Originally Posted by MJ50

moved to W201 forum...

Originally Posted by dolucasi

Hi there!. I know you posted your issues on the Facebook site but this will be much easier to track down here.

Please tell us the model year and engine size and miles on the car. I know the car belonged to your father so must have been in the family for a while.Hello and thanks for your quick reply back. Yes the car belonged to my farther who hardly ever used it, specially after it came back from the garage. The car was manufactured in Germany 1991 , Chasis W201 / 190 E . 2,3 / 8V 250,000km (155.342 miles )

As far as I could gather the history is:

(1) The engine overheated at some point in time and would have trouble running after that, Correct , full compresion but rpms very high on cold start
(2) At some point after this the head was taken off and re-skimmed possibly to plane it.Yes, head replanned
(3) After that there was trouble with cold running (too rich) and the following was changed to remedy the problem
- Replaced O2 sensor. I purchsed a new sensor from the internet . not original bosch( still have the original which seem to be ok but replaced anyway,
- Replaced or re-adjusted the AFM pot (AFM pot adjustment can be tricky, hopefully you did this properly and referred to some on-line videos) Correct , new AFM ( If I remember rightly I calibrated it at 7- V)
- Replaced the coolant temp sensor
- Some measurements were made for the EHA current with inconclusive results, some measurements looked correct, some not, some adjustments made to the plunger screw. Correct, I messed around and measured the sensor plate but not much luck .I must say a lot better than when it came back from mercedes dealers

Did I miss anything important or perhaps I have mistakes? Please correct me if I'm wrong.I also replaced the hose coming from the rubber boot on the air flow meter, injector rubber seals and checked the idle control valve

The first place to generally start with the CSI injection is the fuel pressure test. Because there is a baseline setting there that is required before taking other measurements or making adjustments. Do you have a dual input fuel pressure gauge? That would help.No ,I don't have a pressure gauge but I can buy one .I'm not a proffesional but I just know a little about cars as it's part of my job , But when it comes down to pressures and electrics I'm lost

It is good that you can now measure your EHA current. We will need that ability to resolve your cold running issues because when cold there is no other indicator that will tell you anything useful because the ECU is in open loop mode. Yes, multimeter was on the blink ,got a new and no problem measurring the Amps.

Also, it is best if you copy my text and insert your comments right after each item for easy tracking.

Welcome to the forum!

Originally Posted by dolucasi

While you are at it can you specify if you used dealer parts for temp sensor and AFM pot?

So far, what I'm seeing is that the oscillations at cold start and EHA current stuck at 10mA is a problem (and high fuel consumption).10mA is with the engine running at :80c and 22.3 engine cold ,ignition on

These are most likely unrelated issues:

(1) Cold temp idle cycling - Most likely an AFM pot part or adjustment issue. This is why I asked if it was a dealer/Bosch new part.I will have to take another look at the AFM as I'm not 100% sure I tested it properly (I don't think it was original bosh I would have to check that.
(2) The EHA 10mA current (mostly fixed). You should read your duty cycle to see what it is reporting. It could be reporting a fault code instead of the actual 10mA current (which is 75% or 25% depending on how you interpret the duty cycle). But this is probably why you are also seeing low MPG's.

Originally Posted by dolucasi

Your car is not that aged compared to mine at 185K miles and 210k miles each. Historically though around 150K miles is when some components need replacing like the rubber boot, Fuel Distributor, fuel injectors etc. I assume you found the Vaico rubber boot which is fine. It is a good company. The rubber boot for the 2.6 was available at the dealership a year ago so I bought 2 because one was shot and I figured with 2 cars I will use the second sooner or later.

