This concerns my daily driver Formula of almost 22 years. I started a thread about this problem 13 years ago when it surfaced, and now it’s back with a vengeance. It’s been tormenting me for over 3 months now with no relief in sight. Here is a link to that thread: https://ls1tech.com/forums/pcm-diagn...e-control.html

In addition to everything I mentioned in my first post in that thread, I have disconnected the MAF for several days, installed and ran an open loop tune (to see if bad feedback from the O2’s was causing it), cleaned the TB & intake, done a smoke test, replaced the IAC, TPS, and Evap purge solenoid. Did a TPS reset and idle relearn. The issue manifests in 4 modes:

1. Normal – idles at set rpm of 650. Strong rpm drop and decel on lift throttle. It’s four miles from home to my office and it typically runs normally for the drive, but sometimes manifests just before arrival. Oftentimes it doesn’t do it at all on short trips, and there have been days with a fair amount of driving when it hasn’t happened at all.

2. Semi-normal – idle bumps up and hunts while coasting to a stop, but settles down to about 700rpm. Still has decent lift throttle rpm drop and decel.

3. Moderate – high idle while coasting to a stop, around 1100, but will settle down to 750-800. Some “cruise control” effect, especially 11-1200 rpm..

4. Extreme – idles 900-1100. Will rise to 1400-1500 when shifting into neutral while coasting. Strong cruise control effect – like driving down the freeway at 70mph with my foot off the throttle. Will cruise at 1200rpm all day long without any throttle input.

I thought the problem was gradually going away (like it did before), but it went into moderate mode the other day after a couple of miles, and then after a WOT pull, the car completely freaked out and went into extreme mode. Soon after it threw a P0507 (IAC system RPM too high, duh). I should mention that shutting the car off and restarting it causes the idle to revert to normal.

I’m about at my wits’ end with this. The next step we thought of was installing a friend’s PCM from his rarely driven T/A with my tune loaded on it to see if it may be a PCM issue. I’m open to any thoughts and suggestions. Sorry for the novel!

 

I just read through your old thread and never saw any mention of checking the EVAP system. If the tank vent and EVAP purge solenoid open at the same time, it is the same as any other intake air leak. The purge solenoid is pulse width modulated to prevent problems with this, but if the vent and purge both hang open at the same time it could cause varying degrees of intermittent air leakage. If you have a good scanner I would log the PIDs for both, as well as the normal suspects that you think are already ruled out such as IAC, TPS, etc. Or just disconnect the purge solenoid from the intake and cap the vac port it hooks to.

edit-- If you disconnect the purge valve, leave your gas cap loose temporarily to insure no vacuum or pressure in your tank.

 

Yep, I did that and drove around with it disconnected for a few days and no dice. Then I thought to just replace it anyway, making sure that I had all good parts in the system that could affect this issue.

The tech I was working with said the TPS was bad - it is supposedly spring-loaded and that was not the case with the old one. I thought it might take time for the PCM to adapt to the new one, and for a while it seemed like the problem was diminishing, but that turned out to be a false hope..

 

If you have access to Hptuners, a good datalog while the problem is happening could at least narrow down whether you have a sensor problem or a mechanical issue. However, I strongly suspect you have something that is cracked and leaking air in dependent partially on temperature. With a bunch of 20+ year old hoses and gaskets this could be a pain to chase down. In your other thread you said it was running at one point with IAC counts near zero, so air is coming in from somewhere.

 

I have EFILive and have logged a number of parameters, and yes, the IAC count does go down to zero at idle even with the new IAC. It seems to perform identically to the old one (which I don't believe there was anything wrong with).

I was sure I had a vacuum leak but the smoke test showed nothing. I suppose I could replace the PCV valve; I'm runninng one of the truck style ones with the pinhole rather than a pintel, but it has been on the car for years with no issues.

 

Underhood temps go up dramatically with the hood closed, so if something has a crack that is opening up at temperature, a smoke test may never show anything. Good luck with it. I would cap every vacuum port besides the MAP sensor and see what happens. Unfortunately that will be dangerous as hell without vacuum to the power brakes. Have you checked that the throttle blade is tight to the shaft?

 

Underhood temps go up dramatically with the hood closed, so if something has a crack that is opening up at temperature, a smoke test may never show anything.

Makes sense, but then why would it revert to normal idle simply by shutting it down and restarting it? The EVAP scenario you described was the near-perfect explanation for everything - starting only after being driven several miles, reverting to normal after being shut off. It was so disheartening when the problem surfaced with it disconnected and capped off.

Good luck with it. I would cap every vacuum port besides the MAP sensor and see what happens. Unfortunately that will be dangerous as hell without vacuum to the power brakes.

