M104 stumble

[benzfan]

The car is a 1996 c280. It exhibits the following symptoms;

-cold and hot starting are fine
-cold running is fine
-around town and highway running are fine
-high speed, throttle response, fuel mileage, kickdown, etc. are exhibiting no problems
-when hot, in gear only and stopped, it occasionally stumbles, like the fuel has been shut off for a fraction of a second, then restarted. It recovers to an RPM a couple hundred RPM over normal idle then quickly settles to where it belongs.
-This has gotten worse over the last while, and has not to my noticing started suddenly. In the summer, I would occasionally get a slight backfire after coasting into a corner and goosing it on the exit.
-the idle is smooth, but the tach wanders about half a needle width constantly
-occasionally when the aux. fans come on, the idle briefly drops considerably


The dealer has ruled out the OVP relay, the purge valve and the MAS by substituting new parts (no charge to me yet at all by the way)

Plugs, ends and O2 sensors were replaced 2 weeks ago as they needed work at that time. Coils are known good.

MB scan yielded only 1 fault code:
under TPM,
182 Safety fuel shut-off switch signal for HFM N3/4

The dealer now suspects the throttle actuator (ouch). Does it sound like they are on the right track? I found that if I 2-foot my driving and just rest my foot at a stop on the gas pedal, the stumble does not occur. For the weekend, I have taken a bit of the play out of the throttle by winding out the cable adjuster, and the stumble only occasionally occurs. I have not noticably increased the idle by doing this, just taken the play out until the dealer can get back under the hood Monday for some tests. My car does not have the same external idle microswitch that some of the older cars had, but what have I accomplished with the actuator by doing this? I have no intention of leaving the car like this, I just wanted to make it a touch less dangerous to drive this weekend.
TIA

I'll second the throttle actuator. My E320 with the 104 engine was doing the same thing before I changed my throttle actuator. The fuel safety shut off signal is triggered when the engine computer senses a different throttle plate position as compared to what is actually called for. There are 2 potentiometers in the actuator, one is the drive value and the other is the actual value. If they both don't agree then it trips a fuel safety shut off code. If this code is hard set it will limit RPM's to 1200. If it is an occasional trip then you get the symptoms your describing.

[suginami]

I'll third the throttle actuator.

A common failure part that exhibits the symptoms you describe.

[benzfan]

Well,
The dealer's senior tech poked, prodded and tested this morning. It appears I have a small vacuum leak around cylinder 6, either due to a cracked (plastic) intake manifold or a bad injector seal. Since the precise location can't be determined back there without disassembly, we are going to replace both the seals and the manifold next week when they can get hold of one. Total cost is far less than a throttle actuator.
The tech showed me that, in fact, the actuator is compensating properly, but under high vacuum conditions, apparently air is getting in along with the fluid he used to try to pinpoint the leak. From older vehicles I have dealt with that had vacuum leaks, I never would have assumed that from my symptoms. I guess it is because of the much more complex nature of the modern injection system controls.

[benzfan]

Followup:
The problem has been getting worse daily, particularly as the outside temperature dropped. I had the injector seals replaced yesterday, as I felt the injectors moved a bit too easily. We found one seal that had been pinched and wasn't sealing properly. That however didn't cure my stumble. I was set to have the manifold taken off Tuesday so I started playing around today. I got lucky. The car started running on 5 cylinders after lunch. It luckily did it again when I was at home with my tools out. I immediately started disconnecting injectors one at a time and found that the #2 cylinder wasn't firing. I checked to see if the injector was getting a signal. It was. Now the problem looked like either a clogged injector or ignition. I pulled the front coil, and checked the connector to #2 for proper resistance. It was fine. I put a wrench to the plug and lo and behold, the darn thing was loose with evidence of combustion gases on the plug connector and outside of the plug. I cleaned it up a bit and put it back in with an approximation of the correct torque and buttoned everything back up.
Problem solved. Not a hint of misfire or stumble. Whoever changed the plugs a few weeks ago didn't tighten that one and it has been slowly working looser to the point of not firing properly. On Tues. I am having the whole ignition system reworked and checked.

[benzfan]

An interesting postscript to this. I changed out my plug boots last weekend and the problem immediately resurfaced. Nothing I could do seemed to lessen it, and the only thing I disturbed were the coils. I started back with the ususal suspects, starting with the plugs. All previous tests indicated the plugs were fine, but lo and behold, a plug change cured everything. It appears one of the last set was bad from the start. Perhaps dropped by someone? I know which one it was and I can't for the life of me detect anything wrong, but the car sure can.
I'm just glad it was cheap.

[glenmore]

This proves an old adage that I discovered with regard to my rudimentary wrench skills:

First, check the last thing you screwed around with.

glenmore
1991 300CE