Ultimate Idle Question
 

OK...I think I have read just about every thread on rough idles and ideas on how to fix. However, I have noticed that the rough idle after engine gets warm on an M103, while engaged in Drive, and with break pedal on threads, have not had resolutions to the threads I have investigated (they just end and no post backs saying "Volia!"). There have been all sorts of "have you tried this?" and "have you checked that?", or "I would recommend looking at..." but ultimately the threads end with no closure. As everyone else has, I've replaced numerous parts and components on my 1991 300E but still have the same warm idle issue...when cold, no problem in Park or Gear at a stop (700+ RPM and smooth). When 80 c or higher, RPM drops to approx 500-550 and car shutters and struggles a bit with or without AC on...no diff.
So, has anyone out there actually resolved this issue in their old engine that can be shared? Seems there has to be a common solution as many. many Mercedes have had the same symptom.
Here is what I've done:
1. Replaced O2 Sensor (mileage improved)
2. Replaced Plugs (power improved)
3. Replaced Voltage Regulator (intermittent radio function and speaker output fixed)
4. Replaced Rotor (car starts much faster and stronger when cold)
5. Replaced OVP (No change in anything but needed to for confidence)
6. Cleaned out all Throttle components, ICV, linkages tight.
7. Performed Carb Cleaner spray test on all vacuum lines to no avail.
8. Replaced Fan and Clutch (clutch was frozen) but just works more efficiently now.
9. Adjusted Lambda Air Flow to no avail either way with load on engine and drive engaged.
10. Replaced brakes and car now stops on a dime.

Thanks!

To answer your question yes I have resolved many poor idle quality complaints in a 103 motor. The resolutions are varied and somewhat expensive. First question is how many miles are on the engine and fuel injection. Another consideration is the engine suspension/ motor mounts. If they are weak there will be vibration transfered into the chassis. The transmission mount is equally suspect. That being said the possibilities, that engine is prone to carbon deposits on the intake valves causing all sorts of trouble as they disrupt mixture flow as well as valve closure. They can be inspected with removal of the injectors and cups and insertion of a bend a light to visualize the valve stem. here should virtually no carbon at the valve tulip. Next are the injectors, they are constant flow injectors, not timed or sequential injectors so after time the quality of atomization deteriorates. They need to be tested when removed. The fuel distributor is really suspect, it delivers fuel to individual cylinders through slits cut with a laser, very very small so wear and change in diaphram function cause all types of problems most notably idle quality. The possibility of distributor cap and rotor as well as ignition resistors at the plug wire end are also deserving of attention. Start with Techron treatment for three tanks and see what changes.
Bill

As for the temp sensor I don't remember off hand, the idle air valve gets a max current of 1.0 a. I have seen idle air valve problems but mostly on the v-8 and never had an engine sensor cause idle problems. The control valves would carbon up and stick, a cleaning would usually resolve that.
Bill

I have the same thing on my V8. The idle would slowly drop, then shudder, and jump back up to 650 or so. Are you seeing it change like that, or is it a steady low RPM?

When I hooked up the duty cycle meter, I can see the display values slowly drop, then when the hiccup happens, the numbers suddenly jump back up and the cycle starts over again, about every 30-40 seconds.

Somebody on another thread said he fixed this by installing a new ECU. Apparently the old one was not responding quickly enough to inputs once the system went into closed loop operation (over 80 C). I have yet to bite that bullet.

Sorry guys...I have 170K on this car. It "feels" like a vacuum pressure issue that gets under a load and can't maintain. Here is what happens. Start the car in the morning....fires up pretty quickly for the most part. Idles fine, 750-800 RPM, engaged in gear or not. Drives fine as far as I know with decent power. When engine warms up to right under 80 c or over, when I come to a stop, the car drops down to about 550 RPM and sits there with a little shutter (manual says it should idle from 600-700 RPM). It idles steady but you can tell it just needs a little more not to shutter (50-75 RPMs). If I shift into park or neutral, it jumps back up to 700 RPM and is steady smooth on the idle. When I shift back into gear, it bogs down to 500-550 rpms and shutters. I sprayed Carb Cleaner around the injectors and no change so seals OK? I'm sure the injectors probably need cleaned or replaced and everything else that was mentioned above. If it's too expensive, I'll just live with it but was hoping there might be a fix that everyone seems to overlook. Hope that helps with analysis.

