Engine stumble on cold start

ok, this has been happening for about a month now...when cold starting, the engine fires immediately...tries to set idle at about 700...then stumbles a couple times and then revs to 1100-1200. After about a minute...everything is ok, revs settle back down and everything is smooth. If I try to drive off right away, it will keep stumbling unless I keep the revs way up. I can also smell gas. So I'm pretty sure it's way too rich on cold start. Dirty MAF? I just changed my cats yesterday to Fabspeed 200 cel cats. On the next 3 cold starts, the issue was worst. I'm under the impression the secondary air injection system would help with a cold rich situation no? Or is it only for emissions? 4 of the coil packs are new from last summer, as are the spark plugs. The air filter is 1 month old (oem) The only thing I can think of is the MAF....so I'm going to try cleaning it. Can I use any MAF cleaner or is there a Porsche specific one? (shouldn't be, but you never know with the Germans ) Here is a vid of the stumble on cold start. Note that my foot isn't touching the gas pedal at all until about 30 seconds in.... http://vimeo.com/25356604 Any suggestions would be great. sean

Try cleaning the MAF and throttle body, but I do not think that's the issue. The other 2 coil packs were good I'm guessing because you didn't change them? Any MAF cleaner will work.

Cleaning the MAF isn't going to hurt, standard MAF cleaner is fine. However, if the MAF is going bad, it may not help either. If you have the original MAF, it might be time to replace it.

Quotes source post: 8650700

You can clean the MAF but I don't think that's the cause of the symptom. If the MAF is dirty then you have an engine air filter/intake leak that lets dirty (unfiltered) air into the engine. (Have to mention I've driven my 02 Boxster over 160K miles without even touching the MAF. MAFs just don't get dirty if the intake system is stock and working properly.) From the description of the engine's behavior I suspect the coolant or possibly air intake temp sensor is acting up and supplying an optimistic coolant or air temp value to the DME and the DME is not adding enough fuel for a cold engine. (The intake air temp is provided by the MAF so the MAF may be bad. Which reminds me: Unplug the MAF from the wiring harness. Under the same conditions as before start the engine. If the symptoms are still present probably not the MAF. There is a slight possibility the DME has learned fuel maps based on the input from a going bad MAF and the way to reset this learned mapping is to use an OBD2 code reader/data viewer to erase DTCs. Even if there aren't any this resets the DME's learned maps back to their defaults.) The secondary air injection system is there only to add air after the chambers to help exhaust gases with unburned fuel combust to help heat up the converters. The gas smell is the result of a cold engine, cold converters and an engine that is not running right. The exhaust gases are 'rich' but the air fuel mixture being fed to the engine (so to speak) is lean. If you want to 'clean' something you can remove the throttle body and clean it. If it is dirty (along with some signs of oil wetness) this points to an AOS problem. Sincerely, Macster.

I've wiped out the throttle body before...right at the butterfly. There was carbon buildup, but no wetness. Maybe I'll take it right off and give it a thorough cleaning... I'll also try unplugging the MAF when I go home tonight and see if it happens again. I can erase DTCs with my Durametric tool right? I hope it's not the AOS....I was looking at it when the airbox was out (see my other thread about installing new cats) and I don't see how I could possibly reach it without dropping the tranny. I was going to get that done with the IMS/etc next spring. Thanks for the advice sean

Quotes source post: 8651124

There should be no carbon build up at the TB. Combustion chamber gases should not get that far upstream from the combustion chambers. My guess is what you are mistaking for carbon build up is a very light coating of oil. The oil is dark because oil as it accumulates miles picks up soot which darkens the oil. The oil is carried in vapor form by fumes from the crankcase directed to the intake manifold at a point (at least in the case of my Boxster not sure about the 996 engines) immediately behind the TB. There will be some flowing of these fumes upstream and the TB and its butterfly valve are very close to where these fumes are fed into the intake manifold. If you remove the TB check for oil behind the TB at a hose connection of a hose that runs from the AOS to the intake manifold. Also, after you have the TB removed carefully -- and with watches/rings removed -- put your hand into the intake manifold as far as you can and touch the walls of the intake manifold. If your hand comes out oily... Either you've been running the engine overfilled with oil, or the AOS is suspect. Sincerely, Macster.

ok, thanks. So the inlet to the throttle body...where the plastic hose form the airbox attaches...should be sparkling silver clean after 120,000Kms and 8 years?

Quotes source post: 8651361

Can't say for sure. The only time I had my car's intake hose off and the TB out was when the AOS (as it turned out) was going bad. I can't recall the condition of the TB at its inlet side, but at the outlet side next to where the hose from the AOS connected it was quite dirty, oily. There was a drop of dirty oil hanging from the TB butterfly valve. At the time I didn't know what this meant and I never stuck my hand down in the intake manifold. What I can say for sure is after 160K miles of driving my Boxster (since the 1st AOS failed) I've never had the TB out and cleaned it. In fact the replacement AOS failed -- I can't recall the mileage -- and I never even bothered to remove the TB and clean it before or after the AOS was replaced. I just had the car towed to the nearest Porsche dealer and had the AOS replaced. I'm sure the dealer didn't clean the TB for the labor never showed up on the work invoice. Now regarding the Boxster, I had the MAF out once and cleaned it -- over some white paper towels and didn't get enough dirt off the MAF to even show up on the paper towels -- and ended up replacing the MAF though my diagnosis of a bad MAF proved to be wrong. The problem proved to be a leaking oil filler tube cap not the MAF. However, the replacement MAF eventually (after approx. 70K miles) started acting up and I replaced this MAF with the original MAF which worked just fine and continues to work just fine to this day. At this point, I have put over 160K miles on my Boxster since I cleaned the TB and over 56K miles on my Turbo, and have never cleaned nor had cleaned the TB or MAF. Both engines run just fine. Whether the inlet side of the TB is sparkling silver clean I can't say but if it ain't it ain't dirty enough to cause any symptoms. My WAG would be there might be a very light layer of very fine (very fine) dust, enough to put some color on one's finger but not enough to affect air flow into the engine. Sincerely, Macster.

Updates?

