'02 Carrera: High Idle/poor part throttle performance/CEL codes..

I have posted before on this issue, but here's the latest hoping that any of our members might have some guidance/advice: Mine is an '02 Carrera with 51,000 miles. It ran wonderfully for my first year of ownership. Out of nowhere, I recorded P0507, P1128 & P1130 codes. The idle was high and part throttle performance completely suffered. My first move was to replace the MAF sensor, which did not improve matters. Brought it to the local dealer for assessment, and the diagnosis was the air/oil separator accounting for a vacuum leak. $900 later, the car was claimed to be far better and no CEL. Well, one day later I have a CEL and the part throttle performance is anemic...almost wheezy. Once you apply full throttle, it performs as expected, but you can't drive around full-throttle (at least most of the time). It also seems to be drinking fuel at an alarming rate. During the AOS replacement, the throttle body was cleaned by the technician and the service manager suggested that if things were not completely right with the car, it's likely that the throttle body needs to be replaced. It sure sounds like 'plug and play' repairs hoping something works. Recall that when the AOS is replaced, much of the top-end of the engine is removed to provide access, so I would expect that intake leaks should be eliminated via the work process. I purposely let the dealer replace the AOS (which I could have done myself) hoping that their diagnosis was based on previous like cases. I could have saved myself $700. Looking for any input on next steps. The codes I'm getting now are the P0507 (idle control @ stop) and a 5525 (failed data exchange from the DME). I imagine that replacing the throttle body is the next step (dealer parts are $500 vs. $185 on the open market for the Bosch part)? Thanks in advance, Bob

Try taking the oil fill cap off and making sure the o-ring has a good, clean seal. Might be getting air in the system from a poor seal there.

Sounds like you have an intermittent issue. Seems as though the car passed the self tests after the AOS replacement and TB cleaning, then an issue came back a day later, as an intermittent issue. Intermittent issues are usually electrical in nature as opposed to physical/mechanical but not always. I have seen the plastic vacuum reservoir under the intake that serves as reserve vacuum for the various change-over valves leak on an intermittent basis. The leak caused by a crack at the bottom seam can intermittently seal its self as vacuum pressures and temperature change it propensity to leak. Intermittent issues are very hard to diagnose because test results can not be believed as 100% plausible. All that said, your symptoms does seem to indicate a vacuum leak, but I would recommend a Porsche specialist to diagnose this again, preferably the one who did the previous attempt, as he is most familiar with this cars issue and has done previous tests.

You can try e-gas re-calibration first (free) then ask a shop to do an intake air leak test. From your description it got to be a gross leak that's easy to find (if the root cause is air leak). Then replace the or swap in a known good throttle body. BTW, I can't find any documentation of code P5525, are you sure that's the right code?

Quotes source post: 15625761

I'm not the only one who google searched P5525 and got pictures of panties, right?

Code on my Foxwell was simply 5525 and not P5525. Appears to be a communication issue between the car and the scanner. I will bring it back to the dealer for further diagnosis, which is why I had the AOS replaced by them in the first place should it not cure the issue. My car has run well since I bought it, but I always felt that the part throttle performance was never what I expected as it was decidedly 'choppy' and pronounced, which I chalked-up to the VarioCam activation. Perhaps it was the onset of a vacuum issue. Giving thought to trying the propane test to see if I can locate a leak before bringing it in for further service.

Quotes source post: 15626097

Visually inspect your oil filler neck. They tend to crack and cause the issues you're experiencing. I've replaced mine ones so far and I'm much more careful with it now.

I agree with everyone...you have a massive vacuum leak somewhere on the engine. Like said, carefully check your oil filler pipe for cracks. You may have to smoke test the vacuum system and crankcase. Also, since so much top end stuff has to be removed to replace the AOS, carefully check the intake plenums and rubber joints, make sure they are aligned properly on both sides and the clamps are tight. Your DME is confused because too much air is getting into the engine after the MAF sensor, thus it is enriching the fuel mix to compensate. Also, after checking everything and hopefully finding the leak, disconnect the negative lead of the battery for 15 minutes to reset the adaptations in the DME, perform an e-gas calibration and then drive for at least 30 minutes and one complete heat/cool cycle. Good luck!

First, thank you to all posters with your thoughts and input. Really helpful! To date: - I've replaced the oil fill tube shortly after taking ownership of the car as it was identified as a potential issue at the PPI. - The dealer (Euroclassics in Richmond, Va) just replaced the AOS this week, and in the process removed much of the intake tract in the car to access it. I must assume that all intake tracts are sealed upon their reinstallation. I agree with DBJoe that there is too much air entering after the MAF, accounting for the high idle and I suspect the high fuel consumption. Smoke test is the next logical step before I begin 'plug-n-play' replacing everything. Bob

Might be a good idea to smoke test when the car us up to temp. Good luck!

Ahsai: I did the e-gas recalibration after disconnecting the battery and after some initial idle hunting and rough running, it all came right. Idle is now 700rpm and the throttle run-up is silk. Haven't had a chance to road test it, but in the garage it is a different beast. Thanks for the recommendation! Fingers crossed! Bob

For the record, the e-gas recalibration process is: - disconnect the battery and leave it disconnected for a few minutes - reconnect the battery - turn the ignition on for 60 seconds but do not start the car or touch the gas pedal. - switch the ignition off for a minimum of 10 seconds - start the car and allow the DME to relearn; I let it idle for a good 10 minutes. Bob

Good luck and I really hope it's that simple.

