[spydergear]

Hi All,

I have a high idle problem. Once I drive the car a little ways, the idle eventually creeps up to 1400 RPM. If I turn the ignition off and restart, it will return to a normal idle-speed, even if I rev it, but once I drive it, the idle creeps up again. When it is cold, it seems to start normally with the initial startup idle-flare that settles back down to normal idle-speed. Once it gets warm, the idle gets too high. Once, it restarted with a high varying idle, but only once so far.

I saw an old thread from Peva and that lead me to test the coolant NTC sensor, the TPS, and I also cleaned out the AIC valve with WD-40. The NTC resistance, cold and hot was fine, the TPS had about 13 ohms with the throttle closed and had no continuity with full throttle, the AIC valve had about 10.2 ohms across the terminals.

After I sprayed the AIC valve out with WD-40, I tried to move the little barrel, but could not. Also, the plastic part that fits into the aluminum housing is a little loose.

But if the valve is bad, why is the cold start normal with the programed idle-flare then settling into a normal idle? Wouldn't that require the AIC to be working?

Any clarifications on how this system works and anything else I can do or test is appreciated.

 

[900power]

My 91 turbo had fast idle on several occasions in recent years. Different problems were found and corrected at different times. Luckly, my AMM and AIC valve has been fine. But I often swapped good units to check. That had been a common symptom: idle increasing as the car warms up.


The problems have been:

-TPS (throttle position sensor) wiring connection had corrosion. White dusty corrosion where the connector plugs into the top of the TPS. I used some pot scrubber material and electrical connection cleaner. (I had this type of problem with a fuel injector once)

-Various vacuum hoses deteriorating. Various places such as the charcoal cannister and from the intake near the main fuse box going under the dash . It can be split on the bottom of the hose where you can't see the split.

-NTC sensor. I replaced mine with a good used one.


There are numerous places for vacuum leaks. Check the turbo intake tubing and hose clamps, including the grommet where the PVC hose connects (if it's a turbo car).


Looking at the Bentley manual on page 240-13, I'm not sure you're testing your TPS correctly or you're getting a bad reading. What year is the car, and is it a turbo? LH 2.4.2 = 1991 and later non turbo.

 

[spydergear]

Sorry, I should have mentioned that it is a 1990 SPG Turbo with LH 2.4. The AIC has two terminals. So, not the same as my '85 Turbo, which precludes swapping that unit in.

I did use Deoxit on some of the connectors. Maybe not the TPS connector, I don't recall.

The NTC did measure correctly as per specs in the Bentley manual for cold and running temperature. It is a Chinese aftermarket sensor that I replaced years ago for some reason. I do still have the original from Uncle Robert.

I measured the TPS continuity finding that the center terminal must be ground (most likely), and I think it was the rear-most terminal that was for closed throttle. Those were the only terminals with any continuity. Nothing worked at full throttle regardless of where I probed and I tried every possibility.

I have not cleaned any of the ground connections yet. I think this could be a possible trouble area given that I have had corroded terminals at the distributor connection in the past.

I thought of vacuum leaks, but what really bothers me is that if I shut of the ignition for just a minute the idle returns to normal. This says to me that the AIC works and that vacuum leaks are not a large factor. But perhaps I am using flawed logic...

Is there any direct way of testing the AIC such as hooking it to a 12 volt source?

 

[jvanabra]

Do not put 12v to it or you can burn it out. The valve is controlled by PWM - it's not designed to handle a single voltage. Probably if you briefly applied 12v the gate would move and that would be sufficient, but definitely do not keep it powered for long. A 9v battery might be a safer alternative. TBH, your problem sounds like a vacuum leak, potentially at the bypass valve.

 

[spydergear]

Yesterday, I tested the bypass valve with my Mity Vac. It held 20" of mercury without the gauge moving. Later, after I got home from work with the car idling high, I pinched the AIC hose and the idle returned to normal. Once I let go, the idle was high again. So, when I get some time, I think I will pull off the valve again and clean it out with something stronger than WD-40 and see if I can get that vane to move.

 

[Rodentmaster]

Pinching the AIC hose should've seen the idle drop to approx 600-700 rpm.
Idle valves are meant to compliment the main air path - not be the primary idle air path.
I suggest there may be a vacuum leak or the throttle stop is adjusted too far open......allowing too much air past the throttle butterfly.

