2007 ML350 3.5 A272 - Severe misfire on cold start
Upon cold start it runs like crap. Misfires horribly, it's just dire. You shut it on and off multiple times and voila - it clears up and runs well. So, I observed the cam solenoids live data when the car was running fine and they were indeed bad. All 4 were pegged. Intakes at -30, exhaust at +20. So I changed all 4 cam solenoids with genuine benz ones and this issue was resolved. All solenoids are now actuating and moving from about +4 to -30 as they should. No more codes on that. Misfire still happening. I left the car outside in -19c weather overnight and observed live temps before starting it the next day. Coolant temp was showing +9c, which clearly was false. So I put in a brand new benz coolant temp sensor at the back of the left cylinder head. (This must have been a redesigned part, as the old one used a different kind of gasket.) Anyways, coolant temp is now good, but the car STILL misfires the same way. Connected a fuel pressure gauge to the rail - pressure is 58psig as it should be. I decided to check the fuel pump relay and when I did, I found a surprise. Someone had run a jumper wire between the output of terminal 15r relay to the output of the fuel pump relay. Upon investigating, I found that I am missing a control side ground from the ecu on the fuel pump relay. So someone conviniently ran a wire from 15r relay next to the fuel pump one - to power the fuel pump. This is lucky, I guess, because terminal 15r is powered by terminal 30, which is always live. The fuel pump is also powered through terminal 30 when the ecu decides to energize it. I then decided to investigate the ecu and found out that it has been repaired ones. It's been opened, not resealed, grey glue on every chip's pins etc.
My question here is this. Can the ecu be bad causing this intermittent, yet consistent cold start misfire? Sometimes, it'll also do it warmed, but rarely.
It is pegging specifically cylinders 3 and 4 plus a random mis code. I checked the coils with a spare spark plug and they produce spark. Mind you, I couldn't know for sure if the symptom was exhibiting when I had the coils removed. The car obviously misfired but it could have just been due to the obvious - no coils on top of the plugs. I have not checked for injector pulse yet during misfire. It should be possible to backprobe those and use a small little 3W bulb. As of now I'm not sure what I'm missing during that misfire. My guess is injector pulse on cylinders 3 and 4, because when I pulled the sparks off after testing the coils, I was expecting them to be flooded, but they did not smell at all like fuel. Also, exhaust does not smell like gas during the misfire. I know how it smells when we're missing a spark and this is not like that.
Lambda is damn near a perfect 1 when the car runs normally. Maf reads very good at idle at around 4g/s. Idle air temp is very responsive to speed and reads very closely to ambient temp when driving at 60-70-80kph. All of these lead me to think a faulty ecu.
Do these car suffer from bad injectors? Has anyone else had this kind of insane problem? I'm leaning towards a bad ecu solder joints on the cyl 3 and 4 injector signals. And possibly a fuel pump relay control solder joint. Regarding the fuel pump relay, I also think it's the ecu, because if it would be a wire, the code for the fuel pump would happen all the time, but it does not happen all the time.
Please let me know what your experience with these cars is.
Last edited by Rumen; Dec 4, 2024 at 08:56 PM.




next I would do a compression test just to check that you don’t have a cylinder or head issue.


