Engine diagnostics 300E, lamda adjustment

I am trying to adjust the on/off ratio on my 88 300E. Iwas using a dwell meter as other people have done. I connected the positive side of the dwell meter on pin#3 of the diagnostic connector on the driver's side fender, and the negative side to a good chassis ground. According to the manual I should be getting between 21-27 degrees for a 35% to 45% duty cycle. I get nothing, no indication at all! The needle on my dweel meter does not move at all. Am I missing something here?
Thanks for any help.

You are connecting the dwell meter correctly. Are you sure it's good and are you sure you have good connections at pin #3 as well as ground?

When doing the lambda adjustment, you are looking for 50% not 35 to 45%. This will be midscale on your dwell meter. Proceed carefully with this adjustment, you can get it so far off, you'll wish you'd never touched it. I have tweaked on mine a number of times to get it right.

Good luck,

Thanks for the reply, I am following the MB CD ROM manual which says 35-45%. This is strange. The connections are OK, ground is fine and the dwell meter works fine. I also tried using a Digital voltmeter and it just shows me a constant DC voltage of 13.8 volts. I have read in other posts that I should be getting around 4 volts. What is up with all this?

Public Enemy,

Have you read Steve Brotherton's Engine Diagnostics article in the DIY section. Also if you search for Lambda adjustment there is some information that you need to be aware of.

Good luck,

Yeap, I read everything I could find on the web site. That is why I m confused. I am just doing what everybody else has done, yet it is not working here.

I think that there must be something wrong with your dwell meter. Since you're getting energy to be read by the DVM, there is something there to measure. Have you tried your dwell meter on something else to see if it works?

The discrepancy with your CD about 35% to 45% is very strange. A number of people have done this and reported here about 50%. Mine, once warm and in closed loop, just goes to 50% and fluctuates around that.

Best of luck,

Well, I just got a duty cycle meter and connected it at pin#3. It does not show anything either. However, if I connect it to pin#4 it shows a duty cycle around 20%. Is there any possibility that it should connect at this pin#?

[georgehifi]

This is for a 300e, , get a digital volt meter, pos probe in pin 3 of the diagnostic plug and neg probe to battery negative, adjust the mixture screw to get a reading between 6.8 and 7.2 volts d.c. Remember the mixture screw has to be pushed down against a pressure spring to engage the drive, otherwise your just spinning the allen scerw to nowhere, i found it better to stop the motor do a slight adjustment then start it and take the reading, because if you push down to far on the allen key you are able to flood the motor, there`s a fine line between engaging the drive and flooding the motor. 6.8 is just lean, 7.2 is just rich. Do this properly and you`ll get a perfect idle back again.

Thanks again, I tried the voltmeter again but it did not help either. It shows a constant voltage of approximately 13.8 Volts. I adjusted the mixture screw both left and right to the point that it killed the engine but the voltage remained unchanged. I now left the screw midway between the two points that it was killing the engine. What is happening?

[stevebfl]

It sounds like you have a problem with the controller.

The signal that you see on pin# 3 should look like the ones in my article. They should be a square wave with varying duty cycle. Be carefull as volt meters can read differently. Peak voltage will always be battery voltage, but average voltage is just that; an average over time of the peak voltage plus the zero voltage. That is why a 50% duty cycle looks like 6-7volts. This is because half the time it is at 13v and half the time at zero. (Look at the first scope pattern of my article at 30% duty-cycle).

If you are getting 13v all the time then something is wrong (I don't think there is a fault code for 100% duty cycle - but there might be). To be sure, check the duty cycle with key on engine off. If its a federal car it should have a 70% duty cycle if Ca car 85% duty cycle (unless I'm wrong).

To verify and/or to set up the engine use the current through the EHA method also described in the article. The current is the actual thing that matters. The integrator is only the copntrollers prepared interpretation of what is happening.

Thanks Steve. I checked the duty cycle with key on, engine off. It was 85% (it is a California car). The problem is only with the engine working, that is when the duty cycle is 100%.
I will try the EHA method you mentioned. I have a good precision ammeter and I will connect it to the EHA and monitor the current as I adjust the mixture screw. As far as I understood from your article I should adjust it to get an EHA current of ZERO mA. I assume that this is done with the engine at operating temperature, correct me if I am wrong.

I don't know why it didn't register in my mind that you are in California. It says right there on your signature Los Angeles.

When I first began looking into checking for a fault code, I read a Bosch injection manual that gave a fault code chart. I found out here at mshop that this was only for California cars, Federal cars have a few fault codes that are indicated by duty cycle.