 

[spydergear]

It is possible it went to 700 rpm. Don't know for sure, as I was not looking at the tach, I was pinching...

Yesterday, I did hook up a 9-volt battery to the valve. As I made contact, the valve would rotate to full open. I tried reversing polarity and I got the same result. As the plastic part where the contacts are is slightly loose in the housing, that whole section of the valve would jump to the side of the housing as I touched the lead to the battery. I wish I knew how to make a short gif file from my phone - as it is kind of entertaining.

I also measured the voltage from the connector with the ignition on, as it says to do in the Bentley manual. I got battery voltage on one of the contacts with the DVM negative probe to ground. I really saw very little on this system's operation in the factory manual. There was a mention of a limp-home mode that would keep the rpm at 1200. Mine is around 1400 warmed up. That is until I turn it off and back on as I did tonight and then it goes to a normal idle, or at least it did tonight.

Maybe the valve itself is a source of a vacuum leak? Perhaps I might tape up the loose housing and see if that makes a difference. At this point, I am trying to avoid spending almost $200 on a new valve without first seeing that it is definitely bad.

Thanks, Rodentmaster, for your opinion. If I rule out the valve, I will check for vacuum leaks and look at the throttle. And I will even examine the hooter valve as jvanabra suggested.

 

[jvanabra]

You might be able to use https://cloudconvert.com to create that GIF.

 

[Carfixer2]

I would start with checking for various vacuum leaks. Then I would reset the throttle stop a little more closed. Also I would check resistance of the throttle position sensor cold and hot to see if it changes when hot.

 

[Carfixer2]

Also forget to mention. Clean the grounds on the intake manifold. Also I often will remove the battery terminals from the battery and clean the battery post and the inside of the cable terminal. And last resort is cleaning the were the negative cable attaches to the transmission and even the body.

 

[spydergear]

I like these ideas. Maybe I could perhaps jumper the connector on the TPS, assuming I can figure out which is the idle circuit. May have to peal back the boot to see the color of the wires. I assume that would not hurt the ECU since the TPS is essentially just a switch on LH 2.4?

 

[jvanabra]

Yep, you can definitely do that.

 

[spydergear]

First, the diagram in the Bentley manual of the TPS appears to be wrong for LH2.4. There are no numbers that I can see on the switch. In any case, I measured again with my ohmmeter and found that the two rear-most contacts are for the idle switch. Based on that, I jumpered the two rearmost terminals of the connector.

TPS idle contacts resistance measurement results:
Cold: about two ohms.
Hot: up to about 12-15 ohms (my HF DMM doesn't have the most stable readings).

To try to eliminate the TPS, I jumpered the connector contacts while the car was running to see if the idle would reduce:
No effect, did not bring down idle.
The only way to bring down the idle was to restart the car. It would again go through its idle flair and then settle into a normal idle. I did this with using both the TPS connected and with the connector jumpered and got the same results.

Then with the car running again at a normal idle, I tried to raise the idle by wiggling the plastic insert in the AIC valve. If I moved it a certain (unknown to me) way the idle would elevate again to 1400 rpm. No matter how I wiggled it after that, it would not return to a normal idle. It would not go back down unless I restarted the car again.

So, I'm thinking that either the AIC develops an open circuit internally or sticks open somehow that is corrected with a restart. I am clueless as to why.

Because the plastic insert in the valve moves around in the aluminum body of the valve, and this is clearly not normal, it is my primary suspect. I have not cleaned the grounds yet, nor done a thorough search for vacuum leaks, as it was getting quite cold outside. I will have to do that too, but thought I would share what I found today in case you have any opinions on what occurred today. Thanks for all your help!

 

[spydergear]

Saturday, I made a 120 mile trip, got home, and it was, of course, still idling at 1400 rpm. I decided to give it a whack with a screwdriver handle to maybe unstick the valve (even though I had tried this before). Made no difference while the car was still running. However, when I drove it the next day, and since, the car returned to a normal idle. I've heard that correlation is not causation, and I don't necessarily believe that this will continue indefinitely, but I'm glad it is working again. Any theories about what happened?

 

[jvanabra]

Still kinda sounds like the idle valve is getting stuck, but vacuum leaks, goofy NTC sensor, bad O2 sensor are all on the list You need to test & eliminate variables.

 

[spydergear]

Good point. Today, I got a look at the turbo intake boot and it was quite distorted and old looking, so maybe I'll add that to the list.