I unplugged my MAF and the stumble was still there. I got a PSM and ABS failure notice....but the engine still did the exact same thing, so I don't think it's MAF related. If anything, since changing the cats, it's worst. So not sure why less back pressure would make the situation worst...but the first 30 seconds is really rough, almost stalling a couple times (but it's NEVER stalled) So I guess this weekend I'll be taking the TB off for an inspection and cleaning. Also, I'm thinking of hooking up my Durametric and recording some numbers while cold starting. Specifically temps, AF ratio, timing advance and look for anything that seems out of whack. update to follow on Monday... sean

MAF shouldn't affect idle at all. First and foremost, check for intake leaks. This can really affect idle, but not as apparent on higher revs. For example if the T/B is not completely secure or if a boot in the intake plenum has a leak or is not tightened down properly. Or the oil filler tube has a crack, oil cap ring, T/B clamp or hose broken, brake booster, etc. On older cars the idle air control valve needs to be cleaned every so often. A dirty T/B shouldn't affect the idle that much, the clean side of the T/B should be oil free but could have a very small bit of carbon on it if it's never been cleaned. The plenum (back) side can have a very light coating of oil, normal. Also vacuum leaks can attribute. A more free flowing exhaust might make intake leaks become more evident/apparent since the entire system can "breathe" more freely, however the DME should adjust accordingly, as it sounds like is happening. Although it sounds like you were having the issue prior to the cats... Have you reset the DME by unplugging battery for a period of time to have it relearn? AOS can affect idle as well if it's failing, usually other symptoms follow like difficulty removing oil cap and white smoke. O2 sensors can affect fuel mixture, usually one bank will run richer than the other if a sensor is on it's way out - they can be monitored using Durametric, PST2, etc, also bench tested using a torch. A rich fuel smell is normal for these cars on cold startup, especially if they idle alot or are not driven hard enough, for example city driving makes them use more gas. Also does this happen on a warm/hot start?

Yeah, I was having this isue before the cats. It doesn't happen on warm starts. If I let the car sit for more than 3 or 4 hours, it will happen. I couldn't find anything obvious that was leaking vacuum pressure. I reset the DME when I installed the cats (as you suggested) I will run my durametric and see the values for the O2 sensors at startup and see what they are compared to when I'm warm starting the car. It only lasts about 30 seconds ..then the idle seems correct. My worry is that it's the AOS. I have a new one on a UPS truck somewhere in Canada (by now)

When it first starts...it seems to set idle correctly...for the first 2-3 seconds...then there is a "condition" that causes it to start to stumble. The DME raises the revs to compensate. After about 30 seconds or so, the idle is normal. I'm thinking this might well be an O2 sensor issue as they would not be providing data for starting, just operating. It's only once they (all of them, maybe just one of them) get some heat into them that the idle is correct. hmmm, going to investigate this further... sean

Quotes source post: 8665071

It is almost certainly not an O2 sensor problem since at engine cold start and for some time after a cold start the sensors are not hot enough to work properly and the DME does not rely upon them for mixture monitoring. If you have an OBD2 code reader/data viewer connected to the car you can view the sensor voltages and see the voltages are not 'right'. My experience (observation) is when the secondary air injection pump shuts off or shortly thereafter that is when the sensors are working. The DME will switch from open loop mode to closed loop mode (you can see when this occurs too with the data viewer). I have not gone back and refreshed my memory on this whole thread but from the info you posted in your latest posting I suspect the coolant or air temp sensor is delivering too high of a temp reading and the DME is based on this leaning the fuel mixture when it should be leaving it rich. The bad sensor starts out ok but within a few moments after engine start I believe then goes bad and supplies an overly optimistic temp value. With an OBD2 code reader/data viewer you can view both the coolant temp and intake air temp sensor values at engine start and over the span of time when the engine starts to misbehave. A real good code reader/data viewer would let you log these and view the data either at the reader's display or download the data at your PC and view it on your PC. Have you tried disconnecting the MAF at the wiring harness and seeing if the symptom still happens? The MAF supplies the DME with the air temp value the DME uses (along with other inputs) to properly fuel the engine. Sincerely, Macster.

yeah, based on your recommendation, I tried unplugging the MAF....the car did the exact same thing. I'm going to plug the Durametric in this afternoon before work and log all the sensors that you've mentioned and look for spikes/changes.

I know you're in a world of hurt here with your car, but your mention of a lot of rain (if that is normal there) and hard starting might be an issue with wet or cracked coil packs... might be worth mentioning to the dealer.

Ok, so I got my car back from the dealer. New water pump and serpentine belt installed. Car still does the same stumble on cold starts. So I need to hook up the durametric to check some figures....my questions is which channels/sensors do you guys recommend I record? sean

Did you get a new MAF? Did you change those last 1 or 2 coils you said you didn't get to? Full TB cleaning... O2 Sensors..... If those don't do the trick, then look deeper. But the car is CPO'd, tell them to fix it! You should have nothing to worry about!

Yeah although it's fun to troubleshoot, I would agree to take it back under CPO. Also as Macster pointed out O2 sensors don't play a role in cold start, the DME uses pre-programmed maps until the heaters and exhuast warm up the sensors.

Which is curious, because it's precisely when the DME is NOT using the data from the O2 sensors that it seems to have problems :\ My dealer is 500 Kms away and I need to take a day off work every time I bring it to them. I've taken so many days off so far, that my 911 is becoming quite the joke around the office > I unplugged the MAF and it did the same thing. I've cleaned the TB, it really wasn't that dirty. Same thing. I haven't enquired about the coils, but it it was the coils, it would do the same thing when warm. I guess I'll have to bite the bullet and send it to the dealer again. sean

Quotes source post: 8688216

I would look into the 02 sensors! Also, mention to the dealer that distance and work is an issue for you. Force them to accommodate. Show a little strength and they will help you out! CPO = No worries man! Let them bicker at work... in the end, they know what you drive.

Have you checked if you have a small crack on the oil filler tube? Is your oil filler cap sealing well? Does sound like the fuel/air mixture is messed up a little at start up. Definitely log the intake temp sensor and coolant temp sensor reading at start up. Also, check the long term fuel trims on both banks (to see if there's a consistent lean condition, which may indicate you have a small leak somewhere). Also agree with Macster most likely not an O2 sensor issue since the sensors are not even at their operating temperature and the engine is still running at open loop for the first 1.5 minute or so if you just let the car idle at rest.

