KE Jetronic M103 engine Stall problem

[dhaigh]

I have a 1988 300CE (MY1989) with KE Jetronic fuel injection.
Recently I experienced a stall problem (oh yes, that one!) where the car would stall as the car approached traffic lights, or turn right or left etc. This would happen after the engine was thoroughly warmed up, and after at least 20 minutes of driving (so never when below operating temp). The situation was foot off accelerator, braking lightly, decelerating and engine speed about 1000 rpm, as I approached the lights or to turn (it happened twice on a round-a-bout!). The car would always start immediately, so it wasn't a hot engine start problem. Apart from that the car idled like a kitten.
So here is what I have done: with the assistance of a member of the Car Club Victoria who has the test equipment and an intimate knowledge of the KE Injection system, we tested all the components.
The OVP, Fuel Pump Relay and EHA all replaced in the last 6 months. Also new plugs (Bosch H9DC - non resistor, yes still available!)
Mixture adjustment (Lambda) perfect, crankshaft position sensor OK, O2 sensor signal (under drivers carpet) little high 1.5 - 1.6V but acceptable, Idle Control Valve appeared to be OK (clicked when it had power applied), throttle body switch OK, fuel shutoff (decel) microswitch OK, coolant temp sensor OK.
BUT the Airflow Sensor potentiometer/airflow metering only gave us 3.25V between top and bottom pins - should be nominally 5.0V. However, this may cause a rough idle but shouldn't cause a stall. The metering itself across the resistive material showed a drop-off in voltage as we deflected the sensor plate but nothing too serious - again, the car idles like a happy cat.
SO HERE IS THE ISSUE
1) Is the 3.25V at the potentiometer an issue that would cause a stall?
2) We had heard the ICV can cause a stall if it sticks closed on deceleration, so we disconnected the ICV and have tested the car over the past week and 400kms of driving.
THE CAR HAS NOT STALLED ONCE WITH THE ICV DISCONNECTED!! (All fingers and thumbs are crossed for this simple fix).
Has anyone heard of that before??? Amazingly, over 320kms of a fuel economy test, it needed only 30L of fuel (9.5L/100kms on a combination of freeway and city roads)!!! We expected the opposite with the ICV disconnected but that's the evidence!!!
Any advice re the above would be keenly appreciated.
David

[Ivanerrol]

I assume you have cleaned out the ICV. Good idea to replace the ICV hoses.
Check the throttle body rubber gasket and make sure there's no leak.
What's your economy gauge (vacuum gauge) reading at idle?

Having gone through replacing 2 failed fuel pumps in two cars in a week :whistle: how are your fuel pumps, filter and fuel regulator?
Next tool I'm buying will be a fuel pressure measuring gauge.

Disconnecting the ICV just overides or hides other issues.

There's a circular plate between the distributor cap and the engine. If it has never been replaced I suggest you do it.

[ngruzevs]

You referring to the Suppressor Disc?

[dhaigh]

I assume you have cleaned out the ICV. Good idea to replace the ICV hoses.
Check the throttle body rubber gasket and make sure there's no leak.
What's your economy gauge (vacuum gauge) reading at idle?

Having gone through replacing 2 failed fuel pumps in two cars in a week :whistle: how are your fuel pumps, filter and fuel regulator?
Next tool I'm buying will be a fuel pressure measuring gauge.

Disconnecting the ICV just overides or hides other issues.

There's a circular plate between the distributor cap and the engine. If it has never been replaced I suggest you do it.

If the car doesn't stall for another week I will replace the old ICV for a cleaned change-over unit. At the same time we are replacing the fuel filters (the main beside the fuel pump and the micro one at the FD). Then I will check fuel pressures and consider replacing both fuel pumps as a maintenance job.
The circular plate you refer to sounds like the throttle body switch/fuel pressure regulator?? Is that correct?

[dhaigh]

You referring to the Suppressor Disc?

Does the Suppressor Disc have another name on the M103 KE Jetronic injection???

[Ivanerrol]

You referring to the Suppressor Disc?