You said you checked the spay pattern at the injectors and found no issue, that is great too. Yes ,they seem to be ok. I think if they were faulty I would be having other problems aswell
Sounds like you replaced the AFM pot with possible a $20 part from the internet. That is not good. I have known a case in which these Chinese (potentially) replica's are simply off spec.
They measure the wrong resistances in the terminals.
You are almost better off using the original. My original car at 210K miles still has the AFM pot installed at the factory back in '89. Does is have a minor dead spot, yes. Does that cause any issues. No.
Reason is that the signal is only used during cold running for the initial 3-4 minutes, after that is only used as a third signal to tell the ECU that it should regulate for idle. (the other 2 being microswitch and throttle plate switch). It is looking for that 0.7V you measured to adjust it. And it has quite a bit of slack on that 0.7V BTW. So if the dead spot is nowhere near the 0.7V setting you will never know that dead spot is there. If the dead spot is wide enough, sure you may get hesitation during acceleration when the engine is cold for the first 3-4 minutes. But it would have to be pretty trashed for that. Your original at 150Kmiles is probably not too bad and probably can be used. But of course if you replace it with a $200 bosch unit, it is better, but again I doubt you will see an improvement from your original AFM-Pot. Ahh, yes it was a cheap one I wish I new all this before - . I know I still have the old one but I might of messed with it to check for any irregularities. I'll check it out.

The above is why just replacing parts before testing them is not a good practice because these days often the replacement part is inferior to the original. Not saying this it the case for you but could be. And if this happens, often enough the owner is chasing yet another problem that is created on top of the original problem. I understand , The fact that the problem started after coming back from reparing the head gasket, due to over heating for the second time, Points to be that it must be a bad assembly or a vacumm leak and not any other componets.

Now since the subject matter is measuring, you should measure the coolant temp sensor resistances. I know you replaced it but better make sure the new part is to spec.
Since it is a dual sensor you should make sure both are matching each other (one for the cluster, the other for the ECU) and to spec. They measure ~200 ohms at operating temp but accuracy is important here. Remember the ECU believes you reached operating temp based on this resistance. and it needs ~80c to turn on closed loop, and goes back to open loop somewhere around 75c.
So if it is off spec, your stuck at 10mA EHA current would make sense. The other possibility is that the temp sensor is fine but your ECU is shot or some other component it the culprit.Ok , one mechanic diagnosis was the ECU but hard to believe that a car with only 250,000 km ,Well looked after and always parked inside the garage would have a bad ECU

For the measurement, start and heat up the car to operating temp. Turn it off and unplug the sensor and measure it at 85c. The let the engine cool down slowly and take measurements at various temps at 60c, 40c, etc.It is a linear behavior and goes as high as 2000 - 3000 ohms at 25c. ok will do.

On the fuel pressure meter, you will need one if you intend to keep the car. They should be around $150 these days. It needs to be a dual gauge one to test the FD and its adjustments.
It is not an emergency at the moment since you have a running car, though not running well. I am posting a picture of one that works well and has the required adapters.That's good to know

Since the vacuum leaks have been an issue, I am sure the dealership did a smoke test.I don't believe he did a smoke test , I wassn't present when the discussion took place but I believe he pressed down on the air filter to show my farther at the time the change in rpms . I have not notice any change in rpms when I have pushed down on the air filter. Maybe he touched the micro cut off switch You can but a kit that makes this easy. They cost about $50 and easy to use.
Did you happen to also fix the leak at the CO2 adjustment screw? I haven't noticed any thing wrong with the C'0 screw .Did you go thru the trouble of replacing that tower? NO You would have to tap those rivetted 2 holes for it and replace it with the part listed in the photo you posted. BTW most cars leak just a little at that spot but it is really minor and does not cause an issue. But I would recommend buying a kit and checking it again to make sure the new rubber installation went smoothly I never replaced the rubber boot . . Also this kit is good for measuring exhaust leaks too. So you would use it again sooner or later.OK. I might need it soon .( what's the price ?) I think it would be wise to go over the potentiometer as mentioned,vacummes leaks , temp sensor , and also try to get =0 milliamps on the EAH by adjusting the sensor plate with the 3 mm allen key ?

