[zsigmond.attila]

Just for the records: My problem was indeed caused by a faulty speed sensor. A good sensor will output 12V pulses, 6 times for each revolution of the speedo cable. The ECU will go open loop after 6 seconds of engine load at 2-2.5 krpm, if it doesn't receive any pulses from this sensor, and will remain open loop until the engine is restarted. I also can confirm that a bouncy speedo will not cause the ECU to go open-loop.

All the best,
Attila

 

[H.D.]

I’d like to reemphasize something I already addressed in post 2 of this thread.

In several threads I read that members simply readjusted the duty cycle because it was too high / low when they checked it. … Following my driving-along-the-road metaphor from post 1 that’s like loosening the steering wheel’s locknut and changing the wheel’s position because it’s not straight on a straight road, without caring about possible reasons like uneaven tyre pressure, sticking brake pistons, damaged steering mechanism, ...

Translated to the KE-Jetronic: If the duty cycle is out of tune, there’s a reason for that, and with a probability bordering on certainty it’s not related to the adjustment screw, unless someone fiddled around with it. Maybe there’s a fuel pressure problem, or a leaky CSV, or a false air leak, … … just to mention a few problems which influence the duty cycle ... and which would still be there after simply readjusting it.

The duty cycle should primarily be seen as diagnostic information ... not only when it shows a static error code, but also when it fluctuates !
While that information is too often neglected, IMO, I always wanted access to it (plus other info) at the touch of a button anytime during driving or parking, and the picture below shows a device which provides that. I built it into the ashtray of my 300CE, when it was less than half as old as it is now.

Also, IMO, too often neglected is the information fuel pressure tests (particularly lower chamber pressure tests under specific conditions) can provide … best supplemented by simultaneous EHA current tests (of course, with the EHA adjustment screw not having been fiddled with !) … and, even better, also supplemented by simultaneous closed-loop o2 sensor voltage tests.
These tests in addition to the duty cycle test, can be very informative … of course, with the duty cycle adjustment screw not having been touched, at least not after a problem started ! …

Look at the KE-Jetronic as a playground for diagnostic thinking, instead of replacing parts which are not diagnosed faulty … increases the enjoyment of “golden era“ MBs … and maybe you want to give a device like the one I made some thought … facilitates diagnosing problems with the OVP, FPR, CPS, o2-sensor, AFM-POT, TPS, CTS, and other parts immensely … especially if they’re intermittant. …

H.D.

P.S.: I apologize in advance if I don‘t notice further posts / questions, which might happen because I‘m not receiving (required) email notifications from Benzworld anymore.

 

[luckymike]

I’d like to reemphasize something I already addressed in post 2 of this thread.

In several threads I read that members simply readjusted the duty cycle because it was too high / low when they checked it. … Following my driving-along-the-road metaphor from post 1 that’s like loosening the steering wheel’s locknut and changing the wheel’s position because it’s not straight on a straight road, without caring about possible reasons like uneaven tyre pressure, sticking brake pistons, damaged steering mechanism, ...

Translated to the KE-Jetronic: If the duty cycle is out of tune, there’s a reason for that, and with a probability bordering on certainty it’s not related to the adjustment screw, unless someone fiddled around with it. Maybe there’s a fuel pressure problem, or a leaky CSV, or a false air leak, … … just to mention a few problems which influence the duty cycle ... and which would still be there after simply readjusting it.

The duty cycle should primarily be seen as diagnostic information ... not only when it shows a static error code, but also when it fluctuates !
While that information is too often neglected, IMO, I always wanted access to it (plus other info) at the touch of a button anytime during driving or parking, and the picture below shows a device which provides that. I built it into the ashtray of my 300CE, when it was less than half as old as it is now.

Also, IMO, too often neglected is the information fuel pressure tests (particularly lower chamber pressure tests under specific conditions) can provide … best supplemented by simultaneous EHA current tests (of course, with the EHA adjustment screw not having been fiddled with !) … and, even better, also supplemented by simultaneous closed-loop o2 sensor voltage tests.
These tests in addition to the duty cycle test, can be very informative … of course, with the duty cycle adjustment screw not having been touched, at least not after a problem started ! … :wink_2:

Look at the KE-Jetronic as a playground for diagnostic thinking, instead of replacing parts which are not diagnosed faulty … increases the enjoyment of “golden era“ MBs … and maybe you want to give a device like the one I made some thought … facilitates diagnosing problems with the OVP, FPR, CPS, o2-sensor, AFM-POT, TPS, CTS, and other parts immensely … especially if they’re intermittant. … :wink_2:

H.D.