Quotes source post: 8688216

If the symptoms appear before the sensors are hot enough to work, if the symptoms appear while the DME is still in open loop mode as it would be upon a cold start and some time afterwards, this strongly suggests then one or more sensors the DME relies upon for inputs to determine fueling wtihout the benefit of O2 sensor feedback are bad. At cold start and until the O2 sensors are hot enough to work the DME only has a few other sensor data to go on along with its tables of fuel/timing/valve timing/lift maps to properly fuel the engine. If a sensor is bad, or delivering bad/wrong data... Sincerely, Macster.

Quotes source post: 8688486

Right...I'm asking which sensors those are??

Quotes source post: 8688391

I agree. Also, could this be an impulse sender on it's way out?

Quotes source post: 8688661

Sorry. Two critical sensors are intake air temp and coolant temp. Now the air temp sensor is in the MAF so if you have disconnected the MAF and the symptoms still appear this points away from the MAF and towards the coolant sensor. What you want to do is with everything cold check the intake and coolant temp sensors (reading them via an OBD code reader/data viewer connected to the OBD connector of the car) and comparing their readings to each other and ambient air temperature. Also, keep looking at the sensor readings after you start the engine and while the engine is idling and ideally while it is exhibiting the symptoms you spoke of. The crankshaft position sensor is a critical sensor but what I've gathered is its failure is usually present at engine start and the symptom is the engine cranks just fine but does not fire. The DME is getting an implausible signal/reading from the crankshaft position sensor and because of this the DME can't know where the pistons are and can know when to trigger a fuel injector pulse or sprark. So it triggers neither. Since the engine is not violating emissions no Pxxxx error code is logged. A Porsche error code may be logged but over the counter generic OBD2 code readers/data viewers can't access these. Durametric might. I have no experience using this tool. But I'd eliminate the engine air intake and coolant temp sensors first as I covered above. Sincerely, Macster.

I've read a few posts on renntech about dodgy impulse senders and strange things like stalling. Also something to bear in mind. His car does appear to be operated in the cooooold cooold harsh winters of great white north 450k from Edmonton (props to driving a C4 there Sean, I bet it is fun in the snow!!!!!). A lot of wear and tear on a car in those conditions.

You make good points. If the air temp and coolant sensors pass muster then the cause of the engine's symptoms lies elsewhere and a marginally functioning crankshaft position sensor or from your comments about the driving conditions one that might be loose or damaged from exposure to salt and the elements...well, that possibility needs to be eliminated. Sincerely, Macster.

Man this is frustrating. So hooking the reader up requires the key to be turned on....turning the key on, it takes about 30 seconds for the reader to initialize...than 10 seconds to load...then I try to turn the key to ignition, and the key is locked. It won't turn at all. I have to turn the key off and back on again for it to let me turn it all the way to ignition >| Doing so resets the reader. @#$#@!#@% Anyway, once I get things going, the intake temp and coolant temp are the exact same...which is 2 degrees C warmer than ambient. The coolant temp starts to rise, but that's expected. My MAF and MAP both hover around 7-9 until the system is closed. then they drop to 4-5. None of the reading I could get "look" off or disjointed. I'm going to come back to the secondary air injector for a second...I am pretty sure it does NOT come on right away. In fact, I don't think it comes on until the DME revs the engine up to 1200. Secondly, the MAP and MAF numbers drop when the circuit closes and the SAI shuts down. However, I thought the SAI injects air right into the TB...past the MAF. So why would the MAF numbers change based on the SAI being on or off? With the SAI off, even if I manually rev the engine up to 1200, the MAF numbers do not come up to 9. Anyway, I'm just about done with this whole thing. I go on vacation for 2 weeks in a couple weeks. I think I'll drop the car off at the dealer and tell them to find and fix it. Unless, there are any other suggestions out there? Oh, and Logray....I'm not driving it in winter ever again. It is fun in the snow, but the winters are so harsh here that the bodywork takes a beating. I have two stone chips in the windshield and several paint chips. I'd rather buy a $2000 clunker for the winter than have to repair the paint/windscreen every spring. sean

Quotes source post: 8690993

This doesn't sound right to me for some reason. The DME setpoints for an idling engine Mas air flow and Spec air mass are: 17 kg/h +/- 2.5 Voltage for MAF 1.3 v +/- .2 What are your TRA/FRA numbers? At idle, TRA should be 0 +/- .1 and FRA should be 1.02 +/- .04. Do you have an intake leak, perhaps a cracked oil filler tube, plenum boot damaged/loose/etc? This might be a job left to the pros since they'll have access to a smoke machine.

Quotes source post: 8690993

The SAI should start shortly after you start the engine whether you want it to or not. It is also normal for it to start at other times ( here is a link to a very technical explanation ). It is basically just a pump that dumps oxygen into the exhaust on a cold start (into the top of the block into a channel through the head and out the exhaust manifold). Yes, the secondary air system and DME also operates various check valves around the car including SAI check valve and resonance flap to name a few. At least on the 1999 car, there are potentials for leaks within the SAI system as well, a couple of the air tubes enter the intake plenums and if they are disconnected or cracked there will be exposed holes to the intake, albeit very small. Yours probably looks a little different than the one on my 1999 car, it has a 993's design. In later cars they combined the two check valves into one, among some other changes. Check out this nightmare, my lower check valve was chocked with carbon, and several of the tubes were broken in half. I finally got the spiderweb back together (the big hose in the upper right goes into the pump).

TRA/FRA? Timing advance? I've inspected the oil filler tube before...it's good. My MAP is rated in inHg My MAF is g/s

Quotes source post: 8691135

TRA/FRA = fuel adaption. There are values to monitor in Durametric. " TRA - Time Relative Adaptation; a fuel injection adaptation method used to compensate for engine/ sensor changes at idle, which are short term & small affecting the AFR, e.g. intake air leaks, and which generally varies equally with +/- values under normal conditions " " FRA - Fractional Relative Adaptation; a fuel injection adaptation method used to compensate for engine/sensor changes under load which affect the AFR, e.g. MAFS output error, with some ECMs retaining the old FRA values with battery removal & not updating until closed loop operation returns via the O2 sensors " The oil filler tube might have a hairline crack only detectable with a smoke machine. Also, (not advised) spraying carb cleaner in the intake area to see if the engine stumbles, indicating the cleaner is getting into an area with a leak. OK math time. DME setpoint of air mass with engine idling. 170 Hg = 17 Kg. +/- 25 Hg = 2.5 Kg What is the interval/period for Hg in Durametric, is it per hour, minute, second?

