Mercedes 300CE rough idle

There have been 3 problems with my car (300CE, late 1988, 120150 miles, automatic transmission) for over a year now:

No. 1 (rough idle):
Idle is always rough and shaky – more in “D”, less in “P”- and a lot more when the engine is hot.
The car shakes about once per second (awkwardly noticable by vibrations through the seat) without any noticeable change in rpm. When the engine is still cold (idle speed 20-30% higher) it shakes a lot less.
A bridging wire in place of the fuel pump relay or the overvoltage protection relay brings no change!

No. 2 (engine stalls at idle):
Sometimes the engine just stops and doesn’t want to start again for 5 to 10 minutes. It happened only 5 or 6 times in over a year now and only at or close to idle speed – once it happened rolling towards a red traffic light at very low speed (in overrun mode).

No. 3 (warm engine start):
When the engine is switched off for 15 or 30 minutes and started again, it always needs to be started twice. First it immediately starts, runs for less then 1 second and stops again. Then when you start it the second time it stays on. When the car has not been used for a 1 week or longer it starts perfectly within less then a second and stays on.

Apart from that the engine runs and accelerates very smoothly up from idle speed.

  • Injectors are still the original ones (26 years old). All of them still have a really nice spray cone at rpm’s above
    idle, but at idle cyl. 3 and 6 have a stringy spray pattern. They don’t seem to leak after engine stop.
  • All spark plugs look homogenuous, a little leaning to bright (9 months and 4.500 km old).
  • Ignition cables are new (all 7).
  • Distributor cap and rotor seem to look ok (5 years and 28.000 km old).
  • Engine mounts are from aftermarket (probably not hydraulic version, 9 years / 38.000 miles old).
  • False air can be ruled out.
  • Engine compression: between 185.6 and 195.8 psi on all cylinders (after 10 cranks each).
  • Ignition timing: nicely in the middle of tolerance range (at idle + 3.200 rpm with + without vacuum).
  • Gasoline pressure within tolerance range (control pressure, primary pressure, pressure after engine stop (also
    after 30 minutes).

Looking forward for any suggestion which helps to narrow down the reason(s) for these problems!

“but at idle cyl. 3 and 6 have a stringy spray pattern.”

That’s my guess.

“False air can be ruled out.”

An evap/smoke machine is used to “rule out false air”

Anything else is less accurate

No offense intended to anybody and different methods

Have you used a smoke machine to rule out the false air?

It definitely sounds like you have a fuel problem

Here are some of my suspects, based on my experience with CIS cars

Fuel injectors . . . tend to leak down, cause hard starting, have a bad spray pattern

Fuel pump relay . . . intermittently just quit, then work again for awhile. Only genuine Benz relays are okay, in my opinion . . . based on experience

idle air control valve . . . again, only the genuine Benz parts are any good. Seen brand German aftermarket parts that failed within weeks

Accumulator . . . leak down and cause hard starting

Fuel pressure regulator . . . leak down

Coil . . . tend to fail when hot

What brand plugs and wires are you using? No offense, but even some of the German aftermarket ignition components are junk. Seen it too many times

Honestly, I’m surprised you’ve chosen to hang onto the original injectors. Most CIS cars are on their seconds set . . . at least . . . by now.

CIS cars are extremely intolerant of false air . . .

I’ve worked with CIS a lot and agree with db4690 completely. Air leaks and lousy injector spray patterns are a curse.

While I don’t know if this is common with Benzes, I’ve seen older CIS SAABs have a slightly rough idle simply because whoever checked the engine oil level did not fully seat the oil dipstick in the tube. Not saying this is the cause here; only that CIS can be that finicky.

First I want to apologise, if my reaction on comments is a little delayed. I live in Germany, and when you post a comment - let’s say in late afternoon - I may already have gone to sleep.
Next I want to thank you for your comments, which sound very experienced.

@taxases:

Maybe the injectors are the reason for all of the 3 above mentioned problems.

@db4690:

“…Have you used a smoke machine to rule out the false air? …”
I must admit, I can only rule out false air according to the test method I used. I did not use a smoke machine.
I took off the air filter housing, let the engine run in idle and - equipped with a vaporizer filled with brake cleaner fluid – I thoroughly checked every corner of the hole intake system. Small hoses which are connected to it which use vacuum like the lock of the backrest of the passenger seat, I took off and used my finger to plug the hole. It certainly took me half an hour to do the false air test. The use of a smoke machine is probably more proper for a false air test.

