Mercedes-Benz Forum banner

KE-Jetronic Lambda control (duty cycle check & adjustment)

68132 Views 89 Replies 30 Participants Last post by  Porshapower928
Before I get to my suggestion how to check and how to change the KE-Jetronic’s Lambda control adjustment, I’d like to address some basics in this first post, which might help better understand the matter. I’ll continue with some recommendations and considerations and the detailed procedures in post #2.


Here at I see the Lambda control adjustment usually being called ‘duty cycle adjustment’, or (not really suitably) ‘mixture adjustment’ or ‘air/fuel mixture adjustment’, which can easily lead to (and possibly often reveals) misconception. … By changing the adjustment the position of the fuel distributor’s control plunger in relation to the air sensor plate’s position is changed, which in case of a K-Jetronic (without lambda control) results in a changed ‘air/fuel mixture’, but - aside from the engine’s warm-up phase, or completely floored accelerator, or limp home mode - not in case of a KE-Jetronic!

What is ‘Lambda control’?
It’s the fine-tuning of the air/fuel mixture to a ratio at which complete fuel combustion takes place, in order to minimize pollutants. That ratio is called ‘λ (Lambda) = 1’, which in case of non-ethanol fuel is given at an air/fuel ratio of about 14.7 mass units of air for 1 mass unit of fuel (14.7:1). It’s a compromise between engine torque and fuel consumption. The highest engine torque would be given at a ratio of about 12.5:1, and the lowest fuel consumption would be given at a ratio of about 16:1.

How does Lambda control work?
I like to use a metaphor for illustration. While driving along the road our eye tells our brain to which side the car is about to drift off-lane, the brain processes that information and tells our hand to turn the wheel a little to the left or to the right, which we more or less alternately do all the time. … Translated to the KE-Jetronic the lane is ‘λ = 1’, the eye is the o2 sensor, the brain is the ECU (electronic control unit), and the hand is the EHA (electro-hydraulic actuator).
The EHA is a valve via which fuel flows through the lower chambers of the fuel distributor’s pressure differential valves in order to control the quantity of fuel injection, hence the air/fuel mixture. The EHA’s baffle plate is electromagnetically moved closer to or further away from its inlet nozzle by positive or negative current from the ECU, by which the lower chamber pressure can be changed. And the lower chamber pressure controls the fuel flow through the upper chambers, each of which has a separate injector pipe port.
In order to detect whether complete fuel combustion is taking place, regardless of the type of fuel, the o2 sensor compares the amount of residual oxygen in the exhaust gas with the amount of oxygen in the ambient air. At the ratio which represents complete fuel combustion (λ = 1) the o2 sensor is very sensitive and generates a voltage of 450 mV. That voltage changes significantly at tiny changes of the oxygen ratio. At ‘λ = 0.98’ the o2 sensor voltage is about 800 mV, and at ‘λ = 1.02’ it’s about 100 mV. And as we can not keep the car in its lane without tiny adjustments via steering wheel, ‘λ’ can not be kept at ‘1’ without tiny mixture adjustments either. The air/fuel mixture is either a touch too lean or a touch too rich and alternately has to be enriched and leaned a touch (micro-tuned) in order to keep ‘λ’ close to ‘1’. Lambda fluctuates with about +/- 0.02 around 1, when the o2 sensor voltage fluctuates with about +/- 350 mV around 450 mV, which with a healthy o2 sensor hapens at a cycle frequency of about 0.5 – 1 Hz. That’s the o2 sensor voltage the ECU ‘wants’ to receive, and it adjusts the air/fuel mixture via EHA control in such a way that it does receive that voltage, regardless of the kind of fuel, which in case of non-ethanol fuel leads to an a/f mixture fluctuating with about +/- 0.3 around 14.7:1. In case of ethanol containing fuel it leads to a different (richer) mixture, depending on the percentage of ethanol in the fuel.
Here’s a simplified example of one Lambda control cycle with non-ethanol fuel, which takes about 2 seconds at idle:
- λ ~ 0.98, (a/f ~ 14.4:1), > o2 sensor voltage to ECU ~ 800 mV
- ECU generates more negative EHA current (duty cycle: 45%)
- leaning the air/fuel mixture
- λ ~ 1.02, (a/f ~ 15:1), > o2 sensor voltage to ECU ~ 100 mV
- ECU generates more positive EHA current (duty cycle: 49%)
- enriching the air/fuel mixture
- next cycle: λ ~ 0.98, (a/f ~ 14.4:1), > …..

