inadequate.  I replaced the hose, end of problem. 

Yes, a smoke machine would be helpful for both the vacuum leak(s) and exhaust system check.  Engine turned off and coil wire removed at the distributor cap, you could rotate the crankshaft slowly while the smoke machine feeds the intake and cylinders.  When the exhaust valve opens at the leaking header tube(s), the header or exhaust leak would show up.

In your case, there may be something else going on.  If the idle kick-up speed is precise, especially after all of the parts you’ve replaced, I would check the HVAC controls and air conditioner wiring circuit.  If the speed kicks up to 1000-plus rpm without the A/C compressor operating, this could be a false signal from the A/C circuit.  The HVAC controls could cause the idle to kick up whether or not the A/C clutch engages.  Worth checking, anyway.  If the idle flares up to random speeds, the issue would likely be a vacuum leak or something else.

On the 4-cylinder models, Mopar uses a power steering pump pressure sensor to kick up the idle when steering pressure goes up.  Primitive and mechanical, the system senses the steering effort when the vehicle is stopped and the steering wheel gets turned.  This boosts the pressure requirement and signals the PCM to raise the idle speed and compensate for the added load on the engine.  Your issue could be similar if the A/C wiring circuit is bumping the idle speed to a somewhat consistent rpm.  The A/C circuit may be activating without a compressor load, which would raise the idle rpm.  It’s worth considering the HVAC controls, A/C pressure sensors or a wiring/ground issue.

The PCM is looking for an “Air conditioning select signal” and an “Air conditioning request signal”…These would be related to the HVAC controls and dependent on A/C system high/low side pressures reading normal.  So, when you’re trying to isolate the PCM signal, keep in mind what the PCM is watching for.  Make sure your HVAC controls and pressure switch wiring function properly.  Also note whether or not the A/C clutch/compressor is kicking on simultaneously with your idle speed step-up.  For the rpm to stay at 1000 or so, the compressor would not be running and creating a normal load.

Moses

 

[SomeBuckaroo]

I believe I finally have a resolution to this particular issue. The culprit: an improperly-adjusted throttle-stop screw.

In summary, I went thru all the common idle issues to no avail: IAC valve / housing, TPS, gaskets, vacuum leaks, throttle shaft O-rings, and capped all vacuum ports to eliminate downstream leaks. Still occasionally had high and varying/surging idle issues.

It appears a previous owner backed off the idle-stop screw on the throttle body until there was about 0.020" of gap between it and the fully-closed throttle arm. In other words, the stop-screw was not stopping anything. I re-adjusted the screw, and my idle is now more solid than ever before. Time will tell since this was an intermittent issue.

In more detail: I removed the screw and re-installed with the allen-head upwards for ease of adjustment. I have heard of references to official procedures for setting this screw. However, I found a simple approach which seems to have worked: With the IAC completely closed, adjust the screw so the engine idles around 400 to 500 RPM. This gives the PCM (via the IAC) adjustment range above and below the proper idle (approx 600 RPM).

To get the IAC completely closed, I pulled off vacuum lines to raise the idle without affecting the TPS reading. The PCM detects this and completely shuts the IAC in an attempt to bring the idle back down. I stuck my thumb over the IAC port above the butterfly to confirm no airflow, and then unplugged the IAC electrical connector to keep it in the closed position. I re-connected the vacuum lines and adjusted the set-screw for about 500 RPM. I then re-connected the IAC and verified a proper and steady PCM-controlled idle. Pulling the IAC connector will throw a CEL which would need to be cleared.

My theory as to what was happening: with my idle stop completely backed off, the butterfly would only stop due to contact with the inside of the bore. I suspect this caused a few problems: First, it allowed too little airflow past the butterfly, causing the PCM to open the IAC more and operate in a range outside optimal control. Second, it greatly increased wear between the butterfly and the bore. Third, over time this wear caused inconsistent airflow in the fully-closed position, based on vibration and how well the butterfly happened to seat when it returned to idle.

However, my idle is still a little bit "rough" - almost as if there's a miss every few seconds (but no CEL). I am suspicious that the butterfly and bore are now excessively worn which causes small but rapidly changing differences in airflow at idle. I think the next step is to swap over a known-good throttle body to see if the idle improves.

Let me know what you think!