For a C4S 3.6 liter engine with no loads switched on (no AC), Engine temperature greater than 90 �C, and Ambient temperature 20 �C the values should be: Idle - 16-20 Mass air flow ML in kg/h No load at 3000 RPM - 60-80 Mass air flow ML in kg/h I would want to see what those values are with a warm engine, if yours are different then that might point to an intake or vacuum leak or bad MAF, or problem in fuel delivery system.

ok, I will check those values with the durametric tomorrow morning. I will check the MAF at lunch with my iphone... sean

K...here we go... Here is what Dash Command is reporting (reading from ODBII) at idle: Doing the math, this works out to 14.4 Kg/H At 3200 rpm: At 3200, I'm at 50 Kg/H Also ran the GoLink reader, and it pulled the following info from the O2 sensors. Are they supposed to cycle like that? Bank 1 Sensor 1 Bank 1 Sensor 2 Bank 2 Sensor 2 I missed Bank 2 sensor 1...and I just took screen caps at random times... Thoughts? sean

All your numbers look fine. Note that our cars don't have a MAP sensor so any MAP sensor readings are only derived from the MAF. We do need to see the TRA and FRA for a complete picture. I also vote for small vacuum leak somewhere. Is your CEL on currently? If so, what are the codes? I wonder what that "1" under the engine icon on the Dash Command is...

Did even notice that 1....however, I have no CEL, no codes. Checked for codes with the Durametric last night, with the Dash command and the Go Link today. No codes at all.

[QUOTE=seanmcr6;8691651]Did even notice that 1....however, I have no CEL, no codes. Checked for codes with the Durametric last night, with I see. Do you think your engine starts to stumbles the same instance the air pump is turned ON? If so, that strongly points to (maybe small) vacuum leak there since your said the engine stumbles only when cold so only when the air pump is on.

LOL...so I just found out there is 2 pages under the engine section...that's what the 1 was. Fuel trims.... I don't know what ST and LT are....but LT barely ever moved. ST moved up when revving the engine. At idle...it would bounce around in the negative end of the scale. It is supposed to move around like that? While revving....

LOL... that makes sense. Yes, the short term fuel trim moves quickly to adapt to the A/F ratio in real-time. +/-10% is considered normal. Long term fuel trim moves very slowly and also +/-10% is normal (0 is ideal) These are bank 1 numbers and they look good to me. +3.1% long term fuel trim is not significant to conclude there's a vacuum leak though. Besides, if the leak only happens when SAI is on, that would not show up on these fuel trims at all. Any way to get the Bank 2 numbers?

Yes, with my durametric....so tomorrow morning, while the engine is cold, I'll run it....

Also it would be nice to see the "engine warmed up over 90C" air flow numbers at idle and 3000RPM in the Kg/Hr format... perhaps Durametric will log in that format. Not that I don't think your iphone tool is cool though.

Quotes source post: 8691539

These numbers are a little low, but not too far out of spec. Like I said previously would be interesting what you see once the car is warmed up to 90C and all loads incl. A/C switched off.

God damn this startup procedure with the durametric is a pain in the ****.... So I finally get it working (latest version) and start the car. It's very rough this morning. In fact, the car is shaking side to side when it stumbles. then it revs up and smoothes out. When it goes closed, all is right with the world. Same, same. Unfortunately, I didn't "record" any of the data, I just watched it. The only think I could find was FRAU..adaption range 2 lower Bank 1 and 2 I did that plus an O2 sensor and a couple other things. I really wish I had recorded them And, the interesting parts (maybe?) was that the Adaption Rage 2 lower (FRAU) Bank 2 was noticeably lower then the rest. All the other held values of 1 or .98....this one was .68. So I had 5 coloured lines all stuck together around the 1 mark...and this lone green line down at .68 Once the system went closed...the other numbers all shifted a little. Fanned out as it were...this bank stayed at .68 Then, just to add to the whole conundrum...I always check for codes every time I connect to the DME. I haven't had a SINGLE code in all this time. Even when the water pump was blowing up. Today...just before unplugging, I check codes....big RED error code. P1341 Fault Code 900 - Inlet Camshaft Bank 1 Sigh...so this might be a bigger problem than just a faulty sensor?! That would certainly explain the shakiness at startup if one side of the engine wasn't being powered the same as the other... Is it possible that the AOS could leak into the right side intake....just a little bit? I get NO smoke on start up or at any time. I'm grasping at straws....

Well P1341 is variocam solenoid/actuator or implausible sensor signal. If you reset that code does it come back right away? That might confirm if it is a variocam solenoid or sensor. If it does come back or not it might be worth logging your camshaft deviation and actual angles for bank 1 and 2 as well to confirm/deny that code. If the car seems to run fine except at cold start for the first few minutes, that code might just be a side affect of the stumbling engine. Still though man, you have a CPO warranty, so why worry? I would worry more about fidgeting with the car and having the dealer blame me and not covering it. You wouldn't believe the crap I got away with with one of my CPOs on an old truck I had though... all depends on the dealer I guess.

AS I said before...the dealer is 500Kms away from me. It's a HUGE hassle sending them the car. If I can determine the issue before hand and give them the heads up...they can give me back my car pretty quick. Remember, it took them 5 trips and 6 months to fix my coolant leak issue. They are not the best at diagnosing issues. In exactly 2 weeks I go on vacation for 2 weeks. So I'm planning to give them the car for that time so they can fix it. I would like to track down the issue or at least give them some direction before I do that. I will record the camshaft deviation and actual angles at lunchtime.

What a mystery. Not sure what FRAU is. My understanidng is you will need to log 4 values: Adaption, range 2 (FRA), bank 1 and 2 Adaption, range 1 (RKAT), bank 1 and 2 FRA corresponds to fuel trim at load (long term) and RKAT is for idle (short term). FRA should be close to 1 and RKAT should be close to 0.