P.S.: I apologize in advance if I don‘t notice further posts / questions, which might happen because I‘m not receiving (required) email notifications from Benzworld anymore.

H.D.

I love the idea of the feedback gauge. Do you have records or instructions of how you did it?

As if I'm going to add this to my list of 'to do's'. Well, I'd still love to see the info. It's a great idea.

 

[vdubpower]

can the moderators Sticky this post....i need to read it a few hundred times and maybe understand it

all i know in my car i have some random wire over the firewall that is from a disconnected lamda system

 

[jaymaynard]

@H.D., I'm trying to diagnose my 1987 560SL. I just bought it a couple of weeks ago. It idles rough when it's warm.

The duty cycle wanders all over the place when it's doing that, from the upper 20s to about 54%. That tells me that it's having trouble controlling the fuel flow, but able to do something, since it doesn't give up and set an error code.

Where do I need to look? I don't think the problem's 107-specific, but I have a thread over in the 107 forum if you'd rather discuss there.

 

[H.D.]

H.D.
I love the idea of the feedback gauge. Do you have records or instructions of how you did it?
As if I'm going to add this to my list of 'to do's'. Well, I'd still love to see the info. It's a great idea.

Thanks luckymike … it‘s not very difficult … should be done with care though, so that no components (especially ECU) get damaged.

I might start a separate thread about this device, because it provides info not only about the duty cycle, but also about other things which would be off-topic here in this thread.

But since I‘m not receiving required email notifications from Benzworld anymore (neither on threads I’m subscribed to, nor on PMs), I hesitate to do that, as well as to participate in other member’s threads, because it can easily happen that I don’t notice further posts / questions in such threads, especially if they‘re a few days old.

I added my last post to this thread mainly as emphasis not to neglect the (also w/o such a device) easily available diagnostic information the duty cycle provides … and not to readjust the duty cycle carelessly ! … and not to neglect the other (very informative) tests I mentioned, which I might start separate threads about too when I have more time and when that notification problem is fixed.

I still hope it can be fixed, but unfortunately all efforts and inquiries to BW‘s administration have been unsuccessful so far. … :dunno:
@jaymaynard, I‘ll get back to your post / thread later.

H.D.

 

[Richter Rox]

A thread on your device would be great for all present and future owners of these great cars .

I hope for the best . Good luck

 

[jaymaynard]

A thread on your device would be great for all present and future owners of these great cars .

Amen! I'd love to have something like that.

 

[GreenT]

Unfortunately HD is not getting email notifications and he's a busy dude, I would suggest PMs but he wouldn't see those either without logging in.

Worth a try tho.

 

[H.D.]

[MENTION=560873]... my 1987 560SL ... idles rough when it's warm ... duty cycle wanders all over the place when it's doing that, from the upper 20s to about 54%. ...

In advance I‘d like to say that the diagnosis of your car‘s symptoms (idle & duty ccle) depends on their details.

Basically … the o2 sensor (besides other sensors) tells the ECU what it detects, and the ECU tells the EHA what to do ... and there could be a problem with one of these parts or their connections, or with the FD (fuel distributor) or other CIS parts, like e.g. the injectors. ... The (simultaneous) tests mentioned in post 22 (EHA current, fuel pressure, o2 sensor voltage) would bring more insight. They would show whether the ECU is doing what it’s supposed to do … and whether it’s telling the diagnostic socket the truth about what it’s doing (in duty cycle language) ... and whether what it’s doing has respective effect on the EHA, hence on the fuel pressure in the lower chambers of the FD's differential pressure valves … and, consequently, on the air/fuel mixture … and, consequently, on the o2 sensor … and whether the ECU gets correct feedback from the o2 sensor ...

