Used to idle rough. Now won't idle at all

[bsooley]

A couple of days ago I started my '91 964 C2 and had a rough idle (loping). I had seen this problem a few times previously after she warmed up. This time the loping started immediately and never stopped. I drove a couple of blocks to get a feeling for the symptoms and the car died at the first stop light. Started again easily, but would die at every stop.

Limped it back to the garage and looked on the forum for hints. The Idle Control Valve seemed to be the first culprit. I removed, cleaned with brake fluid (some black but not alot) and reinstalled. Now the car starts but dies immediately everytime. Again, easy start everytime but no idle whatsoever.

Any ideas would be appreciated. Thanks

[bsooley]

BTW, I tried 12 volts on the Idle Control Valve and it closed up.

[Bill Verburg]

cleaned with brake fluid

??

I would clean w/ carb cleaner, there is a rotating valve internall that should be seen to move freely as you rotate the ISV in your hand.

A loping idle(and stalling) is often a sign of lean surge, caused by a vacuum leak.

[bsooley]

Sorry about that. I meant brake cleaner. After it didn't work the first time, I also tried carb cleaner. No difference. If I shake the valve, I hear movement. Nothing visible though.

When vacuum lines might be of concern? I've just replaced the two clamps on the ISV. The original ones were crimp clamps which I replaced with screw clamps. I can't imagine there's a leak there.

[kang]

Originally Posted by bsooley

Sorry about that. I meant brake cleaner. After it didn't work the first time, I also tried carb cleaner. No difference. If I shake the valve, I hear movement. Nothing visible though.

When vacuum lines might be of concern? I've just replaced the two clamps on the ISV. The original ones were crimp clamps which I replaced with screw clamps. I can't imagine there's a leak there.

What do you mean by “nothing visible though?” Do you mean when you look inside the ISV, nothing visibly moves? This is not correct. You should be able to see the internal mechanism move. You can actually poke a screwdriver in there and move it manually.

[tshih]

Does your model year have the OBD1 system installed with the check engine light blinking a code after you switch on the ignition and hold the accelerator pedal down 5 sec (engine not started). If so read the codes (seriesof blinks and then get translation. Or does it accept the Bosch Hammer to see if the trouble code for AFS is triggered as well as O2 sensor.

[bsooley]

I do have the series of engine light blinks after holding the pedal down for 5+ seconds, similar to morse code. Is there somewhere I can find a list of patterns to look for? I also just found out a friend of mine (VW guy) has a reader for OBDI and OBDII connected to his computer. I'll try and track him down. Can you tell me where I can find the OBD connector on the C2?

In the mean time, I connected the clean ISV to its electrical connection but left the pressure lines disconnected. Started the car and the valve closed. Is this correct?

[Fast951]

Check your idle switch. With the engine off, Idle switch should be ON. Check DME pin #52, make sure it is reading fround (continuity to ground).

[tshih]

bsooley,
the Book by Adrian Streather on the 964 probably has the codes deciphered. There is a publication on OBD1 codes "Bosch F.I. System, OBD codes" out there but I don't have the exact reference. Post query on rennlist and they should have a source of the codes. The correct adapter plug for the 964 OBD1 connection to the Bosch Hammer is unfortunately not readily available. I believe a mercedes plug also works but don't know which model mercedes benz. I have had luck with getting OBD2 cables from www.carplugs.com, contact them for leads to OBD1 connectors (worth a shot, good luck).

[bsooley]

Thanks for the rennlist suggestion. I found a link to reading the engine codes using the CEL. The link is;

http://members.rennlist.com/billwagner/CheckEngine.html

I don't have the version of the DME firmware that includes the 1000 code at the end. I checked the codes and I have a 1121 (aka a constant air flow sensor fault). Any idea how to troubleshoot or fix this issue. I just checked the pricing on this site for a new one PN 964-606-050-X-M100 and we're talking about $1279 bucks. Holy crap!

As a note, this is a summer car and sits idle for 6 months every year. Could this problem have occurred through the lack of use?

[bsooley]

I've done a bit of reading on other threads and I'm going to try the following

1. Clean the barn-door style air flow sensor (the one directly connected to the air box). What is the best way to clean this without destroying anything? Seems easy to get off but I've been told that these barn door units are sensitive.

2. Clean the battery terminals. I've been reading that this can affect idle. What the hell? It's cheaper than a tow to the mech

3. Clean my K&N, though it looks pristene.

4. Check all vacuum tubes, though none looked damaged and no clamps were loose. One guy mentioned pressurizing the system to check for leaks. What is the best way to pressurize the system?

If anyone can give me help on how to clean and test the AFS, it would be much appreciated.

By the way, thanks to all for the assistance thus far. Its a whole lot easier to feel confortable doing some of this work with the reassurance that there are people who've been through these problems before.

[cjoenck]

You can clean the airflow sensor with carb cleaner. Just spray it inside and manually move the vane. The vane - barn door - should move freely. No mechanical grinding, etc. and it should close fully. You should also check the electrical connections:

- There should be 5.0V +- 0.3V between terminal 3 and ground with the ignition on
- Use test leads and connect pin 3 of the disconnected air flow sensor plug with terminal 3 and pin 4 with terminal 4. Measure 250 - 260 mV between terminal 2 and ground. Move the plate mechanically, you should not see sudden drops or jumps. At full load voltage should be 4.6V

Did you check the idle control switch? Very easy to do: Use an Ohmmeter between terminal 2 and 3. You should get 0 when idle and infinity when moving throttle.