FML.....this is getting worst and worst. Ok, so I have recordings on the FRA and RKAT. I'll get to those in a minute. While I'm doing all this testing...I notice my oil pressure is sitting just below 2 during idle. Now...the oil pressure sending unit was replaced last Aug...so I doubt it's off. When the dealer replaced my water pump last week, they gave me a jug of coolant and 2 quarts of oil. Which I found strange since they've never done that before. Plus, I understand the coolant, but why were they giving me oil? I checked the levels of both when I picked up the car...and the next day...and 4 days later. (2 days ago) all were good. Coolant was down slightly, but well within range. If you watch the ids I've posted about the stumble recently, you'll see the oil level is full. I also checked this manually the last time, and the oil was right at the mid way on the dip stick. When I rev the engine, the oil pressure shoots up, almost on par with the rpms, just with a slight delay. Usually...the pressure is above 3 at idle..and raises as I get the engine going. It stays up around 4/5 while driving above 3K. I've never seen it below 2. I check my oil level by computer...it's half way?!@ Check it manually and sure enough the oil is way down at the bottom tip. WTF?! Where did my oil go? I'm talking about two days. To work and back twice. That's it. No puddle under my car. man alive this car is driving me bananas. Adaption Range 2 FRA Bank 1 remains at 1.01 consistently. Adaption Range 2 FRA Bank 2 remains at 0.97 consistently. Adaption Range 1 RKAT Bank 1 remains at 0.66 consistently. Adaption Range 1 RKAT Bank 2 remains at 1.55 consistently. Adaption Range 2 Lower FRAU Bank 1 remains at 1.01 consistently. Adaption Range 3 Upper FRAO Bank 1 remains at 0.98 consistently. Adaption Range 1 is the only one that seems to have a deviation between bank 1 and 2. However, I have no idea if it's supposed to be there or not. CAMS MAF

Quotes source post: 8693616

The symptoms do not point to the AOS. The AOS failing is a common failure and both banks' fuel trims would reflect the AOS failing. The error code and the description of the symptoms matches what I experienced when my 02 Boxster's passenger side VarioCam solenoid (and actuator) went out. At idle the engine ran horribly. But above idle it smoothed out and up to about 4K rpms (as high as I dared rev the engine) it ran just fine. I cleared the error code and drove some distance (over 30 miles) with no symptoms and no error code. Obviously the solenoid/actuator failures were intermittent. But I took the car in and the tech confirmed the solenoid was bad. The actuator proved to be bad too. You have or the tech has to be careful. The PST2/PIWIS will almost certainly help him identify the solenoid is bad. But oftentimes, so the techs I rely upon tell me, is that the actuator will prove to be bad too. Often enough they advice the customer ok both items to be replaced. Now what I think happens is the solenoid starts to go goofy and in doing so causes the actuator to be activated and deactivated and from this over time the actuator gets worn out. The trouble is that with the solenoid bad one can't id the bad actuator until the engine is buttoned up. This is what happened in my case. For some reason even though I was told afterwards the SOP is to replace both the solenoid and actuator for some reason I was not given this option/choice. I was told the solenoid was bad and of course I agreed to its replacement. But once I had the car back the error code and symptoms returned after less than 50 miles of driving. Back in the shop the actuator was id'd as being bad. But because I was not given the option of having both the solenoid and the actuator replaced (which I would have done based on the advice of the techs) I was only charged for the parts/labor specific to the actuator R&R. The extra labor came to about 4 hours (to get to the actuator the cams have to be removed and reinstalled and retimed) and the actuator of course has to be replaced once the cams are out of the way. The camshaft cover bolts of course have to be replaced and the camshaft cover resealed but I was not charged again for these parts, the sealer, or the labor. Anyhow, in the case of the Boxster the engine stayed in the car but the exhaust pipe on the passenger side along with the camshaft cover had to come off. The cams had to be removed to gain access to the actuator. All camshaft cover bolts had to be replaced: They come with a thread sealant and are not supposed to be reused. Some of these bolts fit on either side of the bearings that receive high pressure oil and the odds are the bolts will leak if they are reused. Sure the camshaft cover receives a proper application of sealer to prevent high pressure oil from leaving the bearing area but the bolt thread sealant is another barrier to the oil should this sealant not contain the oil from around the bearings. Besides the underneath side of the camshaft cover is subjected an oil storm while the engine is running. The oil pump can pump 60 or more liters of oil per minute at higher engine speeds and a good percentage of this oil is directed to the valve gear which resides directly udner the camshaft cover. All of this oil ultimately is sprayed, slung, splashed about under the cover before it is removed by the scavage pump. The solenoid cost around $700 and the actuator around $800 but I received a discount (a bit more than 10% IIRC) due to my PCA membership and my good relationship with my dealer and its service department. Oh, labor... the labor cost over $1000 and with incidentals (oil/filter service) and a problem with a broken exhaust pipe bolt the total cost of the repair came to around $3000 parts and labor, taxes and so on. Sincerely, Macster.

Quotes source post: 8694354

The FRA's and RKAT's all look excellent to me (0.66 and 1.55 are actaully 0.66% and 1.55% if I'm not mistaken, which means really good). And yes, they are supposed to flutuate. I do NOT know what FRAU and FRAO are though. Re your oil, engine fully warmed up the oil pressure can get a tad below 2 BAR (assuming 0w-40 or 5w-40). If you checked the oil right after you stop the engine (especially a not fully warmed up engine) and don't let enought time for all the oil to return to the oil pan, it could read low. Just try to measure it again 10 - 15 minutes later.

Quotes source post: 8694371

There is a VERY specific procedure in the work shop manual on how and where to apply the beads of sealant to the cam cover. Also, a very thin bead (according to the specification) goes a long way. You don't want excess sealant swimming around in your engine clogging tiny oil passageways. Even this might be too much, even though it follows the Porsche specification/requirement.

Quotes source post: 8694354

There are a lot of threads about where to run your oil level. Full to the brim is usually not recommended. Just over half-way is. There are also TSBs about overfilling and oil level, if your car is a 2002 model year it might require a new dipstick and/or oil level sensor along with instrument cluster re-programming (less oil). Also agree that your oil pressure numbers sound consistent. Perhaps you're used to seeing higher oil pressures in colder temperatures and due to the warmer temperatures and selected oil viscosity you are seeing different oil pressures. The cam readings look normal and it didn't look like one of the actual angles was different from the other, which should be consistent with functional cam mechanisms. Although, did you rev the engine a couple times or was that just from the stumbling startup? Also did the DTC come back after you cleared it? It would be good to see the same graph at idle, then 5 seconds at 1500 rpm, and then 5 seconds at 3000 rpm all in a line. Also would be good to see the camshaft deviation, in addition to the actual angle, I'm not sure that we need to see speculative angle.

