Idle Speculation

[zodman]

The car was driven for an hour down HWY1 down to HWY84 to let all the settings sink into the car's computer since the previous battery was well past its life of service and could no longer hold a charge. Drive was perfect. Initial RPM's at idle were around 800-900. Temp never peeked past the 180° part. Cycled through the heating and cooling to make sure coolant had no air-pockets left in the system. Vented properly with the hinge in the 'up' position.

I let the car cool down after I enjoyed a nice meal at Alice's Restaurant, and then took HWY 35 back to I-280...Just after I left the restaurant, the CEL light came on. No flashing, just solid. I don't have a code reader, but I noticed the RPM's at idle would not drop below 1200. Nothing else seemed out of place.

I know it is just 'Idle Speculation' at the moment without knowing any codes, but any thoughts were to look first on the car? And since I am practically tapped now from the list of repairs(listed below), is there a reader of this thread local to the San Francisco Bay Area with a Durametric available for a read? Otherwise I may have to wait a month or two until I can get an 'Enthusiast Durametric' and someone with a PC laptop since I'm Apple based without any 'PC emulator' software. I will get a Durametric at some point in the near future, I just don't have the budget for one at the moment.

I'm just thinking ahead at the moment and I know it's just a guessing game without knowing the codes for anything, but usually there is someone that has been through the same thing or has known a similar issue.

Will an OBD II Reader work in this case since it is an emission type of issue?

Thanks for any insights...

DIY work performed on my 2001S 986 Manual:

LN IMS Bearing

RMS

Clutch/flywheel/Pressure Plate

160° Thermostat

New Oil

New Coolant

New Transmission Fluid

New Oil Cooler (and seals)

New air filer/oil filter

New Spark plugs

New positive crankcase ventilation valve & seal

New Intake Manifold seals (4-6)

New Intake Manifold seals for Fuel Injectors (4-6)

Through cleaning of the Throttle Body & Resonance Tube

New seal for the Throttle Body

Cleaned MAF

New Battery

[Ahsai]

There are literally hundreds of codes that can trip the CEL so it's difficult to speculate. As a stopgap solution, you can find someone who has an Android phone and get this dongle http://www.amazon.com/Bluetooth-CAN-BUS-Diagnostic-Windows-Android/dp/B009RLB4Z8/ref=sr_1_3?ie=UTF8&qid=1388867860&sr=8-3&keywords=Bluetooth+obdii

and download the free version of the Torque app to read the code(s). The paid version ($5) allows you to monitor and plot lots of sensor readings. It's no Durametric but it's way better than driving the car for a month or two without knowing what caused the CEL. Just my 2 cents.

[Ahsai]

Oh and yes, a generic OBDii reader will also work. Depending on the code(s) retrieved, you can decide if you need Durametric for more in-depth diagnostic. Could be as simple as some sensors you disconnected not properly seated afterwards.

[zodman]

Neighbor had a Matco MD9000a so I was able to retrieve 6 codes:

P1126

P1133

P0102

P0507

P1128

P1130

P0102 points at the MAF, but it was cleaned. P0507 points to the idle Control System. As a whole, it appears to be pointing towards something in the air/fuel induction like a vacuum leak. I did button everything down as tight as tight can be...even throughout this process, bolts were torqued to the Bentley specs....yet obviously something isn't right.

The AOS was replaced somewhere around 70k miles and the car is at 87k now. I didn't clean out the long tube that connects the AOS and the positive crankcase ventilation valve, but I did replace the valve. There was some white smoke initially, but that appeared to go away during the initial test drive. I have not tested the presence of suction in the oil filler cap during this process.

So, I'll recheck all the potential trouble spots for a vacuum leak...Maybe someone knows what the series of codes relates to if there is a single trigger. Could be the MAF finally needs replacing and the AOS is faulty...Still idle speculation at the moment.

[Loren]

Way too many codes - I would clear them all and see what comes back immediately.

[zodman]

After clearing the codes, resetting the boots for the throttle body and the resonance tube, double checking all the vacuum hoses, finding the hose that connects the positive crankcase ventilation valve and the AOS to be slightly askew on the AOS side and then trying to reseat it as proper as I can...Then re-cleaning the MAF again...Then driving for about 10 minutes, the car was able to summon only one code...

P0102

This time the high idle appeared to be gone, but the low idle overall appeared to be rougher than yesterday. Also, testing the presence of suction in the oil filler cap did make the low idle appear to be worse, almost to the point of quitting running.

Idle is around 850.

Edited January 5, 2014 by zodman

[Ahsai]

Have you tried cleaning the idle stabilizer (if your engine has one)? 10 min may be too short for other codes to show up though.

