Boxster idle problem

[davek_964]

Last weekend, my girlfriend's Boxster showed a check engine light. I have Durametric and there were two faults logged - bank 1 and bank 2 oxygen sensors. Now, this is where the problem started :

I cleared the fault codes, but as soon as I did so the car started hunting badly at idle. After a bit of use over a few days it settled down but still sounded like it wasn't firing quite right at idle - idle was fairly steady now but didn't sound right - almost like it was missing quite often.

Today, I checked again. No check engine light this time but both oxygen sensors had logged faults again - in addition there was a code for total misfires and misfires on cylinder 2.

Again I cleared the codes and again it massively upset the idle - to the point it stalled several times. It seems to have settled down now.

I don't understand why clearing the codes of having such a drastic effect on the idle. I don't believe both oxygen sensors are failing especially since it has the same problem from stone cold when the O2 sensors are not used anyway.

I'd suspect the MAF but am reluctant to start dismantling her car to get to the engine compartment - I did start but found a snapped cable at the base of the hood and since she's out I'd rather she knew she has other issues and not think I broke anything!

Misfire on cylinder 2 could be a coil pack I guess and that certainly would explain the 'missing' sound at idle - but I doubt it would throw oxygen sensor errors on both banks.

Any ideas before it goes to a specialist? I am beginning to think something is wrong - hence the fault codes - but that the idle issue is actually just a result of Durametric clearing the codes for some reason.

[anonymous-user]

Personally I'd be wanting proper diagnosis rather than internet guesswork. Changing bits on the off-chance can end up being counter-productive.

However, if it was my car I'd check when things like spark plugs and air intake filters were last changed and make sure the air filters were clear from obstruction. If that was all OK then the car would be off to a dealer/specialist for diagnosis.

[DJMC]

I know nothing about this but find it interesting that having a plug in diagnostic tool seems to have made the problem worse. Add in that you don't believe the error code and I wonder about the usefulness of these tools in DIY situations?

I'd be in exactly the same quandary, but, not having the tool it'd be straight off to my OPC (or indy perhaps if out of warranty) so I'd just have to be happy with a no doubt big bill.

Perhaps there are other situations where the diagnostics would be believed and not "harm" the car further?

Anyway... someone with knowledge should be along soon...

[davek_964]

I've used Durametric for years on a 996 turbo and a Cayenne and it's a valuable tool. Why it's making things worse with the Boxster I don't know.

[davek_964]

For info, I've proved it's not the MAF - unplugged it and idle hunting continues.

[Mogul]

Have you checked for AOS failure? (Try and remove the oil filler cap when the engine is idling. If there is a strong vacuum, the AOS has failed and that can lead on to some idling issues).

[Flyferrari]

I'm not sure your logic is sound on that one. If you unplug the MAF and nothing changes then that would more likely imply that the MAF was u/s as the ECU doesn't see the unplugging as a "change". To be clear, I don't profess to have any expertise on car engine electronics, I'm just questioning your logic. I don't imagine your code reader is causing problems per se but once you clear a fault code the ECU will assume that the sensors are "fixed" and will try to make use of them again. It will take a while before it "realises" that they are giving it rubbish information and locks them out again with the associated code. I also think it highly unlike that two O2 sensors in completely different parts of the engine fail at exactly the same time.

I think it might be time for a visit to your local Indy!

[davek_964]

Flyferrari said:

I'm not sure your logic is sound on that one. If you unplug the MAF and nothing changes then that would more likely imply that the MAF was u/s as the ECU doesn't see the unplugging as a "change". To be clear, I don't profess to have any expertise on car engine electronics, I'm just questioning your logic. I don't imagine your code reader is causing problems per se but once you clear a fault code the ECU will assume that the sensors are "fixed" and will try to make use of them again. It will take a while before it "realises" that they are giving it rubbish information and locks them out again with the associated code. I also think it highly unlike that two O2 sensors in completely different parts of the engine fail at exactly the same time.

I think it might be time for a visit to your local Indy!

If the MAF is unplugged the car sees a fault and uses defaults. Unplugging it and seeing if it runs correctly on default values is a common way to check whether the MAF is giving false readings.