What happens when we change the Lambda control adjustment?
Let’s say the duty cycle is fluctuating like in the above example between 45% and 49% at idle. When we change the adjustment by turning the adjustment screw cw the control plunger moves to a higher position, leading to a richer mixture. That immediately leads to o2 sensor voltage not undershooting 450 mV, upon which the ECU immediately reacts with an EHA current fluctuating around a more negative mean value in order to lean the mixture again, which is accompanied by a duty cycle fluctuating, for example, between 33% and 37%.
And when the control plunger is set to a lower position by turning the adjustment screw ccw, leading to a leaner mixture, the o2-sensor immediately reacts with voltage not overshooting 450 mV, upon which the ECU immediately reacts by sending a current fluctuating around a more positive mean value through the EHA’s coil in order to enrich the mixture again, which is accompanied by a duty cycle fluctuating, for example, between 57% and 61%.
No matter to which position the control plunger is set, unless it’s set too high by cw turns beyond the EHA’s ‘Lambda leaning limit’, or too low by ccw turns beyond the EHA’s ‘Lambda enriching limit’, the ECU always adjusts the air/fuel mixture via EHA control, regardless of the kind of fuel, in such a way that it receives o2 sensor voltage that fluctuates with about +/- 350 mV around 450 mV, which represents ‘λ ~ 1 +/- 0.02’ respectively ‘a/f ~ 14.7 +/- 0.3 : 1’ (in case of non-ethanol fuel).
Or, speaking in terms of ‘duty cycle’: No matter around which mean value the duty cycle fluctuates, as long as it’s above 5–10% (‘leaning limit’) and below 90–95% (‘enriching limit’), if it fluctuates, Lambda respectively the air/fuel mixture fluctuates around the correct ratio, intactness of the system provided, of course. However, around (or close to) 50% it does that more precisely than, for example, around 20% or 80% (I’ll get back to that in post #2).
Conclusion: When we change the duty cycle adjustment we change the operating ranges of both the control plunger and the EHA’s baffle plate … but not the mixture!
- Control plunger higher > EHA more open
- Control plunger lower > EHA more closed

What’s this ‘duty cycle’ about?
Parallel to the fluctuating EHA current the ECU sends a square wave voltage with a corresponding ‘on/off ratio’ to port 3 of the diagnostic coupling X11, where it can be measured in 'duty cycle', 'dwell angle' or 'volt'. This fluctuating duty cycle is an easier to check representative of the EHA current, and the duty cycle check / adjustment is actually an EHA current check / adjustment.
A duty cycle of 50% represents an EHA current of ‘0’ mA, a duty cycle below 50% represents negative current (flowing in one direction through the EHA’s coil), and a duty cycle above 50% represents positive current (flowing in the other direction through the EHA’s coil).
Additionally to the fluctuating duty cycle, a non-fluctuating (static) duty cycle while the engine is running serves as error code.

From the above it may also become clear that the air/fuel mixture can not only be ‘micro-adjusted’ (to ‘λ ~ 1 +/- 0.02’) via EHA control, as often assumed, but also ‘macro-adjusted’ - as long as the EHA’s leaning / enriching limits are not exceeded. Especially the leaning capability via EHA control is significant, and more precise.
Let me illustrate this ‘macro-adjustment’ with my driving-along-the-road metaphor:
As explaned above, the ‘micro-adjustment’ via EHA control is like the constantly done tiny adjustments via steering wheel to the left and right in order to keep the lane, no matter whether we’re driving along a straight road or through a curve. The ‘macro-adjustment’ is like the turning of the steering wheel in order to follow the road’s changed direction. The changed direction of the road represents a changed condition of the system … for example, a changed control plunger position in relation to the air sensor plate, or different fuel, or contamination in the fuel distributor, or a false air leak (of limited size of course), etc. …. conditions, all of which without EHA control would result in more or less significantly too rich / lean mixture.
The more or less far to the left or right turned steering wheel represents the EHA’s baffle plate position more or less far away from the EHA’s inlet nozzle, respectively an EHA current more or less far below or above ‘0’ mA, respectively a duty cycle more or less far below or above 50%.
And like we continue with the tiny adjustments of the steering wheel to the left and right in order to keep the lane, no matter whether we hold the steering wheel in the straight-ahead position on a straight road or turned to the left or to the right in a curve, the ECU continues with the tiny adjustments of the EHA’s baffle plate in order to keep ‘λ ~ 1 +/- 0.02’, no matter whether the baffle plate is operating closer to or further away to either side from its center position (closer to or further away from the EHA’s inlet nozzle).

Continued in post #2 …
See less See more
  • Like
Reactions: 5
Not open for further replies.
1 - 1 of 90 Posts
I‘d like to remind forum members, particularly those who want to give CIS-E related advice, of this (not finished) thread. It holds the key to quick & targeted diagnosis, as a alternative to the time & money wasting guessing & trying that usually takes place in case of CIS-E engine issues.

Based on what I see posted in the relevant BW forums, it is apparently not understood yet. There is more and more wrong info & advice about the KE-Jetronic (also in the form of YouTube videos) posted and spread here at BenzWorld ... and no one objects to it or corrects it. :confused:

See my comments to one example of these YouTube videos in post 20 & 23 in This Thread !
HD, you seem very sharp on Lambda System so here is my dilemma.

I am very stuck with a customer’s 1989 420SEL (California). The engine starts and runs sort of ok! But will not accelerate at all. I put new fuel pumps/filter/accumulator, so fuel pressures are correct. Engine has all new injectors and fuel distributor (Bosch reman), I have replaced the potentiometer and am now on ECU #4 trying to get this thing fixed. The sticking point for me is that the voltage feed from the ECU to the potentiometer and altitude correction capsule is listed as 8 volts (according to WIS/Xentry) and I have 4.3 to 4.8 from my ECU. To me this is a fault that is very important. I have no Lambda function, no duty-cycle response, no response from the X19 Diagnostic socket on the inner firewall (California version) except that the led does light when the button is pressed, but no codes, not even 1-1-1. I recently purchased an ECU that is supposed to be NOS and came in the original MB Dealer parts box. It gives me 4.8 volts at the potentiometer, just like the original and the other three that I have tried. The voltage supply to the ECU is good and the ground is good. Plug terminals and socket pins on the ECU look just fine. I’m pretty much out of ideas here. Big thanks in advance for any ideas on this.
See less See more
1 - 1 of 90 Posts
Not open for further replies.