Does the Suppressor Disc have another name on the M103 KE Jetronic injection???

Suppressor disc. Part of the distributor.
It's coloured orange and is underneath the rotor.
Ala http://www.everythingbenz.com/z/part/10 ... istributor

[dhaigh]

You referring to the Suppressor Disc?

Does the Suppressor Disc have another name on the M103 KE Jetronic injection???

Suppressor disc. Part of the distributor.
It's coloured orange and is underneath the rotor.
Ala http://www.everythingbenz.com/z/part/10 ... istributor

We haven't checked that. Maybe because we assumed if the distributor, or rotor (or suppressor disc?) was breaking down then rough running and hot starting would be the problem. I'll check it out. Thanks

[Mercmad]

Disconnecting the ICV and the car runs better indicates vacuum leaks . How is the central locking, headlight adjuster, brake booster? I had an odd one recently and because i was too busy I palmed it off onto to a mate. The car would stall too, but when starting a puff of smoke would appear from the dash... It was a line off the headlight adjustor. The lean mix would cause a slight cough at start up,sending fumes up the vacuum line. Replacing the line fixed everything. Kjet relys on everything working as designed

[dhaigh]

Disconnecting the ICV and the car runs better indicates vacuum leaks . How is the central locking, headlight adjuster, brake booster? I had an odd one recently and because i was too busy I palmed it off onto to a mate. The car would stall too, but when starting a puff of smoke would appear from the dash... It was a line off the headlight adjustor. The lean mix would cause a slight cough at start up,sending fumes up the vacuum line. Replacing the line fixed everything. Kjet relys on everything working as designed

I'll test for vacuum leaks. Thanks for the tip.
No puff of smoke from the dash as yet!! Brake booster, central locking all OK at the moment.
When the car is decelerating with foot off accelerator, my understanding is that the throttle control switch sends a signal to the ECU (ie. throttle closed) which signals the EHA to cut off fuel to the injectors, but that at approx. 900rpm the speed sensor signals the ECU to start sending fuel to the injectors again as the car returns to idle speed. This is the precise point that the car decides to stall.
Presumably, the ECU must signal the ICV to do something to 'close' (?) the air valve to enrich the mixture for idling. (MY BASIC UNDERSTANDING). Could there be a problem just at the point the ECU receives the signal to restart the fuel flowing again at 900rpm? If the ICV doesn't 'close' the air valve enough, couldn't the car stall???

[Mercmad]

Can you get another ECU to borrow? These are normally bullet proof but a mate bought a car a few years ago which had a problem of no air con and lots of money had been spent on it before i realized that it had the wrong ECU..it was for a non air con car... Such things can be caused by the craziest circumstances.

[dhaigh]

Can you get another ECU to borrow? These are normally bullet proof but a mate bought a car a few years ago which had a problem of no air con and lots of money had been spent on it before i realized that it had the wrong ECU..it was for a non air con car... Such things can be caused by the craziest circumstances.

I could give that a go. However, my view is I should try another ICV first, since (AT THIS STAGE) the car hasn't stalled since disconnecting it. When I reconnect another ICV, if the problem persists, a Club friend and I are doubtful about the reading to the Airflow Sensor Potentiometer of 3.25V when it should be 5.0V nominally. We don't know if that is within tolerance or not. Do you have a view?
Our view is that something is happening at the point when the ECU reaches the set idle position of 900rpm (on deceleration) and something must happen to signal the EHA to pressurise the lower chamber of the FD again, to send fuel to the injector valves (on overrun, the EHA would have stopped fuel flowing so it must receive a signal from the ECU to restart that process again).
If all else fails, then yes we could try another ECU, but surely the readings we are taking should show up a problem somewhere for us to come to that conclusion!!

[Ivanerrol]

Getting a new airflow potentiometer makes a big difference.
However. The OEM ones are $$$$$$.
The elcheapo made in PRC ones are iffy.

I mentioned the suppressor disc above because 99% of cars have never had it changed. It won't help your particular problem though.