[ooid]

DJMC said:

I know nothing about this but find it interesting that having a plug in diagnostic tool seems to have made the problem worse. Add in that you don't believe the error code and I wonder about the usefulness of these tools in DIY situations?

The diagnostic tool is okay, but clearing the codes are not.

Going back to the OP, OPC or anyone with official Porsche durametric would give you a better diagnosis now.

Usual suspects; MAF, o2 sensors (they usually give codes and cel), coil packs, and spark plugs or air leak in the system.

Try if you can borrow a working MAF from another boxster, if the car behaves better you found your solution. Sometimes, unplugging MAF would still give random misfires and rough idle due to the codes in the system.

[Rockster]

davek_964 said:

Last weekend, my girlfriend's Boxster showed a check engine light. I have Durametric and there were two faults logged - bank 1 and bank 2 oxygen sensors. Now, this is where the problem started :

I cleared the fault codes, but as soon as I did so the car started hunting badly at idle. After a bit of use over a few days it settled down but still sounded like it wasn't firing quite right at idle - idle was fairly steady now but didn't sound right - almost like it was missing quite often.

Today, I checked again. No check engine light this time but both oxygen sensors had logged faults again - in addition there was a code for total misfires and misfires on cylinder 2.

Again I cleared the codes and again it massively upset the idle - to the point it stalled several times. It seems to have settled down now.

I don't understand why clearing the codes of having such a drastic effect on the idle. I don't believe both oxygen sensors are failing especially since it has the same problem from stone cold when the O2 sensors are not used anyway.

I'd suspect the MAF but am reluctant to start dismantling her car to get to the engine compartment - I did start but found a snapped cable at the base of the hood and since she's out I'd rather she knew she has other issues and not think I broke anything!

Misfire on cylinder 2 could be a coil pack I guess and that certainly would explain the 'missing' sound at idle - but I doubt it would throw oxygen sensor errors on both banks.

Any ideas before it goes to a specialist? I am beginning to think something is wrong - hence the fault codes - but that the idle issue is actually just a result of Durametric clearing the codes for some reason.

What are the CEL error codes?

You think right: When you clear any error code among other things this clears the long term fuel trims.

If the engine needed extra fuel or less fuel than "normal" this can have the engine running poorly until the DME (re)learns. You can observe DME attempt to relearn by viewing the short term fuel trims. They can be moving from "way too" rich to "very lean" (~+10+ to ~-10%) right after the error code was cleared. Then as the engine idles and as the DME adapts the trims move less.

The poor idle cold is probably due to an intake air leak. 'course insufficient fuel supply or pressure is another possibility but air leaks are much more common.

Unless someone has been at the engine wrenching on the intake the most likely explanation would be the AOS. (I'm on my 3rd replacement AOS with my 2002 Boxster.)

You can try to remove the cap against a suspected bad AOS's excessive low pressure but even if the cap comes off without any real struggle this does not let the AOS off the hook. There is a pressure measurement you can do if you have a spare oil filler tube cap that you can epoxy a hose connection to and then connect a hose from a vacuum gage and actually measure the amount of pressure (it will be lower than atmospheric). I do not recall what a good vs. bad AOS reading is.

However, another source of an air leak can be the oil filler tube cap. The factory cap on my Boxster developed a leak -- even though the cap was properly tightened down -- which affected idling but did not have any noticeable affect at higher RPMs. (This is expected as the amount of air passed by the cap was very small, but still significant at idle speeds. At higher engine RPMs, not so much.)

You can "test" for a leaking cap by with the engine idling and the rear lid open listen for any sounds of an air leak. Even with my lousy hearing I heard this but it was in a quiet place. Move the cap around/about and see if the engine reacts. The cap on my car was so "worn" not only could I move it sideways a bit and cause a leak I could actually pick it up a bit off of the tube and it would then leak. Oh, the car had around 150K miles on it but the cap can have a shorter life. (Believe it or not the replacement cap developed a leak, after nearly another 150K miles. Did I mention the Boxster has now around 312K miles on it? As the miles go up you'd be surprised as to what wears out...)