Bad camshaft?

[ckamila]

I'm in the process of diagnosing a rough idle on my 300D Turbo.

From 700 - 1400 the car has a shimmy/shake that increases as rpm's go up from idle untill about 1200, then decreases until about 1400 where the engine is smooth. So far have done:

Valves (almost all were tight)
Diesel purge
Cleaned & Balanced injectors (pressure was 140,125,130,130,130)
Replaced both motor front mounts
Replaced 1 engine shock
Timed IP at 24 BTDC
Cracked injectors lines at injectors - similar change at all 5 cyl
Tank screen and both fuel filters changed
Checked chain stretch (simple method via cam and crank marks) +2 deg ATDC
Compression 375 or better
Vacuum leak at fuel security latch - fixed the door lock issue
ALDA - hacked and already fully turned CCW - going CW 1/2 turn made performance worse


When i bought the car (my first MB) the idle was set high, which i did not notice but which was pointer out to me by Abe at MBMW Parts here in Seattle. When we tried dropping the idle from 900+ down to 700+ the engine shaking would start which led to the list of fixes and maintenance issue addressed above. But i still have the shakes and what bothers me more, poor mileage (under 20).

Attempted the 2mm timing chain stretch test but could not get the dial gauge to be still, so i lined up to cam and crank marks as listed above - not the greasiest but a good indicator of general health. While there i started looking at the cam too which i found to be pitted on at least 2 lobes - could this be the problem, a poorly ground cam? Engine has 215k, with a rebuild about 70k ago.

[winmutt]

Oh pitting is not a poorly ground issue, that is a time to replace issue. They are hardened at the factory. Just looked at the cam, that is really bad.

[ckamila]

Another thing i noticed when doing the valves, if i slipped the feeler gauge in the center of lobe id get one reading (good), and if at the edge of the lobe feeler gauge would stick - it looks/seems the edge of the cam that sits passed the rocker arm pad/foot has not worn causing a lip to form, making adjusting a little trying.

Also have replace the old IP rack screw with new gold one - no change with adjustment.

I took the car to 2 different places for an opinion and both had different ideas:

1st place - replace IP and retime $$$
2ns place - replace timing chain, Replace ALDA, retime IP $$

Anyone else what to chime in???

[barry123400]

This is the kind of situation that reading the glow plugs output voltage properly should give some answers about the individual cylinder contribution to the overall running of the engine or power balance. You need the firing order in sequential balance for the engine to run in a reasonable manner.
Might even prove the pump is bad by indicating extreme differences in individual cylinder timing from the pump. It displays this by indication of temperatures present in the cylinders while runnng.
You should read up on the subject and get a good grasp of what you are doing. It is possible someone in the past adjusted components on the pump they should not have touched. Or perhaps general aging is involved. If you are really lucky they only messed up #1 injection timing. But if someone retimed the pump later it could be way off for the other four cylinders for example.
The milli volt method will clue you in if properly used. There is a danger that you have one cylinder running too hot already and that can damage a piston at some point. You know your injectors are reasonable and compression seems pretty good so should be worth the attempt. Put another way you can do this yourself and have nothing to loose.
I and others can guide you along the path on site if you desire. It is what I would do rather than throw money at it and just continue to guess what might be wrong. I cannot see your cam to see if it is too far gone or not. Sure it could be the problem but if so the mili volt method will even indicate where the effect of the problem is taking place. Opinions are just opinions. Proof of fault is better.

[ckamila]

Thanks Barry for your insight. I will search as you suggest on the mili volt method - never really paid much attention to it skimming past posts.

Thanks again,
chris