It wasn't... The idle now sits at 1,200 to 1,300 rpm. Off-idle performance seems encouraging, but there's a noticeable flat spot from 2,400 through 3,400 rpm. Once full(er) throttle is applied, it runs strong and pulls hard. On throttle chop in low gears, the car 'snatches' and bucks. It's far better than before, and unless this is how a 996 runs normally (I have no reference point), it's not right. The dealer suggested (when they replaced the AOS and cleaned the throttle body) that it could be the throttle body, which seems plausible...I guess? Does any of these observations align with a troubled throttle body?

Is the CEL ON and has the code P0507 returned?

Here's the latest: Yes, the CEL is back on. The dealership seemed absolutely certain that the issue was a throttle body issue (after replacing the AOS for $1,000 and no better result). Dealer price is $495 on the TB, and I sourced it online for $135 (new Bosch TB). I also changed plugs/coils while I was at it. Recalibrated the e-gas and it fired with a couple of small backfires through the airbox and settled into a 700 rpm idle. I was now thinking that I had it solved! Wrong!! High idle has returned...now 1,400 rpm. Light and part throttle performance is non-existent, almost 'wheezy' for lack of a better term. Full throttle application results in great performance. Given these conditions, the likely culprit is a significant vacuum leak would be the likely diagnosis. I brought the car to a local independent repair shop with Porsche pedigree, and after a day of working on it, they are stumped. There was the slightest of vacuum leaks uncovered during a smoke test (likely when the technician replaced the AOS at the dealership), but the comments are that to get that high of an idle coupled with the part throttle marginal performance, it would take a massive vacuum leak, and that's just not the case. They are now reloading latest codes into the DME and keeping the car overnight. This all happened without any warning. A CEL appeared and was accompanied by weak light throttle performance. It came out of nowhere. Not sure where to go next.... Bob

I smell a rat......literally !! With that low a miles the car sits a lot, I would check the wiring carefully for rat/rodent evidence... How often does it do this? I would monitor the dual/TPS and throttle jacking values for proper function/value while the vehicle is "acting up".

Quotes source post: 15670130

Good suggestion that I�ll pass along to the shop today. Not sure about the possibility of a critter imposing itself unless it occurred prior to my ownership and it just now has manifested itself. Frustrating!

Would a crack in the intake runner do this? Maybe not seen when tested cold but expands when hot, just a thought.

I am heading out to pick up the car as the independent Porsche shop has thrown-up their hands. NOT a good sign. They have exhausted all avenues: no vacuum leaks; all sensors are operating within range. Wondering if plugging-in another DME would be a step to try as the codes indicate that there is an issue with the throttle body (it is new and one week old), and that it is reading a lean condition and pouring fuel into the engine to compensate....accounting for the high fuel consumption rate. I'm running out of options.... The shop questioned whether a reflash was put in the DME as the car's performance is far better than any 996 they've driven. None of my service records or receipts indicate that a reflash was done at any time. Thoughts?

Just wondering have they checked: - the EVAP purge valve (the one that's right above the alternator), which can create internal air leak (and will not be detectable by smoke machine)? - fuel pressure Also, did the 1128 and 1130 codes come back? Have you tried unplugging your MAF and drive the car to see if it makes a difference? I guess trying a different DME may not be a bad idea. However, you said the car has been fine in the past year and it just acted up suddenly.

Quotes source post: 15671534

Full EVAP system checked-out. P0507 is still there, and now getting P1126 and P1133 codes. Can one interchange DME's on this model (2002)? Noticed on eBay that used 996 DME's are relatively inexpensive, and given how reluctant the car is to give-up why it is not happy, it could well be the brain? Thanks! Bob Earlier, I did unplug the MAF, but did not drive it. Car idled down to 700 rpm and ran-up the rev range flawlessly. As a result, I replaced the MAF thinking it was at fault, but as we've come to know, it was not.

Unfortunately the DME is paired with your immobilizer so you would need to reprogram the new DME or you have to get the whole set key+immobilizer+dme How about fuel pressure? Although that's less likely because you said the engine runs fine at higher rev/load.

Btw, did the shop check the wiring between the throttle body and the DME?

Could it be as simple as a bad gas cap?

Quotes source post: 15626080

Sound like your browser couldn't find anything either and was making suggestions based on your search history. edit - not to take away from this thread - I am sympathetic to your plight OP. Sounds frustrating as hell. I have no useful suggestions but I hope you can resolve this.

I couldn't find a fault code P1133. Found fault P0133. Relates to sensor trouble. All your symptoms point to worn out disconnected or damaged oxygen sensors. Have you worked on your exhaust lately? All sensors need replacing sooner or later. Lots of heat, etc. This is where I would focus my attention. Let us know what works out for you.

