Odd Static Duty Cycle Readings... Help Please! - PeachParts Mercedes-Benz Forum

Merkey · #1 08-01-2012, 03:36 PM

Trying to get Peachy, my 1990 190E 2.6 ready for smog...

I am confused by the STATIC KOEO readings I am getting from the X11 connector on the LH fender.
I will be referencing the "Duty Cycle" Holy Grail at the Landis site here somewhat..
BOSCH KE3-JETRONIC MIXTURE ADJUSTMENT

I will try to explain what I see and reading with pictures as I am really confused whether there is an issue or its me!

Car is a CA and Fed smog complient for 1991 MY according to the underhood sticker with the 16 pin diagnostic port near the battery...

Pushing the button on the Diagnostic for 3 seconds or so gives one flash from the LED... No codes.
Holding the button or doing anything else does nothing... I read on Landis site that holding the button in for longer puts the LED in solid and allows the X11 connector to output duty cycle... it does on my son's(Ps2cho) 1988 300TE, but not with this car.

So, KOEO, I plug in my Craftsman DMM...
I heard that the Craftsman DMM's can read backwards, but for a start..
Red into #3 and Black into #2...
I get a 69.9% reading...

Reversing the leads to the DMM so Black is in #3 and Red in #2, I get 30%..
I am gonna leave it like this for the following tests for clarity...

Now I read on Landis that deflecting the Air Flow Sensor Plate should reduce the reading to 10%...
I get no change..

However..
Moving the throttle linkage a small amout does reduce the reading to 10%.

And cranking the throttle to WOT gives me a 20% reading..

Starting the engine from cold it changes the static reading to 50% while its in Open Loop.

Once the engine drops into closed loop after a few minutes, the reading begins to cycle, and I have it set about 50%... so it varies between 45%/55% at idle and about the same at 2500rpm... so thats good!

so a quick recap..
Now aside from the 30% at KOEO, and no change when the air sensor plate is deflected, the 10% and 20% are correct...

Now it gets confusing....
After the engine was warmed and in closed loop...
I did not touch the ignition key, but stalled the engine by deflecting the Air Sensor Plate... the engine stopped, and I was back in KOEO... but now I am reading 70%!

I deflected the Plate more... and I hear a click from the battery area which I assume is the MASS, and the reading drops to !0%!

Now I move the throttle linkage... and I get no change from 70% at either throttle opening.

I go and turn off the ignition and turn it back on again...
Now its changed again...
KOEO I get 50%...

Deflecting the Air Plate I get 40%..

Moving the Throttle off idle I get 10%..

And WOT I get 20% again...

Merkey · #2 08-01-2012, 03:41 PM

I hope I am making sense...

I dont understand what is right and wrong here...
a CA car should read 85% KOEO...
Landis says a Federal car with diagnostic code memory will read 70%...

Factory service manual says that a 30% reading is saying that the temp sender is faulty, but mine checks out fine... leaving it unplugged sets a code, which I erased...

Can someone enlighten me please????

Merkey · #3 08-02-2012, 03:03 PM

Anyone?

I repeated the test again and the results were the same.

JohnM. · #4 08-02-2012, 10:09 PM

It really sounds like a bad engine ECU to me? It can be the cause of some of the weirdest problems with duty cycle readings in my experience. Not sure if any of those functions run through the MAS relay?

Of course you can also measure current flow at the EHA as well.

rob m · #5 08-06-2012, 08:20 PM

I have a 91 300e base I think it is cal and fed compliant. Yellow tag under hood and 16 pin diagnostic with pushbutton and led.
If I push button for 3sec I get I flash. If I push again for 3 sec I get 1 flash. If I push for 9 or more sec I get nothing(unless there is a code to clear. This clears the previous code read if there is one).
Key on eng off. Lambda reads about 65%. ( I use a dwell meter)
Deflect air plate. Lambda reads about 55%
move throttle. Lambda reads about35%
engine running in closed loop lambda reads about62% at idle
Lambda reads about 72% at 2500rpm or so
Eha current fluctuates around 0a at idle
O2sensor reads .3-.6v when warmed up
When driviing shows .1to.9v
Car runs well idles a little rough idling in drive smooth in park
Recentky changed the air duct(rubber boot under air plate assembly)
The o ring between fuel injection octopus and air duct housing and seal for plunger in fuel injection octopus. Measure carefully if you change this. And get it from dealer it is not an o ring it has a shoulder on it
Changing there cured my code 17 problem((I think there was a vacuum leak in there somewhere causing lean at higher rpm

mbdoc · #6 08-07-2012, 08:28 AM

On a California CIS engine, you need to "blink" the LED button 2 times to get the readings to SHOW readings correctly.