Originally Posted by dolucasi

(1) On the AFM pot, try and put the original back in there and adjust it. If you can not find it see if you can source a used but original OEM one. We can also measure your sheepo one for meeting spec before you replace it. With it unplugged see if you can measure 4.7K (from pin 1 to pin 3). At resting the resistance from pin 1 to pin 2 should be in the 1.5Kohm range but that is not precise. This is why an adjustment is needed. It is the voltage that is forced on it at idle that matters (0.7V not 7V you posted which appeared to be a typo I ignored). So when you are adjusting it your affectively changing that resistance of pin 1 to pin 2. I've re-librated the potentiomenter to 0'7 V, I was only getting on 0'2 V
(2) Smoke testers are $50 or so. It is just a heating element that burns up the liquid/oil that is put in it. It has the necessary fittings to connect. Kind of pricey for what it is but worth the money. The diagram you attached marked the tower in the FD. That was probably something the dealers mechanic just marked, I mistook for something they actually sold you. If you want to make sure you.
(3) I am confused about your AFM rubber. I thought you said you replaced it. But later you said you didn't. So did you replace the AFM boot? If not, buy that smoke tester it is even more critical to find the leaks. If the dealer did not do this, chances are there are more leaks they did not find. Some people remove the intake to remove the cylinder head. If that was done, I'm pretty sure they disturbed a bunch of components, ripped rubber, etc that caused issues down the road. I tried a pressure test to by covering the air flow plate with a rubber glove and blocking the ventilation hose, coming from the valve cover, But it wouldn't even hold enough pressure to inflate the glove. I'll replace the upper hoses ans rub a smoke test but that should be the problem. I also ran another EHA test and I am no longer stuck on 10 milliamps but I can't get 03 milliamps it's fluctuated between positive and negative , When I try to make the mixure richer it smells on petrol and I get black smoke. ( oil pressure gauge un and down slightly)
No, I never changed the rubber boot as I can not see any thing wrong with it. I spayed carburator cleaner under the air flow meter to test for changes in revs , took it off and inspected it .And found no aigns of deterioration so i didn't bother replacing it.
The car runns better when waen but it¡s not running 100 % it's still misssing.

If I did not answer all your questions, do let me know.

Originally Posted by 190Efan

Welcome. Just an FYI, W201s are known to be cranky when aftermarket, non-OEM parts are used. This is particularly true with the K-Jetronic fuel system. I'm also not yet convinced that the ECU is bad. ECU's can give the appearance of being bad because they're getting bad information from another source in the system. It might possibly wind up being the EHA but there's still much testing as Dolucassi has already suggested to be done. You might also want to look for moisture in your distributor.Checked distributor for moisture and it seems to be ok. I believe the position of the distributor doesn't really matter as it is the EZL who take care of the timing ? Be sure to source OEM parts if any more parts are determined to be needed. You can also check the parts inventory at the MB Classic Center online parts store. If they don't have what you might wind up needing then try Adsit Co as they're a good source of hard to find MB parts.

Originally Posted by dolucasi

There is quite a bit packed in your responses.

(1) AFM adjustment. If you have not found it to be very difficult to adjust you must be doing something wrong. Basically a fraction of a millimeter adjustment makes a huge difference.Yes, no longer have 0,7 V on the AFM it seems to be all over the place , when I set it to 0'7 V it doesn't maintain the calibration. I've been having other issues with the sensor plate catching when I tried to tighten up the scews at a certaiin V- ,,, it's ok now but not set at: 0'7 V , it changes all the time. it could be broken!
All this said, if you are getting a steady ~0.7V during warm idle now you are probably OK. You should double check it with a cold start to make sure idle fluctuations do not happen. You can probably guess why those fluctuations happen. Say it was adjusted to 0.2V. During cold run the idle controller sees that and decides it is too low and starts to increase the idle unnaturally. Because the opting frequency of the system is too low, it will overshoot to well over 0.7V and then it will go in the other direction causing overshoots and undershoots with 1-2 second intervals. Hopefully you have fixed your problem number 1.
(2) Vacuum leaks. Do not trust the propane/carb cleaner type diagnostics. There are vacuum losses in places you can not imagine. With all my knowledge I found one in my original '89 that explained a lot of second level issues that most owners would not even notice. Once I fixed that one single leak the car basically runs as well as it did when I drove it off the showroom in April 1989. So get the smoke tester, it will be well worth the money. I made a quick devide with a soldering iron and a pump to check for vaccum leaks and they vaccum problem are the hoses coming from the idle controñl valve into the admission and rocker , This is not the problem, this is a minor problem , I'll order the hoses but it won't solve the real problem.
(3) EHA current stuck at some fixed value.The EHA is no longer stuck at :10 mA in fact today i had a good reading appprox -7mA up and down This is a major problem and your emissions control is not working. So please stop adjusting the mixture screw at this point. If it is running and not blowing black smoke out the tail pipe we should start diagnosis and not touch that screw until we are done with the diagnosis. The diagnosis starts in the item below. Yes , I will leave the C0 srew alone now as it's not the problem I had real job getting the car running again today, Had some seroius issues today after trying to get the AFM calibrated I thought I had it and tried to get a fine tuning with the C0 screw and messed it all up. The cars perfomance today was totally irractic , At one point it was stalling , after giving the car a bit of run at 3500- 4000 rpms the revs would not come back down I had to cut the engine , restarted the engine and it wouldn't start ,got it started by altering the mixure , And again it would stall. Basically , it's all over the place - 1 min it's rich the next it's lean , This is deffinatly not a vaccum leak
(4) Read the duty cycle coming out of your ECU and presented in the X11 diagnostic port next to your EZL. Pin 2 (gnd) to Pin 3 (signal) will put out a 100Hz signal with a duty cycle ranging from 0% to 100%. And it will fluctuate +/-5% in closed loop at operating temperature. That is simply the EHA current translated into voltage for mechanics to read it without having to unplug the EHA. 50% duty cycle corresponds to 0mA at the EHA. +20mA EHA translates to 0% duty cycle and -20mA EHA translates to 100% duty cycle. This is why a 2mA fluctuation translates to 10% duty cycle voltage at the X11 port. I couldn't do a duty cycle due to not having a multimeter which reads % , I'll get one tomorrow, I believe you have to have the multimeter set for % . I have no idea about the duty cycle so I will upload the percentages for you to have a look at , if you don't mind.