Quotes source post: 15674325

I've only removed the mufflers to do a plug/coil change, but this is the 2nd set of O2 sensor-related codes I've seen on the car. So, here's the good/bad news. During the drive back from the shop (who, btw, did not charge me for the diagnosis/work because they felt that they did not solve the issue; when was the last time you heard that? The shop is European Cars in Williamsburg, Va.) the ran as it did before all of this began! Off-throttle and light/mid throttle performance was back to how it was; there is an ever-so-slight 'bobble' at 3,000rpm which I have always felt was the transition point for VarioCam; and the idle (though not ideal) was at 900rpm. It was almost miraculous, but it speaks to a larger issue as this condition would seem to be intermittent. We have non-stop pouring rain for the next few days, so test drives will need to wait. Here's the code screen on my Foxwell. As you can see, it calls out a P1133 (oxygen sensing adaption, lower load range, bank 2). Is this a faulty O2 sensor?

No, these are not O2 sensor codes. These codes are complaining that both banks are running too rich, to the point that the DME cannot correct. Since both banks report the same, the problem has to be something common such as high fuel pressure, bad MAF or throttle body or wiring.

Lower load range is just off idle/partial throttle.

My code list doesn't list P1133. Have you traced your leads going to or from all your O2 sensors? There's a possibility that those were crossed when reinstalling your exhaust. Do your tail pipes smell heavily of raw gas? Your DME may be trying to adjust the fuel trims based on faulty info. Humor me and take a moment and drive your carrera for awhile and see if there is an abundance of petrol in the air. Report what you find. I certainly might be wrong, but could be worth a look. Thanks

Check this post out https://rennlist.com/forums/996-forum/1028741-high-idle-and-cel-p0507-p1126-p1128-p1130-p1133.html seems close to your situation. Maybe some carb cleaner is in order and watch the fuel trims while carefully spraying. Trims change drasticlly and you found the leak.

Quotes source post: 15674510

MAF and throttle body are new parts, so we can eliminate them. High fuel pressure makes sense as fuel consumption is higher than it should be. Wiring might be a tough one to trace...

In hind sight I appologize to the OP and all others associated with this thread as I jumped to an incorrect conclusion that the 02 sensors were to blame for rich running. Ahsai was correct in stating that the codes indicate that the DME is trying to make up for a too rich fuel trim or maybe a too lean one. Don't know if the original complaint about rough running has been resolved, but if not maybe look into the vacuum line going to the brake booster. Or you could try and get a fuel pressure reading at the fuel rail fitting. Hope you get things resolved.

Quotes source post: 15681783

Got pulled out of town for a few days, so have not had a chance to drive the 996 and see if there's any improvement. I'll report back on Friday upon my return. The local shop did a smoke test on the car and only found 1 absolutely minor vacuum leak that was an issue of a connection not fully tightened...likely a by-product of the AOS replacement. However, on the drive home from the shop, the car reverted to it's earlier performance. That scares me a bit, as it suggests that the issue might be an intermittent issue as it came-on out of 'nowhere' and seemingly has disappeared in the same fashion. Bob

Smoke machines are awesome. But on occasion they give faulty results. Allow me to explain! An engines intake system is under vacuum, NOT pressure. You could have a spot that is allowing air to be sucked in past a faulty seal or slightly loose connection. When tested using a smoke test it creates pressure and can actually seal the fault until a vacuum is applied again. good example: recently i had a cadilac that had been to several shops and people threw tons of parts at it to fix a p0171/p0174 lean codes. To make a long story short it kicked my butt too. Smoke tested and found nothing leaking in the intake system. We found TSB's talking about a reflash to fix it - tried it - didnt fix it. Took it for a test drive and noted that the fuel trims were adding fuel only during idle and not during load meaning it had to be a vacuum leak (typically). Long story short we watched the 02's and fuel trims and started spraying brake cleaner everywhere. Low and behold when i sprayed under the intake the numbers would pull fuel back hard. On this example caddy there was insanely small bellows boot between the intake manifold and the throttle body (much like the 996). It had the smallest crack that no human could ever see. Smoke tested it again and nothing came out. But spraying brake cleaner at the crack and while running and the car went nuts. not saying its your issue but sometimes the right testing method is not always the right testing method.

Quotes source post: 15683347

+1 ......I've seen my share of this kind of s**t !!!

Quotes source post: 15683347

I sure hope you get it sorted and I can't wait to here what fixed it, you will get it!

Here's the latest update on the issue, and I'm increasingly of the opinion that this is a DME or software issue as the symptoms seem to be all over the place.... At the moment, the light/part throttle performance is back...just as it was before. However, the idle was now at 900rpm, and has slowly creeped-up to 1,100rpm. I went in with my Durametric and cleared all codes. Upon starting the car after the Clear, the car stumbles and runs on 3 cylinders upon start-up! It spits codes on the Durametric to that effect (ignition failure on #6, #2, #3 cylinders) but clears shortly after and all's well. However, despite the codes on the Durametric, they do not trigger a CEL! Now, despite the car presently running well under all throttle positions, it now triggers the same CEL and resultant codes from earlier (as seen on my Foxwell code reader). Like I said, it is all the place. The symptoms arose out of nowhere, got progressively worse despite replacing seemingly at-fault components, and now we are back to where we started except the presence of codes/CEL. Makes no sense. Bob

Quotes source post: 15683347

Back in the day I used to check for vacuum leaks with a can of carb cleaner. Ah, the memories of quick and dirty auto-mechanics of a 19yr old with a Mustang.

