Here it goes:

1987 300TD
Intermittent problem with surging RPMs.

Every so often, it just goes astray and I can pretty much keep it under control by jiggling the wire harness that comes across valve cover.

Obviously one or more of the wires is bad.

Would prefer to have some idea which one(s), before having to trace each and every one.

Any thoughts on that?
Appreciate any comments/suggestions.

Is your surge a stutter (drop) or a rise in RPM, what does tach say?

Is it steady (jumps from one level to another and stays there) or is it jumpy (RPMs are sputtery in the new range)?

How long does a surge last if you touch nothing?

Do you have a functioning engine idle control knob to raise idle up/down? (try it).

__________________
Cheers!
Scott McPhee

1987 300D

It is definitively revving up. It last a few seconds and than may repeat again and again but not at set intervals- just out of the blue.

Do you have a functioning engine idle control knob to raise idle up/down? (try it).

__________________
Cheers!
Scott McPhee

1987 300D

Well, the knob is not doing much to the idle level- I tried different settings and all sound about the same. Again, when it started to rev up at random I just wiggled the harness and it smoothed out right away. Any adeas?

Maybe your engine idle speed control circuit (that knob's wiring) has some poor connections, and the computer can't get a read on the setting. Or, the engine speed sensor is not picking up right. This is a closed loop feedback system, so the computer tries to compensate with speed adjustments until it can read the speed is right. If anything in the loop is wonky, it can't do its job, always hunting. But if you disconnect the idle speed actuator on the rear of the IP, you can tell if this is the problem or not. The car should drop to its lowest possible idle, and stay there. It might even stall if the engine is not fully warmed up. I forget what the number is, but it seems to me like about 550 RPM is the rock bottom.

__________________
Cheers!
Scott McPhee

1987 300D

Yes, I have disconnected the circuit at the back of IP on many occasions and that is exactly what happens: idle drops and stops being erratic. I will try to follow the wire I guess. Thanks for all your input.

So if the disconnecting of the idle speed actuator gets rid of the surging, is it the actuator or the wiring?

If I recall correctly, there was a member on another forum that had the same idle issue.

As it so happened, one of the members has a pump shop over seas, (not sure where but his screen name is Dieselmeken), and he said that the electronic controlled M pumps are susceptible to the Delivery Valves wearing out which results in more fuel in injected at a given rack setting than the computer is calibrated for.

So what is happening is, the DV allows more (unregulated) fuel to be injected causing the engine to rev up, the computer sees this and yanks the rack back causing the engine speed to drop, and then returns the rack to where it should be for the correct idle speed, but the DVs are still causing more fuel to be delivered and the cycle starts all over again.

When you unplug the pump from the computer the idle stabilizes out because the computer is not trying to control the idle speed with DVs that cause delivery rates outside of the computers calibration.

He said replace the DVs and all will be good.

thanks om616- how hard is it to replace the wire harness as I see it as a culprit given the effect that jiggling of the wires leads to normalization of the idle rate?

Well....... And this may sound rude so I apologize in advance.... by asking that question you really should have someone who knows what they are doing trouble shoot the harness to identify the fault. It could be as simple as the metal sleeves in the connector have opened up and doesn’t "grab" the male connector all the time, I am fighting with an oil pressure sensor connector that does just that.

Also, lets take a moment and think about what is going on in an effort to connect engine behavior to events.

If the engine idle starts searching or hunting in a consistent sicklickel pattern, and when IP is unplugged the idle stabilizes, then I look less towards the electrical system and more at the IP. Now it is possible to have a connection that makes when the engine rocks to the right, and brakes when it rocks to the left, creating the repetitive RPM cycles, but, not likely.

If the engine Idle is erratic, every now and then, I would lean toward electrical.

It is possible that you have two separate issues, one mechanical (possibly the DVs) and one electrical (possibly a loose connector or broken wire).

I do agree with you in eliminating the electrical issue first. You will need a wiring diagram, a VOM, and a lot of patience.