The dreaded & irrepressible INTERMITTENT Check Engine Light - ANY NEW SUGGESTIONS?

(I'm desperate and am also posting on Pelican - tired of this chase....) CAR: 2002 Porsche Targa 3.6; about 160,000 miles; always regularly serviced at Porsche dealership prior to my ownership, with records at the dealership. I bought this car over a year ago, and within a couple weeks, the cel came on. The car was running fine � no problems except the presence of the light. We reset the light with our OBD-II tool � the code (P0491) had something to do with the secondary pump malfunctioning (actually �insufficient flow to Bank 1� � see at the bottom of this post for the various interpretations I have gathered). The light stayed off for a few more days � then re-appeared. The car continued to run beautifully. Well, for more than a year now, the light continues to come on every few days to weeks � highly variable, but seems to be related to the car sitting in traffic or with extensive idling. Yes, the car runs through enough cycles for the light to have turned itself off after repairs or resets. We reset it every time. In fact, my wife carries the code reader in the car and resets the light as if it were part of a usual routine. I read many threads on Porsche forums about cel problems � and the thousands of dollars that members have spent trying to track down the causes, and a usual outcome being A TEMPORARY FIX, but inevitably a recurrence of the trouble light. I finally decided to have it checked out by a local Porsche Indy (Lutzo at LMN Motorsports � one of those ace, and completely honest, independent mechanics in the area). Lutzo has all the Porsche factory diagnostics, plus decades of Porsche experience. He went down the Porsche factory diagnostic list and chased the various possible causes, including MAF, piping, O2 sensor, SAI pump checked for operation, even smoke tested for leaks. He called the factory asking about other possible solutions. He swapped out parts on a trial basis (no charge) to see if the problem was solved. I had the plugs and air filter replaced just for general principles. The problem was localized by the diagnostic machines to the first bank cylinders only. The MAF and the secondary pump both checked out OK; the hose that runs under the manifold TO THE SECOND BANK seemed to be leaking (could not get it to seal where it attached), so was replaced. The O2 sensor was over 10 years old, so was replaced. So, listing the chase so far: new plugs; new air filter; new O2 sensor; new SAI tube to Bank 2; MAF checked with a replacement; SAI pump operation checked; smoke test for leaks. In addition, there have been HOURS spent by an expert Porsche mechanic trying to track this down. Every time we thought the problem was fixed, a few days/weeks later, boom, on pops the cel again. I finally gave up � the recurrent 60 mile drive to the mechanic was ending like the horror stories in the Porsche forums. I think most owners just do like we are doing � reset the light and drive on. HOWEVER, HAVE THERE BEEN ANY NEW POSSIBLE SOLUTIONS TO THE DREADED cel repetitious and intermittent recurrence? Below are the various definitions of the P0491 code, as well as a partial list of Porsche forum posts (Pelican and Rennlist) with similar �cel� problems, but highly varied causes and outcomes. http://forums.pelicanparts.com/porsc...ine-light.html http://forums.pelicanparts.com/porsc...ht-popped.html http://forums.pelicanparts.com/porsc...ce-please.html http://forums.pelicanparts.com/porsc...25-errors.html http://forums.pelicanparts.com/porsc...ode-po410.html http://forums.pelicanparts.com/porsc...de-0420-a.html http://forums.pelicanparts.com/porsc...ine-light.html https://rennlist.com/forums/996-foru...d-to-idle.html https://rennlist.com/forums/996-foru...age-fault.html https://rennlist.com/forums/996-foru...-bank-1-a.html https://rennlist.com/forums/996-foru...need-help.html http://p-car.com/996/diy/sai/mainpage.html https://rennlist.com/forums/996-foru...92-solved.html https://rennlist.com/forums/996-foru...mp-issues.html CODE DEFINITIONS GATHERED FROM SEVERAL SOURCES: P0491 Secondary Air Injection System, Bank 1 - Insufficient Flow P0492 Secondary Air Injection System, Bank 2 - Insufficient Flow Possible cause of fault - Electrical fault in power supply or line between relay and secondary air injection pump - Secondary air injection pump relay mechanically faulty - Mechanical fault in electrical secondary air valve - Pneumatic secondary air valve faulty/sluggish - Electrical fault in secondary air injection pump - Mechanical fault in secondary air injection pump or secondary air injection pump blocked - Vacuum system leaking - Air hose to secondary air injection pump slipped off or constricted P0491 Secondary Air Injection System, Bank 1 - Below Limit Possible cause of fault - Fuse F3 faulty - Open circuit to secondary air injection pump - Short circuit to ground in the lead to secondary air injection pump - Secondary air injection pump relay mechanically faulty - Secondary air valve sluggish - Secondary air injection pump (air cleaner) blocked - Air hose to secondary air injection pump slipped off or crushed - Secondary air injection pump faulty/stuck P0491 Secondary Air Injection System, Bank 1 - Insufficient Flow P0492 Secondary Air Injection System, Bank 2 - Insufficient Flow Possible cause of fault - Electrical fault in power supply or line between relay and secondary air injection pump - Secondary air injection pump relay mechanically faulty - Mechanical fault in electrical secondary air valve - Pneumatic secondary air valve faulty/sluggish - Electrical fault in secondary air injection pump - Mechanical fault in secondary air injection pump or secondary air injection pump blocked - Vacuum system leaking - Air hose to secondary air injection pump slipped off or constricted