You still have a vacuum leak or an intake leak after the TB. When your car is running, remove the oil filler cap and note any difference in how the engine runs. You said you have Durametric, so you can log values and do activations of various engine components to see if one is the culprit. Also, how did you check the EVAP system? Use Durametric and activate the EVAP system purge valve for fuel vapor and see what happens. I know you have done a lot of parts replacement to eliminate the problem, you just haven't found the smoking gun yet. Another thing to check is the vacuum line to the brake booster. To check it you can unplug that from the brake booster and plug it. A bad/broken brake booster is a potential source of vacuum leak. Smoke test - did they smoke test both the intake system and crankcase? Thinking AOS bellows leak... Rare, but not unheard of - bad AOS right out of the box. So up next might be a manometer test of the crankcase at the oil filler tube.

Thanks, Joe. I have significant vacuum at the oil filler cap. When I release it, there is a change in engine note, so the AOS that was replaced would seem to be operating correctly without leakage. Of note, when I start the car to check the cap, the idle does settle back at 700rpm and then slowly rises over time. The independent shop that last looked at the car ran the check on EVAP. Having just gotten the Durametric and gaining familiarity with it, I'll go back and try the purge valve activation. Shop assured me that EVAP checked-out, but it cannot hurt to try it. Good Suggestion on the brake booster. Let me check it out. They did an extensive smoke test and found a ever so small leak at a joint that was not tightened firmly enough, likely an oversight by the dealer when they replaced the AOS. No matter, the comment was that it was so insignificant that it could never have created an issue. Bob

DBJoe: Fired the car with Durametric connected, and activated the purge valve resulting in no change in idle. Is this the expected result? The number of inputs and available data via the Durametric is enormous. Is there a guidance somewhere that explains each of the data inputs monitored, what the values should nominally be, and explain some of it in layman's English terms. Oddly, the Durametric is reporting these misfire codes, but the CEL is not triggered.

Yes, that is the expected result....no change in idle. I find it oddly interesting (or unusual) that the misfires did not light up the CEL. I also find it oddly interesting that the misfires are occurring on opposing cylinders 1-6 and 2-4 and the high idle P0507 continues, which indicates an intake leak.

Quotes source post: 15702865

All points in my mind to us trying to have a good conversation with a crazy man, Joe, and he keeps offering-up conflicting answers that we chase-down only to find that they were dead-ends. Codes on Durametric that do not trigger a CEL?...an idle that will come down to 700 only to creep upwards to 1,000+?...light/mid-throttle performance that is severely compromised and then magically back to normal and perhaps better than before? Makes no sense and I think I treat it as such. I can only assume there's a software issue in the DME, and wondering if a scrub and reflash of the programming might be the answer. When your PC goes on the fritz, the best fix is often to power it down and reboot as opposed to trying to 'fix' the issue. I've been looking at Scott at Softronics for his reflash as I've read a lot of good things about it/him. Curious to have a conversation with him and see what his thoughts are. Bob

Maybe the CEL bulb burned out? Does it light up when you turn the ignition to ON? I agree....very confusing set of circumstances and symptoms. As far as the ECU/DME reflash, maybe that is the way to go. Softronics is good, and for comparison try ECUDoctors. At least that will eliminate one more thing.

Quotes source post: 15700811

Nope. CEL does illuminate when these codes are triggered....much to my frustration!

While researching the P1126 and P1133 codes I ran across this thread https://rennlist.com/forums/996-foru...133-p0102.html

And this one with a solution https://rennlist.com/forums/996-foru...130-p1133.html

Old threads but same issues...and everything points to the MAF sensor for the error codes. Early on you said the MAF sensor was replaced. What was it replaced with? Genuine Porsche or aftermarket? Could make all the difference in the world. For these guys, replacing the MAF sensor with OE Porsche solved the same types of problems. http://986forum.com/forums/performan...lat-power.html

This is disturbing: " A contact at BOSCH who make the sensors for Porsche admitted to me that there are differences (even though the part numbers are the same!) between the ones sold in the aftermarket in BOSCH boxes and the ones sold to Porsche in Porsche boxes or bags. A genuine sensor from Porsche costs between �300 and �500 depending on the engine model, and the Bosch ones are much cheaper. However, the Bosch items tend not to last as long, are made/calibrated to lower spec's and fail much sooner than the Genuine parts. Its even been said that the ones sold in Bosch boxes are often ones which have failed Porsche quality assurance testing. Also a Genuine item comes with a 2 year garauntee from Porsche, even if installed by an independent specialist." I replaced the MAF with a Bosch part bearing the exact same part number as the MAF in the car, as shown in the photo of the old MAF and new part box. First, I'm going to take it for a drive with the MAF disconnected. I'm also going to reach out to Bosch, as if this is the case, should they not alert consumers of a potential issue? Wow....

Was your current sensor ever replaced or is the original from the factory?

Quotes source post: 15703635

Not sure if there is a way to determine whether the existing MAF was the OEM part other than the car only has 51,000 miles and the service life of the MAF is longer than that.

Quotes source post: 15703676

Got it. Mine has 127K and I believe it's the original...