No kidding, if you've ever felt these brakes without vacuum, you know - like zero brakes. The only way do it would be to induce the high idle and then cap off the ports one by one. But there are a host of issues trying to do that with a hot engine running. But I couldn't shut it off because then it will revert.

Have you checked that the throttle blade is tight to the shaft? Not particularly, but I'm sure I would have noticed it when I cleaned it. I can double check.

I'll conclude by saying that the car was as docile as a kitten over several short trips yesterday and today. And so the issue continues to mystify me, or more to the point, continues to thoroughly kick my butt.

 

I would log at least Throttle Position, IAC and Ignition Advance. (And of course basics like RPM.)
If the throttle blade is completely closed and the IAC is Zero, that indicates air is sneaking in somewhere, somehow.
The stock Ignition Advance is so conservative that an unexpected increase could significantly increase the idle speed.
Good luck

 

This is an interesting topic to me, I've been dealing with a more minor case of this issue for over 20 years with my '98 Z28. I've been through all the same checks/data logging, attempts, parts replacements, etc. I've had all the same thoughts and theories as you have mentioned above and in the other thread. In fact, I've posted at least one or two threads on this myself over the years.

Basically, my issue is identical except that the idle hang is never that serious, rarely more than 100-150rpm above programmed idle speed (only set a P0507 maybe one time in all these years). But, when it happens, there is also a surge of about 50rpm and IAC counts will be at 0. Perhaps most interesting of all is that I, too, can get the issue to instantly and completely go away by just restarting the engine. Many years ago, I thought I had it solved by replacing the EVAP purge solenoid; this was the final item that could possibly have any effect and it seemed to disappear for a bit after that so I thought all was well. Until it came back.

In my case, this car is not a daily driver. In fact, it's rarely driven at all; just to car shows and the occasional pleasure cruise to the tune of <300 miles per year. The engine is stock, and the tuning is stock other than some changes to shift points and fan settings. If it's going to happen, it will usually only happen during initial warm up after the car has been sitting for long periods (but I don't think it's related to battery voltage because I keep it charged regularly, and sometimes it happens even if the car was driven recently).

My theory at this point is that the IAC is not properly responding to commands. The IAC count is at 0 because the PCM sees that idle speed is too high, but there is nothing the PCM can do if the IAC isn't properly responding. This would explain the source of the air without there being a vacuum leak (or any other signs of a vac leak), and how the issue is able to be cleared so instantly with a simple restart of the engine. It's not a problem with the IAC motor itself, as I have tried several of them (including a known 'good' one from another LS1 car that I used to own), but likely an issue with the signal being sent to the motor (in other words, the signal not being properly received). Not sure if it's a reception problem (harness) or a transmission problem (PCM), but just because the count is shown to be 0 on my scanner doesn't necessarily indicate that that the pintle is truly at 0 nor that it has properly received the signal to be at 0. I don't know how this could be verified with the engine still running and, like you, I can't shut it down to pull the IAC because shutting it down will clear the condition.

For me, it's not that big of a deal to live with it because the idle hang is so small, happens so seldom, and the car is used so rarely. Honestly, a "normal" person who isn't hyper-aware of how their car runs probably wouldn't even notice it in my case. But it did start to drive me nuts when I thought it had to be something simple. At this point, there is nothing "simple" left, so it's got to be a harness or PCM issue. Not worth worrying too much about in my case, but for yours it's more often and more pronounced so I would want it fixed too. I'm just thinking we might have the same problem, and if so I think it's something to do with improper IAC response.

 

One time I replaced the stock IAC with a Duralast IAC and it wouldn't idle down. There must be something different about the aftermarket ones.

 

Edited your comments above to point out my latest theory: the PCM is responding to something it sees and doesn't like. In the absence of a mechanical problem or a vacuum issue, I'm starting to think about possible corrision on the pins to the PCM or a broken or frayed wire belonging to one of them.

The PCM harness has never been disconnected. I'm thinking to pull it and inspect for corrosion and spray it with electronics cleaner. Inspecting and testing every lead that goes into the PCM would be far above my time limitations and pay grade. Does anyone know the pin numbers connected to the IAC and throttle?

Not to bore you guys with unending anecdotes, but I realized something tonight. My brain wants to find patterns in these behaviors, but in truth, it is actually random. Case in point: yesterday through this afternoon it drove completely normally. Then this evening, I drove a couple of miles to a restaurant and it went into moderate mode. Okay, not unusual. After a couple of hours there, when I started it, it flipped out and went into extreme mode. Shifting into neutral when coming to a stop it reached 1600rpm. After cruising into my garage at 1200rpm, I shut it down and restarted it: back to normal idle.