I am no 996 expert (never worked on one, but am looking to buy one), but this code is a very, very common issue with Audi 40 valve motors found in the A8, A6 and S8, all off the older, 2000 model year variant. You have the items listed above that could be the possible cause of the fault. The system basically only works with the engine cold, it turns on a secondary air injection pump (operated by the relay mentioned above) that injects air into the exhaust, on the Audi it's a fan that runs, a solenoid is energizes to port vacuum to a secondary air valve (there is a small vacuum line that runs from the solenoid to the larger port), then the port is opened by vacuum and the air is pumped into the exhaust. On the Audi's, the vacuum line shrinks with age and pulls off the solenoid valve so it can't operate the secondary valve and allow the fan to pump air. The fan just pumps air against the closed valve, you can totally hear the high pitched whine vs. a fan pumping air that is quiet and happy. I documented all of this with pictures many years ago and so many have had the issues. Here is the post with pictures....just look through the 996 maintenance manual and figure out how the system works....I'm sure it's very similar. Best of luck! I bet it's an easy fix.....but doesn't sound like it so far. http://forums.quattroworld.com/a8/msgs/15074.phtml

OP, something does not add up in your post. Knowledgeable Porsche mechanic should have immediately replaced the diverter valves and all of the SAI piping. This problem is caused by the diverter valves in the SAI system, they fail and trigger this code. Let me tell you how P0491 and P0492 appear. You sit in traffic or after getting off the freeway, SAI pump briefly comes on...CEL pops up. Replace the diverter valves and check the piping for leaks. My car developed this problem before I dumped it to the Carmax. I also drove to the Carmax place with OBD II reader, CEL popped up...and I reset it before it went for their "inspection".

Quotes source posts: 11746305, 11746341

PAUL -- Thanks for that info! I can hear the SAI pump coming on when first starting the car. The light comes on ONLY after sitting in traffic a long time, or otherwise sitting at idle,, BUT also will pop on while driving steady at ANY speed (e.g., 30mph, 70mph, etc.) I would think that the CEL would come on almost immediately if it was a problem with the valve not opening. Instead, the car can go for days, and even a couple weeks, before the light comes back on, with the same fault code. I'm thinking that if it were a vacuum leak, it should be constant and always present, so the CEL should come on every single day -- and immediately after re-starting the car after clearing the code. I'll print out and read over your article AK -- I will check into replacing the diverter valves-- can't be expensive and sounds like it would be DIY. The mechanic checked the piping for leaks, as well as replaced one of the pipes where the seal was not good (but it was for Bank 2, while the error was showing Bank 1). BTW, are you certain that the problem with your CEL would have been resolved by replacing the diverter valves? Sounds like you got rid of the car before you got around to replace them, no? Bear in mind that every single one of my "fixes", as well as those of many other owners with their various fixes, resulted in just TEMPORARY resolution of the problem -- but with the long-term outcome being that the light came back on in a week or two.

The valves in SAI and EGR systems get worn out/gummed up/whatever. The EGR valves on my Corvette get stuck every few years, but they're pretty robust, and a shot of PB Blaster is usually sufficient to de-gunk them and get things working right again. I think Paul and Kalashnikov might be onto something...