[zodman]

(idle stabilizer)That's that older part on the 2.5L engines...This engine is the 3.2L S model from 2001(6Speed Manual Transmission). As for the throttle body itself, it was cleaned with the throttle body cleaner through and through. Last time it was cleaned was around 70k when the AOS was replaced. Sticks in the memory as I had broken the J-tube and had to order a replacement part for that. Now at 87k, it was still pretty clean but I gave it a solid cleaning and even replaced the gasket seal between the TB and the tube.

10 minutes out till CEL and 10 minutes back....20 minutes in all. If it was flashing, other than the cursing, I would have turned the engine off and flat-bedded it.

Was following Loren's recommendation of clearing codes to see which little piggy (P0102) was first to go whee whee all the way home. :lol:

[Ahsai]

The matco scanner should be able to read MAF real-time reading. For a 3.2L, I think it should read around 4 g/s at idle (~700rpm) after fully warmed up and no electrical load on the engine (e.g., a/c). The rough idle suggests vacuum leak though. You can also check the short-term fuel trim at idle and see what the DME thinks.

[zodman]

This image is a series of screen shots of the test I made while the vehicle was running...No electrical load was running.

No other tests can be done on the reader that I can see.

First screen was the test selected, next series of shots are the results...I improvised the scroll down by trimming the redundant data.

[Ahsai]

Your engine was not fully warmed up (coolant at 64F) and the fuel system was not even in closed loop. Can you repeat the test with engine fully warmed up and at idle i.e., car not moving? Also any MAF reading?

Edited January 6, 2014 by Ahsai

[zodman]

I can't interpret the data from the "View Freeze Data", other than noticing a few irregularities, the car was warmed up, so it should have read around 180° not 64° - unless that is the ambient temp. I don't have a manual for that OBDII tester, nor can I find one online, even from the MATCO website themselves. The actual rpms at the time of this test wasn't 1400rpms, but more like 900rpms. When the CEL was triggered, the car was in first gear, somewhere in the 4000 rpm range after an initial warmup from a cold start. I recall a post where Loren notes the MAF doesn't start really working until you hit 4000+rpm.

I know there are 3 basic reads for the P0102, but a multitude of different causations depending on the situation. It *might* be possible I have two items that need to be replaced....The AOS I bought a couple of years ago may be butter side down toast and/or the MAF just decided to up and die. I bought the car in '06 with 47k and the car is now at 87k. A new updated MAF would not require an ECU flash.

However...Given that the AOS has maybe only 17k miles on it or less, it should be initially excluded. Since the MAF was working proper before my recent repair, I'd like to exclude it as well...but I can't rule it out either or either item.

Maybe the most likely scenario is a vacuum leak. Since the idle is affected by the AOS when the oil cap is pulled, I'd say that hose that connects the positive crankcase ventilation valve and the AOS may not be fully pressed onto the AOS. It was a bear to disconnect and a bear to reconnect...but the space to get to it is extremely limited due to the cables in the way and that metal tab that comes off of the fuel rail (For the ele. cable tie down). If I had my druthers, I'd replace that plastic hose with a rubber hose with a couple of clamps instead...even weave it around the back of the engine instead of across the engine. I just don't know if doing so would cause problems.

Since the MAF was cleaned & re-cleaned using the proper MAF cleaner, I am wondering "if" it is better to disconnect the battery for a few minutes and let the engine "see" the re-cleaned MAF that way, instead of simply clearing the CEL. Further, as a secondary test to rule out the MAF, is to let the car run for a bit without the MAF connected...I know these cars can run without the MAF, but what exactly should I be looking for/expecting after running it without the MAF for a brief period? (Don't want to hurt the Cats) Should I keep it under 4000rpm and give the car a case of amnesia first(battery disconnection)?

Pardon the novella, I'm just trying to be through in words.

[Ahsai]

Wow, I cannot even imagine how the data can be off by that much. I suggest you use another scanner (Durametric will be the best) since this one will only confuse you more.

A little lumpy idle after oil cap removal is completely normal though since it creates a slight vacuum leak.

I tend to agree vacuum leak is the number one suspect and will definitely eliminate that possibility before replacing the MAF. If you can read MAF data, you can do a snap throttle test, correlate rpm with MAF data and see if the MAF signal is strong enough at higher rpm.

Edited January 7, 2014 by Ahsai

[zodman]

I'm reading a similar Matco manual for a different more current reader since I can't find the right one for this model...The test data on the above screen shots is most likely the data "when" the CEL was triggered...hence the "Freeze Frame" notation doesn't obviously match the real time moment when I was reading the data from the tool...I'll try to run the test again after giving the car some mild amnesia after I recheck that bloody AOS hose.