So we need to see what the ECU is putting out. In your case it is mt today i have a close ost likely a fixed duty cycle. You need that because that translates into an error code. You can look up on the internet to decide what the ECU claims as an error-code/faulty component.

This is your next step until you receive the smoke tester which will trigger other activity.

- Cheers!

Originally Posted by dolucasi

I think you are back in the cycle of messing with the plunger screw on top of your FD.
Once you measured -7mA and moving up and down +/1mA you should have been done and not touched it until everything else is fixed.
I sound like a broken record but do not touch that screw, it is not your problem or solution.
Because you keep adjusting it you are going around circles and also making long distance diagnosis impossible. I Totally agree, I'm still trying to fix the problem by adjusting the C0 srew. There are other issues,

I was hoping to be able to measure your duty cycle and EHA current at the same time just to verify that your ECU is at least reporting the correct duty cycle. Yes, we need to know if the EAH is working or the ECU for the matter. The EAH and duty cycle are now far out. The duty cycle won't come down past 07'65 even when I try to make the mix leaner. (yes, I know stop playing witht the mix )
It is not that critical but more of a verification. If you get that second multimeter with duty cycle capability, we will make that measurement. Not too critical at the moment.

Let me repeat the following if I have not made it in this thread already. This is about idle control. There are 4 components that are needed working for a steady idle:
(1) TPS .TPS is working fine , signal on the second pin for idle and signal on the 3rd pin when full throttle down.
(2) AFM Potentiometer .Put the old one back in . the componets are different but would I say much harder to get the 0.7 V . if fact I can't get it on 0.7 V ( not sure but I think it's working more accurately)
(3) Microswitchare checked it while back and it was ok for signal. i'll check it again ( strange when I push the plastic litle wheel back the revs drop back down ) almost if it's not in it's right position but that's not the case as it can only sit in one place.
(4) Idle controller (early models have this in the ECU, later models had this in a separate box that also include things like fuel pump relay)Checked the idle control by appy a 12 V soure and it's clicking -opening and closing perfectly / it doesn't close completly but I guess it's designed to not close completly otherwise it would be starve the engine for air. ( haven't checked for Ohms. )
The weird thing is when I unplug the ICV the rpms come back down to normal idle , So it's doing the opposite to it's job,Reving high when activated and dropping revs when disconected Now I don't know if it's because of other issues.

Now saying all this you're going to think I'm mad but I think I might of over looked a vaccum leak which I'd looked at before but it's deffinately suspect. The crancase air hose runing from the idle control valve, joins up with another section going to the rocker cover ventilation, has an air intake at the joint. I spayed some carb cleaner on it and the Rpms went up sligtly. I've ordered the plastic conector and the hoses.
So you need to go thru and verify that all 4 components are functioning.
First 3 are easy, last one a bit more involved.
Have you verified that items 1-3 are working? These all involve using an ohmmeter and the sensors electrically disconnected and you actuating either the throttle plate or the AFM plate.

