Pelican Parts Forums

Pelican Parts Forums (http://forums.pelicanparts.com/)
-   Porsche 924/944/968 Technical Forum (http://forums.pelicanparts.com/porsche-924-944-968-technical-forum/)
-   -   86 Porsche 944 idle fluctuation (http://forums.pelicanparts.com/porsche-924-944-968-technical-forum/716376-86-porsche-944-idle-fluctuation.html)

Armon 11-11-2012 10:01 PM

86 Porsche 944 idle fluctuation
 
I need advice on my 944 idle fluctuation problem. It started about 3 weeks ago and now I'm starting to second guess myself. I brought the car in 1999 and I have probably replaced and experienced just about everything that normally goes wrong with these cars expect for idle and fuel problems. About two years I had a head job done and almost everything that was old and needed to be changed was replaced. New injectors, hoses, plugs, wires, cap, rotor, fuel filter, etc...

The car never ran better until about 3 weeks ago I pull in the driveway, turned it off and suddenly remembered that I forgot to go to the grocery store. When I started it back up the engine started fluctuating badly, up and down, between 600 and 800 rpm, back and forth. When cold it does the same thing, just after a minute or so after starting.

Immediately I thought vacuum leak, since most of fuel items should still be good. I couldn't find any leaks, most of my hoses were replaced anyway. Next, after reading the posts on here, I check the fuel pump, fuel delivery, temperature sensor, TPS, fuel pressure regulator, air flow sensor. I also rechecked the injectors, plugs and wires. Everything checks out and is within specs.

Oh, I replaced the ICV and that didn't fixed the problem either, however for some reason I notice that when I cramp the hose coming from the J-Boot the engine idles normally, like it suppose too.

Anyway, I'm almost at the end of my rope! I was going to buy a 02 sensor next, but I have already spent enough trying to figure out this problem. If anybody has any other suggestions or ideas on what to check or might be happening then please let me know. Thanks

Gawernator 11-11-2012 11:01 PM

O2 sensor could help with rough idle.. Have you tried resetting the idle speed per Clark's garage instructions?

Armon 11-12-2012 09:03 AM

No I haven't reset the idle speed. When I cramp the inlet hose coming from the J-Boot to the ICV the engine idles perfectly... but I'll try it and see.

Gawernator 11-12-2012 10:53 AM

Is it possible to install the ICV backwards?

HondaDustR 11-12-2012 11:25 AM

Try unplugging the O2 sensor. The DME will run in open loop mode without it (interestingly enough, the way it was designed for...the O2 system is a crude system tacked on to meet EPA regs, and causes the DME to do weird things if the sensor is defective).

Disable the idle stabilizer and see if the base idle speed is way off, causing the ICV to oscillate trying to correct too much difference.
Idle Speed Adjustment
I've heard the adjustment screw is prone to loosening up as the seal dries up with age. Have you cleaned the throttle body? The idle screw bypass port can get gunked up.

Does the car run fine otherwise?

Gawernator 11-12-2012 02:57 PM

Was thinking about unplugging my on-the-way-out O2 sensor while waiting for the new one in the mail. Open loop mode? I realized that the O2 sensor is what's causing some rough idle and strong fuel smell at higher RPMs. What is open loop mode though? Bad or good...

HondaDustR 11-12-2012 03:11 PM

Open loop mode is where it reads straight off its fuel delivery maps as a function of rpm, air flow meter position, coolant temp, and air temp. There is no feedback loop as the O2 sensor circuitry trims the injection duty cycle depending on whether it is too rich or too lean. The problem is, it is a rudimentary system tacked on after the fact. the DME was not originally designed for it, and it was added on to allow import to the US. It does not constantly monitor the O2 sensor signal, it only triggers an interrupt when the sensor signals too rich or too lean, meaning the mixture constantly oscillates about ideal. As the sensor response speed slows with age and the range reduces, the mixture can vary more wildly and end up offsetting too rich as the sensor sensitivity declines. Unplugging the sensor puts the DME back on track.