Thanks...Steve

Bill...where do you live? I'm in So Cal also. Maybe I can show you the car someday and have you feel it out???

Steve

Maybe its something as simple as raising the idle speed? As you said, 500rpm is too low. It could be just too low and the engine is trying hard to keep it from stalling.

Injectors are definitely suspect though...I had a real bad rhythmic surging problem, but no real power loss. After I replaced my injectors, the idle was still rough, but this rhythmic surging stopped. CIS can be a picky thing.

That's what I was thinking but the idle is controlled electronically in this car. Do you know of a way to mechanically bump it up a bit? I was playing with the throttle linkage but don't want to screw anything up. I know it can idle fine since it does in the morning when cold so why not when warm and what causes it to do this when it warms up? Also, I notice you live in So Cal and had a similar problem you were battling for a while. Did you ever get it solved in your vehicle? I'm thinking the computer is being fooled somehow to cause this low idle issue since it does this at a certain point. However, it can be a combination of things as well due to being old.

Stupid question: Did you adjust the duty cycle to the proper setting in neutral with a meter? Also, is the setting within tolerance in neutral at idle vs. at 2500 rpm? Mark

You just went over my head :) I have no idea what a duty cycle is so not a dumb question. All I know is it idles perfectly until warm and in gear. Idles perfect warm and out of gear. Is there a way I can check that with laymen's tools? I'm not a mechanic but very technically inclined. Thanks!

If the value is fluctuating then the adjusted value is within the cars ability to compensate so the actual adjustment will only effect off idle acceleration. The ECU will be compensating either towards rich or lean if the meter indicates fluctuation in this case
Bill

UPDATE....well today was a cool day outside (different from the recent 100 degree plus days we've had out here for a while) and although my car's engine temp still moved up to 80 c, the idle at a stop, while in gear, idled like it does before it warms up in the morning...smooth. This after plenty of freeway driving (45 miles plus). Does this ring a bell or help to narrow down any of the usual suspects?

Have you checked the coil ? I dont know about this car, but in my car (87, 260E), very surprisingly the rough idling (sometimes at random), random stalling etc were all gone when I replaced the coil. When I checked my coil, the resistance was near the tolerance value, I think on the low side.

The other possibility is the fuel injection temperature sensor.

I haven't checked either. Were you having the same idle symptoms I described above...smooth when cold in or out of gear, rougher when warm and in gear? It doen't stall or go up and down, it is consistent once it decides. How do you check the coil and where is the sensor you mentioned?

You may have to ask Phil on Fastlane about the exact location on your car. On fastlane, the sensor is under the fuel injection category and is called water temperature sensor.

I had the exact same problems years back and was fixed by replacing this sensor.

More recent problems were rough idling, random stalling etc. This was fixed by replacing the coil.

I dont know the range of resistance for the coil on your car. But you can check the resistance by disconnecting the wire that goes from the distributor to the coil (CAUTION: DO IT WITH CAR OFF). Then measure the resitance between the coil's terminal and ground (or batter's negative)

Saumil

sptt: Do a search on duty cycle adjusting and then check yours to see where you are and proceed from there. If you are outside of the design limits closed loop operation esp. idle can be affected. Mark

Thanks Saumil, I did locate it (Ignition Coil) on the right side of the fuel distributor under the Air Intake Assembly. What I'll probably do is replace both since the car is old and everything else I've replaced seems to be original equipment anyway. I'll let you know if it works! Crossing my fingers. Thanks a lot.

Steve

OK...replaced the ignition coil with the Bosh OEM replacement part. A little better idle but not a whole lot of difference. Ordered water temperature sensor, vacuum lines and connectors. If any of those don't make an improvement, I am going to just live with what I've got. Thanks for all of the help. Was hoping I could post a "solve" on that ever eluding idle issue...maybe next week?