 

We might be saying the same thing in different ways, but what I'm thinking is this:

IAC motor is (correctly) commanded to a certain position based on various live data that the PCM interprets, as written in its tune. At this point, everything is working fine. As conditions change and airflow need is reduced, the PCM attempts to send a signal to the IAC to close a bit, but an intermittent fault occurs where this signal is either not properly sent (PCM problem) or not properly received (harness issue), so we see commanded IAC counts eventually reach a "0" (fully closed) position (which the PCM "thinks" it has sent/commanded) as the PCM continues to attempt to bring idle speed under control, but the IAC (for some unknown reason) is not seeing this command (or not seeing it properly) so it does not respond in kind (hence it is not closed and idle speed remains too high). With a restart of the engine, the IAC is automatically reset and it now begins responding properly to commands, as though nothing ever happened.

Once we know that the IAC itself is not a problem (by way of using mutliple ones, "known" good ones, and no change in the condition) then we must assume that the issue has something to do with the signal. So the question is, what makes the signal suddenly weak or corrupted enough that the IAC "freezes" at a position that is no longer consistent with the current command? And then what is it about a restart that clears the condition and restores the signal? To me, this sounds more like a PCM problem vs. an issue with the harness. It's like any other type of computer that starts acting funny until you give it a restart.

To add further info, it my case there is certainly not any sort of corrosion on any connectors or grounds. This is a 19k mile car that is always parked in a dry garage over carpet, never leaves the garage if there are even clouds in the sky, and hasn't seen a garden hose more than once or twice in its life. It has always lived the life of a true show car, and the engine/engine bay has never been apart. As such, I suspect that any issue is more likely an internal circuitry flaw vs. a harness issue, but I suppose something in the harness could have been accidentally pinched or a flawed connector used on the assembly line.

 

I guess I can swap the new IAC for the one on my M6 car (known good) and see if there's any difference, but I doubt it.

Lol my car is the polar opposite of yours. Not only does it have 198k more miles, but the Blizzaks and Mickey Thompsons stacked in my garage are testimony to its wide range of activities. It has seen every kind of weather and most of my driving is short trips, and yet except for this problem, has always performed like a champ.

 

The only time I ever had anything remotely close to this was on my old 93 Ford, and it was the IAC valve. It also threw a check engine light, but I didn't know because the bulb was burnt out Got off the highway one day and I felt like I had to smash the brake pedal to the floor as hard as I possibly could to actually stop at the end of the ramp. Was a little unnerving. And a simple ignition cycle fixed it. Figured out the burned out bulb by paying attention to the dash when I first turned on ignition and found that at bulb check it never lit up. Then I never replaced the bulb anyway.

But of course you've already replaced that.

If it isn't too hard to swap the PCM then try that for kicks and see what happens. Beyond that, I'd literally run a temporary wire to IAC and see if that fixes it, in case the wiring has somehow gone bad somehow.

 

The IAC is working; my logs show that the steps go up and down with rpm. But they're at zero at idle when they should typically be around 30 or so, indicating a bad IAC, an additional source of air, bad signal from the PCM, or other factors discussed in this thread.

 

What's your spark timing at idle? Is it different when the car is doing the high idle thing vs when it isn't?

Also, what throttle body are you using?

 

At mid-700's rpm it averages about 9 degrees, in the 800's 18 degrees, and in the 900's 27-31 degrees. When the idle kicks up, you can see the spark advance jump.

Throttle body is stock, ported by Bo White about 20 years ago. Throttle plate has not been drilled.

 

Does this car have a plastic intake manifold?? It could have a slight crack that only opens up when the hood is closed and temps get right up there

 

That seems like it might be an important clue. Normally the ECU will retard timing to slow down the engine, and increase timing to speed it up.

So I wonder if your closed-throttle table has progressively higher spark advance at higher RPM, and the spark feedback isn't able to retard timing enough to keep idle RPM where you want it. Most days it doesn't matter, but if closed-throttle / idle mode kicks in at 1000ish RPM then maybe it's not able to bring idle down because there's so much spark advance.

For comparison, my car's stock closed-throttle table has:
17 degrees at 800 RPM and lower
20 degrees at 1200
27 degrees at 1600
All at 0.30 load - spark is retarded at higher loads, but those cells seem hard to reach at low RPM so I doubt they matter.
And that's on a manual car, so automatic might have more to compensate for the torque converter when in drive.

 

The Base Spark in Gear table in the stock 2000 tune is much more aggressive than the 2002 table. Yet I've been running my current tune including the stock 2000 table with other minor variations for over 10 years, and I haven't had this problem to anywhere near this extent.

Probably also worth mentioning that the O2 flucuations and variations between the banks are extreme, and LTFT's are through the roof.