“…Fuel pump relay . . . intermittently just quit, then work again for awhile….”
I can not rule out, that problem no. 2 is caused by the fuel pump relay.
A month ago I connected a small 12 volt lamp to the wire which supplies the fuel pump with voltage. I placed it on the center console (with a switch). So next time when the engine stalls I can check whether the fuel pump relay shows voltage output when I switch on the ignition again (it’s supposed to do that for 1 or 2 seconds).
But so far it never happened again yet.
As for the rough idle the fuel pump relay can be ruled out, because there is no change if I pull it off and use a bridging wire instead, so that the fuel pump runs as long as ignition is switched on.

“…only the genuine Benz parts are any good…”
I agree.

“…Accumulator . . . leak down and cause hard starting…”
Ok, a leaking accumulator can of course cause starting problems. But wouldn’t it do that then all the more after the car has not been used for a week or two!? But as I said, after a longer shutdown time the engine starts immediately within a second – and stays on!

“…Fuel pressure regulator . . . leak down…”
Since the fuel pressure test showed flawless results (also 30 minutes after engine shutdown) I think that fuel pressure regulator is also not a suspect anymore.

“Coil … tend to fail when hot”.
I should check that.

The plugs and wires I installed 2.800 miles ago are Bosch.
But here is a confession: The distributor cap and rotor are not Bosch, they’re “Bremi” (one of the German aftermarket ignition components supplier).
I attached pictures of the current plugs, cap and rotor.

I bought new injectors already last week, but haven’t installed them jet. They’re Bosch (with the original Bosch-number and “Made in Germany” engraved on them.
I got them from an aftermarket dealership which offers them for 27 Euro ($ 37,50 acc the current exchange rate). That is half of the regular price at the local Bosch or Mercedes dealer.
The seller accepted that I connect them one by one to a test pipe which I connected to the fuel distributor (injector aiming into a big plastic bottle. (I did that outdoors with the wind blowing any escaping fuel vapor away from the engine – but to all readers: better don’t do that, pulverized fuel is highly explosive!). The seller said I could bring them back, if they don’t spray properly – as long as they haven’t been used in the engine yet.
The spray pattern looks ok, but what puzzles me is that when you release the throttle linkage at higher rpm and the rpm drops back to idle the injectors drip a little. I don’t know – is that normal? Or should I bring them back?
Back at idle they spray ok and atomize the fuel again, but not as nicely as at higher rpm.

@ok4450:

“…slightly rough idle simply because whoever checked the engine oil level did not fully seat the oil dipstick in the tube …”
Never thought about that, but come to do that I think that could be an issue especially for CIS cars with injectors which have air ducts around them which are connected to the valve cover via breather hose. That way of course air can enter the crankcase and cause false air to the intake system. The engine of my Mercedes (M103) has that design.

I checked the pressure under the valve cover of my car this morning. I let the engine run at idle and took off the breather hose between the valve cover and the air filter housing. Then I plugged the breather neck on the valve cover with my finger. Result: There is a slight vacuum under the valve cover! Obviously that is caused by the intake vacuum through above mentioned breather hose between the valve cover and the intake manifold.
I only checked that at idle, I don’t know whether there is vacuum or positive pressure at higher rpm!
Anyway, that test better be done carefully, since air (possibly mixed with dirt particals) could enter the engine through some other spot (maybe through a rotary shaft seal) due to excessive vacuum!

Next week I will also install new motor mounts and a new transmission mount. I ordered them yesterday at the local Mercedes dealer (price incl. tax: Euro 260 / $ 360).

Warm gretings from Aachen, Germany

“…better don’t do that, pulverized fuel is highly explosive!”

I think you mean, “…better not do that; vaporized fuel…” (No offense, but a German I worked with implored me to correct his English so he could learn.)

Back to your problem:

“…what puzzles me is that when you release the throttle linkage at higher rpm and the rpm drops back to idle the injectors drip a little. I don’t know – is that normal? Or should I bring them back?
Back at idle they spray ok and atomize the fuel again, but not as nicely as at higher rpm.”

If your old injectors don’t exhibit this drip, then I think there’s a reason these are half price.

(Also, you usually "bring"something along with you, but “take” something back to where it came from.)

@insightful:

Of course, you’re right! Thank you for correcting my English! Please don’t hesitate to do so, if you see any mistaces!

My old injectors also exhibit this drip, when the engine slows down towards idle with the throttle linkage released. I just wonder, whether that’s normal or not!?

@insightful:

Sorry! I mean “mistakes” (not “mistaces”).

Since you appear to have an excess fuel problem, those little drips bother me (as they obviously bother you). If you plan on driving this car another 120K, I suggest you pay the extra for the factory injectors to take them out of the equation. You may still have issues afterward, but at least you can reasonably look elsewhere for solutions.