Regarding these other involved sensors, and as an example! of the above mentioned duty cycle details … if e.g. the duty cycle fluctuates around a value in the area of 50% and suddenly jumps to a static value close to 30% and then jumps back to fluctuate in a range closer to 50% again, there could be an intermittent problem with the CTS (coolant temperature sensor), or a loose contact between the CTS and the ECU, or a problem with the ECU. ...

H.D.

P.S.: I prefer to dicuss individual cases in indiviual threads (in your case in the 107 forum) ... but the email notification problem should be fixed first, so that it can be done conveniently.

 

[Dan Stratton]

The duty cycle is 99.9%

Tryng to make sense of all this.

I have a 1987 420SEL that has been in the family since it was new. Unfortunately (or fortunately) it has been stored for many years, driven occasionally (stored and maintained properly, no mice eating wires or vacuum lines) with only 24,000 actual miles. It runs beautifully except a stumbling problem when cold. Starts fine, no miss etc, although the idle is low at about 500 RPM. When you first drive off, it's fine, but then it will stumble when you stop and start off again.

Somethings that have been replaced:
Fuel Filter (gas drained - and fresh -non ethanol gas)
Both Fuel pumps (they were leaking)
OVP Relay
Oxygen Sensor

If it check the voltage at the #3 pin it is aprox 7 volts at a cold start, but quickly rises to 14 volts after running a minute. The duty cycle is 99.9%. I've made small adjustments at the fuel distributer (richened) that have helped the drivability, and that doesn't change the duty cycle. What am I missing?

 

[H.D.]

...What am I missing?

First I'd like to be reminiscent of what I wrote in post #2 under “Please note” about touching the Lambda adjustment screw in case of duty cycle deviation from its target value. I suggest to do that only “if the elimination of the reason for its deviation is an option that does not come into consideration”.
And as I said in post #22 ... “if the duty cycle is out of tune, there’s a reason for that, and with a probability bordering on certainty it’s not related to the adjustment screw, unless someone fiddled around with it”.

Based on your description:

The CTS’s (coolant temperature sensor) input to the ECU seems to be OK, although it wouldn’t hurt to quickly check it at several engine temperatures … best by measuring its resistance at the disconnected CIS-ECU connector.

The car having been stored for many years may have (mechanically, hydraulically) affected one or some of the CIS parts (fuel distributor, EHA, fuel pressure regulator, injectors,…).

A quick fuel pressure test would narrow down the number of possible causes ... respectively possibly affected parts.
That test (with the engine at operating temperature) should at least cover:
- System Pressure at idle
- Lower Chamber Pressure at idle and at 2500 rpm (EHA plug connected and disconnected)

An additional lower chamber pressure test (with the EHA plug connected) during rapid acceleration from idle to 2500 rpm at a simulated coolant temperature of 20°C (68°F) would bring more valuable insight. A coolant temperature of 20°C can be simulated at any engine temperature with a simple 2.5 kΩ axial lead resistor (instead of the CTS) connected to the CTS plug.
Simultaneous EHA current measurement would provide further insight into the system.

If a car is only occasionally driven I recommend to use fuel that contains as little ethanol as possible and to constantly use a good fuel additive that protects the entire fuel injection system.

P.S.: Since this thread is about Lambda control and how to check and adjust it, I suggest to start a separate thread for further troubleshooting in your individual case. …

H.D.

 

[Twins2014]

Sent from my iPhone using Tapatalk

 

[plinker17722]

HD Would like to know about the extra gizmo you have under the radio . Would one of these be hard to fit in to the car ?And do you make them ?

 

[H.D.]

HD Would like to know about the extra gizmo you have under the radio . Would one of these be hard to fit in to the car ?And do you make them ?

Installing and connecting it is not difficult. To build one, a little craftsmanship is advantageous.

I made the one in the picture … but I don‘t make & sell them.

I haven‘t gotten around to do it yet, but maybe I‘ll find some time to describe it in more detail in a seperate thread this or next weekend. …

H.D.

 

[H.D.]

Here is the link to a separate thread about that on-board KE-Jetronic test device:

http://www.benzworld.org/forums/w126-s-se-sec-sel-sd/2906786-board-ke-jetronic-test-device.html

 

[Christian Omloop]

Hi,

Thanks H.D. for the explanation! The addition to the explanation of the K-Jetronic.