[Moses Ludel]

SomeBuckaroo...You found a clear issue.  The only reference to this idle stop screw in the FSM for your Jeep reads:

"A (factory adjusted) set screw is used to mechanically limit the position of the throttle body throttle plate.  Never attempt to adjust the engine idle speed using this screw.  [Bold type is just as it appears in the FSM.]  All idle speed functions are controlled by the PCM."

The IAC service references talk about the IAC working with the "throttle closed".  That's a relative position, as you discovered, based on exactly where the throttle plate rests when closed.  The factory sets this screw (stop) to hold the throttle plate at a specific stop point.  If the factory set screw is backed off, the closed throttle plate is not in the correct position.  Unfortunately, for those who mistakenly fiddle with this screw, there is nowhere in the FSM that mentions how to reset the screw or throttle plate's stop point. 

My guess is that the factory sets the throttle plate stop screw on the bench, using an air flow or vacuum meter.  They could measure and adjust the air flow through the throttle body bore with the throttle plate closed.  (The IAC would be nonoperative.)  The "set screw" could open the throttle plate just the amount needed to get the correct flow.  Or maybe the screw is set for the right rpm on a running engine—with the IAC unplugged and its valve closed, using steps like you performed.  Either way, as noted in the FSM, the throttle plate stop is a "mechanical" setting and not a function of the PCM.  The IAC valve, however, is actuated by the PCM.

I like your theory for why this created an idle fluctuation.  The system relies on the IAC to stabilize the idle.  As you note, with the set screw backed off, the throttle plate was closing completely.  This would create an air starve situation.  The IAC's range for compensating would be outside its parameters (PIDs).  Additionally, there would be erratic O2 readings at the upstream oxygen sensor.  That would affect the PCM's idle speed control.

Without a set screw stop, the throttle return spring's tension pulled the steel plate against the aluminum bore whenever the throttle closed.  This would definitely wear the bore.  Setting the idle speed stop screw at 500 rpm may have compensated for the wear.  If the plate is still resting against the bore, I would replace the throttle body.  Was the idle stable at 500 rpm or was it "rough"?  This remaining roughness could be the edgy air flow at the damaged throttle bore.

Did you find this 500 rpm figure listed somewhere or discussed as a known "best practice"?  If it's the workable figure, great news for others.  Keep us posted on whether this permanently fixes the issue.  Congrats on your troubleshooting well done!  Thanks for sharing...

Moses

 

[SomeBuckaroo]

Hey Moses - I got the (approx) 500 RPM figure from a YouTube video posted by a guy who had stroked his 4.0 and was explaining how he was able to adjust his stock throttle body to work with the new larger displacement. This is the video: https://www.youtube.com/watch?v=FvTwFaQxo-E

The vast majority of the information I found online regarding the throttle stop screw is "don't mess with it!" - which is good advice, except for those of us who need to fix the deeds of a PO who messed with it. I can't find any better procedure than what I posted above, which requires no specialized tools and resulted in a stable idle for me (so far). I'd call this a "current best practice" until/unless something more official is discovered.

I think you're on to something regarding "edgy" air flow at the damaged throttle bore. One test on my list is to completely block the throttle body, remove a vacuum hose, and observe the idle quality by modulating the air through the removed vacuum hose manually, playing the role of the IAC myself. If I can get an idle without the "miss" that I am trying to track down, this would suggest the "miss" is caused by my (likely excessively worn) throttle body. I'll keep you updated.

 

[Moses Ludel]

SomeBuckaroo...The 500 rpm is apparently within the parameter window for the IAC to control idle.  Trial and error was the only approach.  The FSM dodges the issue with the disclaimer about not messing with the idle stop screw.

I like your plan to vacuum test the idle quality.  In trying to emulate the IAC, keep in mind that the IAC valve is tapered and actuates from millisecond PCM signals.  Let us know how effectively your vacuum hose test works. 

An approach that could refine the test would be adding an inexpensive, small ball valve at the end of your vacuum hose.  This would enable better control of the air adjustments.  Add an analog vacuum gauge, and you could watch for needle fluctuations.  If the gauge needle wiggles, that's like a vacuum reading on an engine with leaky valves.  Here, you are looking for an irregular gap between the throttle valve and a worn throttle body bore.  In real time, you can see the relationship between the vacuum reading and the engine's rpm fluctuations.

Moses