any help here. I have this really odd problem in that maybe 25% of the time my car will start ok, but run roughly for 1 - 2 minutes. It will idle a bit lumpy, but give it some throttle and it just coughs and splutters, up the road, floor it and it will pick up. After a couple of minute runs perfectly.
Have replaced plugs, coils, MAF, air temp sensor, all throttle etc clean. Done loads of OBD logging and it appears to be that the car goes into closed loop on some starts and tries to read from the O2 sensors, but as they are cold, it struggles, also I replaced both O2 sensors.
If I unplug them I get the exact problem even though it should not read them when cold starting and be in open loop.
Been to porsche who looked for faults, and did intake leak tests, no problems found, They thought it may be because I have intake mods and free flow exhaust, it messes up the figures a bit and the car gets a bit confused. Has been going for a year now. Sometimes even on hot starts, like after getting some petrol.

 

With no CEL and no pending DTCs generally a engine that exhibits mild troubles upon a cold start is running lean.

The coughs and sputters -- if it is from the intake side -- also points to a lean condition. Generally spit back or coughing through the intake is a sign the fuel mixture is too lean, while backfires from the exhaust are due to an overly rich fuel mixture.

One of the first explanations is there is an air leak. Anytime the intake has been messed with an intake air leak is always a concern.

That this cold start behavior doesn't happen all of the time though tends to eliminate that possibility, but you have to keep it in the back of your mind just in case.

You have replaced the MAF which is responsible for supplying the intake air temp.

Have you looked at the coolant temperature? What I advise is to upon cold start note the ambient air temp, the intake air temp and the coolant temp. They want to be about the same.

You want to ensure both the intake air temp and the coolant temp, and since you have replaced the MAF especially the coolant temp, do not suddenly show an implausible increase in temperature.

The coolant temp will after a moment or two of engine running go up a degree (F) every 2 seconds or so at idle. Intake air temp will also go up as as engine heat heats the intake but the intake air temp will not go up very much. And this is important for reasons I'll get to below the intake air temp should not move around, drop or rise suddenly.

I can't keep the various after market intakes straight so I will speak in general terms. With an aftermarket intake system there's the risk of an intake leak. The only times I've had an intake air leak on any engine was when installing an aftermarket intake system. These can be a bear to get sealed.

There is another issue with an aftermarket intake system in that it may mess up the air flowing into the intake and the air flow past/through the MAF. The MAF requires laminar air flow to derive air mass accurately. If the air flow is not right the MAF can be all over the map regarding air mass and as a result the fueling of the engine is not quite as good as it should be. A cold engine is more sensitive to this.

Now because the engine is cold no error codes may be generated. And since the engine's behavior improves as it gets warmer the situation may not be bad enough to trigger a CEL.

You can look as I mentioned above at the various temperatures readings.

You can also look, log if possible, the short term fuel trims. It has been a while since I have done this with my cars and I can't trust my memory 100% as to what normal behavior is other than they will start out rich (on the plus side of the number line) and this will drop as the engine warms and fuel is subtracted.

What you should not (probably should not ) see is wild swings

But I note the symptom doesn't occur all the time so monitor/log all starts and capture the readings when the engine behaves and when it misbehaves and analyze for any clues as to what's going on.

There is another possible explanation though, and I don't want to be too alarming so take this as just something to consider, and that is the VarioCam system is not working right sometimes and the engine behaves the way it does because of this. Now even though the DME is not in closed loop mode I would have to believe that if the camshaft position sensors delivered bad news the DME would note this and set the CEL and log one or more DTCs.

But if you can monitor the camshaft position/activity (whatever there is available using the OBD logger (I think Durametric may be the only system with the ability to log this info unless you have a Porsche System Tester 2 or something newer) and compare the data obtained when the engine is acting up and when it isn't for a hint as to what is going on, if anyting in this area.

Added: Probably should add some caution about the AOS being the source of the symptoms so you'll have to consider that which falls under the air leak category of possible causes of the symptoms and even once I replaced a MAF when the root cause of a symptoms (mainly CELs) was a leaking oil filler tube cap, which is just another air leak possibility.

Sincerely,

Macster.

 

Thanks for the detailed answer, I sometimes get an error on the air intake temp at startup, not a CEL, but logging of -40C for a few seconds, before it sets to ambient.
This is part of the MAF which is new, but the old one was the same. Odd one.
I added a poor warm start for info

 

Can you smell anything from the back when it's idling rough? Any smell of gas at all?

What do describe is exactly the same issue I've had for over a year and the dealer could not find the issue either. They've replaced injectors, 2 coils and all spark plugs. Checked for air leaks, etc. Still does it.

I was running sport cats...and removed them and put the stock cats back on, just in case. Now the car is 100% stock (except for clutch/flywheel) and it still happens.

Mine has a definitely gas odour however, so I know I'm having a rich condition.

