being a diesel novice and this one being my first and only diesel, I don't know all of the sounds or really exactly what a good 601.921 engine should sound like so I apologize if I'm mixing up terms, but I'll be as descriptive as possible.

My '84 190d 2.2 has had a very shaky idle since I've owned it and I've been trying to turn it into a well running daily driver. So far, I have rebuilt and balanced the injectors, replaced the lift pump and replaced the return valve (at the IP) This made the car drive smoother and gave me some extra power, but the rough idle has not been solved.

When listening to it today using a long screwdriver as a stethoscope I found that the only abnormal noise I could hear was from the injector on #4 (cylinder closest to the firewall). The only way I could describe it is a very distinct "pinging" sound that did not go away at any engine speed. Even though the engine smooths out 1200rpms or so, the noise remains at all revs. It also DOES NOT travel with the injector. After switching the injector for the one in #3, the pinging stayed in cylinder #4. I can only hear this with the screwdriver directly on the injector.

I have to believe that whatever is causing this is also causing my poor idle, however, when cracking the injectors, I can't discern much of a difference of impact on the engine between cylinders.

I'm looking for any ideas on what my next troubleshooting steps should be.

It should also be noted that it seems the overflow (fuel return) valve had been installed on the inlet side of the IP instead of the back (return) side possibly starving the IP for fuel... for an unknown amount of time. (at least 20k)

Thanks all for your ideas and input. Hopefully my description is alright... I'm learning every day :-)

With you having only four posts it is difficult to suggest the milli volt method in the archives. It is about the only way you are going to establish if that cylinder is running at a different temperature than the others though. It has to be done carefully with some consideration.

If a compression check shows basic equality with the cylinders first it it is where I would go. Sequential injection quantity of fuel or timing might be off with the injection pump on one injection pump element..

. If you have delivery valve seals I would replace that one cylinders seals if the milli volts indicated something was wrong. I suspect that cylinder may be running substantially hotter than the others. If so it is not a seal issue.

It should be proven before any course of action is attempted. This again takes a careful milli volt approach. There is nothing to lose by a proper approach as a cheap pick and pull injection pump could be substituted if the present one proves to be out of calibration and cannot be tweaked for one cylinder.

One cylinder tweaking usually works out okay once established properly that it is needed. Multipal cylinder tweaking is not practical .You do need a strong guide of where things are at. Thats where the milli volt method comes in. Not rocket science but consideration must be used when applying. It can and does work. If nothing else you can learn a lot about what is going on in that engine with it.

If you did not put the relief valve in the wrong place. A previous owner may have also done other things that are causing some of the present grief.

The relief valve was not improperly installed in the injection pump at a pump shop I would hope. If it was the pump was initially calibrated wrong. It would have still been sequentiall accurate though as long as the operational pressure was the same. Even at normal current presures it would not be this far off I would think.

That was a good find by you incidentally. The milli volt method is about the only sensible way I can think of in examining all this. There is something going on out there I do not totally understand. A few people have found that they did not have internal parts in their relief valves and I think there was one other oddball situation like you experienced more or less mentioned as well.

Now the hard part. When you get this sorted out I would drop the partial pan if there and plastigauge the number one rod bearing. Thats if you felt the injection pump had run at basically no fuel i pressure for a very long time. I see no reason that your newer engine would basically not be subject to the same issue as the 616. More a precaution than anything else. I only hope your car also has an acess panel on the oil pan like the 616 has.