I agree it was odd to get better mileage with the lower octane fuel and have no explanation. I took a look at the kinds of miles - pretty much the usual combination of freeway cruising and neighborhood errands. On long trips, the mileage soars, but that's probably because I'm good about travelling at a consistent speed.

I ain't sayin' what that speed is, but it's not legal and it's fun....

__________________
David

His:
2003 Dodge Durango SLT
2002 BMW E39 530i Sport
1988 Mercedes 300TE
Hers:
2003 Chevrolet Suburban
1999 E430

Having read all the posts on rough idling engines of the 300E, I read nowhere the item that effects idle speed and smoothness. When my '87 with the 103.98 engine idles rough I've been able to smooth the idle by adjusting the Lambda Control of the output voltage from the O2 sensor. I use a dwell meter to read the output from the sensor and using that long allen wrench, down through the hole in the airfilter, adjust the idle mixture until I get the right dwell at 2000 rpm. The smart box controls the speed, you adjust the mixture. Check it out. Happy Trails Beep Beep from Houston.

Donald.
and all little Benzes is smuv.

Donald, thanks for the post. Just last week when I was watching the last diagnosis (I have another scheduled this coming Wed - the idle is still unacceptable), my tech was adjusting what I think you're talking about. It's just to the right of the fuel distributor... and the tech looks over at me and says "Never adjust this. If someone comes up offering you a ??mm Allen wrench, just say no". I never circled back in the conversation to confirm WHY that was sacred... I think it may just preserve the gulf between Tech and regular humankind... like "don't eat the fruit from that tree" kinda stuff.
I'm a sinner: What size Allen wrench do I need?

The reason you shouldn't adjust the mixture adjustment is that normally you won't have any way of seeing the response. Donald is using a dwell meter. I would suggest a multimeter that reads duty-cycle but the two are similar functions the dwell meter is just marked wrong.

The problem with making this adjustment without the meter is that it DOESN'T change the mixture and will make NO difference in the idle if everything is working correctly. With this adjustment you are changing the base mixture. With a properly warmed lambda system the mixture will always return to lambda (about .5-.8% CO or 14.7 to 1 airfuel ratio). If you adjust richer the system will lean it back to lambda. If you adjust leaner the system will enrichen. A properly adjusted system runs at a duty-cycle of 50/50. In this position it will have equal ability to correct lean or rich. If the system is way rich (say 90/10)it will still be at stoiciometry (lambda) mixture because it is almost out of bounds but not quite. At 90/10 there is just enough range to correct to .5-.8% CO (before cat - real mixture). If you then take this system and adjust further rich you WILL change the mixture. Once the system is overwhelmed small adjustments make BIG changes in the mixture and can not be compensated anymore. This of course will set off the check engine light as the control system is now non-functional.

Basically the point I am trying to make is that you will make NO difference in the running by adjusting this screw unless you overwhelm the system and defeat it. All slight mixture corrections are almost instantly compensated for and the mixture remains the same.

In one post recently (maybe this one) I stated that one should remove the aircleaner and slightly depress the plate on the airflow meter. If one does this an instantaneous result of mixture change results. Almost instantly it goes away as the system compensates. In this brief instant if the engine smooths out one can infer that the roughness is mixture related. This does not mean the system is running at the wrong mixture, it means that one or more cylinders are not rich enough under these uniform conditions. Possible reasons are restricted injectors or vacuum leaks.

__________________
Steve Brotherton
Continental Imports
Gainesville FL
Bosch Master, ASE Master, L1
33 years MB technician

Steve, thanks a lot for keeping me straight. I appreciate your techy response. Of course, I'll need to read it a couple of times before I get it all... But I got the point: I won't screw with the mixture (at least until I know much more about the underlying and related systems than I know today).

I can't emphasize enough the value in making the simple, non-intrusive test I mentioned above. Not only does it give the first step in a real diagnostic test, but it also gives a feel for what is happening in the system.

Don't be afraid of it, go try it. Press slightly, more slightly, heavily, more heavily. It will kill the motor. Obviously at a point between slightly and more slightly you overcome the feedback electrical control and as you go heavy you flood the motor. You won't hurt anything doing this at idle a number of times. You can overheat the catalyst by continuously overwhelming the feedback control, though.

There is a sensitivity to be learned here and if your feedback system is already overwhelmed you won't see the compensation. If single cylinder mixture defficientcies are responsible for your rough running (and this is VERY common)you will see a smoothening instantly, when pressed, which goes away if the plate position is held ridgidly (as the system corrects your movement). My first step in diagnosing a K-jet car is to do this process. I WANT to see if basic mixture either richer or leaner affects the problem. The answer is just a start, but, this feel needs to be felt by anyone who wishes to understand these systems with test equiptment.

