Mercedes 300CE rough idle

@ wesw

“…and I know nothing of Mercedes …”
Mercedes puts it’s pants on the same way as you – or as Germans say: “They also use water for cooking”.

fuel screen:
When I tested the fuel pressure 3 weeks ago, I also disconnected the fuel return pipe from the pressure regulator, to see whether it could be blocked somewhere. For that I had to connect a hose to the regulator and hold it into a glas jar. With the fuel pump running the flow was as strong as the flow from an open water tap. And the pressure was perfect.
So, I’m positive, clogged screens are not an issue in this case.

sensors:
I checked all involved sensors and switches: intake air temperature sensor, both coolant temperature sensors, air flow sensor (potentiometer), throttle valve switch, throttle linkage switch, automatic transmission shifter switch, crankshaft position sensor.
They’re all in order.
But I’m not 100 % sure about the crankshaft position sensor, though it’s voltage and resistance are correct.

oh well, this blind squirrels going hungry tonight.

@‌HD

Remember that link I sent you some weeks ago?

Under group 49, programmed repairs Combustion 1, there’s something interesting

It’s titled “Engine vibrates when idling” and it looks like something you could easily check into

@ db4690

Thank you db4690 for taking the time to check that link so thoroughly!

Besides renewing the engine mounts, which I did, there are 3 things to be checked according to this instruction:

  1. “adjust idle speed”
    It refers mainly to the ignition timing (a) and gas mixture (b) adjustments at idle.

a) I checked the ignition timing a few months ago - it’s perfect:

  • at 3200 / min without vacuum hose connected: 29° (target: 27°-31°)
  • at 3200 / min with vacuum hose connected: 42° (target: 40°-44°)
  • at idle with and without vacuum hose connected: 8,5° (target: 6°-11°)
    (all values: before TDC)

b) I adjusted the gas mixture after the renewal of the injectors 10 days ago.
You, as an ex-Mercedes-insider certainly know that, but for those who don’t:
The basis for the mixture adjustment is a square-wave voltage which is generated by the CIS-ECU (based on the input from the o2-sensor, at which’s frequency it fluctuates around a center value). It can be measured as dwell angle or duty cycle at the car’s diagnostic socket. I use an analog Bosch duty cycle meter (see picture from several weeks ago), which displays the duty cycle in “percent” with a fluctuation of +/- several percent points.
I tried a digital meter - that was not really practical with the constant fluctuation.
Now, the instruction in that Mercedes link says that for my engine the duty cycle at idle should be in the area of 35-45%, which means that the mixture is rich and the o2-sensor fluctuating around a voltage above 450 mV.
The opinions about the mixture adjustment are divided.
Some Mercedes people I talked to say according to their experience a duty cycle around 50% is better, because than the o2-sensor works around it’s optimal voltage of 450 mV and the mileage is better.
Others say that it’s better to go as low as 35%, although that means quite a rich mixture with a little lower mileage, but it’s better for the idle quality, and above idle the mixture would become leaner again by respective input from the ECU to the EHA.
Before I renewed the injectors I adjusted the mixture down to a center value of 38% at idle on a trial basis. And it’s true: the duty cycle at 2500 rpm was a lot higher then (46%), which means leaner again. But idle quality wasn’t better after that.
After renewing the injectors I adjusted the mixture to about 50% +/- 1.5%.
Maybe with the new injectors I should try that rich idle mixture adjustment (low duty cycle) again!?

  1. “perform fuel quantity comparison measurement”
    I didn’t finish to build the measuring devise yet which I want to connect to the fuel distributor for that purpose.

  2. “test camshaft timing”
    That’s one of my suspects, which I want to check when I renew the lifters, the valve stem seals and the rubber seal for the camshaft sprocket cover, since then the camshaft drive will be accessible.
    However for that purpose you need a rocker arm equiped with an adjustment srew. I will either have to borrow one or prepare an old rocker arm myself. The best would be if I can find a used camshaft bearing cap with rockerarms, so that I would just have to tack-weld a hex nut for the adjustment screw on top of the intake rockerarm.
    Btw: in this link it reads group “05-215”, it should be “05-2150”.

@‌HD

Unfortunately, you’re looking at “irregular idling”

That’s not what I meant. Please look at my message again. All the other links will cause you to waste more time.

You were in the right section, but looking on page 1. Try page 19. There you will find a brief article titled “Engine vibrates when idling”

@ db4690

Sorry db4690 - you’re right! Thank you for reminding me!

I will check that cable which goes from the ECU to coupling X26 with my Ohmmeter tomorrow.

The next suggestion “untie cable at X26 and connect to ground”, which increases idle speed in “D” by 50 rpm, I could also try tomorrow.

I tried something similar once: in “D” I pushed the brake pedal with my left foot, and the gas pedal a little with my right foot, so that the engine speed was a little higher. The engine was less shaky then.
I know - that can not really be compared with the described procedure, because after moving the gas pedal the switch of the throttle linkage and the throttle valve switch are already released with respective input to the ECU.

Okay – it’s really worth a test. I think I could live with a slightly higher idle speed, if it would not be shaky anymore then.
I hope the car can still hold it’s uphill position in “D” without moving forward then.

@H.D.

This was one of those things that was in the far recesses of my mind

There’s another thing that I remembered . . . sometimes the o-ring for the fuel distributor goes flat, causing a rough idle. It is a bear to find that leak, and of course the fuel distributor would have to be removed, if that was the case.