Unplugging the O2 sensor is also a useful troubleshooting trick, as a properly functioning sensor can mask other problems such as dirty injectors and vacuum leaks. If the engine doesn't run well without the sensor plugged in, something is definitely wrong.

Another interesting tidbit...the DME also goes into open loop mode at wide open throttle (assuming your throttle switch is working and adjusted correctly), but WOT open loop mode also disregards the airflow sensor as well. If you have any vacuum leak problems that will cause the AFM to read low, you'll notice the car runs poorly at part throttle, but wakes up and runs fine if the throttle is floored. This reaction will also diagnose a defective O2 sensor. If the car still runs badly at WOT and you've confirmed the WOT switch is properly activating, then the problem can be narrowed down to fuel pressure, crank sensors, fuel injectors, or ignition (or wiring to those systems...:eek:), as o2 sensor problems, vacuum leaks, and air metering issues are irrelevant in WOT mode.

Armon 11-12-2012 03:34 PM

The car doesn't run right, the rpm surges back and forth. I did a lot of cleaning this past weekend, around the throttle, pull the J-boot off and cleaned it and the valve. I didn't take it off though but now the engine or the surging sounds a little better. I also notice a coolant leak on the ground today, I hope this haven't broken something else.

Anyway, the only way I can get the car to idle normally is cramp the inlet hose that goes into the ICV...why does this work, I don't understand? I even drove it down the road and back, almost no acceleration problem... When I got back home I remove the pliers from the ICV hose and I heard air, then when tightening the pliers the air stop. This is strange because this isn't a sucking sound, its more like air coming out or being released, like a hole in tire. I spray carb cleaner all over that car this past weekend and I couldn't find any leaks... the engine rpm never changed. What could that be, crack intake manifold? Maybe I need to take it off again...

HondaDustR 11-12-2012 03:37 PM

Did you try disabling the idle stabilizer as explained on clarks garage? What rpm does it idle at with the diagnostic port jumpered?

Gawernator 11-12-2012 03:57 PM

Quote:

Originally Posted by HondaDustR (Post 7087900)
Open loop mode is where it reads straight off its fuel delivery maps as a function of rpm, air flow meter position, coolant temp, and air temp. There is no feedback loop as the O2 sensor circuitry trims the injection duty cycle depending on whether it is too rich or too lean. The problem is, it is a rudimentary system tacked on after the fact. the DME was not originally designed for it, and it was added on to allow import to the US. It does not constantly monitor the O2 sensor signal, it only triggers an interrupt when the sensor signals too rich or too lean, meaning the mixture constantly oscillates about ideal. As the sensor response speed slows with age and the range reduces, the mixture can vary more wildly and end up offsetting too rich as the sensor sensitivity declines. Unplugging the sensor puts the DME back on track.

Unplugging the O2 sensor is also a useful troubleshooting trick, as a properly functioning sensor can mask other problems such as dirty injectors and vacuum leaks. If the engine doesn't run well without the sensor plugged in, something is definitely wrong.

Another interesting tidbit...the DME also goes into open loop mode at wide open throttle (assuming your throttle switch is working and adjusted correctly), but WOT open loop mode also disregards the airflow sensor as well. If you have any vacuum leak problems that will cause the AFM to read low, you'll notice the car runs poorly at part throttle, but wakes up and runs fine if the throttle is floored. This reaction will also diagnose a defective O2 sensor. If the car still runs badly at WOT and you've confirmed the WOT switch is properly activating, then the problem can be narrowed down to fuel pressure, crank sensors, fuel injectors, or ignition (or wiring to those systems...:eek:), as o2 sensor problems, vacuum leaks, and air metering issues are irrelevant in WOT mode.

Very useful info! Luckily my car has been running pretty well at part and wide-open throttle, and very well with my Wilk's Chip turned on (which bypasses the O2 sensor, the Race version). Is the throttle switch the same as the throttle position switch? I have been trying to figure out if mine is on the way out or not.