You have to remember that many many codes will not turn on the CEL on the first drive cycle, they will be set as "pending", then if they are "confirmed" on the next drive cycle the CEL will be illuminated. This is a normal function. If the codes are not "confirmed" after 20 drive cycles they will be erased, without ever illuminating The CEL. As Ahsia mention codes 1133 1126 are for mixture adaption near idle, if using PIWIS there will also be a subset description that tells if the rich limit or lean limit was reached. Either limit will set the same code, without the subset description you will have to figgure out if the rich or lean limit was reached. Since you also have high idle and code p0507, common sense would say you have un-metered air entering the system AFTER the MAF, and that the rich limit has been reached by the DME trying to compensate for the un-metered air. I don''t think you have a software or DME problem, but an unusual and intermittent vacuum leak.. EDIT: you should also rule out lazy 02 sensors. Good 02 sensors will switch lean/rich about 5 times per second. This may be difficult to read on many scanners depending on sample rate, but an oscilloscope will work perfect. If the 02 sensors are so lazy that the switch rate has slowed to less than once per second, it should be able to be seen by just about any scanner. If the 02 switching has slowed so much that the DME has time to reach rich limit it can set these codes. 02 sensors are a maintenance item and should be replaced at 100k regardless.

Great insight on the CEL illumination. Did not know that (actually, a lot that I don't know....). My nose meter tells me it's more rich than lean, along with a higher than normal fuel consumption. I have had both a dealer and a local independent shop searching for leakage, and I feel confident that they're right. As much as I hate to do this (don't forget that we're operating at a home workshop here...), I'm now open to trying the spray brake cleaner (carb cleaner?) at the engine to find that leak. Note: I unplugged the MAF and drove the car. First, PSM and ABS warning lights illuminated. However, the idle remained rock steady at 700rpm. The 'crispness' of the throttle was not there, but the performance was good and linear. Might this signal that the MAF is at fault, and the newly installed unit is either defective or the earlier commentary of Bosch aftermarket vs. Bosch OEM does have merit? Thanks! Bob

Spent the day checking, once again, for vacuum leaks. I used starting fluid spray, aiming it at every possible leak area that could create this issue. Not one single blip in engine note. Sprayed it at throttle body area, plenum connections and intake manifold to cylinder head area. I feel very confident in sharing that there are no vacuum leaks on the engine. I also checked the vacuum hose at the brake booster and all was good there, too. (also realized that there was only 10lbs of air in my space saver spare!....which I hadn't checked until now). We can safely say the issue is not one of air leaking into the intake tract past the MAF. Once again, I disconnected the battery and allowed it to drain-down. I reconnected and went through the calibration process. The car fires and idles at 700rpm and holds there for a few minutes, then begins its inexorable rise upwards to 950-1,000rpm. The erratic nature of this is really where the discussion lies. I ran it yesterday, and apart from the 1000rpm idle, it ran fantastic. Out of the blue comes the CEL, and the same codes.

This reminds me of when I had my 02 sensors wired backwards. I'd get an occasional surge at idle. I then discovered my error and swapped the wires around. I have not had the surge again.

Quotes source post: 15717098

Haven't touched the O2 sensors, but wondering if they are the issue? The CEL codes do not indicate faulty O2 sensors.

Quotes source post: 15717765

Read the edit at the bottom of post #55 again, slowly. Your 02 sensors may be working (just enough not to set a code) , just lazy.

Quotes source post: 15717834

Yup, that's what I had, a slow sensor and it would surge at idle occasionally. Replaced them and it has not surged in a couple of months. G, get them checked out!

Quotes source post: 15717834

Interesting that the tech piece on Pelican Parts suggests that O2 sensors are 30,000 mile replacement items. I'm not even going to mess around and ordered 4 Bosch sensors arriving Monday.

[Image-only post; no author-written text. Original image references:] https://rennlist.com/forums/images/smilies/popcorn.gif

I wish that I had better news, but this car is getting the better of me. All 4 O2 sensors were replaced with Bosch pieces; relatively easy project, and they are plug and play. Upon installation, I again recalibrated the throttle by disconnecting the battery, pressing the brake pedal a number of times to ensure that all power is out of the system. I reconnected the battery, turned the key to the 'on' position for 60 seconds, then 'off' for a minimum of 10 seconds, after which I'll start the car. The engine runs rough for the first 2 minutes and then begins to smooth-out. The idle then settled at 1,100rpm and after about 30 seconds, retreats to 750rpm. As it was pouring outside, I did not road test it, but I did shut it down and re-fire the car with better results: a smooth idle at 1,100rpm initially and then retreated to 750rpm. I allowed myself...for a moment...to think that I'd solved the dilemma. That was a mistake. I road tested it today, and within 60 seconds after leaving the garage, the initial 800rpm idle was now 1,400rpm. The car ran fine, but the idle would not come down from 1,400rpm. And then it ignited a CEL. Good news is that I no longer get the O2 sensor related codes, but I (again) do get a P0507. When I shut down the car and re-fire it, I initially get a 770rpm idle, and then it slowly climbs to 1,400rpm. - I connected my Durametric, and checked the air flow over the MAF and the readings at idle and 3,000rpm were within the correct range, though on the low side of the margin (14ML at idle and 52ML at 3,000rpm). - while checking the air flow, I erased the code and the engine immediately dropped down to 750rpm. Of course, it then slowly crept-up to 1,350-1.400rpm. - the car's performance is good aside from the high idle and corresponding CEL, though there seems to be an off-throttle surge when you let-off in gear. - we have smoke tested the engine, and I then sprayed starter fluid on all of the possible suspect areas with no change in engine note. Removing the oil fill cap reveals plenty of suction there and no real engine note change. I feel very confident that vacuum leaks are not present. Plus, aside from the high idle, the car runs fine. A leak would create some of the symptoms I encountered earlier in this goose chase. - the car's strong performance, particularly at full throttle tells me that fuel issues are not present. It has to be the DME or the software, no?? Bob