No, the DTC did not come back after I reset it. My car is a 2004. Yes, those spikes are me revving the engine. So the FRA/RKAT are excellent...the MAF is bang on....the cams look good....and my oil pressure is fine. Great. So I'm no further ahead I guess...hopefully the dealer can figure this **** out, cause I'm done. Nothing more I think I can do. Thanks for all the suggestions and feedback everyone, It's greatly appreciated!!! sean

The only suspicious thing I saw is your bank 1 cam reading. It has lots of "hash" and appears much noisier than bank 2.

Quotes source post: 8696931

Good call out, I didn't notice that on first glance!

Sean, any update on this? My 996 TT has the exact same problem and it's driving me nuts!! So far I've changed the ignition coils, the spark plugs, the MAF sensor, the fuel filter and the air filter but the problem persists. I also inspected the cam sensor brackets to see if they are broken but both are in good shape. Hopefully the durametric tool I ordered today will help to shed some light on this. I'll let you know if I find anything.

Quotes source post: 8712897

You've eliminated quite a bit. Hopefully with the Durametric tool you can verify the coolant and air temp sensors are providing reasonable readings, that one or both are not supplying a too low or too high a temperature which causes the DME to supply an air fuel mixture that is wrong for the actual coolant (engine) and air temperatures. Also, if the temp sensors appear to be supplying reasonable values then hopefully you can verify the variable valve timing and lift is working properly. I can't think off hand what else could be the cause of your Turbo's symptoms. Sincerely, Macster.

I checked my temp sensors and they were giving normal readings that didn't fluctuate. I'm down to the air/fuel ratio. Bottom line is the car is running way too rich on cold starts. Period. There is a very pronounced gas smell. only on cold start, never on warm ones. Once the system loop closes, it idles just fine. I'm not sure what else I can do to pin point the exact cause, but I'm still pointing at the secondary air injector. It only runs on cold starts (during open loop)....so either there is a leak with that system or it's just pooched. The only other thing I've found wrong(?) is the "actual angle for intake cam" on bank 1 shows much more deviation (in frequency, not amplitude) over bank 2. This is the same intake cam bank that threw a CEL too. I leave for vacation next week. This weekend it gets dropped off at the dealer, with all the printouts and video recordings and hopefully they can find the root cause. If their computer/diagnosis finds nothing, I'm asking them to check the SAI and the variocam mechanism (in it's entirety including the solenoid). Seeing as I've eliminated most of the usual suspects, I'm hoping they can find the issue rather quickly. If they find the issue within the first week, while I'm in Ottawa, I'll update this thread. If not, I'm in Cuba after that, with no internet access, so updates will wait until I get back. Sean

I was able to read the temp between the two sensors with a cheap OBDII reader and the difference was 2-3 deg. Is that enough to cause this issue? Also, my car runs smooth as glass once it's warmed up so if it was a defective variable valve timing would it only affect the engine when it's cold?

Quotes source post: 8713403

It is the difference the two sensors readings have compared to each other and to the actual temperature of the coolant and intake air. If the two sensors disagree with each other that's one thing. One sensor is probably bad, going bad. If the two sensors read very close together but do not report close enough to the actual temps of their respective coolant or air that's another thing. To avoid being fooled by legitimate differences in temps it is important to check these temp sensor their readings after the car has sat for 12 hours or more and not in the sun. The assumption is that after 12 hours or so the coolant and air temp have reached equilibrium with each other and the ambient air temp. Now you'd think intake air temp would be ambient but not as long as there is heat in the car/engine compartment. After you verify both sensors produce temps that are close to each other and also reflect reality you want to continue to monitor them as the engine runs/warms up to confirm while they read ok at from the start they do not start delivering wrong or incorrect numbers as the engine gains heat. While the engine is warming up from cold the coolant temp reading can jump by 0F to 4F degrees per approx. every second. Intake air temp can climb a bit as the engine and engine compartment heats up and this heats the intake air. Even if the car is moving down the road the air intake temp will always (my experience) read higher than ambient. The reason is that the intake air is made up from air that in part comes from the radiators and this is heated air. The very closest I could see the intake air temp and ambient air temp (measured with a digital thermometer stuck out side the window of the car) get is within 10F to 12F of each other. The worst is of course when you restart a hot engine that has heat soaked the air intake system. In this case the intake air temp can be 30F or more degrees hotter than ambient. But I have never observed either of my cars' engines reacting in any way to this difference in air temperature. Oh, the variable valve timing system may not work properly from cold but begin to work once the engine warms up. Another thing I have noticed with some failing components (O2 sensors) is they go bad, the check engine light comes on and the error code points to the sensor. But the engine gets shut off and upon hot or warm restart the component is working just fine. Lastly, you have other possible causes to suspect before you start condemning the variable valve timing hardware. Sincerely, Macster.

Quotes source post: 8713047

Sean, that's cold start fuel enrichment, part of the ECM program during cold start so I'm not sure that is not normal. "The ECM software controls fuel enrichment during cold-start by altering the timing of the main injectors based on engine temperature, no 5th injector is required. The idle speed is also fully controlled by the ECM, including fast-idle during warm-up (no thermo-time switch is required)." from http://en.wikipedia.org/wiki/Motronic I smell gas too at cold start.

sigh....well, the dealer has it now. we'll see what they "find"....