Have you asked at a Mercedes forum? I would think there would be folks there that know these cars inside and out.

@‌HD

You certainly seem to pretty thorough, based on your second posting

Now here’s an interesting question . . .

You said the fuel pressure holds when the engine is shut off.

Okay. Good.

But is the pressure correct at idle?

It’s not too high?

I believe there’s a vacuum hose attached to the regulator . . . is it getting good vacuum at idle?

Have you verified that the engine coolant temperature sensor is accurate?

I believe you have 2 sensors . . . one for the cluster and one for the engine control module. Don’t bother checking the one for the cluster

@db4690:

You’re obviously very experienced, judging from your postings.

“…But is the pressure correct at idle?…”

I carried out all issues of the fuel pressure test procedure which Mercedes specifies in their workshop manual:
Here is what I tested in detail with results (target range in brackets):

  1. control pressure: 79.32 psi (76.8 – 80 psi)
  2. primary pressure: 73.23 psi (4.3 – 6.5 below control pressure)
  3. pressure after fuel pump is switched off: 42.05 psi (> 40.6 psi)
  4. 30 minutes after fuel pump is switched off: 36.54 psi (> 36.2 psi)
  5. primary pressure at overrun cut-off (engine temperatur: 185°F): 78.30 psi (close to before tested control pressure)
  6. primary pressure at accelaration load enrichment (engine temperature: 68°F, simulated with a 2.5 kOhm test resistor in place of the coolant temperature sensor): 66.70 psi (dropping, but not below 55 psi)

Point 1 to 4 are done with only the fuel pump (not the engine) running.
For that I pulled the fuel pump relay and connected a jumper cable with a cord switch I prepared for the pressure tests (long enough to a actuate it from wherever I’m standing around the engine compartment).
So, I can’t say what neither the control pressure nor the primary pressure is at idle. Plus, Mercedes does not specify any data regarding the pressure at idle.

“…I believe there’s a vacuum hose attached to the regulator . . . is it getting good vacuum at idle?…”

Yes, there is a thin vacuum hose attached to the pressure regulator, but that one only sucks off fuel which leaks through the membrane of the regulator when it’s damaged. The other end of this hose is connected to the crankcase breather hose between the valve cover and the air filter housing. It’s clean.

I also checked the vacuum hose for the ECU that drives the ignition. It’s connected to the throttle valve housing, clean and obciously working well (see initially mentioned ignition timing data). Of course a throttle valve which is crusty and smudgy inside could cause problems with the vacuum input to the ECU.

”… Have you verified that the engine coolant temperature sensor is accurate?…”

I checked the coolant temperature sensor at 2 temperatures (at 68°F and 185°F) and it showed correct resistance (2.500 Ohm at 68°F and 300 Ohm at 185°F).

”… I believe you have 2 sensors . . . one for the cluster and one for the engine control module. Don’t bother checking the one for the cluster …”

Yes, there is a double temperature sensor for the engine coolant with 2 cables plugged to it - one connected to the ignition trigger box, which affects the ignition timing and one connected to the CIS-ECU, where it’s input is processed for control signals to the electrohydraulic actuator, the idle air control valve and - via fuel pump relay - also the cold start valve.

Another temperature sensor is attached to the intake air duct of the air filter. In my car that one is only connected to the CIS-ECU (up from late 1989 also to the ECU for the ignition).

Yes, there is a thin vacuum hose attached to the pressure regulator, but that one only sucks off fuel which leaks through the membrane of the regulator when it’s damaged
I do not think this hose is to drain off leakage, there should be NO fuel in this hose! If there is fuel in this hose the "membrane"(diaphragm) is damaged and the regulator needs to be replaced. This hose supply's vacuum only to the "membrane"( diaphragm), and controls fuel pressure.

@PvtPublic:

“…I do not think this hose is to drain off leakage, there should be NO fuel in this hose! …”

I’m sorry to correct you, but that’s exactly the purpose of this hose thin plastic hose: lead fuel which passes through the damaged diaphragm in the pressure regulator to the crankcase breather hose where it would be sucked of by the intake vacuum. Mercedes installed this to the pressure regulator in order to prevent fuel from just spilling out into the environment when the diaphragm of the regulator becomes leaky.
You can take this tiny hose off, it doesn’t have any effect (at least not any noticable effect) on the function of the pressure regulator.

Look at the attached drawing of the regulator, then you will see!
This small hose (it shouldn’t even be called “vacuum” hose) is connected to a rubber fitting which is plugged into the regulator from the bottom (on this drawing).