Can you drive the car for a day or so then check the fuel trims FRA and RKAT using Durametric? That may tell us if you are dealing with a rich or lean condition

As Ahsai said. Drive the car for i little bit. Atleast 20 miles. Then at idle check the fuel trims. If they are lean (adding fuel) at idle, raise the rpms to 2500. If the trims go to zero you 100% have a vacuum leak. If the fuel trims stay out of spec at 2500 rpm then that rules out a vacuum leak and will give some direction for diag.

Will do. So I understand, fuel trims are the adjustments the ECU makes to correct for rich or lean conditions as reported by the O2 sensors, correct? What it FRA and RKAT? I've only read about Short Term Fuel Trim (STFT) and Long Term Fuel Trim (LTFT). Thanks!

Quotes source post: 15731470

Thanks! Great guidance. I'll report back. Bob

Quotes source post: 15731481

Fra and rkat are porsche nomenclature for stft and ltft. They dont report "exactly" the same. But very similar. Hopefully once you report your finding the guys who are a little more familiar with FRA abd RKAT can help explain how lean or rich she is.

...... and you ruled out vacuum leak at the brake booster how?

Quotes source post: 15731608

Did a visual at the booster in the front trunk. Hose was intact. Did a visual at the rear. Dealer did a smoke test that indicated no leaks, and I also used starting fluid spray at the booster line elbow and intake manifold attachment point with no change in idle speed.

Quotes source post: 15732520

Booster can and most commonly do leak internal. If any leak shows it shows in cabin not external. Remove the hose going to it and plug it off. then recheck the issue after clearing codes,

Quotes source post: 15732520

I would suggest clamping off the vacuum hose to the booster with a pair of needlenose vicegrips when the RPM goes up to 1400 to see if it drops back down to target. Seems like at diagnostic leads are dead ending, may have to start back at square 1.. something may have gotten missed..

Quotes source post: 15732537

Just plugged the line at the intake manifold junction and fired the car. No difference.

Quotes source post: 15731470

Drove the car for a good 25 miles, during which the idle remained steady at an indicated 1,400rpm. Returned to the garage, connected the Durametric and checked air flow at MAF: 21 @ 1,200rpm and 52 at 3,000rpm. Both values are within range. Bank 1 FRA is 1.26 and does not fluctuate as you raise rpm to 2,500+ Bank 2 FRA is 1.25 and does not fluctuate as your raise rpm to 2,500+ Bank 1 RKAT is 6.61 and does not fluctuate with rpm Bank 2 RKAT is 6.84 and does not fluctuate with rpm. Given this data, the car is running rich, correct? Of note, when I start the car, the idle is now 1,000rpm. Still high, but better...and unexplained. Also, the Durametric shows the P0507 code (no other codes...) but the CEL hasn't triggered.

To review process thus far... - the oil fill tube has been replaced as a crack was suspected at the PPI - the AOS has been replaced by the dealer, suggesting that this was the cause of a vacuum leak. - the throttle body has been replaced as the dealer suggested that this was the likely culprit - plugs and coils have been replaced as a 'while I'm in there' measure - all 4 O2 sensors have been replaced and codes 1128 and 1130 (oxygen sensing adaption, idle range) are no longer appearing - no detectable vacuum leaks; validated by smoke test and starting fluid spray - car runs well throughout the rpm range; full throttle performance is not compromised. Bob

Quotes source post: 15732592

RKAT is the fuel trim near idle, the 6.61-6.84 means the DME is adding 6.61% (bank 1) and 6.84% bank 2 to the Ti injection time to achieve Stoich (14.7 AFR).. Basically means a moderate vacuum leak..

And FRA is the fuel trim under load 1.26 and 1.25 means the DME is adding 25~26% MORE fuel, which means the engine tends to run lean. Have you checked for fuel pressure and delivery rate? I asked that in post #21 but didn't get an answer. Combined with running lean at idle, I suspect fuel delivery problem.

Since you have a Durametric, watch the throttle position sensors voltage, there are two of them, one will start near 5v and go down as throttle opens, the other will start near 1v and go up as throttle opens. one should never move without the other moving the same amount. This is the redudancy the DME uses to make sure it can believe the data it's seeing...If there is a problem on one wire (resistance from connection or rodent nibbling) it will show here..

Quotes source post: 15732647

Yes !! Very good point !!

Quotes source post: 15731407

I think you meant 14kg/hr, which is very low for a 3.6 if taken @700 rpm. Perhaps you should also try a different MAF.