[Ahsai]

Ok, that makes a lot more sense. You need it to show real-time data. If the above is accurate, that means the code was tripped at cold start, even before the system goes to close loop. Are you sure the MAF electrical connection is seated properly?

[zodman]

MAF is seated correctly, paying attention to the nubs where they connect, there is only one way it connects. Will run freeze data down again from a warm engine after rechecks of hoses etc.

[zodman]

Like a bad habit I'm back...I drove like I had a high deductible as I warmed up the engine, keeping the rpms under 3k, then I parked and turned off the engine for a few minutes as the CEL didn't trigger, then I started again and within a minute or two, with the engine still crispy hot I triggered a CEL while still under 3k rpms. P0102 code. I checked the freeze data and it showed virtually the exact same info as before except it showed the coolant temperature 5 degrees "cooler." That's pretty strange considering the previous result was a read "cold" engine and this was a "hot" engine.

I cleared the code, and then I drove like I had a low deductible. No code thrown....yet.

From the previous reply until the drive now, I rechecked and reseated that lower hose for the AOS that I was concerned about. It is now seated firmly after I took off the Oil hose for a bit more leverage and room. The upper Oil Hose was also inspected and reseated (it is also new), the MAF connection was also re-verifed as to being properly connected.

So aside from reading and clearing codes, seeing how the freeze data is erroneous with this car I may have to either find someone with an unlimited Durametric here in the Bay Area that'll be kind enough to give me a read, or throw some more funds at things in the dark sometime next month.

[Ahsai]

Yea, you should have a much better chance diagnosing it correctly with Durametric. The freeze frame data maybe referring to a pending P0120 code that got tripped when the engine was still cold but CEL not turned on yet. Some codes require multiple tripping before the CEL is turned on.

[zodman]

Yeah, I know with some readers some codes get tripped and others are pending, but this device doesn't state pending events, and, after each code event the data is purged from the reader and the system so no pending event could have occurred (looking at it as cause and effect as a best guess). The other item of note is that the FD shows both events occurring when the car is motionless...Yet both times the triggers on the dashboard occur when the car is in motion. If the FD was operating correctly, it would have recorded the speed as well.

Guess now is the time for some fermented wheat flavored water...

Edited January 7, 2014 by zodman

[zodman]

Quick question for anyone that's had to futz with the P0102 code, I recall reading about an error code mystery that was solved after finding the gas cap had cracks in the rubber seal and after replacing the fuel cap the error code hadn't returned, but I can't relocate that post now to recall the error code in question, but I thought it may have been the P0102...Any thoughts on that...e.g., would a worn out rubber seal on the Gas cap trigger a P0102 event?

[Loren]

Quick question for anyone that's had to futz with the P0102 code, I recall reading about an error code mystery that was solved after finding the gas cap had cracks in the rubber seal and after replacing the fuel cap the error code hadn't returned, but I can't relocate that post now to recall the error code in question, but I thought it may have been the P0102...Any thoughts on that...e.g., would a worn out rubber seal on the Gas cap trigger a P0102 event?

Not the P0102 fault code.

P01012 is clearly an MAF fault. Usually cause is that the MAF it disconnected or bad.

[zodman]

Guess I won't futz with it further then. Before your response I had bought a new gas cap (it needed replacing) at the dealer and indeed a P0102 event occurred shortly after that. Rpms were around 1100 at the event, and this time the freeze data showed a higher coolant temp...Ordered a new MAF from Pelican for a Friday delivery, 986.606.125.01 (replacing a 986.606.125.00) and I'll have to pay the piper next month.

Till then really short trips and no where near 4k rpms.

[Loren]

If the gas cap needed replacing that is good that you replaced it.

As far as the P0102 - it could not really have anything to do with the gas cap - just a coincidence I think.

Here is the troubleshooting for a P0102 from the Porsche Service Manuals. Either the wiring is bad or the MAF is bad - period.

(click to enlarge)

[Ahsai]

That's good to know, Lauren. A simple and short diagnostic tree.

[zodman]

This maybe slightly OT...

Sunset sent a track num to pick-up the transmission fluid they incorrectly sent to me....Yesterday I ordered the MAF from Pelican. The MAF was dropped off on the front porch by the FedEx driver today with nary a knock or a ring on the door...The UPS driver later arrives with nary a knock or a ring on the door...

I guess you can guess what happened next even without any Idle Speculation...