Hope this helps.

[bsooley]

OK, me again. I've done a bit of troubleshooting tonight and have the following feedback;

1. Removed, cleaned and tested the AMF based on cjoenck suggestions. All checks out.
2. Cleaned battery terminals
3. Checked K&N. No need to clean
4. Checked vacuum tubes. Didn't see anything.
5. Checked the idle control switch. Resistance is good. 0 at idle, infinity otherwise

The verdict, no idle. After reconnecting the battery terminals, CEL light still on, so checked the code again. Different code. hmmm.... New code is Code 1X15 - Idle speed contact break.

I don't have a clue what this one means. Any ideas?

[ischmitz]

It means the DME detected an open on the signal line to the ICV. If you had it disconnected and turned the ignition on could be an old stored fault. You might want to clear the fault memory (keep accelerator pressed longer than 10 seconds when switching the igntiion to ON) and re-test.

If the fault comes back your wiring might be bad. If the ICV is not controlled properly by the DME the engine dies.

[bsooley]

Tried to reset using the pedal. No luck. It just keeps blinking the 1115 at me. Note that this is a California car. I'm not sure if the DME reset was in this version of the software.

If this is a real fault, how would I troubleshoot it? There are only 2 pins on the ICV connector. I'm assuming a control signal and a ground. After cleaning and reinstalling, I turned the ignition on and it closed (with both tubes disconnected). It seems reasonable that both pins are making contact. Is this a valid assumption? Could it be the DME?

[bsooley]

Did some more searching on Rennlist. Two threads mentioned that the 1115 code refers to the throttle valve switch, which seems to be another name for the idle switch. I've checked the idle switch at the connector in the engine bay and all is fine. Next I'll check it at the DME using the following method I found described on Rennlist.

"You can quickly check the idle switch with an ohmmeter between pins 52 & 30
on the DME connector without going in the engine compartment. You should
measure less than 2 - 3 ohms with the throttle pedal at rest. A small depression
should "open" the switch."

Getting there slowly...

[ischmitz]

You are right. A 1_15 means the DME has not seen the idle contact closed after the ignition turned on. This leaves the CEL on. It could be a simple mis-adjustment of the accelerator cable or microswitch. Before you even start to measure things there is an even quicker test:

Open the engine bay with the engine not running. Gently push the throttle to the closed position and listen. If you hear a faint click that means the micro-switch got engaged. The return spring is not strong enough to overcome the excessive tension of you thtottle cable. If that is the case simply slacken the cable. There is an adjustment where it comes out of the tunnel under the car. You need to crawl underneath.

If the throttle closes all the way against the stop and the switch does not engage properly you might have to adjust the position of the microswitch itself.

Also lube the throttle assembly slightly. Years of grime could take it sticky.

Last, there could be an electrical problem - such as a broken switch. I doubt that since your problems worsened over time. This is typical for a switch that is getting out of adjustment over time. And without the switch engaging idle is hunting and rough.

Good luck,
Ingo

BTW: The ICV is a 2-terminal device driven by a H-brigde. This allows the motor to run either forward or reverse. The DME switches polarity in quick fashion to achieve a certain opening angle. Both contacts are signal.

[cjoenck]

Checking the ICV control voltage is actually very easy. Leave the ICV installed and disconnect the electrical plug. Start the engine. You should be able to measure battery voltage, i.e approx. 13V between terminal 2 of the plug and ground. To check voltage on terminal 1, you need an scope - you should see about 100Hz oscillations. If you get 12V on terminal 2, I would borrow another ICV and install to see if that makes a difference.

Idle speed contact is between 52 and 24 with ground bridged to 30 - as per workshop manual.

BTW, is this a Tiptronic? The 2 micro switches are different on a TIP - the WOT switch is integrated into the throttle position poti...

[tshih]

If you are going to work on a 964 at home you should definitely get a copy of Adrian Streather's 964 book. My 1993 RSA had the same problem as your's except it only has 11000 miles and the AFS was broken despite it running fine intermitently. I took it apart and checked everything and couldn't find any problems. After putting in a new unit (at the time it only costs $600) the codes were cleared and it ran perfect still today.

The K&N filter with the oils are not recommended for AFS (especially later 993 hot wire versions) in porsche models. The filter allows excess oils and dirt to gum up the AFS and hot wire leading to problems. Sticking with the OEM papaer filter and AFS has been just fine. Putting holes in the airfilter housing to increase air intake has no added benefits since the AFS barndoor is the main restriction to more air flow. The 993 on the other hand does benefit (about 8 HP gain) from opening up the air filter housing to let more air in quicker. The OEM papaer element is efficient and will protect your expensive engine much better than the porous K&N or after market stuff being peddled out there.

[tshih]

Also I don't know if there is a rebuilding service for that model AFS similar to our host's service for the 914-4 1.8L AFS ($150 for service versus $1600 for NOS). You should contact Pelican and ask. Also the alternative is to get a larger throttle body aftermarket AFS (switching to hot wire/film AFS) that the british tuners have success with. I would go that way if you are spending $1600-2000 for the OEM parts since you supposedly can get about 10 more HP.