P0507 potential causes: – Throttle valve is sticking – Leaks in intake air system Sounds like air leaks, dirty throttle valve and/or bad MAF . Sorry, I'm not going back to read everything and recall you have a new throttle body right? Again, the little voice in my head keeps repeating something I once read by JR...Vacuum leaks can go undetected via smoke testing and other common methods. Did you use carb cleaner at idle or raised RPM and at full temp?

Well, I just saw your previous post recapping it all.

BTW, since you have Durametric, you can log the IAT and MAF (HFM) and do a few 2nd gear to redline runs. If the MAF is good, it should report maximum value of 230-240g/s (or 828~864kg/hr. Durametric reports in kg/hr). See more details of MAF reading in this thread by Porschetech3.

Quotes source post: 15732647

Being that full throttle performance remains strong, I thought we had passed over a fuel delivery issue. I'll need to get a pressure tester to check. Any recommendations?

Quotes source post: 15732691

Yes, new throttle body and I forget to mention a new Bosch MAF (not Porsche Bosch).

Quotes source post: 15732684

Actually, starting fluid (with extinguisher at hand) on a warm engine at idle.

Quotes source post: 15732730

You need a pressure tester with a metric fitting (M14 if memory serves). There's a hex cap on each fuel rail that you can remove and hook up the pressure gauge. Also remember to test for flow rate by jumping the fuel relay at the footwell fuse/relay panel. Or you can try a different/new MAF first if that's easier. I know you said the WOT performance is good but seeing the DME is adding 26% of injection timing, it's really suspicious. Also, 26% is either the MAX or close to the MAX that the DME can compensate for and is imminent to trigger those codes such as 1128, 1130, etc. Your fuel system should retain 3.8+/- 0.2 bar (around 55 psig) with the engine off, and sow 3.3 +/- 0.2 bar (48 psig) running at an idle. With the engine off, your fuel pump should deliver 850 ml of fuel in 30 seconds.

Quotes source post: 15732730

Based on your FRA (+26%) and RKAT (+6%) on both banks, the DME is adding fuel both during idle and during load big time, reaching the max it can handle. Therefore, these are the most likely scenarios: 1) The MAF is under-reporting the intake air so the actual intake air is more than expected (both at idle and under load) 2) The fuel flow/pressure is lower than specs (both at idle and under load) Note if your problem is only intake air leak, the FRA should not that much larger than 1.0, it would be much closer to 1.0 +/- 0.06 on both banks because under load, there's not much vacuum so air leak will not have much effect. Note your engine is running fine but that's due to heavy DME corrections.

To test fuel pressure in theroy without a pressure tester. Watch ONLY the upstream O2 sensors with a graph if possible. During WIDE OPEN THROTTLE the O2 sensors should go full rich which is about 1.0 volts and stay put. Find a road where you can watch the data. At about 10mph put it to the floor and stay on the floor till about 50 or 60mph. Your O2 should stay flat lined at full rich. If they go lean or stay switching during this test you have a fuel deluvery issue.

Bought a pressure tester kit on Amazon that seemed to be comprehensive and was said to include a 14X1.5 adaptor. Nothing in the kit came close to fitting. Can anyone confirm the 14X1.5 fitting dimensions? Putting a caliper inside the cap, I get 14.5mm (as shown). If that is the correct size, this seller offers a tester in 14mm: http://www.arnnworx.com/catalog/inde...8#.XKX_hPZFzVg Also, the testing port does not have a Schrader valve inside, which I find surprising. I assume that I'm at the right port which required a 19mm socket to remove.

I believe this is a 16mm fitting, but the 'no Schrader valve' has me puzzled.

Can't help with the size but that's the right cap. There is one cap on each fuel rail and there are no schrader valves on the 3.6 engine.

I just started having this problem as well. I just purchased my car and it was shipped from CA to CO. It is a 2000 C2 Cab with 87,000 miles. I cleared the codes on Thursday, passed emission in CO on Friday, and the codes returned on Sunday. I took pictures of the codes and the live data and video of the codes being cleared and the idle creeping up. If this will help anyone troubleshoot this issue Here is the link to the photos . Alex

At least your fuel trims are where they should be. I reset mine by disconnecting the battery for 20 minutes. The short term fuel trim came 'out of the box' at 1.0, but after a minute, it began to creep upwards ending ultimately at an alarming 1.32! Idle sits at 1,050rpm... Bruce at ArnnWorx ( www.arnnworx.com ) is putting together a fuel pressure gauge for me that will have the correct adaptor for the 996 fuel rail test port. It is a 16mmX1.5 and the sealing is called a 'globe seal'; there are no O-rings. I'll post photos when it arrives.

I'll read through this post again and see if it sparks any ideas. Let me know if you want me to run any more tests. Between work and school I'll see what I can get done.

Bruce Arnn produced a wonderful fuel pressure tester that proved to work flawlessly and I obtained the 55psi figure that I expected to see. I've flat run out of ideas and the car, as of today, is in the hands of Checkered Flag Porsche in Virginia Beach where they have a Porsche Classics technician to work on earlier model cars. I've provided him with a full chronology of the work done to the car and the erratic behavior; in fact, though the idle is 975rpm, and the fuel trims are high, the car drove flawlessly to the dealership some 50 miles distant. I'll post their findings, but I fear they'll come-up empty handed.