[dhaigh]

Getting a new airflow potentiometer makes a big difference.
However. The OEM ones are $$$$$$.
The elcheapo made in PRC ones are iffy.

I mentioned the suppressor disc above because 99% of cars have never had it changed. It won't help your particular problem though.

It would be worth changing the airflow potentiometer if the tests showed up a drop in voltage to zero as the sensor plate is slowly deflected. But the test was OK. The question whether 3.25V is within tolerance does need to be conclusively checked but an electronics whiz tells me that whether the voltage is 3.25V or 5.0V shouldn't affect the way it operates.
I agree about the suppressor disc - I'll make it a maintenance item when I renew the distributor cap and rotor (around 200K kms).

[Ali_Boz]

I was having the same issue on my '91 300TE. My car would often stall only when warm just like you described. After changing the OVP (red cap), plugs, distributor and cleaning the rotor I changed the ICV for a brand new one (and one of the hoses). Like you when the ICV was disconnected the car was run perfectly.

The problem didn't stop until bought a multimeter and leaned the mixture out. The car stopped stalling in traffic and was able the actually accelerate from its 2nd gear take off. I found that my air flow potentiometer was cooked and the engine was hunting for idle speed so i bought an elcheapo from china just to see if it would stop the idle, which it did. Now the car idles 100 fold better than 6 months ago and plan to do a full KE replacement and rebuild the fuel distributor as preventative maintenance.

I hope this is some help even though you've had the luxury of having most of your components tested.

heres the thread to when I posted about this problem
viewtopic.php?f=6&t=22341

[dhaigh]

I was having the same issue on my '91 300TE. My car would often stall only when warm just like you described. After changing the OVP (red cap), plugs, distributor and cleaning the rotor I changed the ICV for a brand new one (and one of the hoses). Like you when the ICV was disconnected the car was run perfectly.
The problem didn't stop until bought a multimeter and leaned the mixture out. The car stopped stalling in traffic and was able the actually accelerate from its 2nd gear take off. I found that my air flow potentiometer was cooked and the engine was hunting for idle speed so i bought an elcheapo from china just to see if it would stop the idle, which it did. Now the car idles 100 fold better than 6 months ago and plan to do a full KE replacement and rebuild the fuel distributor as preventative maintenance.
I hope this is some help even though you've had the luxury of having most of your components tested.
heres the thread to when I posted about this problem
viewtopic.php?f=6&t=22341

I read your original post - I note you were having a hot start issue after stalling. This normally points to distributor, rotor and plugs. Interesting to hear you changed your airflow sensor potentiometer and the problem was resolved. My pot appears OK but I will re-check my lambda ratio (fuel mixture adjustment) as the O2 sensor is registering 1.2V which is a little on the rich side. Maybe the O2 sensor needs to be cleaned. I'll try a change-over ICV first and post on this forum how it goes - stay tuned!!
For everyone's interest here is a link to the Bosch manual on the "KE Fuel Injection system" as used on R107, R129, W124, W126 and W210 for that period. A scintillating read of 23 pps even if I say so myself - a complete novice, but learning a lot!!
http://www.berlinasportivo.com/Technica ... ic-OCR.pdf
Ignore the warning message you may see and click OK - you receive this message because the file is PDF). After the PDF opens in your Browser, you can save it to your computer. Ignore Page 1 (advertising for the site) and start at Page 2 – Officeworks will print this PDF for you, starting at page 2, if you take it to them ($7.50).
Enjoy

[Ali_Boz]

I forgot to mention that I replaced the o2 sensor as well because the old one was filthy and didn't know how old it was.

Good luck!

[dhaigh]

I forgot to mention that I replaced the o2 sensor as well because the old one was filthy and didn't know how old it was.

Good luck!