Merkey · #7 08-07-2012, 11:44 AM

So you press the button twice or press and hold for 3 sec twice in close succession?

mbdoc · #8 08-07-2012, 12:23 PM

Yes, press & hold for 3 seconds with key on & wait 1-2 seconds & do that again.
At that point the ECU should be in the diagnostic mode.

Merkey · #9 08-07-2012, 12:39 PM

Ok...
Just went out and tried that..
Pressed button for 2 to 3 sec... and after release it takes a sec or so to flash the LED... pressed it again for 2 to 3 sec and the LED gives one brief flash as before... No change in the readout from the X11 connector... still reads 30%.

Tried pressing for 2 to 3 secs and then pressing again before the LED flashes... no flash occurs...

On my Son's 88 300TE which has the smaller diagnostic port with I think 7 pins you do the one press and then another and the LED come on solid signalling you are in diagnostic mode...
This 16 pin car is not doing it...
What am I doing wrong?

Merkey · #10 08-07-2012, 12:46 PM

Underhood Emission sticker...

Merkey · #11 08-07-2012, 04:37 PM

Well, upon some suggestions via PM, I figured a few things out, but not home yet...

When I checked the codes from the onboard LED before I erased them I had codes: 7, 17, and 27.

I read up on the codes here... page 24 reading off pin 3.
http://www.autolib.diakom.ru/CARS2/pribor/cs1000/manual/cs1000_mb.pdf

These codes are Emission related code and the same as reading the codes from Pin 3 of the Diagnostic plug
Code 7=TNA signal(rpm) read by CFI control module
Code 17=Oxygen sensor is shorted to positive or ground
Code 27 =Data exchange between CFI control module and ignition control module interrupted

Now, when I bought the car, the electrical plugs into the EZL were disconnected, so I felt that would explain code 7, and 27, as I never reset them.
I also replaced the O2 sensor with a Bosch Mustang unit, and got the heater cables mixed up, at first, so that would explain the code 17...

Now the plot thickens....
Using my home-made LED code reader, at Arthur Daltons suggestion, I tested pin 14 which is the MAS/CFI pin...
I had a bunch of codes stored!
I recorded then looked up the codes on page 28 of the doc link above...
I had 2,3,4,5,6,7,12,and 13 stored!
I then went thru the codes one by one and cleared them...
What was interesting was each time I cleared a code, there was not a long LED flash as you get clearing the codes from pin 3... ther was no LED flash, but there was a loudish click from somewhere on the firewall behind the battery, and also a stereo click from the front of the engine... I think the smog/air injection pump.

I then restarted the engine and left it idle until it dropped into closed loop...
Then i turned it off and KOEO checked the pin 14 for errors...
Code 3 had reset..
Code 3= TN/TD signal (RPM) interrupted.

I did some searches and found this thread, which details that the 1990 diagnostic is a rarer beast... also Athur Dalton describes a memory clearing reset procedure, which I followed, but the code returned as soon as the car is started!

1990 300te Code 3 on Pin 14

Now searching the FSM for a test procedure for the TN/TD sensors....

Merkey · #12 08-08-2012, 11:49 AM

After looking thru the FSM, I could not find where the TN/TD sensor or sensors are or how to check them...
I know there is a sensor above where the timing marks are on the front of the engine, but could not follow the cable...
Can someone enlighten me?

mbdoc · #13 08-08-2012, 04:59 PM

Does the engine run?
If so, then the flywheel sensor is good.
At that point the EZL converts that signal to square wave & that is the out going TD signal.

Merkey · #14 08-08-2012, 05:45 PM

Quote:

Originally Posted by mbdoc

Does the engine run?
If so, then the flywheel sensor is good.
At that point the EZL converts that signal to square wave & that is the out going TD signal.

It runs pretty good....
Starts first time, and shows no sign of issues aside from the reason I started this thread!
I did notice on the FSM in Tech mods that the KE 5 Control Module software was modified to "Stop the storage of fault code 3, and eliminate it" on 07/90...
My car's production date is 10/90.

I feel lost and like I am chasing my tail...