Armon 11-12-2012 04:09 PM

Oh, yeah, I followed Clark Garage instruction, when B & C terminals jumped it got about 1000 rpms. without it the car idles around 900.

HondaDustR 11-12-2012 05:41 PM

Quote:

Originally Posted by Armon (Post 7088043)
Oh, yeah, I followed Clark Garage instruction, when B & C terminals jumped it got about 1000 rpms. without it the car idles around 900.

But you said before it fluctuates from 600-800. If it idles around 900 without the jumper, that's about right. Please explain. Is that with the ICV blocked? Try adjusting the idle bypass screw shown in clarks garage so that it idles around 850 or so with the terminals jumped and see if that helps the ICV work properly. Have you tried unplugging the O2 sensor for the time being to see if that changes anything? Is your throttle cable adjusted correctly so that the throttle body is allowed to close completely reliably against the idle stop every time the throttle is released? Is the throttle position switch adjusted correctly so that you hear the click of the idle switch just as the throttle reaches the idle stop? Check all of the other hoses that are connected to the intake manifold for condition and tightness of the clamps. There's one directly under the "log" part of the manifold that connects to the brake booster venturi and so on. Make sure nothing is cracked. Check that your oil fill cap is tight and the rubber o-ring is still there and intact. Is it possible the fuel injector seals or the intake manifold gaskets are leaking? I had a manifold gasket blow out all of the sudden once awhile back (probably from an intake backfire) and that wrecked havoc on the idle IIRC. Any vacuum leaks will make the car run lean, which will increase the chance of intake backfire.

Quote:

Originally Posted by Gawernator (Post 7088013)
Very useful info! Luckily my car has been running pretty well at part and wide-open throttle, and very well with my Wilk's Chip turned on (which bypasses the O2 sensor, the Race version). Is the throttle switch the same as the throttle position switch? I have been trying to figure out if mine is on the way out or not.

They're the same thing, although for the NA engines all it is are two switches. One is the idle switch, which clicks, and the other is the wide open throttle switch. It's pretty easy to check. Just unplug and use a multimeter set on ohms. The center pin is common and each of the outer pins goes to each switch. The idle switch should read at or near 0 when the throttle is closed, and the WOT should read at or near 0 ohms at about 90-100% throttle position. I have had problems with mine in the past where the WOT switch was unreliable. Eventually had to take it apart, found that the actual switch contacts were just fine and there had to be a bad connection between the switch and the pin, dig through the plastic with a razor blade until I got to where the WOT pin connected to the internal switch connections (it was cast into the plastic base!) and soldered a jumper wire from the exposed pin to the switch, as the spot weld on the original connection had failed. It beats paying $80 for a new one.

Armon 11-13-2012 05:28 AM

The put the new ICV in the same direction as the old one, with the arrow pointing toward the intake manifold. I also check the voltage at the plug, 12v, and the resistance was good. The car is still doing the same thing so its not the ICV.

Yes, when ICV hose blocked its normal at 900, however now rpms is fluctuating 1100 to 1600. I adjusted the idle and it didn't work. The hose from my brake booster goes to side of manifold, near the throttle switch, no hoses are underneath that I can see. I'll checked out the 02 sensor next and then the other stuff you mention, thanks.

Armon 12-17-2012 08:03 PM

My Porsche mechanic found a burned out terminal in the engine computer and bad throttle switch. Don't know if one cause the other to fail or not but replaced both of them and now the car is idling normally.


All times are GMT -8. The time now is 11:24 AM.

Powered by vBulletin® Version 3.8.7
Copyright ©2000 - 2020, vBulletin Solutions, Inc.
Search Engine Optimization by vBSEO 3.6.0
Copyright 2020 Pelican Parts, LLC - Posts may be archived for display on the Pelican Parts Website


DTO Garage Plus vBulletin Plugins by Drive Thru Online, Inc.