I tested the O2 sensor signal (under drivers carpet) and it was a little high at 1.5 - 1.6V but acceptable. Besides, if the sensor was breaking down, it would not cause a stall on it's own (simply may run a little rich).
The Bosch manual on the KE Injection system (link above in an earlier post) is very informative. Here is another link that will give you all the necessary measurements for the different components of the injection system - this link is for models with the M103 engine/KE Injection, but I have the link to other KE Injection measurements if you have the M104 or M119. Just PM me for details.
http://w124-zone.com/downloads/MB%20CD/ ... 3-0121.pdf

[Bradley]

I have a very similar problem with my w124. The symptoms are virtually identical: Great cold start but intermittent lousy warm idle with stalling as the engine drops down to idle when stopping the car such as a stop sign etc.

My research indicates the ICV is 'on the fritz' but I am eagerly awaiting the next instalment of your diagnosis before I get my hands dirty.

And yes, the link to the description of the KE fuel injection system is a bewildering but interesting read for an amateur.

Mark

[dhaigh]

I have a very similar problem with my w124. The symptoms are virtually identical: Great cold start but intermittent lousy warm idle with stalling as the engine drops down to idle when stopping the car such as a stop sign etc.

My research indicates the ICV is 'on the fritz' but I am eagerly awaiting the next instalment of your diagnosis before I get my hands dirty.

And yes, the link to the description of the KE fuel injection system is a bewildering but interesting read for an amateur.

Mark

** UPDATE **
We have now installed a new ICV, since the car didn't stall once since disconnecting the old one. When we fitted the new ICV, everything came back perfectly - lambda ratio 45-55%, idle 550rpm, starting and acceleration perfect. O2 sensor and CPS fine. Fuel pressures fine. EHA fine. And, of course, it idles perfectly as I coast to a stop at lights etc. I also fitted a new fuel filter (at the rear) AND replaced the micro fuel filter at the FD (at the fuel inlet to the FD).
One clue I hadn't noticed that may be of interest to others, was that before I installed the new ICV the engine would tend to 'chug..chug..' or 'kangaroo hop' a little while reversing out of the garage when the engine was cold, and the accelerator pedal was being feathered. Since fitting the new ICV, the car reverses so sweetly and smoothly when cold that it underlines how important the ICV is in the overall system. The ICV is very sensitive to small changes in current, and all I can assume is that the old one was simply worn on the inside rotating valve and/or sticking - time for a new one!!!

[AMG]

Pity I didn't see this earlier...
9/10 times it's the IACV gummed up with crud and has never been properly (read thoroughly) cleaned.
a good overnight soak with appropriate cleaner (threebond engine conditioner) or using aggressive solvent type cleaners like brakeleen or carby cleaner / heavy aerosol degreaser / kerosene can sometimes free up the IACV operation. What it doesn't do is fix a coil with hig resistance or a dead short.
the IACV is controlled by a PWM signal, so if the electrical connection is suss, or the resistance is not within spec, then it simply cannot be fixed (returned to operational specs) other than by replacement.

First port of call on all KE engine diagnostics is to follow the test / diagnostic procedures in the service manual - aka RTFM. It's in the reference library for those reading who have an interest in KE-jet function in their cars.

As Ron mentioned before - KE is a system that requires everything to be functioning correctly and in sympathy - when one area is substandard or not functioning as intended, then there is a potential money pit waiting, unless the correct diagnostic procedures are followed.

KE is not a system where you 'just throw stuff at it' in the vain hope it will fix everything. Indeed, this is where the majority of complaints come from owners on here - and it's almost 100% down to not properly understanding how the systems are integrated or how they function, or because of a lack of mechanical knowledge and a blind faith in a mechanic - are suckered into throwing stuff at it just to endure a huge bill and not have the issue fixed - yes, it happens, and it's seemingly common occurrence - the suggestion in those casess to take the car to a renowned specialist independent MB mechanic, where the poor owner will have to fork out to have all the previous misguided attempts fixed properly.

It's actually another good reason to seek expert advice when something does suddenly go bad - and also why it is important to have a good working relationship with your mechanic - because a lot of 'little' things can easily cause similar symptoms of much bigger issues, with bigger price tag fixes... when it could be simple stuff like vacuum leaks. It's all too common for an issue to be a " just throw a ***** on it and see if that fixes the issue" or that old chestnut " Sounds like the Battery son........ Under the bonnet son" if you get my drift.