Do you have an Android device or Durametric? Sounds like your problem is confined in bank 1. My take is your bank 1 sees insufficient air from the pump, hopefully persistently and not intermittently. Your car, MkII has the shut-off valve part #11 below but not #9. Air goes from the pump via #11 then into part #8 to split the flow into bank 1 and bank 2. Your mechanic must have fixed the leak on #16 based on your description. 108-05.gif (courtesy of Atlanta Porsche) To verify bank 1 sees less air than bank 2, you can monitor and compare the four O2 sensors during COLD start. If the above theory is correct, you should see the bank 1 O2 sensors reading LESS oxygen (i.e., >0.44V) during the time when the SAI pump is running (engine runs very rich during cold start). I had plotted the voltages and rpm for bank 1 below. Bank 2 is similar. The left vertical axis is rpm and the right is voltage (in V). The horizontal axis is time in mm:ss. Keep in mind O2 sensors don't operate and they can't detect oxygen till they get up to temperature (assisted by the built-in heaters in the O2 sensors). The DME supplies a 0.44V voltage to the sensors at the beginning and the rest you can read. So you can say these are the "signatures" of a healthy O2 sensor reading during COLD start. When the sensor starts to operate (10s into cold start for the pre-cat thanks to the hot exhaust gas and about 70s for the post-cat ones, being farther downstream from the heat). Afterall, the DME detects the SAI operation via what it measures from the O2 sensors so if you can see the O2 sensor readings, you can see what the DME sees and why it's complaining about bank 1. Porsche-cold-start-SAI-O2-reading.jpg

Ahsai, if Rennlist had a rep system, I'd +1 you for that post. Awesome.

^Thanks! I've plotted it before for a while for occasion just like this

Ahsai- Interesting picture.....it's basically an exact replica of the Audi system.....or the Audi system is a replica of the Porsche system.

Quotes source post: 11746623

Yes and not surprising. Most SAI systems operate the same way. Basically the SAI system guarantees that the air pump injects excessive oxygen to burn with any unburnt fuel in the exhaust gas flow. The DME verifies the correct SAI operation by monitoring the pre-cat and post-cat O2 sensor readings. Basically looking for 0v where excessive oxygen is expected.

Quotes source post: 11746570

Yah, Ahsai -- MIT or Caltech? Thanks for the great explanation and figures. Helps understand the process. The mechanic had Durametric, I'm pretty sure. And yes, he stated that the problem codes all indicated Bank 1 -- but it must be INTERMITTENT, right? I can go for days and even a couple weeks without the light coming on. He also checked the O2 sensors and said that they all were performing correctly, except that one was "borderline", so he replaced it. He noted that the car "ran better" -- probably a judgment based on his instruments, as I can tell no difference in engine performance while driving. Given that this is so intermittent, what would you suggest? a) replace all O2 sensors? b) work on parts 9 and 11, thinking that they might get gummy, and therefore occasionally stick? (Are these the diverter valves to which AK was referring -- or is #8 the div valve?) c) Other? BTW, while I am a strong believer in human fallibility, Lutzo is no dummy mechanic, and I know he checked many other items in the course of his diagnosis, which flew over my head during his explanation. I even think he mentioned squirting something -- possibly a valve that could gum up?? Thanks.

Quotes source post: 11746673

Haha, really no need to be an MIT/Caltech grad to solve this. The intermittent nature could be because bank 1 is marginal. Also some codes will need to occur in both consecutive trips before it will trip the CEL and this code may be one of those. Still the same recommendation from me. Check the O2 reading during COLD start and plot them like in my post. Then once it's confirmed the problem is only confined to bank 1, you will need to at least remove and check that shut-off valve, which combined the functionality of two valves - one way flow so exhaust don't blow back to the air pump and open/close the air path between the pump and the exhaust. http://p-car.com/996/diy/sai/00modelsai.jpg This thread is for a boxster that has the same symptoms as yours. Basically its SAI ports on bank 1 are clogged due to high mileage. Cleaning the ports solve the problem. Looks very possible you have the same problem as you also have a high mileage engine. http://986forum.com/forums/general-d...1-%96-3-a.html

Quotes source post: 11746810

Ahsai, Excellent possibility that carbon build up is the cause (160K+ miles). Also, yes, marginality of the defect could cause it to be intermittent -- and the combination of errors needed to trip the cel is interesting. I'll have to figure out how to proceed next -- don't want to tear down the top end, but..... Maybe I need to use my bore-scope.

With 160K You can look forever and not isolate it,in my case.I lived in Montrose,Co. and the altitude and fuel quality caused mine to do it all the time.I finally started to buy turbo blue in 55 gal drums.It cured my intermittent check eng. light.My Mercedes had an altitude sensor in it and it really changed the air/fuel mixture.