You are not alone...by my count, there is 5 of us on this forum who seem to have the same intermittent rough idle on start up. (usually cold starts)

I've been down the MAF, O2 sensor, intake system, vacuum leaks trail for a while now. For me, I'm starting to think the Secondary Air Pump might not be pumping enough air on start (?)

I've tried cold starting the car with the SAP unplugged to make sure I wasn't getting a lean condition as someone on here suggested to me at the time, but it made no difference. So maybe I'm just not getting enough air?

Good luck and if you find the solution, please post!

sean

 

I'm short of time today but I have printed out the PDF and will analyze.

Briefly, the fuel system status values are contrary to the specs. 1, 2, 4, 8 and 16 are valid. Odd numbers are not valid. There are two explanations that come to mind: the DME is not 100% in agreement with the spec in this regard or the test equipment is for some reason or another supplying bad data.

I have never observed a zero (0) or -40C temp from any sensor.

This suggests the sensor signal is bring grounded thus the voltage reading is 0 volts.

Unfortunately that is all I have time for right now.

Sincerely,

Macster.

 

Also, looks like bank 1 sensor 1 O2 sensor is lazy. During frames #6-#18, that's 50+ seconds and the voltage of B1S1 stayed around 0.1v (lean) and the short term fuel trim on B1 also was quite high at 10~20+%. Compared that to B2S1, which flutauted between 0 and 0.7 a few times already.

Granted, these measurements were taken 4-5s apart so the sampling is not frequent enough to represent the real O2 readings but still, hard to believe within 50s, all the samples collected were around the low end.

It would be useful to continue recording the same AFTER the hiccup period, then you can have good numbers to compare with when the engine smooths out. For example, now the sensor 2 on both banks seemed not moving but that may change once the engine smooths out.

 

Six. And counting. Coils replaced at 60K, no change.
q

 

Come to think about it, could also be extra oxygen from missfiring in bank 1 that caused the lean bank 1 reading.

 

Please ignore my post from the other day. I was all mixed up. (Sloppy due trying to rush things and got the column labels and numbers off by one.)

Short term fuel trims that approach +/- 10% are signs of trouble.

I think Ahsai may be on to something regarding the sensors.

But I note or seem to recall you replaced the #1 sensors so there can be other explanations as well. The sensors in this case are signaling a problem they are not the problem.

Maybe.

Let me continue.

The #1 sensors want to swing between 0.1 volts and 0.7 volts (or a bit lower and a bit higher) as the DME varies the fueling from rich to lean (by small amounts on either side of ideal).

The #2 sensors want to read around 0.7 volts though they will react some to the varying mixture though they will not react as nearly much as the #1 sensors.

The #2 sensors get the exhaust gas after it has been processed by the converters and it should have a dearth of oxygen the oxygen having been consumed in the processing of the exhaust gases.

Now a reading of 0.435 volts is special in that it is the reading a non-functioning sensor generates. The sensor is too cold perhaps or is fouled are 2 possible explanations.

If the #2 sensors are bad the DME can go crazy trying to get a reaction out of them. It will adust the fueling as far as it can until it reaches a threshold or until the engine misfires.

With mods to the exhaust and intake the first thoughts that come to my mind are there are exhaust leaks which if ahead of the #1 sensor really mess it up and this causes the DME to go crazy with the fueling. Short term fuel trims are all over.

Because of this the fueling is so bad the engine can run poorly and the #2 sensors deliver chaotic readings that arise from the fact because the fueling is so bad the converters have no chance of operating properly.

Then there is the intake leak possibility.

Another possibility is if the engine never ran right with the mods they are just not suited to the engine.

The intake system does not provide the type of air flow the MAF needs to work properly.

On the other end of the engine the exhaust may not feed the exhaust gases to the #1 sensors right. These need to be bathed in the exhaust gas flow from all cylinders of that bank.

Then the after market converter may not have sufficient catalytic material or may be too open and the exhaust gases do not come in contact with with the working surfaces of the converters so exhaust gas processing is sub par and sporadic.

Another problem is the #2 sensors are not kept warm enough and their functionality drops off in this case.

So what does this all mean?

First I can't keep track of all the mods but you need to determine as best you can that the mods work. Just because something's for sale doesn't mean it works as it is supposed to. I do not know anything about companies that made the mods that are on your car I'm speaking in general terms.

If you can find others who have the same mods and have met with success that's a good step in the right direction.

You have to be sure the mods are installed correctly or if the symptoms are not concurrent with installation that the new hardware has not loosened up developed a leak.

If you are sure the intake/exhaust system is leak free then you have to suspect a component problem.

You replaced the #1 sensors but not the #2 sensors. The #2 sensors may be bad. This is not a recommendation to replace these but you must keep this in mind as you work through and eliminate other possible causes.

If the symptoms are conncurrent with the mods there is always the risk something was upset during the installation. You have to suspect fuel lines, fuel pressure, fuel supply or something with the secondary air injection system.

Sorry I have not been able to give you anything specific but the mods bring with them more things to consider vs. a stock engine acting up.


Sincerely,

Macster.