Originally Posted by P.Bat

Originally Posted by P.Bat

The black wheel on my swich does not go as far back .I looked at altering it's position because it effects the rpms when pushed far back.. Does it have to be pushed back or is it ok just to be toching while idling ?

Originally Posted by dolucasi

You need to go thru the first 3 items carefully with them unplugged. Read up on the details on the internet like you did with the microswitch.

DO NOT cut corners, like you did with the vacuum leak. I did to save time ,I made a home made smoke machine with a soldering Ion I know you have vacuum leaks. Buy the smoke machine and test it properly.
Only a $50 investment. There should be no smoke coming out of anywhere in the intake and injectors when you do this test.I put in new seals in for the injector, so they should be ok. Also changed the hose coming out of the ruubber boot. The T. shape vaccum lines going into the intake are also good and main line coming out in...
You may notice a very minor seeps right at the FD plunger screw. If so, you can either ignore or put a cap over the tower. Smoke coming out of anywhere should not be tolerated. some smoke will come out through the sensor plate and ventilation valve cover ? By the way the valve cover has no pvc valve as it broke when I was cleaning it ( but that has nothing to do with the issue we're solving here ,I have found a used valve cover on line with the pvc valve)

I'm going to trust your TPS verification. It is simple and you believe it is working. Yes , definitely working now . Tested for signal and Ohm's ( it wasn't working ) Now I need to adjust the mixure again to keep the car idling.
On the AFM pot: You can see that the original is working and the new one was just junk.Correct , TOTAL RUBBISH - I tested for ohms and it's not working. The old one appears to be ok. A slight jump when pressing down on the sensor plate but I guess that's never going to be 100% ( I will calibrate properly tomorrow as it's off spec due to the micro swich and not getting it exactly right ( fingers crossed) The way to adjust is to tighten only one screw of four. Then start the car and let it warm up . Yes , it's very sensitive it alters when you tighten down different screws ( very sentive )
Then try to adjust it to spec. You can only do it by tapping the AFM-pot ever so lightly to the right position not with your hand just very slight taps. yes, i'm aware of the tatic.There is a reason why Mercedes only sells the AFM pot together with the entire fixture pre-adjusted. They do not want people like you and I to mess with it. The reason why it is so sensitive is because it is not just a voltage divider like a slot car controller. Not only does the ECU provide 5 volts to pins 3 to 1, but it also forces current to pin 2 depending on the voltage it sees on pin2. This makes it highly INSTABLE due to positive feedback. It is also why the idle goes unstable when running cold if it is adjusted incorrectly.

This cycle you went thru with the AFM pot is a prime example why throwing parts at a problem never works because the swap actually caused you more headaches. You ended up with a AFM pot that lost its adjustment. Ask me how many times I have adjusted or replaced my AFM pot at 210K miles. The answer is NEVER. People sometimes replace the AFM pot because it was a tiny dead spot in its working range. Wrong. That hardly every causes issues. I cheap Chinese replica particularly misaligned will cause you a lot of headache.I agree

But you can only properly adjust AFM pot AFTER all your vacuum leaks are fixed. Because if air is escaping thru means other then the AFM plate, your adjustment is off. Yes, I realised that I just need to get the mixure better as it's now all over the place again due to the micro swich now working ( I think &#129300

It is clear that your ECU is working, not sure if it is 100% but we will never know until you fix all the other issues around it. The ECU's are solid, they rarely break. I have 3 extra verified working ones for my model year in my parts cabinet. With two W201's at ~200K miles, I figure I will need a new one at some point but I suspect they will never get used.

I think you are starting to see the picture and why I channeled you here from Facebook. I always chuckle when I see posts there. There is no way one can properly diagnose a CSI system on Facebook.
You will just see random comments and can not converge on a solution. Sure .making progress and understanding more about the entire ke Jettronic fuel injection system.
Happy New year BTW. Yes, Happy New Year , all rhe best for 2026

Cheers!