Plus – think about the low level of vacuum in the crankcase breather hose! When you take this tiny hose off the breather hose and block the 0.15 inch hole in the breather hose with your finger, you can hardly feel any vacuum.

Ok I guess the MB regulator operates differently than what I am used to. How is he fuel pressure regulated?

@PvtPublic:

Okay – let’s try! But it is a bit of a headscratcher.

First the description of the connections:

“1”:
Connection to the lower chambers (one for each cylinder) of the differential pressure valves in the fuel distributor. The fuel coming via this connection has a slightly lower pressure than the “primary pressure” in the lower chambers, because before it leaves the distributor, it passes the non adjustable restriction passage (diameter: 0.012 inch), which creates the purely hydrodyynamically predetermined pressure difference of 4.3 – 6.5 psi between the upper and the lower chambers of the distributor.

“2”:
Return pipe to the gas tank.

“6”:
Connection to the fuel distributor prior to the electrohydraulic actuator (EHA). Via this connection the “control pressure” of the distributor - the pressure which comes into the distributor directly (of course via fuel filter) from the fuel pump - is connected to the regulator.

With the running engine, fuel enters the dark colored chamber of the pressure regulator through (6).
With it’s pressure it pushes the diaphragm (8) against the load of spring (9) down (“down” due to the orientation of the drawing).
That plus the load of the “upper” spring (4) causes the valve body (10) with it’s valve plate at the top to follow and open connection (2) to the return pipe. That allows access fuel coming from the distributor via (1) to return back to the gas tank (remember: there is always more fuel delivered by the fuel pump than the engine needs, even at the highest engine load).
When the “control pressure” of the distributor coming via (6) rises further, the diaphragm (8) is further pushed down, but this time without valve body (10) following further. Why doesn’t it follow any further? It can’t, because it rests against it’s cone shaped block below it’s valve plate.
That way the ball valve between the diaphragm (8) and the valve body (10) opens and allows access “control pressure” to be released through the now open passage in valve body (10) and the still open return pipe connection (2).
Thus: when the engine is running, the pressure regulator releases access fuel and access pressure into the return pipe of the gas tank.

When the engine - and the fuel pump with it - is switched off, the pressure in the distributor drops via (1) and - due to the pressure in the dark colored chamber plus the load of the upper spring - the still open return pipe (2) down to the “shut down pressure”.
At that point the pressure in the dark colored chamber is in balance with the load of spring (9) which keeps the valve body with it’s valve plate and it’s seal (5) closed (i.e. the return pipe to the gas tank is blocked).
(As soon as the engine is switched off the upper and the lower chambers of the distributor are permanently connected to each other via the EHA, which is in it it’s open center rest position without input from the ECU.

I hope I didn’t make any mistake in this explanation under the pressure of my wife pushing me to finally come to dinner!

@‌HD

Based on your postings, you might be more qualified to fix your car than the local Benz dealer

I meant that as a compliment to you, not as a criticism of the local dealer

Okay – let’s try! But it is a bit of a headscratcher.

Head scratcher indeed. Because by the looks of it that membrane serves no purpose that a fixed solid partition would. But thank you all the same for the explanation, now go eat dinner!! LOL

@HD, it appears to me the membrane serves the same purpose as in most fuel pressure regulators, i.e., as pressure builds, it moves down to bleed excess (not “access”) fuel back to the tank. I’m used to seeing the bottom tube under significant vacuum, thus pulling the membrane down and reducing fuel pressure. Then, when the throttle is opened and vacuum is lost, the membrane moves quickly up, reduces fuel bleed back to the tank, and increases fuel pressure to the injectors to provide added fuel for a power boost (somewhat like the function of a carburetor accelerator pump). Diagnosis is the same, of course; if you pull the bottom tube off and there’s fuel in it, you need a new regulator. I’m pretty sure you know all this. I’m just trying to keep it straight in my mind.

Have you installed new injectors yet?

@db4690

Thank you very much for your compliment!
I appreciate that very much – especially when it comes from someone who is as knowledgeable and experienced as I believe you are.

I wanted to “tap” someone’s experience with symptoms like the ones my car displays, which can be caused by many reasons.
So three days ago I was googling for an expert on the Mercedes W124 model (built from 1984 to 1997) around my hometown here in Germany. Then I stumbled over this link “Car Talk”. And I thought: That sounds like the show on the car radio I like to listen to when I’m driving. There are these 2 funny and very bright hosts who make me laugh just by listening to their laughing already – very enjoyable plus very informative. This show runs on AFN, a radio station which I can receive here in Germany.
I couldn’t resist to click on that link, that way I landed on this forum.

Warm greetings from Aachen, Germany!