@ db4690

Do you mean the O-ring between the distributor housing and the air flow sensor housing, which if damaged can cause false air?
Or do you mean the axial sealing ring, on which the control plunger sits after the engine is switched off and which if damaged causes fuel to leak into the intake manifold ?
I know both problems are not very easy to find, and for both the distributor would have to be removed.

@H.D.

The first one

@ db4690

That one is probably only testable with a smoke machine. I tried to reach it with brake cleaner fluid, but there’s probably too much friction for the fluid in the tiny gap between the distributor and the air flow sensor housing to have an effect on idle speed.

I also wonder about the condition of the axial sealing ring of the control plunger in the distributor.
Did you experience that one often causing problems during your Mercedes time?

I found one of those flat o-rings once, with carb cleaner

But it wasn’t easy

One guy was saying “Yeah, the idle’s a lot better”

Another guy was saying “No, nothing’s changed”

As for the second o-ring, not that I can recall

But your car is 26 years old . . . and the dealership mechanics don’t often work on 26 year old cars

Yeah, that’s a problem with keeping cars like that on the road – there’s only a limited source of experience with and knowledge about cars of that age.
A problem which many W124 model owners seem to have. Up to 5 years ago you could still see many of these cars on the road here in Germany, but since then they have become quite rare.
However among the residual W124 models the percentage of those in good condition has risen since then.

did the disappearance coincide with the rise of scrap metal prices?

I face that problem with my truck. they were everywhere but now they have all been crushed for scrap. I have switched to jeep Cherokee now. they are fairly reliable, simple and plentiful.

I like to find a car I like and stick with it.

just when I had my truck and carbs figured out, I have to learn about fuel inj. and jeeps

@ wesw

“…did the disappearance coincide with the rise of scrap metal prices? …”

As a mater of fact, something related to that indeed played a part in the disappearance of older cars. In 2009 the German government – as an answer to the financial and economic crisis – came up with the so-called “scrappage bonus” limited to one year.
Everybody who had a car scrapped and registered a new car on his name received Euro 2.500 ($ 3.400) from the tax office.
And guess what! I saw cars disappear from my neighborhood due to that arrangement which could have easily been sold for twice as much on the second-hand car market.

But the disappearance of the Mercedes W124 model happened more steadily over the past five years.

hey! we had cash for clunkers too! small world.

@ db4690

I followed the instructions under the title “Engine vibrates when idling” in that internet service manual today.

First I tested the cable between the ECU and connector X26 (in the fuse box) with my Ohmmeter: It’s in order.

Then I suppressed the recognition of the selector lever position, by untieing that cable from connector X26 and connecting it to ground, which is supposed to rise idle speed in “D” or any other drive position by about 50 rpm.

In reality idle was about 125 rpm higher, i.e. 675 rpm (more or less the same as in “P” or “N”).
And idle was indeed less shaky, but not completely gone.

I think it’s not worth the disadvantages, which are (starting with the most annoying):

  • more drag in slow overrun
  • car moves forward after releasing the brake in “D” in an uphill position
  • engine is a little louder while standing in “D”
  • more brake wear
  • lower milage (at least in urban traffic)

So I connected that cable back to connector X26.

But it was worth the test.

“Immediately after starting the cold engine it’s a “tap”, but not each time.
When I go slightly uphill after starting the cold engine (higher load) it’s more a “clack”, but not each time either.
When the engine is warm and you keep your ear close to it it’s a “tap”, all the time.”

That “clack” under load when cold could be piston slap. Do you know if this car was run at high speed much before you got it? (I assume that would not be unusual in Germany.)

There is that word I was trying to keep out of my mind: “piston slap”.

It was there this morning again, when immediately after starting the cold engine I drove slightly uphill: “clack, clack, clack, clack…” - but after about 150 - 200 yards it disappeared.

Two days ago, same situation: no clacking.

My “wishful” thinking tells me, it’s a failing lifter of an outlet valve.

I don’t know if this car was run at high speed much before I got it.
I’ve had it since January 2000 (60.000 miles ago).
The first time I heard this “clack” (less then now) in the same situation, was less than 10.000 miles ago.

But, wouldn’t piston slap show lower compression?
With the engine cold the lowest compression is only 4.7 % lower than the highest compression.
With the engine warm it’s 5.2 %.

Good compression is certainly to your advantage, and anyway, piston slap shouldn’t cause a shaky idle. My nephew’s Civic developed a cold-start clack like yours and his mechanic called it piston slap. That’s the only car in which I’ve ever heard it. He put many thousands of miles on it since then and it hasn’t gotten any worse. Any schedule for your lifter and valve seal work?

@ insightful

That cold-start clack in your nephew’s Civic: does it disappear after a few hundred yards too?

Regarding a schedule for my lifter and valve seal work:
I haven’t mentioned yet, but when I had my head under the hood close to the fuse box today - for the test which db4690 suggested yesterday - I saw a wet spot of half an inch in diameter on the cylinder block at the rear left corner. It’s coolant!

Since the installation of the new radiator 6 weeks ago I haven’t refilled any coolant, but I saw today, that the level in the reservoir has dropped about a quarter of an inch since then.

So, I guess I should get used to the thought, that I have to take the head off after all.