Ah, bummer. You have not tried another new MAF yet, right? That seems to be the last possibility.

Also, have you verified the flow rate of the fuel? You may get enough pressure but not enough flow.

Ahsia is right, your next step should probably be installing a known good MAF sensor. There are a couple of things that come to mind that could possibly turn out to be the problem since I hope the smoke testing and spraying for vacuum leaks has proven to be a good sealed system, and fuel pressure has been verified. What about your fuel quality, since the problem came on suddenly, has that tank of fuel been ran out? I had a car once that had been to three different dealerships for poor performance, no codes, just very poor performance especially before the engine temp got up normal. Turned out that someone had put a few gallons of diesel fuel in the tank accidentally. Car ran terrible cold (very lean) and ran decent hot just down on power. Drained gas and fresh gas cured the problem. Another possibility is that very dirty fuel has been in the tank and injector filters are clogged. There are some very small/fine filters on top of each injector that if partially stopped up will run lean, and will be more lean in the upper RPM's or quick snaps.

Did you check the plug which plugs into the connector on the throttle body. I got once really strange idle behaviour but with different fault codes. Maybe replace it to rule out any electrical issue with functioning of the throttle body.

Quotes source post: 15807848

Hi Ashai: Full throttle performance is not compromised whatsoever, which is why I have dismissed fuel flow rate. The independent shop that tried diagnosing the car felt it was the best performing 996 they had worked on in that respect. Bob

Cautionary note: Many a failed diagnosis has come from trusting the test results and assertions of the tech or shop that went before you. Trust but verify fits here. That may not be true for you, but I like to make sure that I do all of the diagnostic steps in the logical order to isolate or diagnose any anomaly. It may seem like a waste of time to redo a fuel pressure or volume test, but I always feel better knowing for sure. I was a software developer for a long time and this is just as useful for debugging as it is in auto mechanics. YMMV, etc.

Quotes source post: 15808196

I understood what you're saying BUT as I explained before, your engine is running fine because the DME is adding 26% of injection time (i.e., ~26% of more fuel) to compensate on BOTH banks. We just need to find out why. Possibilities are fuel pressure/flow lower than spec, MAF reading biased, O2 sensors biased, etc. Only the first two are common for both banks.

I'm pleased to share that a long and tiresome episode has seemingly come to a close. Just got an email from the tech at Checkered Flag in Virginia Beach. He found that inside the air box, the resonator chamber had come apart and allowed unmetered air enter the intake tract after the MAF sensor! I never would have thought to look in there, but it makes total sense. I cannot thank all of you who offered very considered thoughts and ideas to chase after in trying to find the culprit. It just had to be something ridiculous given all the steps taken to date. As another surprise, a new air box via an on-line Porsche dealer is only $188! Frankly, I'm shocked.

Nice! Glad you finally got it sorted after this long saga. We should all go check our airboxes!

No kidding!!

We were on the right track that the engine is running lean (per the FRA fuel trim). However, without a visual inspection like they did, even a smoke leak detector would not have found that kind of "internal" leak. Kudos to the shop.

WOW !! With that big freaking hole by-passing the MAF, I can't believe it ran as well as it did. Glad you found the illusive culprit.. That just goes to show, sometimes the simple things can be the hardest...

Quotes source post: 15811698

Exactly! Who would have thought of looking inside the box? I looked for every possible post-MAF place for air to enter and there right in front of me was the resonator tract. I mean, what could possibly go wrong with an air box???? Now we know.... Again, thanks for the help!

Quotes source post: 15811720

And now we learned a new correlation. If engine runs lean under load (reflected by FRA), fuel pressure is good, MAF is good, no intake air leak => check the air box!

Quotes source post: 15811738

With the car at home and a new air box winging it's way to me, I pulled the airbox for a look around. Joe Finkle (Porsche Classic certified tech at Checkered Flag in Virginia Beach) had graciously plugged the resonator port and glued the resonator chamber back together to enable my drive home. Joe had mentioned that someone had cut the resonator box apart, which I thought to be too far fetched, and assumed that he meant someone had pried the chamber apart at the seam. No sir...someone had taken a Dremel (?) to it and CUT it in half. Go figure....you just never know what priceless treasures a previous owner will leave behind for you!

Ok, that makes a lot more sense. This air leak potential crossed my mine before when I studied the orange cap mod long time ago. I was convinced the construction was sturdy enough.

Is the blue circle around a brake booster vacuum line? I have two small holes in it from rubbing on inlet air snorkel.

Yes, that's the brake booster vacuum line. It should be 90 degrees vertical coming out of the intake manifold so that it doesn't rub on the snorkel. I released the hose from the adaptor and plugged the adaptor when I was trying to diagnose the idle issue, and confirmed that there was no issue with the booster.

Wow, I'm glad you got it sorted. But how did you not see that airbox seam, it's right there in plain site?

The resonator is inside the airbox and not visible unless you remove the air box and take it apart. It never occurred to anyone that someone would actually cut the resonator apart.....and why they did it?

Oops, I thought I was looking at the airbox, no wonder it looked different. Anyway, glad the saga has finally ended for you!