ragpicker said:
My friend (who is my Indy) had this issue on a car he had in a few weeks ago.
He spent literally days trying to figure it out - changing lambdas, crank sensor, throttle sensor, plugs, packs, earth points, engine out looking for air leaks etc. I'm not joking he spent 3 straight days on it (this is the reason my car took so long to get in to have my problem sorted).
He has also got a 3.4 carrera. In the end he took his ECU and loom from his car and put it into the car with the problem. Bingo! Problem of rough idle sorted instantly. Symptoms were exactly the same as you have described.
When he plugged the faulty cars ECU into his car - the problem was there again.
This is the 3rd time he has seen this problem apparently.
Fortunately ECU's with keys can be found on eBay for about £100 but it must be the same engine model number as yours. There was a bit of an ECU changeover at the end of the 3.4 production run apparently.
Give Nelson Porsche a ring (stockton on tees) - say Ragpicker sent you :wink:
I got as far as this then stopped reading .. its also what i was thinking and i posted before reading this which makes it even worse im afraid as 2 of us have come to the same conclusion independantly.
Mine was a bitch of a problem to find as well but i did have thottle body codes , random ones but a pointer at least.
Heres what i tryed to try and prove it , which i want you to try please ..
You know where the dme is .. it gets hot .. very hot , stupid place to put it if you ask me .. remove the carpet , remove the bracket and have the dme face up , so hopefully it stays cooler , then try a drive and see if any difference.
A reprogram is a possible but my one we just replaced it to fix it.
Couple of other things you may want to consider , i made some notes as i was reading from the beginning ..
Tank venting , happens at operating temp , need to find a way of blocking the pipe , vent pipe left open i think would be safest and seeing if that makes any diff .. not sure the best way to do it .. but you know what i mean.
Lambdas .. they work from about 1 min after cold .. fault would be there before operating temp i feel.
Vario cam is a possible , it will affect idle .. but ive always seen rough running or misfires with that fault , it wont stop the car from starting either .. possible but im unsure .
Erm the pict of the AOS .. thats an earth lead there next to it .. might be an idear to clean it or at least check its a good contact as it doesnt look like it lol ... not sure off hand what its for but it will be sensor related .. probably nothing to do with the fault .. but even so
The fault itself .. dme monitors crank sensor and so rpm , if it drops below 700 it will increase the throttle to raise the rpm .. this is a constant thing so it will continually alter the throttle to maintain idle.
Your car is not reacting untill it drops almost to stall then bangs open the throttle to catch it .. over reving to a point and then repeating .. crank sensor is main input .. dme is in charge , throttle pedal position and throttle position all have inputs to decide idle .. hence i wondered if you could see any of these values ..
Pedal can cause something like this .. its yet another possible .. but im also back at the DME as well ..
Very hard without spending a fair bit of time on the car to help .. but hopefully something ive posted will be usefull .. please keep posting though of what you do / decide .
As i say to some of the guys with a bitch of a problem .. Good Hunting .
EDIT one other thing i did think of which you may already have tryed is check the fuel pressure at the rail .. low pressure wont give fault codes and can cause varying faults.