Hello,
I have a 1996 SL500. About 5 or 6 years ago, it developed a problem – the engine was surging at idle and often it stalled at the red light. I have replaced the MAF sensor, a remedy I found on this forum, which fixed the problem until recent couple years. Now, I observe that there is a slight surge and RPM is bouncing up and down by about 50 at the idle again. This seems to be less pronounced when the engine temperature rises about 90 deg Celsius. In addition, there is an occasional but annoying drop of RPM by about 100 when driving at a constant speed at low engine load. This goes away as soon as I push on the accelerator pedal slightly. No error codes. I feel that I hear a slight puff sound coming from the exhaust pipe when the car is idling in garage, like if there was a misfire.

Couple more things, live data on CarSoft 7.4 shows fluctuation of Lambda values of one of the upstream O2 sensors, dropping down to -1%, and the other sensor showing 0% all the time: not sure if this is supposed to drop to negative value. Not even sure how to interpret the readings, I thought that Lambda = 1 is the air/fuel mixture at 14.7:1, then perhaps the % of Lambda given by the CarSoft is a % of change from 1, so the -1% would actually mean Lambda of 0.99. Both sensors read about 1200 mV, dropping down to 600 mV and back to 1200 mV, is this normal? The downstream O2 reports rock steady 900 mV. The throttle position is open at 0.6 degrees, jumping up to 0.9 degrees, corresponding to the RPM oscillation at idle.

Recently, I have changed the transmission fluid. Right after the change, the engine shacked violently when it was first fired up, and it is still shaking very slightly. It might have been there before but I have not noticed it

Any experiences with similar behavior? I am a bit puzzled at the moment and not sure if I should chase MAF, crankshaft position and O2 sensors, or start checking the transmission converter.

Thanks in advance for any feedback or suggestions.

 

Hallo,
Test the crankcase ventilation lines, they harden and break with age. Check whether the tank ventilation line is still connected to the valve on the driver's side behind the headlight. Test the air mass meter. Start the engine and blow into it, the engine has to react and stop.

 

Thanks for response. I tried to do a smoke test today, no leak from engine, and no leak from the ventilation tank. However, I found a smoke coming from underneath of the monovalve at the passenger side. There are few fat black vacuum lines. I could not pinpoint the exact location of the leak as it is quite crowded there. I also could not find and part numbers on those hoses, any idea what they are called, part numbers or diagram?
Just googled what those hoses are, they are supposed to be hearer lines. It seems a bit strange, why a smoke impelled into engine through vacuum line would come out through a heater line?

 

First thing---the 'fat lines' connected to the MONOVALVE AER heater-related lines--to/ from, each side of the heater.

I suggest you remove and clean the connector to the MASS_flow sensor. A flakey connection will cause wileckd drifting fluctuations of the idle speed and smoothness. Also, check ALL the vacuum hoses and especially connections and joints of any kind
I once had an 89-560 SEL It had the multi-port continuous-flow fuel injection.' which meant 64 connections (metal-plastic, plastic-plastic) that fed the idle air and fuel into the intake ports. Worked wonderfully as long as everything was leak-tight. It was an annual PM to replace everything. Parts were cheap and it took maybe an hour for me to do it. The air quality here in Houston at that time was horrible with high ozone levels that attacked nearly every piece of rubber on the car leading to the annual replacement of lots of parts.

 

I repeated the smoke test and found a crack on the vacuum block that is located beside the monovalve. The issue with the vacuum block is quie well described here: Misfire at and above idle M119 | Mercedes-Benz Owners' Forums (mercedesclub.org.uk. I have checked all other vacuum lines and connectors on the engine top while doing the smoke test, but I have not detected any other leaks. I sealed the crack on the vacuum block with epoxy, cleaned the MAF sensor and connectors. This has improved the engine performance a bit, the oscillation is less frequent and is about 10 RPM, barely noticeable. In spite of that, I still feel some engine shake or a misfire, so the problem is not solved yet. The engine and transmission mounts were replaced last year.
I am thinking that the problem is the torque converter, the engine shake is more pronounced when I let the car creep forward very slowly on red traffic light, while the brake pedal is not fully depressed. Any advice on how to check the torque converter?

 

Did you checked the fuel pressure regulator?