I had a similar issue once, the culprit turned out to be a couple of cracked air hoses. When the car got hot, the cracks opened up letting air in and causing a low and rough idle.

Easy test is to spray (sparingly) carb cleaner around suspected areas. Watch for idle surges. On that motor, pay attention to the cold-start injector hose, (under intake manifold), idle air hose, and all the rubber parts that you can't see such as the boot on the bottom of the air meter. Also, the vacuum hose to the valve cover can crack and leak on the underside and look fine from above.

Also wouldn't hurt to check the idle air valve, if it's unable to change as the car warms up, that could give a similar symptom.

Thanks TP. I sprayed the heck out of the hoses with carb cleaner but could not get a surge or change. I can try again per your areas to test but how about this? It has cooled off her in So Cal, so now, after driving on the freeway for a while and exiting to a stop, my temp gauge reads about 72-76 degrees and it idles nice. As I sit there and wait, it warms up to around 79 degress and boom! The idle drops to 500 and gets rough. I take off and go down the street, stop again, temp drops, good idle. Sit there and gets hotter...77, 78, 79 boom! Rough idle! Doesn't this sound like something is sending a false signal or something, or causing a richer run condition? Do you think this would be an air leak issue being so precise on reaction?

72-76C isn't the normal operating temp range on a M103. It's too low. If the correct 87C thermostat is installed, it may be stuck open, not allowing the motor to reach optimum operating temp. Or it's not the correct thermo. It's also possible you're getting faulty resistance readings from temp sensors. You may already know this, but the main coolant temp sensor is actually two identical sensors contained in a single assembly. One sensor sends a resistance reading to the EZL controller and the other sensor sends a reading to the CIS controller. Those signals are supposed to be identical readings, otherwise the two control units are getting conflicting information and attempting to make counteracting adjustments to the motor. It's easy to check the sensors if you have a multimeter. The temp/resistance table is listed in the FSM. There's also an air temp sensor located in the air intake cowl that can be checked and uses the same table of specs as the coolant temp sensor.

Now that sounds like what the problem might be. I have ordered that sensor and am waiting on delivery. My driving temperature is below 80 c. When I come to a stop it slowly moves up to 80 c. If it's hot out and I sit there, it will eventually get to 90 c and hold. If I take off and drive again, it will drop back down to 80 or lower depending on drive distance. I hope that is the problem and the sensor makes a difference. I certainly noticed a difference (improvement) when I replaced the O2 sensor. Thanks!

Just wanted to post back with an update. I changed out the water temp sensor in my 300E. In order to do that, I had to take off the manifold cover which was easier than I thought. Be careful if you do this and remove the water pressure if possible by taking off the radiator cap first as water sprayed all over the engine and onto my uncovered engine parts. Put everything back together and started it up until operating at normal temperature. Did not solve the problem. Then, one day, I was playing with the sensor connector and taking it on and off, on and off, while the car was idling (I was turning the connector to see if it mattered or acted differently when I connected it at a different 1,2,3,4 pattern) and since then, it has been idling fine even when temp exceeds 80 degrees and at a stop. Not sure why it's been running better, maybe the computer adjusted after running it for a while? I also replaced all of the rubber vacuum hose ends I could see and a sucking noise I used to hear down below the engine got much quieter. Anyway, so far, so good. RPM is 600 vs 500 when idling.

Hurray! After all of my parts replacements and thread reading, I have finally got my 300E with 180,000 Miles to idle nice and smooth!!! For my car the Lambda setting was misadjusted by someone else for I don't know what reason. I had been trying to adjust it with the air housing off (because I didn't have a long enough hex wrench) until I read a thread last night that said it had to be on due to the air cover temperature sensor sending signals to the computer. I did the adjustment (richer for me..clockwise) and now the thing feels like it's off when I'm in the car (well, maybe a very slight little tingle when at a stop). I hope when the warmer weather returns, it stays the same. I did wait until it reached operating temperature of 80C. And I adjusted without equipment so it might not be a perfect mixture setting. For now, I'm pumped! Thanks for everyone's help! Great forum!!!