I wasn't "condemning" the VVT hardware at all...just looking for possible answers. My car is parked in an underground lot. All of my tests are after the car has been sat 12+ hours overnight. The temps are fine. The all read what they should read, as per your suggestion. O2 sensors aren't even used when the engine is cold. So I fail to make the connection on how their status could effect cold starts. I am not saying there was a faint gas odour...I'm saying it reeks of gas on cold starts. That's not right...for any car. If you can smell it...it's NOT being burned. If it's not being burned...it's doing all sorts of damage to your engine. I've had two small backfires actually (if I try to rev it higher when it's stumbling)....tell me that's normal! Anyway, I'm done. The dealer has it. I'll post an update when I have one. sean

HEY! back from vacation. Tip for you all...don't break up with your Cuban gf in the middle of the Cuban country...with no transportation of your own. Lesson learned xD Anyway...get back to Porsche dealer to get my baby. The service advisor gets the tech that worked on my car and says she wants him to talk to me directly as I seem to know more technically about the vehicle than she does. Nice guy. Brings me through his 5 day diagnosis and repair of my 911. Found a vacuum leak in a PCV crankcase hose coming off the throttle body. It was rubbing again the top of the engine and wore a hole. This was sucking dirt from outside the system into the throttle body, which explains why I found my TB to be dirty. Replaced system. Retest...issue still there. Check fuel pressure...58psi, good Remove fuel supply line, drive link fuel pump, fuel pump output is good. Compression test. All cylinders are equal on initial stroke and even on final stroke. All values almost identical across all cylinders, around 260 psi. Check cam timing. Actual versus spec. Cam timing good. Inspect spark plugs....2 and 4 are wet from fuel. Check roughness values. 2 and 4 are higher than other, even when engine is warm. Using his PIWIS (or whatever it's called) he was seeing 12-18 misfires on cylinders 2 and 4 when the cold stumble would happen. They decided to replace all my plugs. Issue still there. Swapped the injectors from 2 and 4 with 3 and 6. Misfires during stumble moved to 3 and 6. They replaced those injectors and emptied/cleaned the fuel rails. reset everything and the stumble as gone. Engine now runs smoothly. (Awesome, so it was the injectors. Although the work order lists injector valves...not injectors themselves, so maybe he just repaired them? Although he did tell me he replaced them.) BTW...we noticed 2 small oil leaks. One is either RMS or IMS...the other is the transmission shifter shaft assembly. Took photos, sending them to Porsche to ask permission to replace seals under warranty. He tells me the transmission leak is an expensive job as he's not only got to drop the tranny, but he has to open it up. So I asked him...if Porsche says yes to that work....could I pay for some new syncros and have him replace them while he's in there? (if there is any significant wear). He's cautious about it and said he didn't think he could without asking them first. Whatever, I got her back and she's fixed. I pay for my new sparkplugs (parts only) and my oil change and head out. I get in the car, which has been parked for two days, and guess what? It does the exact same stumble on start I took her home as I was late for a shoot already. This morning, she started better, but there was still a hesitation and an rpm dip. Also, as with before, I cannot simply drive away...if I try to drive right away, the engine bogs and I have to over-rev and run the clutch to get it moving. So I don't think it's fixed. Sadly. She does idle more smoothly now (once warm) So when I bring it back for them to fix the leaks, I'll let him have another kick at it I guess. sean

Ouch Sean! Both on the Cubanita situation, as well as your car! Good thing you've got such a sense of humour, I'd be pretty frustrated about this. You're lucky that your "Veronica" is still under warranty! Keep us informed... Maybe you need to send the question to the guys on "Car Talk"?

Well, I have no one to blame but myself for the cuban princess I just went out for groceries after letting it sit all day and the stumble was less pronounced. I'll leave it for a week and then decide if they need to see it again. BTW...while I had the mechanic's attention, I asked him about putting in my brand new, bought 8 months ago, still in the box, Stage 2+ clutch kit with LWFW...and if he thought there was any risk to my crank/engine and he said it should be fine, he was going to asked another tech who races and works on several track machines to get some feedback, but he's never heard of a 996 that went down due to LWFW. I also asked him what he thought about doing the LN retrofit. He said they've only ever seen 2 996s that had any kind of IMS issue and those were many years ago. He said he think the internet blows things out of proportion and said if I don't hear anything coming from it, don't worry about it. He said you can really obviously hear it when the IMS is on it's way out. I'm not preaching or evangelizing, I'm just forwarding his thoughts to the group. YMMV.

Sean, this IS frustrating. Sorry to hear that. I was hoping to see a happy ending in the middle of your post. Any updates?

I think the fact the problem only happens in the first minute or so after a cold start really limits the time window for diagnostic. I wonder if the missfire was the cause or a consequence.

well, like I said...there is way too much fuel flowing somewhere. It's not being burned and on one occassion, attempting to rev the engine produced a backfire! Since he found the plugs on 2 and 4 were fouling/wet and replaced those injectors...I'm hoping it's just another injector that decided to give up. When the stumble happens now, the engine really chakes from side to side...so now it's unbalanced, leading me to believe the issue is with only one injector. Heck, maybe he didn't install one of the injectors right? unlikely but possible. They are getting the car at the end of the moth to repair the RMS and transmission leak..so he'll have to look at it then. Good news is that Porsche said he's allowed to do extra work for me while he's got the transmission out....so if any of the syncros are worn, I can save 90% of the labour charge to have him replace them. I'm happy about that! I will update....in early Sept when I get her back. Cheers sean

Sean, Still experiencing the stumbling problem? Mark

Any update?

Yes. still happening....no, no update. Was supposed to give the car back to the dealer a couple weeks ago but I couldn't be without my car. It's hard for me since the dealer is 5 hours away and won't give me a loaner. So I'm faced with renting a car for a week just so the dealer can tinker with it.

Quotes source post: 8944877

Are you under CPO? Bummer no loaner and 5 hour drive...yikes that's not convenient. I won't tell you how close my dealer is.

Are the misfires on the cylinders that still have old coil packs? I know everyone has been focused on afr issues but bad packs will cause similar behavior. S.

Regarding the transmission leak... My Turbo's tranmission (6-speed) developed a leak around a selector shaft seal and the short version of the story is the old leaking transmission was removed and replaced with a factory supplied new (remanufactured) transmission. Not only did the 'new' transmission not leak -- and has remained fluid tight in more miles than the original one -- the 'new' transmission shifts better than the original one. It may be different for your car but under warranty the leaking transmission may not be fixed at the dealer, it may be replaced. Sincerely, Macster.

Given two injectors were bad I would suspect more than 2 are bad and replace all 6 with new injectors. However, the dealer is handicapped in that it can't just throw parts at the symptom, it must adhere to whatever guidelines the factory has in place to justify the injectors' replacement. I mean you have eliminated just about everything else and that the symptom was cured albeit temporarily by replacing 2 suspected injectors, but all this means now is that the 2 worst injectors have been replaced and this gives the next worst injector or injectors a chance to act up. Sincerely, Macster.