__________________
Steve Brotherton
Continental Imports
Gainesville FL
Bosch Master, ASE Master, L1
33 years MB technician

Steve, back to the fuel dist. If it's isolated as the problem (not getting fuel to #2 injector)... IF that's bad, what's your preference for replacement on limited budget? Rebuilt, off-brand, bone yard, new OEM at all cost? I've got a Bosch rebuilt on it. I'm convinced it's bad. At 11k miles in (and having had the car for 5k miles), I bet it has always been bad. My cheap and honest MB mechanic said he never uses rebuilt. He much prefers boneyard parts. He KNOWS the rebuilts didn't work at one time and can confirm the bone yard parts do work. He feels more confident.
What do you say?

We would use a Bosch rebuilt. I presume thats what you would also get from MB. Watch the prices, the Bosch one lists for about 50% more than the same thing through MB. Any good shop should know this.

We would also have a differential flow meter to absolutely prove a defficient fuel distributor. If you have had two fuel distributor problems in a short time I would check your tank for rust. Once it starts the tank must be replaced. Look at the neck where you put in the fuel. If there is white powdery deposits or, worse, red rust showing, its history.

__________________
Steve Brotherton
Continental Imports
Gainesville FL
Bosch Master, ASE Master, L1
33 years MB technician

The fuel input looks pristine. But I can only see a little shiny steel/aluminum around the hinged plastic guard. Could one get a full survey of tank condition by removing the fuel level sender unit and peeking around?
Why would rusty tank consume fuel distributors? Sending rusty bits of tank would just clog up the workings? Could the fuel dist be retro flushed instead of replaced?

Steve is right relative to the dwell meter than a digital volt meter. I suggested the dwell meter for I thought more owners would have one than the digital. The Factory Service Manual Engine 103 section 07.3 states "IF NO LAMBDA CONTROL TESTER IS AVAILABLE, A DWELL ANGLE TESTER MAY BE USED". The manual lists the procedure as well. The best I've seen was with a Fluke 98 Scope Meter mdlII.

The specifactions per M/B are,
idling speed: rpm 650 +/- 50
control range: 35-45% this is of peak battery voltage.
dwell angle 21-27 degrees
I don't mean to be oposite with Steve, this process was taught to me my my M/B dealer's Technican. And if I may this is my technique and it has worked excellently for 12 years.
You can obtain the Allen wrench from your M/B part Dept.
I believe it is 3mm.x 6" long with a "T" handle.
Hood open and Vertical.
You will need a Dremmel Tool and a carbarundium cut-off disk.
After removing the air filter and covering the air intake; using the cut-off tool, cutoff 10mm from the Lambda Tower, NO MORE than 10MM. This removes the safety ball. You can cover the hole when your finished. If you are a purest you can purchase a Lambda kit when you buy your Allen wrench.
Insert your "A" wrench and push to feel the spring of the adjustment screw, it's very sensitive a little move makes a terrific change. Be careful, don't turn it yet.
Locate the round diagnostic connector on the left inner fender panel. Connect the positive test lead of your dwell meter or what ever you have(DVM or SCOPE)to pin #3 and the negative lead to a GOOD chassis ground.
Set your test device to the proper range.(dwell mtr 20-30- degress, scope to 30-50%, DVM dcauto).
Check for2.5-5.0 vdc or 21-27 degrees dwell, scope 35-45%.
Disconnect and plug the purge valve at the throttle valve assembly.
Install the air cleaner.
Start and run the engine until it reaches operating temperature.
Snap the throttle then run the engine at 2500 rpm to warm the o2 sensor.
Check your meter readings.
Return engine speed to idle.
If the readings are off, inseert your "A" wrench through hole in air filter. Feel for the adjusting screw.
Make your adjustment, very slightly. Remember you are making a fuel discharge adjustment. Wait for the O2 sensor to respond. You want middle of the spec range.
Vehicles and climate vary, tweaking may be required for that smoooth idle, but stay in spec.
Apply emergency brake and put Transmission in Drive.
Idle should be smooth and speed right. If not smooth enough repeat your tweaking.
Your DVM on pin #3 should read the average voltage, approx. 50% of peak voltage and the dwell meter approx. 50% of range.
Unplug the O2 sensor under the passenger carpeting. KEY OFF, ground the lead to the control unit. Connect your meter to the single wire from the O2 sensor, now start the engine. A reading of 0.45- 1.0v(full rich) means the sensor is responding properly.
Remove and replace the air filter, unplug your meter and return the cover cap.
I use a tubing cap to seal the lambda tower.
I hope I havent left anything out, possibly Steve can help us both.

Donald
odd ball out