Could somebody tell me if the EHA is able to fully cut off fuel to the injectors? For example if the O2 sensor is broken?

I could do some test like explained in this thread, the problem is that the problem not often occurs. From time to time my SL-300 24V stalls after a period of driving, not able to start the engine for a certain period. Last time I had the time to check the sparks, the CPS signal and system fuel pressure, they are all ok. But no fuel at the outputs of the fuel distributor or injectors. A brand new EHA has been installed before the last diagnoses (it was leaking fuel).
Beside the fuel distributor I was thinking that the O2 sensor could cause this stalling of the engine.

Thnx in advance!
Christian

 

[H.D.]

... Thanks H.D. for the explanation! The addition to the explanation of the K-Jetronic. ...

You‘re welcome! … But this thread is about one of the functions of the KE-Jetronic, the Lambda control function. It is not about the K-Jetronic.

... Could somebody tell me if the EHA is able to fully cut off fuel to the injectors? For example if the O2 sensor is broken? ...

Fuel flow is fully cut off via EHA control during overrun and when the engine‘s speed limit is reached. The o2 sensor has nothing to do with full fuel cut-off. Based on input from the o2 sensor EHA control takes place between the enriching limit & the leaning limit, which is a control window that is significantly smaller than the EHA‘s overall control window.


Christian (and other readers), since this thread is meant as a tutorial about the KE-Jetronic‘s Lambda control function, which I‘m planing to add considerably more information to when I have time, I would appreciate if you post in this thread only for general discussions about that topic and start a separate thread about your specific issue(s).

Thanks in advance! …

H.D.

 

[H.D.]

Wrong idea about “duty cycle adjustment“

I’d like to remind CIS-E car owners to an essential, but largely unknown, fact that I addressed at the beginning of this thread already.

Still many posts here at Benzworld reveal the belief that the air/fuel mixture is changed when the duty cycle is changed/readjusted … that the engine is running leaner when the duty cycle is fluctuating at a higher range and that it is running richer when it’s fluctuating at a lower range. I also saw documents posted that reveal or support this belief. Some of them looked like original MB documents at first view, but are not. Even the real MB FSM from Stuttgart can mislead to believe that, also the German version. … But it’s a misconception ! … In fact it is the most widespread misconception about the KE-Jetronic … and the also widespread habit of calling the KE-Jetronic’s Lambda duty cycle adjustment “mixture adjustment” will probably continue to feed it.

For the sake of better diagnosing/troubleshooting I really recommend to leave this misconception behind and to think of the Lambda duty cycle check/adjustment as what it actually is, namely an “EHA operating range check/adjustment”. That’s the range within which the EHA’s baffle plate fluctuates during Lambda control in relation to its currentless rest position. … That is not to be confused (!) with “EHA adjustment” via the small hidden adjustment screw on the backside of the EHA, by which the EHA’s, or to be more specific, its baffle plate’s currentless rest position itself is changed, which is an option that should only be taken into consideration in very few specific cases, verified by specific tests … and carried out properly, again verified by specific tests … and not be discussed in this Lambda control thread, please. :wink_2:

I usually do Lambda control checks/adjustments by directly measuring the EHA current instead of by measuring the on/off signal (in ‘volt’, ‘duty cycle’ or ‘dwell angle’) at the X11 diagnostic coupling. Afterwards I check if the CIS-ECU is telling the X11 diagnostic coupling in form of the Lambda on/off signal (which is actually better called off/on signal) the truth about what it is telling the EHA. That can reveal possible ECU problems that would otherwise be overlooked and lead into wrong troubleshooting directions ! … Only when I’m on the road in my (wife‘s) 300CE I occasionally monitor Lambda control via the off/on signal in the ‘ashtray’ (see post 36 ), which also allows me to monitor during driving if there is any (intermittent) problem that is represented by one of the CIS-ECU‘s error codes.

