After disconnecting and replacing the battery the check engine light came on. I had it diagnosed and it indicates it has misfires on 3 cylinders, but the engine runs fine. I had the light reset and it stayed off for a couple weeks. Then I started it and it ran rough and the check engine light came on again. I shut it off and restarted it and it ran fine. It was diagnosed again that it was misfiring on 3 cylinders. Had the light reset and it happened again. It appears to be something in the electronics since if you restart the car it runs fine even though the check engine diagnosis is that it is misfiring. Any thoughts on the cause????? would be most appreciated. This is getting very frustrating.

My wifes ML had a simialr series of events about a year ago and was covered by warrantee. The problem as I saw it was that the controllers software for adapting its missfire counter was malfunctioning or just inadaquate. The adaptation involves something called sensor ring adaptation. To understand, one must understand how the controller IDs a misfire.

It does this by watching the acceleration rate of the crankshaft. The highest acceleration is after each firing. Extraneous considerations are the driveshafts (as harmonics can decelerate the crank), harmonic balancer, flywheel, etc. The original ECU had 5 adaptation zones (engine speed versus load). The new one had 18 (newer software for the same year -1998) zones.

Before the ECU was replaced the sensor ring adaptation was redone and the problem went away for almost a month.

The misfire strategy is to shut off the fuel after a misfire has been IDed, so if once IDed the engine continues to misfire till the key is cycled.

Engine management ECUs are covered under a 8 year 80,000 mile emissions warrantee BTW.

__________________
Steve Brotherton
Continental Imports
Gainesville FL
Bosch Master, ASE Master, L1
33 years MB technician

I've seen many instances like this on several 112 equipped vehicles. In the cases where the ECU was the problem, the mis-fires resulted in cylinder shut-down, which is unmistakable. There are two types of mis-fire codes possible. A given number of mis-fires per 1000 revolutions will set the fault code, but will not shut the cylinder down. A given number of mis-fires per 200 revolutions will result in cylinder shut-down because it can damage the cat. You didn't say which cylinders were mis-firing. If they're all on the same bank, and no cylinder shut-off occurs, you may just have a bad O2 sensor before the cat on the side that's affected. I've seen it happen a few times, usually on vehicles with 40,000+ miles. Good luck and let me know how it goes.