My manual tells me how to read fault codes on the California car. It is done by removing the fuse on the fuel pump relay and counting the blips at the diagnostic socket. To adjust the lambda, you PROBABLY have to use the current method, but I'm not at all sure. You need a manual for a Cal. car.

I'm sorry I didn't pick up on this earlier to keep you from chasing your tail on this.

Good luck,

[stevebfl]

Boy did Larry hit it. I have only seen a couple CA cars and for years I ignored the lack of intergrator values and used only the EHA current method.

A few years ago I found that if you prepared the controller properly one could get the integrator value at the X11 with on-off ratio as the Federal cars do. I think the proceedure might be to energize the fault code sequence at the X11/4 connector.

When doing the EHA current test and adjustment a couple things should be noted. Current during cranking can be as high as 70ma when combined with the current used for cold running. All forms of current control should stop within less than a minute, probably about twenty seconds. After all starting and cold running enrichment is gone the nominal value of zero ma (milliamps) will proceed till the O2 sensor activity is recognized. An important system check can be to disconnect the green O2 sensor lead under the pass carpet and verify that zero ma is nominal value with engine warm.

O2 sensor control (closed loop - lambda control) has the ability to increase fuel mixture (+ma) by about 12ma. The control can subtract to -12ma if the natural mixture is too rich. The idea is to get the system in closed loop control with its base mixture such that the elctronic system has equal ability to control rich/lean; at zero.

The real value to adjusting mixture this way is that feedback control is also monitored (actually all electronic control is monitored as this IS the way all electronic control is affected). If you are stuck at either end -12ma or +12ma you will have to correct the mixture untill the value moves. If it won't move verify zero ma with O2 sensor disconnected. If the system has been rich for too long the O2 sensor will be fouled. I often disconnect the O2 sensor set the mixture to best running (moved from lean to rich) and drive it around the block to clear the sensor. Reconnect and try again.

Once the system starts to follow the mixture it should eventually gain closed loop. This means that the ma value is seen to go rich, stall, go lean, stall, go rich , stall, go lean etc. etc. This will take place very much faster with the added heat of 2000rpms. It also should center around the same relative mixture position (4ma is a lean condition, -3ma is a rich condition) whether at 2000rpm or at idle - just the speed of transition will be much greater at the faster speed). Setting the mixture to zero means setting the fuel distributor to the mixture that gives the average position of closed loop control is at zero. The ma values will swing one to two ma both sides of zero (remember average - a constant ma will mean you are not in closed loop).


BTW although zero is the preferred book value, I always set my cars to about 3-4 ma negative (rich). While this makes no difference to warm, steady state, closed loop, mixture. It does leave a rich bias to all non closed loop states such as full throttle, rapid acceleration, and cold operation.

Larry and Steve, again thanks for your input. I should have mentioned that I was working on the car according to the guidelines mentioned in the factory CD ROM. I was aware that there is a special step that needs to be followed for the CA cars and that is to press the button on the X92 connector before proceeding with lambda adjustment. I actually did all these things but still was getting no read from my duty cycle meter. Anyway, I came to the conclusion that there must be something wrong with my duty cycle meter as well as with my trusted dwell meter (which by the way works just fine with measuring dwell angles in point ignitions). The problem is probably related to the way the two instruments integrate the square waveforms. Lesson learned: Not all dwell meters will work with lambda adjustment.

So, I went ahead to adjust the screw based on EHA current method as described in Steve's DIY article. I fired up the engine (which by the way was completely cold) and got a cranking current somewhere around 40 mA. I let it warm up while observing the EHA current. After a few seconds, it started droping quite rapidly and within a minute it had approached 10 mA. It continued dropping and within the next 10 minutes or so when the engine had heated up it had gone down to approximately -9 mA. I corrected this rich condition by turning the screw counterclockwise and observed the current going up. It actually went up a bit too much, (this screw is very sensitive to even the slightest turn adjustment). I played with it for a couple of minutes and eventually brought the current at ZERO mA. As Steve mentioned it fluctuates a little bit, it is NOT steady at ZERO mA, in my case it was fluctuating between -1.8mA and 1.6mA, but on the average it was settled around ZERO mA. So, I left it there. Worked like a charm!!

One more thing. I also noticed that setting the mixture a little rich at around -3 mA makes the car start easier when the engine is warm, and corrected the need for cranking it twice in a row in order to start it (only when warm, cold was always starting on the first try). So, it seems as though some slight hot starting difficulties may be corrected by setting this lambda adjustment screw slightly on the rich side. This maybe of help to some people who are trying to sort out a slight hot start problem.