Anyway … no matter how often or where you read or hear about “adjusting / changing the mixture“, I assure you that on the KE-Jetronic you can not adjust or change the air/fuel mixture ! … unless you turn the Lambda adjustment screw cw beyond or (close) to the system’s leaning limit (duty cycle = 0%) or ccw beyond or (close) to its enriching limit (duty cycle = 100%), or Lambda control is not active (warm-up phase, WOT, or due to a problem).

With its Lambda control function the CIS-ECU tries to keep the air/fuel ratio (via EHA control) in the range where it receives O2 sensor voltage fluctuating around 450 mV, which represents “stoichiometric air/fuel ratio” … provided that there are no other fuel combustion affecting problems ! .. which is why instead of saying “which represents stoichiometric air/fuel ratio”, it’s more precise to say “which the ECU takes as representative for stoichiometric air/fuel ratio” .. a mostly overlooked, albeit for CIS-E diagnosis essential, difference I’ll get back to later.

When the duty cycle is fluctuating … (with amplitude & frequency as described under “Check procedures“ in post #2 !) … Lambda control is active. No matter whether it fluctuates around a low value or around 50% or around a high value .. if it is fluctuating, the (intact) ECU is receiving O2 sensor voltage fluctuating (in case of a healthy o2 sensor with about +/- 350 mV) around 450 mV.
Thus, provided that the O2 sensor input to the ECU is reliable and that the fuel combustion is not impaired (in any way !), when the duty cycle is fluctuating, Lambda is fluctuating with about +/- 0.02 around ‘1‘ (‘λ = 1 +/- 0.02’) … the air/fuel ratio is fluctuating with about +/- 0.3 around ‘14.7’ in case of non-ethanol fuel (or, for instance, around ‘14’ in case of E10, even though a wideband a/f meter would still read 14.7 ) … hence, the air/fuel ratio is stoichiometric.
That ratio is not changed when the range of the fluctuating duty cycle is changed ! … unless it is changed beyond or (close) to 0% or 100%.

I explained all that and what happens when the Lambda adjustment screw is turned in more detail in post #1 of this thread.

Don‘t ignore the purpose of the KE-Jetronic’s “duty cycle” !

Two of the most commonly made CIS-E-related mistakes are also still largely made here at Benzworld … ignoring the Lambda off/on ratio (duty cycle) and, IF it is checked and it deviates from the target value, rashly readjusting it via Lambda adjustment screw … or even worse, tampering with the Lambda adjustment screw without caring about the duty cycle.

That‘s why I‘d like to reiterate what I addressed already in posts 2 & 22, namely that the purpose of the duty cycle is to provide comfortably available diagnostic information … which should neither be ignored nor wiped away by simply turning the Lambda adjustment screw ! … Since it was likely that the latter would often be done by DIYers or incompetent mechanics at workshops, there was a plug put into the Lambda adjustment tower.

Duty cycle deviation should first induce a search for its cause, not for the adjustment wrench !

A fluctuating duty cycle is an easy to check representative of the EHA current which determines the EHA’s operating range that the ECU has to provide in order to keep receiving o2 sensor voltage fluctuating around 450 mV (which the ECU takes as representative for stoichiometric air/fuel ratio).
A static duty cycle, while the engine is running, is an error code. With ’ignition on, engine off’ it is either an error code or it shows, as far as it is detectable by the ECU, that the system is in order.

Although this diagnostic information, that the (fluctuating & static) duty cycle provides, can be very valuable for troubleshooting and is so easily available, it is downright unbelievably often neglected. There are CIS-E cars it has never been (properly) checked on. Many unnecessary part replacements could be prevented already by properly checking & interpreting the duty cycle !

Do not simply readjust a deviated duty cycle via Lambda adjustment screw !

Simply readjusting a deviated duty cycle via Lambda adjustment screw without caring about what‘s causing the deviation is a bad idea and can lead to engine damage ! … The further away from 50% the fluctuating duty cycle deviates, let alone if it‘s static 0% or static 100%, the worse a readjustment via adjustment screw can be for the engine.

Regard the duty cycle as a messenger that tells you whether there is a fuel combustion affecting problem or not. If it tells you (by low, or high, or static readings) that there is a problem, listen to it and follow its message, instead of shutting it up and simply readjusting it via adjustment screw ! … As I mentioned in post 2, besides after switching between non-ethanol & ethanol containing fuel, readjusting the duty cycle should be the last thing to do after fixing all causes for its deviation, respectively if fixing the cause(s) deliberately does not come into question. I emphasize “all” because I often see duty cycle readjustments done after something has been fixed/replaced without first checking for all other possibly still existing causes for deviation.