Quotes source post: 11747130

If the car in the avatar is 96+ and it's yours, you should be the expert of SAI already Highly doubt your boroscope can snake through the SAI ports as they are pretty small.

Quotes source post: 11747182

Ah yes, the "ZEN-BXX" is mine -- but it is a mixed up hybrid project car, which has numerous concurrent initiatives being performed on it, hence has been a garage queen for over a year. Too many other cars, too many other car problems, too many other projects besides cars, etc. etc.

Looks very nice. I wish I could afford a 993 haha

Quotes source post: 11746570

Ahsai, Paul Waterloo, and others.... I wanted to give final (hopefully) follow-up to my CEL that showed intermittent and frustrating left bank and SAI codes. After a $1K chase-down in Virginia with no change in the intermittent appearance of the CEL, I brought the car to Stuttgart Specialists in Louisville, KY, where Tony Proasi and Russ (primarily) again tracked it down to the left bank. I mentioned to them what Ahsai and Paul had said in the past regarding carbon buildup in the air passages (in ref. to Audis), and Russ seemed aware of that kind of problem -- although he stated he then researched it again. Russ removed the exhaust manifold and, lo and behold, he found basically what had been suggested in this thread in the past ref. blockage of the air passage in the left head! ...EXCEPT, the blockage was due to a marble-sized plug of media blasting material, not carbon! Same diagnosis, but surprising etiology. Russ said when he looked at the heads, he thought they were much too clean for a car with 175K miles. He thought someone had worked on the engine fairly recently. This correlated with the car's history suggestive of a blown head gasket (around 160K miles) that was repaired -- at which time the heads were media blasted to be cleaned. On further examination, Russ saw this PLUG of material blocking ONLY the main feed to the left bank air injection ports! (See attached pic showing on a similar head which passage was blocked -- finger points to area.) He removed the plug and cleaned out the passages, which showed no other obstructions. Obviously, the mechanics who had replaced the head gaskets had not done a thorough job of cleaning out residual media blasting material, and it congealed into a glob that occluded the main air port -- sort of like shooting an embolus from the heart to the "head", a.k.a a stroke or cerebro-vascular accident (CVA) in human medicine. Thanks to Ahsai, the light has not come on YET -- but the diagnosis of this blatant obstruction seems like it would explain all my past problems with the CEL. So, I want to lend STRONG support to being very suspicious of blockage of the Secondary Air Injection ports in the heads as a cause of the CEL and this error code 0491/92. Thanks to all, again. Now we might keep the car another 50K and have the IMSB and motor mounts replaced after all.

Wow, it's nice to hear that you got it sorted finally! It's been a while so I had to re-read the whole thread to remind myself your case. If you had shown us the O2 voltage plot, it would have been even easier to diagnose The SAI clogging is not a wide spread problem for 996s AKAIF (may be more for Boxsters) but your symptoms strongly support that hypothesis. I'm glad your mechanic finally removed the exhaust manifold to check for blockage. Yes, your head looks extremely clean the only way to get that clean is either a brand new head or it's medium blasted. Everything makes sense.

I hate to resurrect an old thread like this, but I'm having exactly this same issue, and wanted to keep my thread inline with the super useful info in this thread already. I recently got a 2000 boxster S and I'm having trouble with smog. It's failing the Secondary Air Intake system (SAI) reporting as FAIL in Durametric, and everything else reports Ready. When I activate the secondary air pump through the activations secition of Durametric, I hear the pump turn on. When I activate the swithover valves and other SAI related switches, I hear them click. Based on the information in this thread, I grabbed a realtime voltage trace of the bank 1 and bank 2 pre-cat O2 sensors, but I realized I forgot to grab the post-cat ones. I can get those this week if needed to diagnose. Here's what my bank 1 and 2 O2 voltage graphs look like on cold start. Other than the bump upwards before it drops down, it looks very similar to the "expected" graphs on this thread. Would appreciate any help you guys can offer on sorting this out so I can pass smog. Thanks!

This could be one of several things, and it can't be diagnosed without looking at the vacuum lines and the pump.

A P 0501 code is causing the CEL message to scare me to death, accompanied by the CEL light illuminating on my '03 996 coupe with manual transmission. I am hoping to find how to remedy what might be a Variable Speed Sensor issue in my garage - it seems it might simply be case of putting it on the lift and inspecting & chasing wires.