Good to see a problem resolved in a straightforward way, without the excessive hype, bs and conjecture.

If your O2 sensor hasn't been replaced, then consider buying a new one - as they do not last forever, and the O2 sensor does control the EHA to regulate idle fuelling. I noticed you said it was out of spec. Did you happen to test the resistance as well as the voltage?

M103 is a pretty darn reliable engine. Even with a few decades of "really average" service history, they seem to be impervious to general neglect, and it's not hard or expensive to give them a complete new lease on life. Really underrated powerplants. Pity they don't make them like that anymore.

[dhaigh]

Pity I didn't see this earlier...
9/10 times it's the IACV gummed up with crud and has never been properly (read thoroughly) cleaned.
a good overnight soak with appropriate cleaner (threebond engine conditioner) or using aggressive solvent type cleaners like brakeleen or carby cleaner / heavy aerosol degreaser / kerosene can sometimes free up the IACV operation. What it doesn't do is fix a coil with hig resistance or a dead short.
the IACV is controlled by a PWM signal, so if the electrical connection is suss, or the resistance is not within spec, then it simply cannot be fixed (returned to operational specs) other than by replacement.

First port of call on all KE engine diagnostics is to follow the test / diagnostic procedures in the service manual - aka RTFM. It's in the reference library for those reading who have an interest in KE-jet function in their cars.

As Ron mentioned before - KE is a system that requires everything to be functioning correctly and in sympathy - when one area is substandard or not functioning as intended, then there is a potential money pit waiting, unless the correct diagnostic procedures are followed.

KE is not a system where you 'just throw stuff at it' in the vain hope it will fix everything. Indeed, this is where the majority of complaints come from owners on here - and it's almost 100% down to not properly understanding how the systems are integrated or how they function, or because of a lack of mechanical knowledge and a blind faith in a mechanic - are suckered into throwing stuff at it just to endure a huge bill and not have the issue fixed - yes, it happens, and it's seemingly common occurrence - the suggestion in those casess to take the car to a renowned specialist independent MB mechanic, where the poor owner will have to fork out to have all the previous misguided attempts fixed properly.

It's actually another good reason to seek expert advice when something does suddenly go bad - and also why it is important to have a good working relationship with your mechanic - because a lot of 'little' things can easily cause similar symptoms of much bigger issues, with bigger price tag fixes... when it could be simple stuff like vacuum leaks. It's all too common for an issue to be a " just throw a ***** on it and see if that fixes the issue" or that old chestnut " Sounds like the Battery son........ Under the bonnet son" if you get my drift.

Good to see a problem resolved in a straightforward way, without the excessive hype, bs and conjecture.

If your O2 sensor hasn't been replaced, then consider buying a new one - as they do not last forever, and the O2 sensor does control the EHA to regulate idle fuelling. I noticed you said it was out of spec. Did you happen to test the resistance as well as the voltage?

M103 is a pretty darn reliable engine. Even with a few decades of "really average" service history, they seem to be impervious to general neglect, and it's not hard or expensive to give them a complete new lease on life. Really underrated powerplants. Pity they don't make them like that anymore.

Wise words Mr AMG!! Do the tests according to the specifications and track the problem methodically.
At idle, the O2 sensor voltage was seen to oscillate up to 1.6V, which is higher than specification (0.45V - 1.0V) indicating the mixture is rich, ie. the oxygen content is lower than ambient air and the sensor voltage is greater. However, we tested the Lambda ratio and that was fine at 45% - 55% and my fuel economy is fine (actually on a run recently, I achieved 7.5L/100kms whereas factory spec is 9.8L/100kms - Combined cycle).
Specifically, we didn't test the resistance of the O2 sensor, which I gather is 0 ohms (which sounds odd to me, but that is what I gather).

Over time, the sensor gets a build-up which delays the information to the ECU. Some people recommend it can be cleaned using mineral turps.