Most duty cycle readjustments that are done, even if they bring subjective improvement about, only mask the deviation’s actual cause(s) which continue(s) to exist, with negative effects !

Here‘s a simple example (of many):
An intake vacuum leak causes low O2 sensor voltage. The ECU immediately reacts by sending higher current (in positive direction) through the EHA’s coil in order to increase the amount of injected fuel as an attempt to compensate for the vacuum leak and raise the O2 sensor voltage again, which works if the leak is not too big and the Lambda enriching limit via EHA control is not reached or exceeded. ... In that case this higher EHA current is represented by a duty cycle fluctuating in a higher range. It can, of course, be lowered again by raising the position of the FD‘s control plunger in relation to the AFM plate’s position via Lambda adjustment screw … which most DIYers would probably do … with the consequence that during each warm-up phase (before Lambda control is active) the engine will suffer from way too rich a/f mixture. … It would be much better to leave the adjustment screw alone and lower the duty cycle again by fixing the vacuum leak, wouldn’t it !? …

The duty cycle (respectively the EHA current) should always be checked after fixing a fuel combustion affecting problem, respectively after replacing a fuel combustion relevant part. If (besides FD / AFM replacement which usually does require duty cycle readjustment) after a repair / replacement the duty cycle is still out of line, there is either a problem with the repair / replacement, or there is another problem, or the adjustment screw has been unprofessionally messed with.

Do not blindly tamper with the Lambda adjustment screw !

Tampering with the Lambda adjustment screw without even checking or caring about the duty cycle (or the EHA current) is an even worse idea than simply readjusting a deviated duty cycle without caring about what‘s causing the deviation … with even higher risks for the engine’s health !

Plus, done without at least knowing exactly how far in which direction the adjustment screw has been turned, valuable diagnostic information is gone for good (!) and everything that has an effect on fuel combustion (see examples under “Please note“ in post #2) has to be checked first before further touching the adjustment screw, in order to reliably restore a proper condition of the system.

Maybe this post helps a little to understand how justified the difficult to remove plug in the Lambda adjustment tower was/is. ... Without that plug, the adjustment screw would have been unprofessionally tampered with a lot more often, making diagnosis more difficult for MB workshops & Bosch service stations from the beginning. That plug may not exist anymore in most CIS-E cars, but the reason for it continues to exist !

So don‘t tamper with the Lambda adjustment screw … and make it a habit to occasionally, or better frequently, check the diagnostic information the duty cycle provides !

In case of engine problems checking the duty cycle should be one of the first things to do ! ... That alone can already point to many possible suspects as well as rule out many possible suspects.

Check it under the following conditions:
1) ignition on (engine not running)
2) ignition on (engine not running) & air sensor plate halfway deflected (throttle closed)
3) ignition on (engine not running) & throttle completely opened (air sensor plate not deflected)
4) engine warmed up to full operating temperature & running at idle speed
5) engine warmed up to full operating temperature & running at ~ 2500 rpm
6) engine warmed up to full operating temperature & running at idle speed again

In steps 1, 2 & 3 the off/on ratio (duty cycle) is static. … In steps 4,5 & 6 it is supposed to fluctuate (with amplitude & frequency as described under “Check procedures“ in post #2) !

For detailed instructions about with what meter & how to check the off/on ratio (duty cycle) see what I said under “Measuring device”, “Preparations”, “Check procedures” and “Static duty cycle” in post #2 !

And if the duty cycle deviates from what it should be, interpret the deviation correctly and check for its cause(s) ! … Ask someone who‘s familiar with the KE-Jetronic to help you with that if necessary.

H.D.

 

[luckymike]

Two of the most commonly made CIS-E-related mistakes...H.D.

H.D., Can't thank you enough for this info. I'm doing a lot of studying these days to get my head wrapped around this system and your comments always add to my education.

Thanks so much for taking the time and making the effort to share.