See, that's what happened. He had misfires on 2 and 4. Swapped those injectors for 3 and 5. Misfires moved to 3 and 5. So he figured the injectors were bad and replaced them. However, problem is still there. As for the coilpacks. That would be nice to know...unfortunately, the dealer I bought the car from (across the country) never told me which coil packs they changed. Nor did they provide a service history they said they would, etc, etc. No, the dealer just took the transmission apart and fixed the seal. At least, that's what the service writeup says.

Sean, i have maybe the answer to your cold start stumble. I was having same problem and yesterday i installed all new coil packs. I waited for this morning to verify problem resolved because forst start of the day usually the worst and i just came in from doing that.............started right up and rpm immediately went to just a tad over 1000rpm..........smooooooooth smooooooth smooooooth no stumble whatsoever!!!! I think you'll solve yours too changing them out, Good Luck and let us know. Tim

That's good news Tim, hopefully replacing coil packs will eliminate Sean's problem as well since the injectors didn't.

I sure hope so cause he's been round and round with the dealer. I thought about it and for mine, I first acquired the car in may this year, very hot temps and no issues but then it started to cool down @ a month ago hence more moisture in the air. That's about when the problem presented itself. When I changed out the spark plugs I inspected the coil packs and 4 of them had fine hairline cracks. So after reading and deducing moisture now with cooler temps was invading the coil packs common sense for me was change them all cause they looked like the original ones still and besides better to do all at once then piece-meal a few. What a difference it made! So I guess I was on to something afterall with my reasoning cause she's purring like a kitten again.

When I first purchased my car, 5 years or so ago, I had a slight stumble when idling. The misfires were only on two cylinders, but as soon as I replaced those coil packs the stumble moved to two other cylinders. I couldn't find any cracks on the coilpacks, but I replaced all six. This resolved the problem and it's been five years with no issue. If you don't have a way to read the misfires I'd start by getting yourself something. That should help you track down which coilpacks are old. If nothing else I'd start with #2 and #4 based on the mechanics original findings. Worst case scenario you replace all six for ~$360 and move on with your life. -Shawn PS--I just realized you could probably pull the plugs and figure out where the misfire is as well. I didn't look back through the thread, but didn't you have two plugs that were wet with unburnt fuel? Sounds like a cylinder not firing....

Hi seanmcr6 For your vehicle to encounter such stumble during only cold start only in the morning is basically two things you need to check: Firstly- check for a confirm air leak in hoses if there's if not then I'll narrow down to intake manifold gasket. Secondly- do a re- learn process for your issue remove the negative terminal on your battery and allow it to rest for 10mins or so, there after step on the brake for afew time before reconnect back your terminal on the batt. And before you start the engine operate all the electrical equip and accessory in full round of operation including power window(in key on position), from this stage start your engine and let it idle at least 10 min. If you encounter engine stall just loosen your trottle sensor screw and do a slight adjustment clockwise or anticlockwise. And start again till it idle. At time you'll face engine idle too high which is normal cause of the trottle sensor was being adjusted. Just do an re-adjusted till it idle go back to normal. Hope this help you

Well here is my update to my cold start issue. After replacing a while ack all coil packs, plugs, tubes it seemed fine. Then it started again but not as bad. The other day I was using my durametric farting around and afterwards thought why am I still getting this stumble. Long story short I thought maye the TB removed before for cleaning and not properly tightend. So out with a 10mm socket and sure enough each bolt needed tightened. Every time I start it now it is rock solid and instantly goes to @1100 rpm. No more dipping to @700-800 and stumble.

Nice to find out that sometimes it's the simple things that can cause a problem. Good troubleshooting on your part.

Quotes source posts: 9053619, 9053453, 9053453

Several posts ago.... his car was at the dealer for quite a while several times an they've already fixed many if not all of the intake leaks. More could still be present I suppose... He's done this as well already several posts ago. Huh? Could you elaborate a little more, he has a 2004 e-gas car.

Quotes source post: 9053832

Absolutely and thanks!

Oh one more thing is that do check on your injector as well which I suppose you've done so, but just to be sure. By the way intake manifold gasket leak is something can't be determine but one dismantle replace is needed. This is what I've done on my and indeed it does really solve all my issue just like wat you've there. My issue was at early morning cold start at engine temp less than 30 it ramp up to 1200 and within 1-2s it start to stumble and the iac valve start to kept on self adjusting so it cause the engine to stumble, and the injector Also spraying different pluse of fuel timing which also the cause of stumble. But once it warm up to around 60 everything is back to normal for the rest of the day, I've been trying to tackle this problem for about a mth and finally it solved. By the way about my background: I've been working in automotive trade for 18yrs from a technician to an technical supervisor trainer. BMW, Volvo, VW, Benz, bentley, jaguar, Cheryl, Reanult

I haven't seen Sean here in a while but I read an interesting possible solution to this. Try a tank of lower octane gas, perhaps mid or even low grade.

I guess a loose throttle body is a vacume leak Oops !!! I wonder if the porsches have a gasket that has to be oriented correctly for thermal insulation. What kind of gasket was it if any?

I have read this chain hoping for an answer.. none. I have the exact same problem, occasional stumbling starts hot or cold, runs like a pig for 30 secs then clears and is perfect. Changed all plugs, all packs, maf, air temp sensor. Been to dealer, they checked for air leaks, nothing. Done the durametric all looks ok, fuel trim ok. What is really odd though. If I unplug the lambdas it is is worse, but they shouldnt be read at start up. Also according to another OBD reader, sometimes my car starts in closed loop ( when it runs poor) and then is fine when it either starts in open loop or when it goes to into open loop. Surely it should always start in open loop and ignore the lambdas, so why would unplugging them affect starting? Please help as gone on for a year now and dealer stumped. 1998 996 3.4. Funny about recent post, I had the feeling it was worse with high octane fuel, so I will run it on 95 and see if its any better. Why would this be?

IT should only run in open loop when the engine/exhaust/o2 sensors are NOT up to temp. So if you ran your car for a while and turned it off...it will start in closed loop if everything is still warm enough....

thks, what is odd that it always starts in closed loop, hot or cold for the first 10 seconds, then switches over in open loop when it seems to be better. Is this normal? Also odd that the air temp sensor always reports -40C for the first 30 seconds before it shows the correct temp. I have changed this for a new one, but its the same. like most things if your car is ok, you dont obsess with figures, but as soon as it plays up everything is a culprit. It hasnt been too bad recently, what I need is the laptop connected when its being really difficult.